101 — Sunday, October 1, 8:00 am - 9:30 am,...

CME 1: Cardiology: Practical Issues in Nuclear
Cardiology: “How to…..”
of left anterior descending coronary artery (LAD), lateral region of left circumflex artery (LCX),
and the apex was referred depending on the CA result. Sensitivity, specificity and accuracy were
calculated in all cases. The comparison between each pair of images for the same artery was
carried out by means of the Kappa index (KI), with it being considered that a lack of agreement
between the methods was considered when KI<0.42.Results: See table. Significant differences
were found in the specificity in RCA between NCG and AC (KI=0.41) and in patients with nonsignificant artery stenosis (NSAE) (KI=-0.03). In 7 cases the AC induced a hipoactivity in the
anterior region, considering these cases as hyperattenuation, specificity in LAD artery increases
from 74.1% to 96.7%.
How to implement guidelines for imaging of perfusion and
Birger Hesse (DK)
How to report a cardiovascular nuclear test
Claudio Marcassa (IT)
How to use attenuation correction in myocardial perfusion
Wolfgang Burchert (DE)
Se=96.1% Se=84.6% Se=84.6%
Sp=38.2% Sp=47%
Se=79.3% Se=79.3% Se=82.7%
Sp=93.5% Sp=93.5% Sp=74.1%
Se=55.5% Se=55.5% Se=55.5%
Sp=92.8% Sp=92.8% Sp=100%
Number of patients
NSAE Sp=21.4% Sp=28.5% Sp=78.5% 14
Conclusions: The ability of MPS to detect RCA disease increases by the use of X-ray attenuation
correction. In the anterior region corrected and non-corrected images must be evaluated to avoid
misreading due to hyperattenuation.
How to establish myocardial perfusion imaging in the
emergency department
Alber Flotats Giralt (ES)
102 — Sunday, October 01, 2006, 8:00 am - 9:30 am, Trianti Hall
Cardiovascular: Myocardial Perfusion - Methods
How to decrease
perfusion SPECT
S. Dabiri Oskouie, B. Mahmoodian; Tabriz University of medical scienses,
Tabriz, Iran (Islamic Republic of).
Aim: Extra-cardiac activity on myocardial perfusion scans (MPS) with Tc-99m- sestamibi
(MIBI) may influence scan interpretation. Increased liver, intestinal, or gastric activity can ceate
a major problem in the visual and quantitative interpretation of the inferoposteroseptal walls,
particularly at rest. The aim of this study was to compare the effect of milk, high fat content
meal, and cold water to minimize this extracardiac activity.Methods: Ninety patients (pts) had a
series (triple) of planar scan at rest, which were performed 10, 45 and 60 minutes after injection
of Tc-99m-MIBI, and also performing rest SPECT study 75 minutes after injection of this
radiotracer, without attenuation correction. They were divided into 3 groups. Preparation
consisted of no action taken (30 pts, group1), 150 mL whole milk and 450 mL cold water
respectively 15 and 50 minutes after administration of Tc-MIBI (30 pts, group 2), and high fatty
content butter and some bread instead of milk to third group (30 pts). The presence of activity in
liver and stomach was determined visually and quantitatively per each pixel, comparing to
myocardial activity on planar images. Activity was defined as interfering when it might result
misinterpretation of myocardial uptake in SPECT images, visually.Results: Milk induced
significant decrease of liver activity comparing to fatty meal (p=0.01 vs p=0.17), however both
of them induced significant decrease in intestinal activity (p= 0.0024 and p=0.024 respectively).
Cold water causesd lesser-decreased hepatic and intestinal activity (p=013 and p=0.07
respectively) comparing to milk and fatty meal. Interfering activity was seen visually in 74%
ingroup1, in 33% in group 2, and in 40% in group 3.Conclusions: The interpretation of MPS
SPECT images are facilitated by having the patient drink both whole milk and water before data
acquisition. In conclusion, ingestion of milk may have better influence in reducing hepatic and
intestinal activity compared to high fat content meal. Key words: myocardial perfusion SPECT;
Tc-99m-MIBI; liver, and bowel activity.
Contribution of attenuation correction by low resolution tcderived attenuation maps to myocardial perfusion spect.
A. Rodríguez-Gasén, M. T. Bajén, Y. Ricart, J. Mora, M. Roca, R. Puchal,
D. Carrera, N. Ferran, A. Domenech, E. Quintana, C. Pallarés, J. MartínComín; Hospital Universitari de Bellvitge-Idibell, L'Hospitalet de LlobregatBarcelona, Spain.
Objective: The aim of this study was to calculate both sensitivity (Se) and specificity (Sp) values
in myocardial perfusion SPECT studies (MPS) in the management of coronary artery disease
using three types of images: the non-corrected images, attenuation corrected images and noncorrected images jointly with gated-images. Patients and methods: 60 patients (p) with known
or suspected coronary disease were studied. In all cases stress and gated rest MPS using a hybrid
low resolution TC-gamma camera, and contrast coronary angiography (CA) -within 3 months
interval- were obtained. Three groups of images were analysed: non-corrected stress/rest (NC),
non-corrected stress/gated-rest (NCG), and attenuation corrected stress/rest (AC). It was assumed
that inferior region was responsibility of right coronary artery (RCA), anterior and septum region
Attenuation Correction by the Use of CT Maps in Normal
Hearts at Rest: Comparison between 180 and 360 Data
D. J. Apostolopoulos, T. Spyridonidis, C. Giannakenas, T. Skouras, N.
Papandrianos, C. Savvopoulos, P. Barla, A. Paschali, P. J. Vassilakos;
University Hospitla of Patras, Department of Nuclear Medicine, Patras,
Aim: To examine the impact of CT attenuation correction (AC) and of the angular range of
SPECT acquisition in myocardial perfusion studies of normal subjects at rest.Materials and
Methods: Patients referred for parathyroid scans, with negative cardiologic history and normal
resting ECG, were recruited. After exclusion of technically inadequate studies, 12 men aged
55±18y with BMI 25±2.4 and 33 women 58±11y with BMI 27±5.1, were analyzed. The
myocardial perfusion studies were performed 45 min after injection of 25 mCi of 99mTcSestamibi at rest, with a SPECT/CT hybrid system (Hawkeye, GE). Emission data was acquired
over 1800 and 3600, followed by a CT session. AC was applied using CT derived transmission
maps. The 4 protocols compared were: 1) 1800 no AC, 2) 1800 with AC, 3) 3600 no AC and 4)
3600 with AC. Summed rest score (SRS) was automatically calculated from a 20-segment polar
map model. Homogeneity of count distribution was expressed by the coefficient of variation
(CV) in the corresponding polar maps.Results: The average±std SRS in men with protocols 1, 2 3
and 4 was 4.5±3.7, 1.3±1.5, 1.3±1.5 and 0.8±0.8 respectively. 84%, 89% and 93% of protocol-1
segmental defects were improved with protocols 2, 3 and 4 respectively, while new defects were
detected in 2.6%, 0% and 2.0% of protocol-1 normal segments. The respective SRS figures in the
female group were 2.1±1.7, 1.2±1.6, 0.5±0.7 and 0.7±1.1. The respective improvement rates with
protocols 2, 3 and 4 were 75%, 99% and 91%, and the deterioration rates 4.7%, 3.3% and 2.8%.
SRS with protocol 1 was significantly higher than all other protocols in both sexes (p<0.005). In
men, SRS was not significantly different between protocols 2, 3 and 4. In women, protocol-2
SRS was higher than protocols 3 and 4 (p=0.02 and p=0.001 respectively). The two latter
protocols did not differ. The average CV with protocol 4 was significantly lower than all other
protocols in both sexes (p<0.01). CV with protocol 2 differed from protocol 1 only in men
(p=0.02), while it was not different from protocol 3.Conclusions: AC via CT maps with either
1800 or 3600 emission data collection significantly reduces attenuation artifacts. The 3600
protocol seems preferable, particularly in women, because it results in more artifact reduction
and more homogeneous count distribution. Non-AC 3600 acquisition also yields a very low
defect score in both sexes, but produces less homogenous appearance of the myocardium.
The diagnostic and prognostic value of combined supine and
prone myocardial perfusion SPECT
C. Giannopoulou, E. Zaromitidou, N. Papathanasioy, P. Karampina, E.
Skoura, N. Pianou, G. Keramida, I. Datseris; Evangelismos Hospital,
Athens, Greece.
Introduction: The frequency of artifactual inferior myocardial wall perfusion defects at
Myocardial Perfusion SPECT (MPS) is quite high and - in the absence of gated SPECT
acquisition - necessitates the repetition of the acquisition after 3-4 hours for Tl-201, or an
additional radiopharmaceutical injection at rest for sestamibi or tetrofosmin protocols. The aim of
this study was the assessment of the practical usefulness, diagnostic accuracy and prognostic
value of combined supine and prone MPS (cMPS), in a large teaching hospitals’ one-year
practice. Methods The MPS studies (stress- rest 99mTc-sestamibi, or stress- redistributionreinjection Tl-201 protocols) from January 2004, to January 2005, were retrospectively reviewed.
Patients with inferior wall defects underwent a consecutive prone MPS and the study was
considered as negative if the inferior wall defect disappeared. Patients with persisting inferior
defects on prone images completed the imaging protocol. Patients with negative cMPS studies
were surveyed for cardiac events (major= death, infarction, unstable angina, or any=major
events, or coronary angioplasty, or revascularization surgery) at 18.5± 6.0 months. Results MPS
101 — Sunday, October 1, 8:00 am - 9:30 am, Friends of Music Hall
was performed in 1217 patients by Tl-201 dipyridamole, or 99mTc-sestamibi dipyridamole.
Thirty two percent of the patients or 386 patients (318 men, mean age 62.7 years, 68 women,
mean age 63.1 years) without known coronary artery disease (CAD) had inferior wall defects on
the stress supine image and underwent consequent prone imaging. In 367/386 patients (95%),
inferior wall defects were absent on the prone imaging (negative cMPS), and in 19/386 patients
(5%) defects persisted. Follow-up information was available for 344 of 386 patients with
negative cMPS. No major cardiac events were observed in any of the 344 patients with a
negative combined supine-prone study at 18.5± 6.0 months of follow-up. One patient with
negative cMPS had an angioplasty performed 16 months post MPS. In conclusion, performing a
combined supine-prone MPS in patients without known CAD and inferior wall defects reveals a
high rate of artifactual inferior wall defects, shortens the study time duration and prevents the
patient from an additional radiopharmaceutical injection. Patients with negative combined supine
and prone MPS have low risk for subsequent major cardiac events.
Impact of base delimitation on sectorial quantification in
myocardial SPECT.
F. Cachin1, J. Lipiecki2, C. Thouly3, A. Kelly1, D. Mestas1, N. Boisson1, C.
Merlin1, J. Maublant1; 1Jean Perrin Cancer Center, Clermont Ferrand,
France, 2Gabriel Montpied University - Hospital, Clermont Ferrand, France,
Segami, Paris, France.
Aim: Sectorial quantification of myocardial perfusion is rarely used in clinical practice and
visual reading is largely preferred for coronary artery disease diagnosis. In most quantification
software, base delimitation remains operator dependent. We assessed the impact of this manual
selection on sectorial perfusion quantification.Materials and Methods: A new myocardial SPECT
software, Cardiogam, was applied to ten 99mTc stress - 201Tl rest studies. The reconstructed
transverse sections were automatically reoriented, centered and zoomed using a least square
fitting method to an ellipsoid. The myocardium was then segmented thanks to a spherical
sampling in the apical part and cylindrical sampling in the basal part. Following automatic base
delimitation which was considered as reference, the resulting bull’s eye was used for perfusion
quantification in 20 sectors. In order to assess the effect of base mispositionning, bull's eye
analysis was performed after manual displacement of base location by -2 , +2, +4 pixels from the
reference along the left ventricular long axis. Correlation coefficients and a Bland Altman
parameters were calculated in order to assess the impact of manual intervention.Results: Overall
sectors correlation coefficient between automatic processing and -2, +2, +4 manual processing
were respectively 0.95, 0.97, 0.93. Correlation coefficients limited to the six basal sectors and six
apical sectors of the bull's eye were respectively 0.79, 0.85, 0.87 and. 0.97, 0.99, 0.99. The mean
+/-standard deviation of the difference calculated between manual and automatic processing for
overall, basal and apical sectors were:
All sectors
-0.7+/- 4.2
0.8 +/- 3.3
2.0 +/- 4.7
Basal sectors
-2.1+/- 6.6
2.3 +/- 4.7
6.7+/- 5.4
Apical sectors
0.2 +/- 2.7
-0.1+/- 1.9
-0.2 +/- 2
Conclusions: A misalignment of edge bases introduces high variation in quantification of basal
sectors but not of apical sectors. Disagreement between visual reading and automatic perfusion
software could be explained by the difference in bases delimitation from stress and rest studies.
Identification of multi-vessel coronary artery disease by
quantitative SPECT with the measurements of absolute
myocardial uptake of
C. Matsumoto1, T. Morishima1, T. Hatano1, T. Chikamori2, K. Takazawa1;
Tokyo Medical University Hachioji-Medical center, Tokyo, Japan, 2Tokyo
Medical University, Tokyo, Japan.
With the use of myocardial imaging, the correct identification of individual coronary stenosis is
limited in multi-vessel disease due to its relativity in visual image. Accordingly, not only visual
myocardial imaging but also quantitative assessment of SPECT was performed in 81 patients
with suspected CAD. To analyze absolute count of 99mTc-MIBI, SPECT was performed with
two-day protocol using 740MBq of 99mTc-MIBI each day. Normalcy levels of absolute count of
Tc-MIBI in individual coronary arteries derived from normal coronaries were applied to
identify a significant coronary artery stenosis. To detect the presence or absence of a significant
CAD, visual SPECT analysis revealed sensitivity of 81% and specificity of 42% (p<0.05). To
identify multi-vessel CAD, visual SPECT showed sensitivity of 39% and specificity of 78%
(p=NS). In contrast, respective sensitivity and specificity were 73% and 51% (p<0.05) using
quantitative SPECT with absolute count measurements of 99mTc-MIBI. These results suggest that
visual assessment of myocardial imaging is specific but insensitive to detect multi-vessel CAD.
However, quantitative SPECT with absolute count measurements of 99mTc-MIBI may help better
identify multi-vessel CAD since this measurement is highly sensitive in the detection of a
significant individual vessel stenosis even in multi-vessel CAD.
The clinical value of a novel index of lung thallium-201
uptake in the determination of advanced coronary artery
disease with exercise myocardial perfusion imaging.
E. Moralidis1, T. Spyridonidis2, G. Arsos1, C. Anagnostopoulos3;
Hippokration Hospital, Thessaloniki, Greece, 2Hippokration Medical
Center, Larissa, Greece, 3Royal Brompton Hospital, London, United
Aim: The precise clinical utility of lung thallium-201 uptake in exercise SPECT myocardial
perfusion imaging remains debatable. This study determines an optimal index for lung thallium201 uptake measurement in the prediction of higher-risk coronary artery disease
(CAD).Methods: 398 patients with suspected or known CAD, but no coronary interventions,
referred for routine exercise SPECT myocardial perfusion imaging and submitted to coronary
angiography within 3 months from scan date, were enrolled. Another 56 individuals with low
probability of CAD comprised the control group. Patients were separated into sub-populations of
lower- (n=156) and higher-risk (n=242) CAD. Non-significant coronary lesions (<50% diameter
stenosis) and 1-vessel disease, other than proximal LAD, was classified as lower-risk CAD.
Higher-risk CAD was considered to be present when there was a proximal LAD lesion, when
multiple vessels showed • 50% luminal narrowing and in cases of LM stem disease. From a
planar, short, anterior, post-exercise acquisition, the Lung (L) to Heart (H) maximal (L/Hmax),
total (L/Hmean) and background-subtracted total (L/Hnet) count density ratios were calculated.
In 30 consecutive patients measurements were performed twice by the same operator and once
by a second operator. ROC analysis (area under curve (AUC)) was used to compare the L/H
ratios and define thresholds of abnormality from the coordinate points of ROC curve (the cut-off
value providing the best sensitivity at an associated specificity of •90%).Results: Bland-Altman
analysis showed very good intra- and inter-observer agreement in the calculation of all lung
thallium-201 indices. L/Hnet (AUC 0.780) had a better correct ranking probability for higher-risk
CAD than both L/Hmax (AUC 0.692, p<0.001) and L/Hmean (AUC 0.715, p<0.001). At the
defined thresholds of abnormality, the sensitivity of L/Hnet, L/Hmax and L/Hmean in the
determination of higher-risk CAD was 50%, 34% and 37%, respectively, at an associated
specificity of 92%, 93% and 93%, respectively. After applying models of covariables of the L/H
ratios, as derived from the control group, the adjusted L/Hnet was a better discriminator of
higher-risk CAD than both the adjusted L/Hmax and the adjusted L/Hmean. However, no
significant difference was attained between L/Hnet and the adjusted L/Hnet.Conclusions: In
patients with CAD and no coronary interventions, the lung to background subtracted myocardial
count density ratio is superior to other widely accepted techniques of lung thallium-201
quantification in the identification of advanced CAD with exercise SPECT imaging. This index
is simple to calculate, highly reproducible and its adjustment for confounding variables is
Neural networks: An efficient way to reduce erroneous
interpretation of myocardial perfusion imaging
K. Tagil1, B. Hesse2, M. Lomsky3, J. Marving4, M. Ohlsson5, J. Richter6, L.
Department of Clinical Physiology, Malmo, Sweden,
Department of Clinical Physiology, Rigshospitalet, Copenhagen, Denmark,
Department of Clinical Physiology, Gothenburg, Sweden, 4Department of
Clinical Physiology, Rigshospitalet, Copenhagen, Denmark, 5Deptartment
of Theoretical Physics, Lund, Sweden, 6WeAidU in Europe AB, Lund,
Sweden, 7Department of Clinical Physiology, Lund, Sweden.
Aim. The aims of this study were to use a neural network based technique to improve the quality
interpretation of visual evaluation and the quality of a database of already classified myocardial
perfusion scintigraphy (MPI) images. Methods. Neural networks were trained to detect
myocardial ischemia in five (ant, sep, inf, lat, ap) myocardial segments using a database of 316
MPI classified by one experienced physician. The neural networks were used to select from these
316 studies 20 cases in each myocardial segment (n=100) most likely to have an incorrect
interpretation. A second visual evaluation of those 100 cases together with 100 control cases
were presented in random order to 3 experienced physicians for a visual re-evaluation. A second
neural network was there after trained using the database with the interpretations from the reevaluation. The performance of the different neural networks were measured as the area under
ROC curve. Results: The neural networks trained with the initial interpretations showed
performances, measured as area under the ROC curves of 0.82-0.91 for the 5 myocardial
segments, compared to 0.89-0.99 with the second re-trained neural network, p<0.05. The
interpretation was changed in 53/200 images at the visual re-evaluation. 46/53 (87% ) of the
changes showed disagreements between the original and the first neural network interpretation,
while only 13% were control images (p<0.001). Conclusions: This study demonstrates that
neural networks can be used to detect suboptimal interpretation of MPI studies, which need a
second look. The consistency and interpretation quality of MPI will thereby be efficiently
improved with very limited costs.
103 — Sunday, October 01, 2006, 8:00 am - 9:30 am, Banqueting Hall
Oncology: Malignancy of Unknown Origin
The Comparison of early and late phase FDG PET scanning
in Malignant Lesions: Which one is better in delayed
Y. M. Chen, G. Huang, X. G. Sun, J. J. Liu, Y. P. Shi, L. R. Wan; Renji
Hospital, Shanghai Jiaotong University, Shanghai, China.
Objectives To investigate the diagnostic efficacy of early and late delayed standardized uptake
values (SUVs) in three phase PET/CT scanning and to explore the optimized time for delayed
imaging. Methods 80 patients who had pathologically or clinically confirmed malignant lesions
have been received three phase FDG PET/CT(GE Discovery LS PET/CT) scanning, which were
performed at 72±12min,118±18min,241±29min after FDG was injected. There were totally 215
malignant lesions analyzed in this study. The maximal SUVs of every malignant lesion was
recorded and compared by paired t test in every two group among the three-phase scanning.
Increase in tumor marker CA19-9 level without evident
radiological findings is insufficient for justification of FDGPET examination
K. Inoue1, K. Okada1, Y. Taki1, R. Goto1, S. Kinomura1, T. Kaneta2, H.
Fukuda1; 1IDAC, Tohoku University, Sendai, Japan, 2Graduate School of
Medicine, Tohoku University, Sendai, Japan.
Aim: Carbohydrate associated antigen 19-9 (CA19-9) has been used for the screening of cancer,
because the increase in its level has been associated with several types of cancer. We aimed to
find whether 18FDG-PET provides additional evidence for cancer when it was prompted by an
increase in CA19-9 level without an evident lesion in the usual clinical workup. Patients and
Methods: We retrospectively analyzed the records of 1566 18FDG-PET scans underwent
between October 2003 and December 2004. Inclusion criteria were as follows: 1) PET study that
was prompted by an increase in CA19-9 as well as in other tumor marker levels, with or without
a history of cancer, and 2) patients who had radiologically occult or inconclusive findings.
Patients whose blood glucose level was > 200 mg/dl immediately before the tracer injection, and
patients without confirmation of diagnosis by at least 1 year follow up were excluded. Twentythree scans were selected; 12 patients did not have a history of cancer, and 11 patients had a
history of cancer that had been treated with curative intent.Results: The levels of CA19-9 of the
patients ranged from 42.9 to 3043 U/ml. Four scans had positive findings; 2 patients had a colon
cancer, and 1 patient had a lung cancer (true positive; 3/23). One patient showed no pathologic
state, and the case was considered as false positive. One patient who had detected small lung
nodules in a computed tomography examination had been reported to be inconclusive in a PET
study, and this case was also considered as false positive (2/23), because no malignant lesion has
been detected. Eighteen scans had negative findings; 15 patients had no malignant disease (true
negative; 15/23). For the remaining 3 patients, a local recurrence of hepatocellular carcinoma, a
local recurrence of pancreatic cancer, and peritoneal dissemination were detected within 6
months (false negative; 3/23).Conclusions: Although an in crease in CA19-9 level has been
shown to precede months before the detection of a malignant lesion by the usual radiological
examination, we found only 6 cases that had a cancer, and 3 of them were properly diagnosed
with 18FDG-PET. 18FDG-PET in this study showed high negative predictive value, but it was
considered to be due to the low prior probability of cancer in the present patients, because many
non-malignant conditions cause an increase in CA19-9 levels.
F-FDG PET scan in the diagnosis and localization of
primary tumours in patients presenting with brain
T. K. Au-Yong, C. P. Wong, K. S. Chu, W. C. M. Tong; Department of
Nuclear Medicine, Queen Elizabeth Hospital, Kowloon, Hong Kong.
Aim: We studied the value of FDG PET scan in locating unknown primary cancer sites in
patients presenting with brain metastases. Methods: 25 consecutive patients with radiological
presumptive diagnosis of brain metastases were referred to a single PET centre for the diagnosis
and localization of unknown primaries. FDG PET/CT scan was performed from head to upper
thigh. Dedicated brain PET scans were also performed. Scan findings were correlated with final
diagnosis, either histologically or clinically. Results: Of the 25 patients referred for scanning,
PET scan detected extra-cranial foci of abnormal FDG uptake in 15 patients. Of these 15
patients, 7 patients had histological proofs of the FDG-avid extracranial foci. Of them, 6 were
true positive primary sites (5 lung cancers, 1 kidney cancer). The remaining 1 patient had colon
lesion detected but was proven to be dysplastic polyp by histology (false positive). Another 8
patients had clinical confirmation of extracranial primaries (6 lung cancers, 1 renal cancer and 1
colon cancer). The remaining 10 patients did not show suspicious extra-cranial FDG uptake
abnormality. 6 of these 10 patients were proven to have primary brain cancer (4 glioma
multiforme, 1 chordoma and 1 primary cerebral lymphoma). Of them, one of the patients with
glioma multiforme had low FDG uptake in cerebral lesions while the other 5 patients had
markedly increased FDG brain lesion uptake. 3 patients had non-malignant brain disease (1
multiple sclerosis, 1 inflammatory pseudotumour and 1 toxoplasmosis) and all of them had low
FDG uptake. Remaining one patient had high FDG uptake of the brain lesion but no histological
proof was available because he refused brain biopsy. Therefore, 24 of the 25 patients had either
histological or clinical evidence of final diagnoses. If the scintigraphic diagnosis of extracranial
primaries of these 24 patients with the presumptive diagnosis of brain metastases was based on
the presence of an extracranial focus of abnormal FDG uptake, FDG PET had 100 % (14/14)
sensitivity, 90 % (9/10) specificity and 96 % (23/24) accuracy of diagnosing extracranial primary
tumours. Conclusion: FDG PET is a valuable diagnostic tool in the localization of primary
tumour sites in patients presenting with brain metastases.
The role of whole body F-FDG PET/CT in the management
of unknown primary tumors
Z. Wu;
The PET Center of Union Hospital,Tongji Medical
College,Huazhong University of Science&Technology, Wuhan, China.
[Aim] To assess the role of 2-[fluorine-18]-fluoro-2-deoxy-D-glucose(18F- FDG) positron
emission tomography/computed tomography (PET/CT) in the management of primary cancer
with metastases of unknown primary origin.[Material and Methods] Thirty-four patients with
metastatic lesions of cervical lymph node (n=10), malignant pleural effusions(n=5),malignant
ascites(n=4),malignant pleural effusions and malignant ascites(n=1),the brain(n=2),the
marrow(n=1),the vertebra (n=1),the thighbone(n=1), thyroid(n=1),the liver(n=2),the
kidney(n=1),axillary lymph node(n=3) and inguinal lymph node(n=2) of unknown primary sites
were studied after unsuccessful conventional diagnostic work-up.Patients received 5.55MBq/kg
body weight 18F-FDG intravenously,and whole-body images were obtained 60min after
injection.Surgical, histopathologic findings,and/or clinical follow-up were used to evaluate
PET/CT results.[Results] In 20 of 34 patients FDG PET/CT showed focal tracer accumulations
corresponding to potential primary tumor sites located in the lungs(n=9),the colon(n=3),the
rectum (n=2),the pancreas(n=1),the right aryepiglottic wall (n=1),the esophagus(n=1),the ovary
(n=2) and the breast(n=1).In 17(50.0%,17/34) of these 34 patients FDG PET/CT was truepositive,identifying the primary tumor in the lungs(n=8),the colon(n=2),the rectum(n=1),the
pancreas(n=1),the right aryepiglottic wall(n=1),the esophagus(n=1),ovary(n=2) and
breast(n=1).In 3 of 20 patients FDG PET/CT was false-positive[In the lung(n=1),the
colon(n=1),the rectum (n=1)].In the remaining 14 of 34 patients,FDG PET/CT did not reveal
lesions suspected to be the primary tumor sites in 13 patients,and it was impossible to identify
one lesion as the most likely primary tumor in one patient,due to the presence of multiple hot
spots in several organs.However,in 1 of these 14 patients,the primary tumor site was found at
clinical follow-up,In the remaining patients at clinical follow-up,the primary tumor sites were not
found.As a result,18F-FDG PET/CT identified the primary tumor in 17/34 (50.0%) of
patients.The false positive rate of FDG PET/CT was 3/34 (8.8%). FDG PET/CT was shown to
affect the management in 17/34(50.0%) patients because of the findings of the primary tumor
and(or) the additional metastatic lesions.[Conclusion] 18F-FDG PET/CT is effective in detecting
the primary tumor sites in patients with unknown primary tumors.In addition,18F-FDG PET/CT
assists in guiding endoscopic biopsies for histologic evaluation and has relevant impact on the
therapeutic management of patients with unknown primary tumor.It is recommended that 18FFDG PET/CT be performed in all patients with unknown primary tumor after unsuccessful
conventional diagnostic workup.
Diagnostic accuracy and prognostics value of 18FDG-PET/CT
in patients with carcinoma of unknown primary. An analysis
of 190 investigations.
P. Fencl, O. Belohlavek, M. Jaruskova, M. Skopalova, I. Kantorova, K.
Simonova; NNH Homolka, Prague 5, Czech Republic.
Aim: Based on follow-up data we evaluated diagnostic accuracy and prognostic value of
18FDG-PET/CT in the search for primary as well as for presence of malignancy at all in patients
with Carcinoma of Unknown Primary (CUP).Materials and Methods: Between June/2003 and
July/2005 we investigated 190 consecutive patients by 18FDG-PET/CT originating from 2
different groups. Group HP (Histologicaly Proven) comprised 82 patients with metastatic disease
histologically proven before 18FDG-PET/CT and no primary neoplasm found by conventional
imaging modalities (CIM); group CS (Clinical Suspicion) comprised 108 patients with clinical
suspicion of malignancy and no signs of malignancy by CIM. The histology and/or follow-up
data were used as a gold standard. Sensitivity (Se), specificity (Sp) and overall survival were
calculated.Results: In the group HP Se=54.5%, Sp=75.0% for the search of primary and
Se=92.5%, Sp=82.8% concerning the presence of malignancy at all. In the group CS Se=76.5%,
Sp=85.9%, for the search of primary and Se=96.0%, Sp=86.7% concerning the presence of
malignancy at all. The sensitivity of exact determination of primary was significantly lower
(p=0.0012) in patients with multiple hypermetabolic lesions in comparison to the patients with
the only one lesion. We didn’t find significant difference in survival between groups HP and CS
(p=0.77); on the other hand we found the difference in survival between 18FDG-PET/CT
negative and positive patients (p=0.00001), based on log-rank test.Conclusions: 18FDG-PET/CT
is a contributive diagnostic tool in the search for malignancy even in case when other imaging
modalities failed. The fused imaging of glucose metabolism and body structure provided by
18FDG-PET/CT cannot determine primary with the same accuracy as histology can. The
negative 18FDG-PET/CT finding was a good prognostic factor for higher live expectance in our
group of 190-CUP patients.
Efficiency of FDG-PET-CT in the diagnosis of primary tumour
in patients presenting with a suspected paraneoplastic
syndrome (PNS): a comparison with conventional CT.
S. Banayan, J. Ninet, M. Claret, M. Janier, C. Billotey; Hopital Edouard
Herriot, lyon, France.
Objectives : The purpose of this study was to compare the performance of FDG-PET-CT and
conventional CT in the diagnosis of primary tumour in patients suffering from suspected
PNSMethods: We investigated 18 patients (male:female= 9:9; mean age 62; range: 45-77 years)
with a suspected PNS (cf table). Each patient underwent both a thoraco-abdomino-pelvic
enhanced-CT and a whole body FDG-PET-CT.We compared the conclusions of these two
exams, to reach the etiologic diagnosis of PNS. We obtained a histologic proof of a cancer for 6
patients (pts 1-6). A non-neoplastic cause was concluded for 5 patients (pts 7-11). We still have
no etiologic explanation yet for 5 patients (pst 12-16). 2 patients (pts 17-18) died before
diagnosis and without autopsic investigation.Results: Three of the 6 proved cancers were
detected by FDG-PET-CT versus 2 on the CT. Among the 3 cancers seen on the FDG-PET-CT, 2
were not described on the CT. On the contrary, among the 3 cancers not seen on the FDG-PETCT one was described on the CT; and 2 were not either shown on the CT (table). Among the 5
patients for whom a non-neoplastic cause was concluded,FDG-PET-CT was helpful in two cases:
for the diagnosis of an infection of hip prothesis and a pulmonar vascularitis of
Wegener.Conclusions: In our serie, the number of FDG-PET-CT true positives was higher than
CT. But, since false-negatives are also possible with FDG-PET-CT, we believe that FDG-PETCT should be performed in patients with high clinically suspected PNS when conventional
Retention index (RI) was calculated according to the formula: (SUVdelayedSUVearly)/SUVearly*100%. Correlation analysis was used to analyze the two group of data
acquired through 2 delayed imaging. We select RI=0% as cut-off, the diagnostic efficacy of this
two kind of delayed imaging were compared using chi-square test. Results The maximal SUVs
(mean±SD ) of malignant lesions in three phase imaging was 9.9±5.7,10.9±6.2,10.9±6.8
separately. The SUVs of malignant lesions in two delayed phase were significantly higher than
they were in the first phase (P<0.01), but there was no significant difference (P>0.05) between
the two delayed imaging. The RI of two delayed imaging was positively correlated, (R=0.51,
P<0.01). SUVs increased in early delayed imaging in 164 malignant lesions, and 133 in late
delayed imaging. The positive rate of early delayed imaging is significantly higher than that in
late delayed imaging(76.3% v 61.9%,P<0.01) ConclusionsThe RI value in early delayed
imaging(118±18min) is correlated with that in late delayed imaging(241±29min). The later is
traditional used in most centers. Our study shows that the early delayed imaging had higher
diagnostic efficacy, and is more convenient for patient by reducing the waiting time. So early
delayed scanning is recommended in clinical practice
imaging fails to identify a tumour. Indeed, it decreases non-diagnosed cases of cancers and
sometimes can also give arguments for a non-neoplastic etiology.
Lung carcinoid
veinal thrombosis
Peritoneal mesothelioma
identifying UPC of the most common histologies, namely adenocarcinomas, squamous cell
carcinoma and poorly differentiated carcinoma. It is well known, however, that 18F-FDG PET is
of limited value in tumors that usually do not show enhanced FDG uptake, as neuroendocrine
(NE) tumors: for this reason alternative PET tracers have been proposed for studying NE tumors
and 18F-DOPA was among the most effective. Although unusual, it is possible to have UPC of
biopsy-proven NE nature: the aim of our study was to evaluate the potential role of 18F-DOPA
PET-CT for the detection of UPC in patients with proven secondary lesions of NE tumors. Our
population consisted of 5 patients (2males/3females) with at least one biopsied secondary lesion,
and all examinations and imaging procedures negative for detection of primary cancer, including
physical examination, abdominal and thoracic CT, abdominal US, whole-body somatostatin
receptor scintigraphy. In three patients secondary lesions were diffusely localized in the liver and
in abdominal lymph nodes, while one patient was previously operated for a single liver
metastasis and one patient had a positive biopsy at an inguinal lymph node. Whole-body scans
were carried out 1 hour after i.v. injection of 370 MBq of 18F-DOPA. In 2/5 cases 18F-DOPA
PET-CT detected the primary occult lesion. In a patient with diffuse abdominal secondary lesions
the primary cancer was correctly identified at pancreatic level; in a patient with doubtful findings
at CT and SRS the primary cancer was confirmed to be located in the ileum and unsuspected
bone lesions were documented. In the remaining 3 cases 18F-DOPA PET-CT did not allow to
identify the UPC; in one case, however, unknown bone and limph nodal lesions were
demonstrated.Although very preliminar, our data indicate the feasibility of detecting occult
primary tumor of NE origin using 18F-DOPA PET-CT.
Cushing syndrome
Lung carcinoid
peripheric neuropathy
Pancreas adenocarcinoma
veinal thombosis
Pancreas adenocarcinoma
Stomach adenocarcinoma
peripheric neuropathy
Hip prothesis infection
Cushing syndrome
Wegener vascularitis
peripheric neuropathy
Cryoglobulin vacularitis
10 hypercalcemia
11 polyarthralgy
Rhizomelic pseudo-polyarthritis
Combined morphological and functional imaging consisting
of gated SPECT & angiography in patients with suspected &
known CAD
13 peripheric neuropathy
Marcus Ludwig Hacker (DE)
14 dermatomyositis
15 periostitis
16 phosphated diabetes
Sybille Fischer (DE)
17 peripheric neuropathy
18 veinal thrombosis
Perfusion tracers & stress modalities – which test for which
patient ?
104 — Sunday, October 1, 2006, 8:00 am — 9:30 am, Mitropoulos
CTE 1 Cardiology
Cardiac SPECT – State of the Art
A. Gagnon (CAN)
Diagnostic value of FDG-PET in Paraneoplastic Syndromes
J. R. Garcia, M. Simo, M. Soler, G. Perez, S. Lopez, F. Lomeña; Cetir
Grup Medic, Esplugues Llobregat, Spain.
AIM: Paraneoplastic syndrome (PS) includes a wide range of heterogeneous clinical symptoms
and signs which are associated to malignancy. Diagnostic is difficult with conventional
diagnostic techniques and it is crucial to differentiate PS from autoimmunitary diseases, in light
of the therapeutic options. We aimed to evaluate the usefulness of FDG-PET in the clinical
management of these patients.Materials and Methods: Twenty-six patients with PS suspected (15
men, 11 women; age range 44-73 year-old) with a FDG-PET study performed in the last 4 years
were included in the study. Clinical diagnoses that forced to the PET study were as follows:
polyneuropathy (N=9), encephalitis (N=3), cerebellar syndrome (N=3), Eaton-Lambert syndrome
(N=3), Stiffman syndrome (N=1), carcinoid syndrome (N=2), Cushing (N=2), and
thromboflebitis migrans (N=3). Whole-body PET study was obtained with GE Advance Nxi
equipment 1 h after FDG administration (0,125 mg/kg). PET study results were compared to
histopathology (N=13) and clinical-imaging follow-up (range 8-16 months).Results: PET study
was negative in 15 patients. Follow-up confirmed non evidence of malignancy, except for 2
patients in whom lung cancer was detected respectively 5 and 7 months later. In 11 patients PET
showed images suspicious for primary malignancy. This was confirmed by histopathology in 9
cases: small cell lung cancer (N=6), rectum (N=1), ovarian (N=1) and oropharynx (N=1). One
patient was diagnosed with adrenal adenoma and another with lung infection. In 3 out of 6
patients with bronchogenic carcinoma, lymph node infiltration was detected ipsilaterally. These
patients were treated with chemotherapy/radiotherapy (N=6) and surgery/chemotherapy (N=3).
To sum up, PET identified 9 out of 11 malignancies (sensitivity: 82%) and rule out malignancy
in 15 out of 17 patients (specificity: 88%).Conclusions: In our study, the PS was associated to a
malignancy in 35% of patients. PET was able to detect tumour uptake of FDG in lesions not
previously detected by the conventional techniques. It allowed accelerating the diagnostic
process of PS, with direct implications in subsequent therapy.
Whole-body F-DOPA PET-CT for the detection of unknown
primary cancer of neuroendocrine nature.
S. Fanti1, P. Tomassetti1, D. Campana1, C. Nanni1, P. Castellucci1, M.
Farsad1, V. Allegri1, G. Montini1, V. Ambrosini1, F. Nori1, S. Pandolfi1, A.
Palucci1, D. Rubello2, A. Alavi3, R. Franchi1; 1S.Orsola-Malpighi, Bologna,
Italy, 2Ospedale S.Maria della Misericordia, Rovigo, Italy, 3University of
Pennsylvania, Philadelphia, United States.
Unknown primary cancer (UPC) is defined as a biopsy-proven secondary lesion with no
detectable primary tumor after physical examination and conventional imaging tests (X-Rays, CT
and US). Whole-body PET and PET-CT using 18F-FDG have been successfully used for
105 — Sunday, October 01, 2006, 8:00 am - 9:30 am, Scalcotas
Oncology: Lymphoma
FDG uptake in untreated follicular lymphoma
G. L. Cascini1, F. Oliviero1, M. L. Vigliotti2, F. Frigeri2, L. Aloj1, C. Caraco'1,
E. Squame1, A. R. De Chiara3, A. Pinto2, S. Lastoria1; 1Medicina Nucleare,
Istituto Nazionale Tumori, Fond. "G.Pascale", Napoli, Italy, 2Ematologia,
Istituto Nazionale Tumori, Fond. "G.Pascale", Napoli, Italy, 3Anatomia
Patologica, Istituto Nazionale Tumori, Fond. "G.Pascale", Napoli, Italy.
Follicular lymphoma (FL) is the second most frequent type of non-Hodgkin's lymphoma (NHL),
being 20-25% of all lymphomas. The role of FDG-PET, well established in staging and
monitoring patients (pts) with Hodgkin's disease and aggressive NHL, it is still to be clarified in
FL. No data are yet available as to the prognostic significance of SUV in FL. We studied 62
consecutive pts [median age 55 years (range 27-82 yrs)], newly diagnosed FL, who underwent
pre-treatment FDG-PET scan before bone marrow biopsy. The highest SUV value for each
patient was statistically correlated with known prognostic factors in FL. FDG-avid sites of FL
were demonstrated in 52 of 62 pts (83.8%). The rate of FDG-PET positivity did not correlate
with histologic grading (G1, 76.9%; G2, 90.9 %; G3b, 85.7%; p=0.4); tumor cell proliferation
rate (Ki-67 staining), extranodal involvement, bone marrow infiltration, B symptoms and bulky
disease. Conversely, the rate of PET positivity appeared to progressively increase among pts with
higher FL International Prognostic Index (FLIPI) score (low, 66.6%; intermediate, 93.7%, high,
95.4%; p=0.016), even though a direct correlation with single IPI components (age, stage,
hemoglobin, LDH) was not statistically significant (p=0.49), except for the presence of > 4
involved nodal areas (p=0.04). The mean SUV was 6.34 (r 1.2-24.5) and no significant
differences were found in relation to histologic grading (SUV: G1, 6.42; G2, 5.76; G3, 7.08). By
stratifying the pts in 4 different groups with increasing SUV (i.e. 12) statistically significant
associations were found with tumor cell proliferation rate (Ki-67; p=0.02), B symptoms (p=0.03)
and bulky disease (p=0.0003). Finally, analysis of the ROC curve against prognostic factors
confirmed that a SUV <11.7 is strongly associated (97.7%; p< 0.00001) with the absence of
bulky disease. The intensity of FDG uptake in FL is independent from histologic grading but
correlates with some biologic features: tumor proliferation rate, host-tumor interplay, or bulky
disease. The rate of PET positivity well correlates with FLIPI risk, being highly linked to nodal
dissemination (> 4 sites), which might increase the probability of FDG uptake. These data
suggest that serial FDG-PET evaluations are warrented in pts with FL.
Whole-body fdg-pet compared to conventional imaging
methods for staging patients with hodgkin´s lymphoma in
J. Puskacova1, I. Makaiova2, A. Foltinova1, A. Hraskova1, J. Jindrova1, J.
Vesely2, S. Kovacova2, P. Banki2, V. Procka2, E. Kaiserova1, D. Haviar1;
University Children´s Hospital, Clinic of Pediatric Oncology, Bratislava,
Slovakia, 2Clinic of Nuclear Medicine Medical Faculty UK and St.
Elisabeth´s Oncologic Institute, Bratislava, Slovakia.
Objectives. Improvement of survival and reduction of long-term sequalae of patients with
Hodgkin´s Lymphoma ( HL) in children relies upon precise initial staging and appropriate
treatment. 18FDG-PET became efficient method routinely used for staging with it´s possibility to
detect the disease at metabolic level. In Slovakia we perform 18FDG-PET in children since
October 2000. Methods. At the time of diagnosis were concomitantly used 18FDG-PET scans
and other conventional imaging methods ( CIMs) as well. We performed retrospective analysis
of staging, comparation of these modalities and we evaluated discrepancies and influence on
staging. Results. Since October 2000 till March 2006 we had 41 patients with newly diagnosed
HL, 14 boys, 27 girls, median age 172 months ( range 95-214months). Most frequent histologic
type was nodular sclerosis (35), then mixed cellularity (4), nodular paragranuloma (2). One
patient was stage I A, 20 st. IIA, 5 st.II B, st. IIIA, 2 st.IIIB, 7 st.IV A, 1 st.IV B. Patients were
treated according POG protocols 9425 and 9426. All patients but one are alive, 2 of them in
second remission after autologuous transplantation. In all patients USG and CT scan were done.
Staging 18FDG-PET scan (using SUV for semikvantitative assesment) was concomitantly done in
35 patients (85%). Concordance in these modalities was in 17/35 cases (49%). 18FDG-PET
caused upstaging in 8/35 (23%) cases ( 1x from I to II, 2x from II to III, 5x from III to IV, 1x
from II to IV). Only in 3/35 (9%) cases downstaging was indicated (1x from III to I, 2x from III
to II ). In one case was 18FDG-PET completely negative. 18FDG-PET was more sensitive for
detection lymph nodes bellow diaphragma ( 4x), above diaphragma (1x), lungs (1x) and spleen (
3x). In 6 cases 18FDG-PET revealed bone marrow infiltration, only in 1 of these patients
concomitantly done trephine biopsy was possitive. Conclusions. Whole-body 18FDG-PET has
become routinely used staging method for our pediatric patients with HL. It enabled more
appropriate staging and influenced the management in some of these patients. 18FDG-PET scan
performed at the time of diagnosis is essential for regular assesment of the treatment response (
correlation of SUV values).
The challenge of immunohistochemical data and FDG-PET
staging as prognostic factors in non-Hodgkin lymphomas at
M. C. Nollevaux, C. Nogarède, A. Sonet, B. Krug, A. S. Pirson, J. Jamart, J.
George, M. Delos, A. Bosly, T. Vander Borght; Mont-Godinne Medical
Center, Université Catholique de Louvain, Belgium.
Last year we reported that the initial staging by FDG-PET appears to be a major risk factor on
survival of NHL patients among other factors, such as histological aggressiveness; conventional
staging; SUVmax and the International Prognostic Index (IPI/FLIPI) (Eur J Nucl Med Mol Imag
2005;32:S23). Aim: In the current study, we review the node biopsies to measure different
histological parameters of aggressiveness as well as to correlate them with the patient survival
and the FDG-PET study.Materials and Methods: From 03/2000 to 12/2004, 82 NHL underwent
FDG-PET for primary staging; 55 were aggressive and 27 indolent. Ki67, Bcl2 and p53 were
measured semi-quantitatively on node biopsies. Differences between aggressive from indolent
NHL as well as risk factors on survival were studied by univariate and multivariate analyses. The
later were also correlated with SUVmax using Spearman’s correlations.Results: Although FDGPET staging was a major prognosis factor on survival by multivariate analysis (p=0.034),
conventional staging (p=0.044), IPI/FLIPI (p=0.030) and Ki67 (p=0.056) also owned some
prognostic information. In contrast, histological classification, Bcl2, p53 and SUVmax were not
significant prognostic factors. Aggressive NHL were more extended by FDG-PET staging
(p=0.054), avid (SUVmax 6.7±5.8 and 15.8±9.1, respectively; p<0.001) and proliferative (Ki67;
p<0.001) compared to indolent NHL. Only Ki67 correlated significantly with SUVmax (r=0.043;
p=0.001).Conclusions: These data confirm the major prognostic information of the initial FDGPET staging in NHL. They also support a similar role for Ki67. The latter correlated with
SUVmax, but not for Bcl2 and p53. Analysis of the prognostic value of combined FDG-PET and
immunohistological data are underway.
Prognosis value of positron emission tomography using F18fluorodeoxyglucose (FDG-PET) in the setting of ASCT in
patients with high grade non-Hodgkin lymphoma (HGNHL).
C. Bodet-Milin1, V. Rolland2, A. Oudoux1, C. Aansquer1, S. Le Goille2, F.
Valette1, C. Rousseau3, B. Dupas4, P. Moreau2, J. Harrousseau2, F.
Kraeber-Bodere1; 1Nuclear Medicine, University Hospital, Nantes, France,
Hematology, University Hospital, Nantes, France, 3Nuclear Medicine,
Cancer Center, Nantes, France, 4Radiology, University Hospital, Nantes,
FDG-PET is currently performed to evaluate therapeutic response in HGNHL. The aim of this
study was to evaluate retrospectively the prognosis value of FDG-PET performed before and 3
months after autologous stem cells transplantation (ASCT) in patients treated for
HGNHLMaterials and Methods: Forty-nine patients (26M, 23 F), median age 51 [17-65], with
diagnosis of DLBCL (n=44), mantel cells (n=4), and anaplasic T lymphomas (n=1) were treated
between 2002 and 2005. The International prognosis index (IPI) was 0-1 in 18 cases (39 %) and
2-3 in 31 cases (61%). All patients with 0-1 IPI had a bulky disease, stage IV or less than
complete response after the initial chemotherapy regimen. ASCT was performed as part of first-
line therapy in poor risk patients (n=39, 79.6%) or at time of relapse (n=10). FDG-PET imaging
was performed 60 to 90 min after iv injection of 18FDG using an hybrid FDG-PET/CT. The
median follow-up of living patients was 12 months [3-43 months].Results: -/- : FDG-PET
negative before and after ASCT -/+ : FDG-PET negative before and positive after ASCT. +/+:
FDG-PET positive before and after ASCT. +/- : FDG-PET positive before and negative after
FDG-PET Results
- /- -/+ +/+ +/• Total
23 1
• Evaluation after ASCT (CHESON’s criteria)
Ø Relapse or progression
22 0
Ø Alive in CR
20 0
Ø Alive in PR
Ø Alive in relapse
Ø Death from disease progression
Ø Death from other causes
• Last follow-up
Conclusions: A negative FDG-PET before ASCT has a good prognosis value and a positive
FDG-PET before and after ASCT is associated with a poor outcome. Interestingly, patients with
positive status before ASCT and negative status after ASCT have the same prognosis than those
negative before and after ASCT (under reserve because of the low number of patients and the
short follow-up).
Respective Value of PET using 18F-fluorodeoxyglucose and
Lymphangiography in Staging Hodgkin’s disease patients
with negative abdomino-pelvic CT.
F. Valette1, S. Querellou2, A. Oudoux1, C. Bodet-Milin1, T. Carlier1, B.
Dupas1, J. Chatal1, O. F. A. J. Couturier1; 1University of Nantes, Nantes,
France, 2University of Brest, Brest, France.
Objective: Lymphangiography (LAG) and Positron Emission Tomography (PET) using 18Ffluorodeoxyglucose (FDG) may identify small tumor deposits in lymph nodes not enlarged at
computerized tomography (CT). We evaluated the respective role of PET and LAG in staging
Hodgkin’s disease (HD) patients stage I-II (supradiaphragmatic), i.e. with negative contrastenhanced infradiaphragmatic CT.Methods: 28 HD patients underwent FDG-PET and LAG at
initial staging. Concordant positive findings at both tests were regarded as actual HD locations
and concordant negative findings as true negative. In case of discrepancy, the reference was
biopsy or magnetic resonance imaging (MRI). Results : Concordant results were obtained in 26
patients (24 negative; 2 positive). In two of the 24 negative patients, PET showed additional
lesions in the spleen and one celiac lymph node (1 patient) or in the right kidney and the right
iliac crest (1 patient). Discordant results were obtained in two patients. Both methods indicated
infradiaphragmatic involvement in different locations (1 patient). PET was falsely positive in the
last patient (PET done within 24 h after a negative LAG), confirmed by biopsy (inflammation
due to LAG medium). Conclusion : FDG-PET and LAG gave comparable results for staging HD
patients with negative infradiaphragmatic CT, making not necessary invasive LAG. Furthermore,
LAG when performed before PET, can be responsible for false positive PET results.
FDG-PET/CT in staging of patients with lymphocytepredominant hodgkin disease (paraganglioma of poppema
T. Hervouet1, T. Lamy1, C. Bodet-Milin2, S. Le Gouill2, A. Devillers1, F.
Gaillard2, E. Frampas2, F. Kraeber-Bodéré2, C. Ansquer2; 1University
hospital, Rennes, France, 2University hospital, Nantes, France.
Lymphocyte-predominant Hodgkin disease (LPHD) is an indolent disease, with a majority of
patients at clinical stage I or II. The prognosis is usually good despite a tendency to relapse but
fatal complications can occurred and are mostly attributed to treatment. This study was designed
to evaluate retrospectively the usefulness of 18FDG-PET/CT in pre-treatment staging of patients
with LPHD. PATIENTS AND METHODS Ten patients (5 M, 5 W, median age: 34 years) with
LPHD were studied before treatment for initial disease (n = 7) or recurrence (n = 3). All patients
underwent a whole body PET-scan 60 min after injection of 5 MBq/kg of 18FDG on an integrated
in-line PET/CT system and conventional explorations (CE): computed tomography, clinical
examination, bone marrow biopsy and ultrasounds. FDG-PET images were analyzed visually and
tumor SUV max determined. The Ann Arbor stage assigned on the basis of FDG-PET was
compared to the stage determined by CE. The standard of reference for the disease was imaging,
pathological and follow-up data. RESULTS A total number of 25 sites (1 to 8 per patient) were
involved. 3/24 nodal sites were totally removed before imaging. All the 21 remaining nodal sites
(100%) were detected by PET/CT and only 13 (62%) by CE. The extra-nodal lesion (a lung
nodule) was only detected by PET/CT. SUV max ranged from 1.75 to 15.6 (median value: 6.1),
with variable values in the same patient. Interestingly, the highest SUV max was found in a
patient in which a transformation to a high grade NHL was suspected. Modifications of staging
are summarized in the table. Based on PET/CT findings the radiation fields were modified in 1
patient and a localized radiotherapy was performed in an other patient.
Restaging after PET/CT
Number of patients
I 3
Stage II
Stage III to II 2
Down staging
Stage III to I
Up staging
Stage II to IV 1
107 — Sunday, October 01, 2006, 8:00 am - 9:30 am, MC 3
Stage I to II
Clinical Science: Gastroenterology
CONCLUSION Our preliminary results indicate that PET/CT is accurate in the staging of
LPHD. PET/CT is particularly useful to confirm localized disease that require minimal treatment
and could have then an impact on the therapeutic strategy. A multi center study is on going to
confirm these results in a larger series.
The value of Positron Emission Tomography (PET/CT) in
staging of diffuse large B-cell lymphoma.
S. Fuertes Cabero1, X. Setoain1, A. López-Guillermo2, F. Lomeña1, D.
Fuster1, J. Ortín1, F. Pons1; 1Nuclear Medicine Department. Hospital Clínic.
University of Barcelona., Barcelona, Spain, 2Hematology Department.
Hospital Clínic. University of Barcelona., Barcelona, Spain.
Aim: The aim of this study was to evaluate the role of fluorodeoxyglucose positron emission
tomography (FDG-PET) in the staging of lymphoma patients.We wanted to study whether
PET/CT improves staging and changes the management of aggressive lymphoma patients in
comparison with the conventional imaging modalities (computed tomography (CT), Ga-67
scintigraphy).Materials and Methods: Forty consecutive patients (25 men and 15 women, median
age of 57.3 years) with histological proven diffuse large B-cell Non Hodgkin lymphoma
(Aggressive lymphoma), were prospectively evaluated. All 40 patients underwent a whole body
FDG PET/CT and conventional staging techniques (chest and abdomen CT, Ga-67 scintigraphy)
were studied before therapy. Sixty minutes after the intravenous administration of 370 MBq
FDG, a whole body PET/CT was acquired. We hypothesize that PET/CT improves the diagnostic
staging of lymphoma and changes the clinical management of patients.Results: PET/CT and CT
were concordant in 26 patients (65%). However, PET/CT detected more lesions than CT in 12
patients (30%). Only in one patient, CT revealed more extensive disease than PET/CT.
Additional information of PET/CT had lead to a change in staging (upstaging) in 6 patients
(15%), in turn leading to a change in treatment strategy in 1 patient. PET/CT and Ga-67
scintigraphy were concordant in 23 patients (60.5%). PET/CT detected more lesions than Ga-67
scintigraphy in 14 patients (42%). PET/CT results changed staging (upstaging) in four patients
(15%), leading to a change of treatment strategy in none patients.Conclusions: The impression is
that PET/CT detected more lesions than conventional examination, but this rarely translates into
changes of staging and treatment strategy in diffuse large B-cell Non Hodgkin lymphoma
Recommendations of the imaging committee of the
International Harmonization Project (IHP) for FDG-PET (PET)
use in patients with lymphoma
M. Juweid , S. Stroobants , F. Mottaghy , O. Hoekstra , A. Guermazi , M.
Dietlein6, G. Wiseman7, K. Scheidhauer8, A. Buck3, R. Naumann9, R.
Hicks10, S. Reske3, M. Schwaiger8, V. Diehl6, B. Cheson11; 1University of
Iowa, Iowa City, IA, United States, 2University Hospital Gasthuisberg,
Leuven, Belgium, 3University of Ulm, Ulm, Germany, 4VA University
Medical Center, Amsterdam, The Netherlands, 5SynarcInc., San Francisco,
CA, United States, 6University of Cologne, Cologne, Germany, 7Mayo
Clinic, Rochester, MN, United States, 8Technische Universitat Munchen,
Munich, Germany, 9Universitatsklinikum Dresden, Dresden, Germany,
The University of Melbourne, East Melbourne, Australia, 11Georgetown
University, Washington, DC, DC, United States.
Aim: International guidelines for PET use in lymphoma are desirable to standardize PET
imaging and interpretation in clinical practice and compare results among studies. The IHP was
convened to discuss harmonization of clinical trial parameters in lymphoma. Six committees
were formed, including imaging. We present the imaging committee’s recommendations
endorsed by the IHP.Materials and Methods: These recommendations were based on published
PET literature. Consensus was reached regarding controversial points. Results: PET should be
routinely used for response assessment at the conclusion of therapy (tx) in pts with Hodgkin
Lymphoma (HL) and diffuse large B-cell lymphoma (DLBCL), in which case a pretreatment
PET is strongly encouraged but not obligatory for response assessment as these lymphomas are
routinely FDG-avid. In contrast, pretreatment PET is mandatory for variably FDG-avid (VA)
aggressive NHLs if PET is used to assess response. PET should only be used for response
assessment of indolent NHL and mantle cell lymphoma (MCL) if response is a major endpoint in
a clinical trial, in which case a pretreatment PET is strongly encouraged but not mandatory in
MCL and follicular lymphoma (FL) but required for all other indolent NHLs. For all VA NHL
histologies, PET should only be used to assess response if positive prior to tx. PET at tx
conclusion should not be performed prior to 3 wks post-chemo-/chemoimmunotherapy,
preferably at 6-8 wks, and 8-12 wks post-radiation/chemoradiation. Visual assessment alone is
adequate for interpreting PET findings as positive/negative when assessing response at tx
conclusion. Mediastinal blood pool structures (MBPS) uptake is recommended as a "reference
background tissue" to define PET-positivity for a residual mass (RM) > 2 cm in greatest
transverse diameter (GTD), regardless of its location; such RM is only considered positive if its
uptake intensity is greater than that of MBPS. In contrast, a RM 1.1-1.9 cm in GTD or a normalsized lymph node (i.e., < 1 cm) should be considered positive if its uptake intensity is above
surrounding background. Specific criteria for defining PET positivity in the liver, spleen, lung,
and bone marrow are also recommended. Non-attenuation-corrected PET is acceptable for
response assessment but is discouraged in favor of attenuation-corrected PET. “Mid-tx” PET,
performed after 1-4 cycles of chemotherapy, should only be done in a clinical trial. There is
currently no justification for PET as routine post-tx surveillance. Conclusions: We recommend
these guidelines be adopted by the imaging community to facilitate patient care and comparison
among studies.
Esophageal diverticula:scintigraphic evaluation.
V. Valenza, D. Di Giuda, M. Pizzoferro, M. Totaro, D. Mattioli, R. Urso, P.
Granone, V. Porziella, G. D'Errico, A. Giordano; Catholic University of
Sacred Heart, Rome, Italy.
Aim: Esophageal diverticula (ED) are often defined by their anatomic position. Cervical
diverticula (CD) (Zenker’s diverticula) occur in the area of the cricopharyngeous muscle. Midesophageal diverticula (MED) develop from the middle third of the esophagus. Epiphrenic
diverticula (EpD) arise from the distal esophagus proximal to the lower esophageal sphincter. ED
are a rare condition and are generally found in middle-aged-elderly patients. The majority of
patients are asymptomatic or have minimal symptoms (dysphagia, noisy deglutition, chest pain,
regurgitation of ingested food, aspiration, halitosis). Malignancy represented by squamous cell
carcinoma, is a rare complication of esophageal diverticula. The aim of this study was to assess
the role and effectiveness of oropharyngo-oesophageal scintigraphy (OPES) in the diagnosis and
follow-up of symptomatic patients affected by ED.Methods: We studied 38 patients (16/22
male/female, mean age 66±18 years; 15 CD and 23 MED or EpD), and 17 healthy volunteers
(10/7 male/female, mean age 53±25 years) as control group. All patients were submitted to
manometry (MM), radiologic esophagogram (RE) and OPES. OPES method: we administered 37
MBq of 99mTc-colloid diluted in 10cc of water and acquired 480 sequential images (0,125
sec/frame for a total 60 sec) with patient standing in front of the gamma camera in 80° right
anterior oblique position. Two static images, before and after ingestion of 100ml of water
respectively, were performed at the end of the dynamic phase to evaluate persistence of
radioactivity in esophagus as an indirect sign of inflammation. Sequential scintigraphic images
permitted qualitative evaluation as diverticulum visualization, multiple deglutitions, refluxes and
inflammation; T/A curves permitted quantitative analysis: Oral, Pharyngeal and Esophageal
Transit Times (OTT, PTT, ETT) and Retention Indexes (ORI, PRI, ERI).Results: OPES
qualitative analysis showed: diverticulum visualization in 92% of patients (100% at RE),
multiple deglutitions in 40%, refluxes in 26% (31% at RE), inflammation in 84% (60% at RE);
moreover 10% of patients (all CD) showed tracheal aspiration. Quantitative analysis
demonstrated an increase of the following parameters: OTT in 52% of patients and ORI in 42%;
PTT in 60% and PRI in 31%; ETT in 58% and ERI in 71%. MM showed hypertone of upper
esophageal sphincter in 15 patients (40%), 8 of whom had hypotone of lower esophageal
sphincter.Conclusions: OPES demonstrated a good correlation with other diagnostic techniques
usually used in the evaluation of oro-esophageal disorders associated with ED. Furthermore,
OPES is a low dose sensitive, well tolerated and simple technique.
Improved the gastric emptying in diabetes patients with low
dosage erythromycin
S. L. Chen, Y. S. Gu, W. G. Liu, X. F. Chen, B. L. Li; Zhongshan
Hospital,Institute of Nuclear Medicine,Fudan University, Shanghai, China.
Objective:The aim of this study was to investigate the effect of low doses erythromycin on the
gastric emptying of diabetes.As a chronic complication of diabetes mellitus, there are few
effective treatments against gastroparesis.Methods: 30 male diabetes patients were determined
gastric emptying rates before and after erythromycin administration. All patients were fasted at
lest for 6 hr before the examination. The gastric emptying rates were determined following a
standard meal consisting of 99mTc-sulfur colloid bound to eggs, mixed with 150g of noodles. A
series scans were acquired for up to 3 hr after ingesting the meal and stomach radioactivity was
measured. After the first examination, the thirty diabetes patients were divided into three groups.
Group A take orally erythromycin lactate 0.125mg/day for 3 days, Group B received 0.25mg
/day, Group C 1mg /day. The gastric emptying rates were determined again in the same
condition. Another 10 health volunteer men treated with a control group (Group D).Results: The
gastric emptying of diabetes patients was obviously delayed. In control group (Group D), the T1/2
of gastric emptying was 62.0±27.3min. And in diabetes patients the T1/2 of gastric emptying was
116.9±33.3min. Compared with normal health volunteer, the T1/2 of gastric emptying was
dramatically delay (P<0.01). All of the three different erythromycin doses improved gastric
emptying significantly (T1/2 84.8±29.6min, P<0.01). There were no differences in Group A,
Group B, and Group C. No dosage-effect relation was found. After low dosage erythromycin
treatment for 3 days, the gastric emptying accelerated significantly (84.8±29.6min vs
116.9±33.3min, P<0.01 ) in patients with diabetes.Conclusions: Gastric emptying was prolonged
significantly in patients with diabetes. Low dosage erythromycin treatment can improve the
gastric emptying.
Evaluation of the relationship between the metabolic status,
the gastric emptying and the diabetic neuropathy
Delayed gastric emptying scintigraphy in Cystic Fibrosis
patients before and after lung transplantation.
M. Papós1, T. Várkonyi2, É. Börcsök2, R. Takács2, C. Lengyel2, M. Lázár1,
P. Kempler3, E. Máté4, J. Lonovics2, L. Pávics5; 1Eurodemic Diagnostics
Szeged Ltd, Szeged, Hungary, 21st Department of Medicine, University of
Szeged, Szeged, Hungary, 31st Department of Medicine, Semmelweis
University, Budapest, Hungary, 4Institute of Informatics, University of
Szeged, Szeged, Hungary, 5Department of Nuclear Medicine, University of
Szeged, Szeged, Hungary.
C. Bodet-Milin1, S. Querellou2, A. Oudoux1, A. Haloun1, D. HoreauLanglard1, T. Carlier1, Y. Bizais2, O. F. A. J. Couturier1; 1University of
Nantes, Nantes, France, 2University of Brest, Brest, France.
The possible correlations between the metabolic status, the gastric emptying and the neuropathy
are still not clearly explored by the studies of the past decades. The aims of this study were to
perform a continuous glucose monitoring during the determination of gastric emptying and to
evaluate the severity of neuropathy in patients (pts) with type-1 diabetes (DM). Patients,
methods: 13 pts with type-1 DM were included (HbA1c: 8.4±0.7%, age: 35.5±2.5 years,
duration of DM: 15.1±2.9 years; mean±SE). The emptying of the stomach was evaluated by a
scintigraphic gastric emptying procedure. The subcutaneous glucose levels were determined by a
CGMS method during the total interval of the gastric emptying. Cardiovascular reflex tests were
applied for the assessment of autonomic neuropathy (AN). During the investigations the pts were
free of beta-blockers. Sensory nerve integrity was studied with a Neurometer. Results: There
was a longer gastric emptying in diabetic patients than in healthy subjects, but this difference did
not reach the level of significance (T1/2: 84.4±12.4 vs 49.6±5.5 min., p=0.06, diabetic vs control).
The glucose levels recorded during the gastric emptying (the lowest and highest, the mean, the
highest difference in glucose) did not exhibit correlations with the gastric motility. The gastric
emptying of groups of pts created by the different glucose parameters did not differ significantly.
The HbA1c values did not correlate with the gastric motility as well. Moderately severe AN was
found in diabetics (AN score: 2.9±0.5 vs 0.3±0.2 p<0.001, heart rate response to beathing:
16.4±1.9 vs 26.3±2.3 beats/min, p<0.01; 30/15 ratio: 1.04±0.02 vs 1.21±0.04, p<0.01; Valsalva
ratio: 1.34±0.06 vs 1.71±0.1, p<0.01; handgrip: 15.6±2.8 vs 28.3±3.5 mm Hg, p<0.05). The
current perception thresholds (CPT) on the peroneal nerve at 5 Hz of the patients differed from
controls (CPT: 3.12±0.9 vs 0.68±0.07 mA, p<0.05). Conclusions: Any of the parameters of the
continuously measured actual glucose levels did not correlate with the gastric emptying. Slightly
slower gastric emptying was found in the presence of a moderately severe autonomic and sensory
neuropathy in pts after a 15-year-long duration of DM. These data may strengthen the hypothesis
that the current glucose levels have a less important role than neuropathy in the pathogenesis of
delayed gastric emptying.
Role of esophageal scintigraphy in evaluation of esophageal
motor dysfunction in patients with chronic renal failure
M. Roshdy, S. Salem, M. Fawzy, M. El khateeb; Cairo university, cairo,
Introduction: Esophageal motor dysfunction (EMD) has been reported in patients with chronic
renal failure (CRF). However, a little data have been reported about how to relate patients’
symptoms to qualitative (Qul) & quantitative data (Qun). Esophageal scintigraphy (ES) provides
safe, accurate and non-invasive method to detect this abnormality for further treatment and
follow-up. The aim of the study: To assess EMD using ES. Materials & Method: The study
included 31 patients with CRF (group I), 10 patients with renal impairment (group II), and 10
healthy control (group III). Their age ranged 23-60 years (44+10). Exclusion criteria included:
patients with amylodosis, diabetes mellitus, collagen vascular diseases and patients on drugs that
affect gastrointestinal motility. The study was performed using dual head gamma camera
(ADAC) equipped with all-purpose collimator and connected to Pegasus computer. The patient
was fasting for a period of 4-6 hours. 1 mCi of Tc -99m labeled semisolid meal was administered
orally to the patient. The data were acquired using dynamic frames (0.5 sec/frame) for a total of
240 frames. Then anterior & posterior static images were acquired at 10 min. Four ROI’s were
drawn on the proximal, middle, distal third of the esophagus as well as the whole esophagus.
Images were reframed and time activity curves were generated for the four (ROI’s). Condensed
images (CI) were created for all patients. Then the data were qualitatively assessed by reviewing
the cine of the images & the CI. Quantitative assessment of the studies was done by calculating
the total esophageal transit time (ETT), the global esophageal emptying time (EET), esophageal
emptying rate (EER %) as well as residual activity (RA %). Results: Qul assessment of cases
revealed esophageal abnormalities with abnormal curves in all cases of group I, while 3/10
patients in group II (33.3%) had esophageal abnormalities.The ETT, EET, EER and RA of the 3
patient groups are demonstrated in the following table:
Group I
Group II
Group III
23.4±16.5 19.9±7.1
35.4 ±11.2 68.5±16.3 72.3±18.3
RA %
26.6±15 % 4.2±1.7 % 3.1±0.8 %
A statistically significant difference was found between group I & II and group I & III in the 4
indices (ETT, EET, EER and RA) (P<0.05). While insignificant difference was found between
group II & III.Conclusions: ES provides an accurate method to detect EMD in patients with CRF.
Objective : Evaluation of the rate of gastric emptying (GE) in cystic fibrosis patients scheduled
for lung transplantation.Methods: 30 patients (20 males, 10 females, 22.6 ± 6.4 years) were
evaluated by GE scintigraphy before (1.58 ± 1.11 years) and early after (5.8 ± 2.6 weeks) heartlung transplantation (n=13) or lung transplantation (n=17). Solid retention rates at 2h and 3h
(RR2-RR3), half-emptying times (T50) of solids and of liquids obtained before transplantation
were compared to those after transplantation. Results were also compared to those obtained in a
control group of 53 healthy volunteers.Results: Before surgery, 20 patients (67%) displayed a
delayed GE (T50 of solids = 160.86±59.21 min. vs. controls 75.43±15.13 min., p<0.0001), and
four of them displayed also a delayed T50 of liquids. After surgery, T50 of solids were not
reliable (too important stasis) in 24 patients. Thus, analyses were done on the basis of solid
retention rates. 29 patients (97%) displayed a very delayed GE when compared to controls
(p<0.0001), and 20 of them displayed also a delayed T50 of liquids. RR2 and RR3 were
significantly higher after surgery than before (RR2=86±17% and RR3=77±22% vs. 50±24% and
27±24%, after and before surgery respectively, p<0.0001). However, there was no correlation
between pre- and post-transplantation scintigraphic results.Conclusions: Delayed GE of solids
was a frequent abnormality in patients with end-stage cystic fibrosis, and dramatically delayed
after surgery in almost all patients. These results underline the need of an early management of
such patients by dietary manipulation and/or prokinetic medications.
Imaging Chron’s disease by 99mTc-HMPAO-labeled
leukocyte: usefulness of SPECT/CT with a hybrid camera.
L. Filippi1, C. Petruzziello2, C. Manni3, R. Barone1, C. Bruni1, P. Nicoli'1, R.
Danieli1, L. Biancone2, O. Schillaci1; 1Department of Diagnostic Imaging,
University of Rome "Tor Vergata", Rome, Italy, 2Department of Internal
Medicine, University of Rome “Tor Vergata”, Rome, Italy, 3Department of
Diagnostic Imaging, University of Rome, Rome, Italy.
Aim: Scintigraphy with 99mTc- HMPAO -labeled leukocytes is known to be a useful tool to
diagnose intestinal lesions in Crohn’s disease (CD). The aim of this study is to evaluate the
clinical usefulness of single photon emission computed tomography (SPECT) and transmission
computed tomography (CT) performed simultaneously with a hybrid imaging device for
functional anatomical mapping of inflammatory intestinal lesions in CD. Materials and Methods:
Twelve inflammatory bowel disease patients assessed by conventional endoscopy or radiology
were enrolled. Leukocytes were labeled with 99mTc-HMPAO. Planar images were acquired at
30 min and 2 h after injection. SPECT/CT was obtained at 2h using a dual-headed gammacamera coupled with a low-power X-ray system. Results: Both planar and SPECT images were
negative in one patient and positive in the remaining eleven patients showing active bowel
inflammation. SPECT/CT provided precise anatomical localization of all positive foci. In 3 out
of 11 patients fusion imaging enabled the characterization of inflammatory lesions near sites of
physiological uptake (i.e. bladder, bone marrow). In two patients with severe perianal disease,
SPECT showed an area of leukocyte accumulation in the pelvis; the fusion with the anatomical
(CT) data allowed to demonstrate that the uptake was localized along a perianal fistulous tract.
Conclusions: our results suggest that SPECT/CT with hybrid device may improve the accuracy
of Tc-99m HMPAO labeled leukocyte scintigraphy to detect CD inflammatory lesions,
especially within the pelvic floor.
Red blood cell scintigraphy in occult gastrointestinal
bleeding: usefulness of SPECT/CT with hybrid camera.
L. Filippi1, R. Danieli2, C. Manni1, P. Nicoli'1, P. Romano1, O. Schillaci1;
Department of Diagnostic Imaging, University of Rome "Tor Vergata",
Rome, Italy, 2Department of Diagnostic Imaging, University of Rome,
Rome, Italy.
Aim: Red blood cell (RBC) scintigraphy is often applied to detect lower gastrointestinal (GI)
bleeding sites, when other studies have failed to detect the source of the blood loss. Nevertheless,
nuclear imaging is lacking of precise anatomical landmarks, so it might be difficult to obtain an
anatomical localization of the site of bleeding. The aim of our study was to assess the usefulness
of SPECT/CT with hybrid camera to discolse the source of GI bleeding.Methods: Eight patients
(5 male, 3 female, age 65 + 12 years) with positive fecal occult blood test were included. All
subjects underwent endoscopy and angiography, which were unable to detect the source of
bleeding in all cases. Scintigraphy was performed by using the in vivo method of RBC labeling.
After tracer injection, patients were studied using sequential sets of dynamic images. If no
bleeding source was evident after 90 min of observation, study was stopped and stored. When the
bleeding source was identified, a SPECT/CT was acquired by using a dedicated hybrid camera
(Millenium VG & Hawkeye; General Electric Medical System, Milwaukee, WI, USA) to localize
the blood site.Results: In three of eight patients, RBC imaging was unable to identify the source
of blood loss. Five out of eight patients showed focal RBC uptake: in particular, three subjects
presented activity in the central abdominal region. In such cases, SPECT/CT localized the site of
bleeding in the small bowel. The remaining two patients had RBC uptake in the pelvic region; in
these cases, the fusion of SPECT and CT images was able to disclose a rectal site of bleeding,
dicriminating rectal bleeding from bladder activity. All patients with RBC positive scan were
submitted to specific therapy with a complete remission of the bleeding in all cases.Conclusions:
our results suggest that SPECT/CT with hybrid device may improve the accuracy of RBC
scintigraphy for the imaging of occult GI bleeding.
The diagnostic value of
Tc-DTPA lung clearance and
esophageal transit to evaluate the severity of complications
in systemic sclerosis
N. Prandini1, R. La Corte2, M. Casali1, S. Fabbri3, S. Volpinari2, S.
Panareo1, M. Giganti1, C. Cittanti1, A. Costanzo1, F. Trotta2, L. Feggi1;
Nuclear Medicine, Ferrara, Italy, 2Rheumatology, Ferrara, Italy, 3Health
Physics, Ferrara, Italy.
Background: Lung and esophageal involvements are common in systemic sclerosis (SSc). Aim of
this study is to assess if there is a correlation between the involvement of the respiratory and
gastrointestinal systems in the same patients.Materials and Methods: We studied 280 patients
effected by SSc (259 females and 14 males), with a mean age of 55 years (from 20 to 78 years
old) and a mean lenght of disease of 106 months. All pts underwent 99mTc-DTPA clearance using
a commercial available jet nebulizer (Venticis, CIS) aerosol inhaling 500 MBq of 99mTc-DTPA
aerosol for 2 minutes at normal tiding breathing. Images were recorded dinamically with a dual
head camera (Philips Vertex) and two ROIs were defined around lungs and the activity/time
curves were generated. The best exponential fit of both curves in the first 15 minutes were
calculated: the results were considered abnormal if half time was < 60 minutes. All patients
underwent to esophageal transit scintigraphy with the assessment of esophageal transit time of
liquid and solid swallows in upright and supine position (10 ml of water or fruit juice labelled
with 40 MBq of 99mTc-colloid). The results of this examination were classified in five degrees of
severity from 0 (all normal swallows) to 4 (all pathological swallows) using a qualitative analysis
of summed images.Results: Our data were summarized in table I:
Acute pulmonary embolism is a frequent disease with high mortality. Conventional diagnostic
strategies have relied on ventilation-perfusion imaging complemented by venous imaging. New
strategies and techniques have recently been developed i.e. spiral computed tomographic
angiography and new nuclear medicine procedures. Single photon emission tomography (SPET)
is taking place in lung perfusion and ventilation imaging. We compared the visibility of small
defects in planar and SPET imaging with an anatomical physical phantom.Methods: The physical
phantom consisted of plastic containers filled with plastic pellets to imitate the three dimensional
shape of the lungs. Inside the lung phantom there were 22 cubic defects (10 at left and 12 at
right). The number and size of them were 3 x 2.5 cm3, 3 x 2.0 cm3, 4 x 1.5 cm3, 6 x 1.0 cm3 and 6
x 0.7 cm3. Phantom was imaged with two headed Siemens Ecam+ gamma camera. In planar
acquisition anterior image was acquired with 600 million counts and the rest of the projections
were imaged at the same time. SPET was imaged with 64x64 and 128 and 128 matrix, 128
projections over 360 degrees and 10 s per projection. Reconstruction was filtered back projection
for 64 matrix and iterative for 128 matrix. Nuclear medicine physicians used a computer monitor
to evaluate visible lesions from all planar projections and SPET slice sets.Results: Visibility was
63 ± 41 % for all defects, 100 ± 0 % for 2.5 cm3 cubes, 89 ± 19 for 2.0 cm3, 85 ± 17 for 1.5 cm3,
39 ± 19 for 1.0 cm3 and 0 ± 0 for 0.7 cm3. Average visibility for a range of defects 2.5 - 1.0 cm3
was from planar images 63 %, while it was 85 % from SPET (64x64) and 88 % from SPET
(128x128) images. For defects between 2.5 - 1.5 cm3 planar imaging revealed 78 %, SPET
(64x64) 96 % and SPET (128x128) 100 % visibility.Conclusions: Small defects were difficult to
see as expected. All the defects over 1.5 cm3 can be seen with SPET. According to these results
tomographic imaging is superior to planar imaging with small defects.
Quantitative Image Analysis
pertecnetate scintigraphy
M. Lyra, C. Skouroliakou, N. Toubanakis, A. Antoniou, C. Chatzigiannis,
G. Limouris, L. Vlahos; Athens University, A' Radiology Department,
Athens, Greece.
Lung clearance
half time > 60 min
31 18 33 25 115
half time < 60 min
35 40 59 25 165
66 58 92 50 280
We compared the results obtained in the same patients by lung clearance and esophageal transit,
finding that there isn’t a significant correlation between the groups of patients (y = -0,0008 x +
2,391, P = 0,13). Discussion and conclusion: the involvement of lungs permeability and
esophageal motility increases with lasted of disease but there are not significant correlation
between the involvement of lungs and esophagus in SSc patients. So we must use both
techniques to assess correctly the involvement of the two systems.
106 — Sunday, October 1, 2006, 8:00 am — 9:30 am, Allegro
Symposium: Molecular Imaging in Nuclear Cardiology
Monitoring stem cell transplanation in cardiology: from
bench to bedside
Winifried Brenner (DE))
MIBG-imaging for evaluation of sympathetic innervation of
the heart
Norbert Johann Czech (DE)
Radiolabeled Oligonucleotides for imaging of acute heart
transplant rejection
W. Uwe Kampen (DE)
109 — Sunday, October 01, 2006, 8:00 am - 9:30 am, Meeting Room 1
Physics: Varia
Visibility of defects in planar and tomographic lung imaging:
a phantom study
J. O. Heikkinen, T. Tarkiainen, P. Ylikangas; Etelä-Savo Hospital District,
Mikkeli, Finland.
Aim: This study attempts to characterize thyroid tissue by automatic texture analysis of thyroid
scintigrams by Tc99m pertechnetate. Thyroid scintigraphy is a routine examination for assessing
the disorders of the thyroid gland. However, the current diagnosis practice is based mainly on
qualitative evaluation of the resulting scintigraphic images, therefore depending on the
physician’s experience. Computerized texture analysis could widely be employed in scintigraphic
images of various organs (thyroid,liver, breast) as a scintigram is created by digitally received
data, which could also be interpreted or reconstructed by the gamma camera processor. Our aim
is to indicate an increase of the sensitivity of diagnosis by providing semiquantitative parameters
that improve scintigraphic images of the thyroid gland.Methods: The texture features, that are
calculated, are based on co-occurrence matrices. The sample consisted of 50 patients. For each
patient two scintigraphic images (2 time intervals post injection) were recorded by GE Starcam
4000 gamma camera, in interfile format and transferred by Procyon Starlab program to a
conventional PC for further processing. The lobes were manually delineated in each scintigram
and the co - occurrence matrices for six separation vectors were calculated.Results: The texture
features extracted from each one of these matrices are: contrast, correlation, energy and
homogeneity. Stepwise logistic regression was used to evaluate the ability of the calculated
features to differentiate between normal and pathologic thyroid tissue.Conclusions: The results
indicate that quantitative analysis of thyroid scintigrams can provide an objective
characterization of thyroid tissue and facilitates diagnosis semi-quantitatively. Helping clinicians
to make effective use of information in the structure of thyroid gland is a key aim of medical
informatics research.
A new approach for processing
Tc-sestamibi Parathyroid
J. Isidoro, D. Freitas, G. Costa, A. Ferrer-Antunes; Coimbra University
Hospital, Coimbra, Portugal.
Parathyroid scintigraphy has been used in the localization of hyperfunctioning parathyroid tissue.
Several methods are used ranging from dual tracer, dual phase and subtraction techniques. Dual
tracer studies give higher doses to patients and dual phase requires prolonged acquisition
protocols. Aim: We propose a new approach for processing 99mTc-sestamibi Parathyroid studies
base on the determination of the washout or uptake ratios for each pixel.Materials and Methods:
A dynamic study with 40 images, 1 min each, is acquired immediately after the injection of
Tc-sestamibi, with a matrix size of 128x128. The proposed methodology performs, to each
pixel, the following calculations: 1) correct the dynamic study for radioactive decay; 2) apply
robust straight-line fit to the logarithm of the time series for each pixel, excluding the values
corresponding to the initial uptake period of the thyroid. 3) from the previous calculated slope we
generate a parametric image showing the T1/2 values. The data processing is done by a java
plugin that we developed for ImageJ (http://rsb.info.nih.gov/ij/). For better localization we can
use ImageJ to visualize the fusion of the parametric image with the 99mTc-sestamibi dynamic
study.Results: This methodology was used to process 99mTc-sestamibi Parathyroid studies and the
results were in agreement with those of the traditional dual phase 99mTc-sestamibi and the
subtraction techniques. Further clinical validation is required to validate this approach and
eliminating the requirement for late image acquisitions.Conclusions: The proposed methodology
for processing 99mTc-sestamibi Parathyroid studies is simple and robust. With this approach only
a 40 min dynamic post-injection acquisition is required to collect the necessary data and thus
shortening the acquisition protocol. This software may also be used to process other dynamic
studies were different uptake or washout times are expected.
Experimental and clinical validation of energy windows for
double-isotope parathyroid scintigraphy
G. Petyt, F. Raynaud, F. Martini, Y. Zieba, T. Prangère, M. Steinling, D.
Huglo; Nuclear Medicine, Hôpital Huriez, Chu of Lille, Lille, France.
Energy of 99mTc (140 keV) and 123I (159 keV) are close. So, for parathyroid double-isotope
acquisition (99mTc-MIBI-123I), asymmetric windows are recommended. The aim of this study
was to optimize these windows to our gamma-camera and collimator.Materials and Methods: A
DST-Xli (SMV-GEMS) with a LE-UHR collimator is use for parathyroid scintigraphy and this
study. For experimental data, 1.9 MBq of 123I and 3 MBq of 99mTc in 3mL syringes were placed
at 8cm of collimator, behind 25mm-Plexiglas. Symmetric +/-10% energy windows were the
references. Acquisitions with various windows and/or centres were performed, for 99mTc: 140
keV +/-3% to 7%, and from 138 to 142 keV +/-6%, and for 123I: from 159 to 167 keV +/-6% and
163 or 165 keV +/-7% or 8%. For each frame, ROIs were drawn on syringes to define the
sensitivity (99mTc in the 99mTc-window, 123I in the 123I-window) and the contamination (99mTc in
the 123I-window, 123I in the 99mTc-window). The best compromise was obtained by ROC curves.
For clinical validation, 3 double-isotope acquisitions for parathyroid imaging were performed, A
(beforehand defined): 133-147 and 154-170 keV, B (recommended by literature): 130-150 and
152-174 keV and C (best experimental choice): 132-148 and 153-176 keV. ROIs were drawn on
thyroid and myocardium (with only 99mTc-MIBI) and the relative gains B/A and C/A
calculated.Results: For experimental data, the relative sensitivity varied from 55.2% to 91.2% for
Tc and from 68.0% to 92.9% for 123I. The contamination of 123I in the 99mTc-window varied
from 6.2% to 16.8% and for 99mTc in the 123I-window from 1.0% to 12.9%. The best compromise
was 140 keV +/-6% for 99mTc (relative sensitivity: 85.0%, contamination: 12.5%) and 165 keV
+/-7% for 123I (relative sensitivity: 85.5%, contamination: 3.1%). For clinical evaluation, 10
patients were studied. For 99mTc-windows, the mean gain for thyroid (B/A: 1.29[1.11-1.65], C/A:
1.08[0.96-1.15]) was near the one of myocardium (B/A: 1.31[1.17-1.45], C/A: 1.13[1.03-1.25],
so 123I contamination can be neglected in both cases. For 123I-windows, B had a higher gain on
thyroid (B/A: 1.33[1.18-1.59]) than C (C/A: 1.18[1.01-1.36] but higher 99mTc contamination
(B/A: 1.53[1.14-2.09], C/A: 1.26[1.06-1.39]).Conclusions: Because energy response varies
according to collimator and gamma-camera, it is necessary to research the best windows fitted to
double-isotope acquisition in each case, in experimental and clinical conditions. For us, chosen
windows will be now 140 keV +/-7% (130-150 keV) and 165 keV +/-7% (153-176 keV).
Effect of CT-based attenuation correction on uptake ratios in
skeletal SPECT
V. Schulz1, W. Römer1, A. Noemayr1, J. Hornegger2, H. Herzog3, W.
Bautz4, T. Kuwert1; 1Clinic of Nuclear Medicine, Erlangen, Germany, 2Chair
of Pattern Recognition, Erlangen, Germany, 3Institute of Medicine, Jülich,
Germany, 4Institute of Radiology, Erlangen, Germany.
Aims: Hybrid cameras consisting of an X-ray computed tomography scanner (CT) and a singlephoton emission tomographic camera (SPECT) offer the potential to correct SPECT images for
artefacts caused by attenuation. To date, data on the effect of attenuation correction on uptake
ratios quantified by SPECT are scarce. We therefore compared ratios of skeletal uptake of Tc99m-polyphosphonates (DPD) between attenuation-corrected and uncorrected SPECT images in
38 patients. Furthermore, we sought to determine the effect of misalignments between SPECT
and CT on these variables.Methods: SPECT/CT examinations were performed using a dual-head
gamma camera in conjunction with a dual-slice spiral-CT scanner installed within the same
gantry (Symbia T2, Siemens Medical Solutions, Erlangen, Germany). Measurements were
performed in 38 consecutive patients studied for clinical reasons. Using regions of interest (ROI),
raw counts were determined in the vertebral body of the fifth lumbar vertebral body, its facet
joints, both anterior iliacal spinae and of the whole transversal slice. These measurements were
performed in uncorrected (NATT) and attenuation-corrected (ATT) images. Furthermore, the
ROI measurements were also performed in ATT scans gained from SPECT and CT images that
had been deliberately misaligned by one centimetre in one of the three dimensions beforehand
(ATTX, ATTY, ATTZ).Results: ATT whole-slice ratios of DPD uptake differed significantly
from the values determined in uncorrected images in all regions studied, the mean percentual
differences ranging from 31% for the left facet joint to 40 % for the vertebral body. Left-to-right
ratios were not affected by more than 5 % by ATT. Attenuation correction using misaligned pairs
of data sets led to significant differences in uptake ratios when compared to ATT images from
well-aligned SPECT and CT: for ATTX 5/ 5, for ATTY 3/ 5, and for ATTZ 1/ 5 regions had
either significantly lower (n = 3) or higher (n = 6) relative DPD uptake. The differences in uptake
ratios between aligned and misaligned data sets ranged from 1 to 33 %. Furthermore, left-to-right
ratios were also significantly affected by prior misalignment between SPECT and
CT.Conclusions: Attenuation correction affects DPD uptake ratios significantly. Furthermore,
misalignment between SPECT and CT by one centimetre introduces significant errors in
quantitation. Clinical evaluation of attenuation-corrected scans should therefore be performed
carefully; in particular, special attention should be given to possible misalignments between
Affine co-registration of thorax Ga-67 SPECT and CT images
L. C. Freire, F. Godinho; Atomedical, Laboratório de Medicina Nuclear,
Lisboa, Portugal.
Aim: In this work, we propose a dedicated method specifically designed for the co-registration of
thorax Ga-67 SPECT and CT images acquired in different scanning sessions. Although
commercial SPECT-CT scanners are widespread, there may be some studies where CT data is
available prior to the acquisition of the SPECT data, reducing the justification of an extra CT
irradiation. Clinical applications of thorax Ga-67 SPECT include the detection of Hodgkin
lymphoma.Materials and Methods: The proposed method uses an affine, mutual informationbased registration method, which incorporates high-order information into the calculation of the
similarity measure and an image apodization scheme (Gouveia et al., 2004, Proceedings EANM
Congress, 31(2), S236-7). The computation of the similarity measure may be restricted to the
thorax through the use of a recognition program capable of identifying the lungs and surrounding
regions. This necessity is due to the fact that the patient may, for instance, have its arms in
different positions in both scans, or to differences in the FOV between both images. The
evaluation of the method was done using real images, acquired on a SPECT-CT scanner, which
enables a goldstandard assessment of results. SPECT image was duplicated 32 times using affine
geometric transformations, whose parameters varied random and uniformly between -5 and 5
mm for translations, -5 and 5 degrees for rotations, and -5% to 5% for scaling and shearing
parameters, yielding a test dataset of 32 deformed SPECT images.Results: Mean absolute coregistration errors were calculated for each registration parameter, in two situations: I) full CT
image and; II) 25% top slices of CT image turned into 0, in order to simulate a great variation in
FOV's. For each situation, results are given for a) conventional MI-based registration method; b)
MI-based method restricted to lungs and nearby regions and; c) same as b including high-order
information. Results are presented in table below. One can see that when FOV's are identical, all
methods provide identical results (situation I). However, in situation II, methods b and c provide
better results than conventional MI. The inclusion of high-order information generally improves
the results of method b.Conclusions: This works shows that this method dedicated to the problem
of registration of thorax images, in an affine search space, can provide better results that a
conventional approach. The work will be further developed in order to incorporate a non-rigid Bsplines deformation model.
Mean absolute co-registration errors for the 12 affine parameters for situations I and II (n=32).
I - tx ty tz [mm]
rx ry rz [deg]
scx scy scz [%]
shx shy shz [%]
3.7 0.4
1.4 0.6
1.6 1.8
3.0 0.6
0.7 0.9
1.8 1.1
2.9 0.6
0.6 0.9
1.7 1.2
II - tx ty tz [mm]
rx ry rz [deg]
scx scy scz [%]
shx shy shz [%]
a - 2.0 10.3 6.5 1.9
1.2 2.0
0.4 1.6
2.0 0.6
0.8 1.0
1.3 1.3
1.9 0.7
0.6 1.0
1.5 1.3
Anatomical accuracy of automated linear and non-linear
software registering CT to FDG-PET
G. Wolz, A. Nömayr, T. Hothorn, J. Hornegger, W. Bautz, W. Römer, T.
Kuwert; University of Erlangen-Nürnberg, Erlangen, Germany.
Aim: The purpose of this study was to compare the anatomical accuracy of automated linear and
non-linear registration software for aligning data from separately performed computed
tomography (CT) and positron emission tomography (PET).Methods: Analyses were performed
on independently acquired PET and CT data from 40 tumor patients. Linear as well as non-linear
automated fusion was performed using the commercially available Mirada 7D platform (MIRL
and MIRNL, respectively). In addition, the data underwent a second automated non-linear
registration with an as yet not commercially available software (SCR; Siemens Corporate
Research, Princeton). SCR is based on supervised registration by joint intensity learning,
maximization of the mutual information, and maximization of the correlation ratio. The accuracy
of the registration was evaluated on 105 malignant lesions, clearly visible in both CT and PET.
For this purpose, the distances between lesion representation on PET and CT were measured in
X-, Y-, and Z-direction (X-, Y-, Z-distances). Statistical evaluation was by mixed effect analysis,
comparing separately MIRL with MIRNL and SCR with MIRNL.Results: The percentage of
lesions misregistered by less than 1.6 cm was for MIRL 88% in X-direction, 74% in Y-direction,
and 70% in Z-direction; for MIRNL, 82 %, 85 %, and 70 %; and for SCR 93 %, 89 %, and 87 %,
respectively. The average X-, Y- and Z- distances ranged between 0.59 +/- 0.57cm for SCR (Xdirection) and 1.28 +/- 0.97 cm for MIRL (Z-direction). MIRNL was significantly more accurate
than MIRL in Y- direction. Furthermore, a significantly better alignment was found for SCR
compared to MIRNL in the X- direction, however, only for thoracic lesions.Conclusions: The
accuracy of both linear and non-linear automated image registration can be expected to be within
a range of 1.5 cm for the majority of malignant lesions. Alignment tended to be more accurate
with non-linear registration.
Evaluation of spatial resolution of planar detector PET
systems: different strategies show different results
N. Matela1, M. V. Martins1, H. Cordeiro1, M. Correia1, P. Rodrigues2, A.
Trindade2, N. Oliveira1, N. C. Ferreira3, J. Varela2, P. D. Almeida1;
Universidade de Lisboa, Faculdade de Ciências, Instituto de Biofísica e
Engenharia Biomédica, Lisboa, Portugal, 2LIP, Laboratório de
Instrumentação e Física Experimental de Partículas, Lisboa, Portugal,
IBILI, Instituto Biomédico de Investigação de Luz e Imagem da Faculdade
de Medicina da Universidade de Coimbra, Coimbra, Portugal.
Aim: This work evaluates possible differences in evaluating the spatial resolution in a planar
PET detector by calculating the PSF with a phantom made of several point sources instead of one
source positioned in different locations.Materials and Methods: To support the existence or
absence of any kind of difference between the two methods of measuring the PSF during the
development stage of a new Positron Emission Mammography (PEM) detector, we simulated the
emission from an activity point source in the center of the field of view (FOV), reconstructed the
data and calculated the FWHM of the gaussian fit. We repeated this procedure along 13 different
locations along the orthogonal axis. After this, we have merged all simulations, considering them
as one digital phantom with 14 sources (chain phantom). Emission data was obtained using the
Monte Carlo Geant4 toolkit. We have simulated the ClearPEM detector, which consists in two
14.2cmx16.1cm plates 10cm apart at two orthogonal projection angles. Simulations included
Depth of Interaction (DOI) information, detector Compton scattering and no background activity.
The image reconstructions were performed using 3DOSEM, 2DOSEM, 2DART and 2DMLEM.
The 2D image reconstructions were performed after rebinning data with SSRB or FORE. Finally
we compared the results with all algorithms and rebinning methods using the two different PSF
measuring approaches, sources alone (SA) or chain phantom (CP), and evaluated the differences
between PSF values.Results: The spatial resolution results obtained with 2DOSEM+SSRB,
2DMLEM+SSRB and 3DOSEM for the more relevant sources are presented in table I. Images
obtained with other reconstruction algorithms and with FORE are currently being evaluated.
Discussion: The results with 2D reconstruction algorithms show significant differences in the
PSF measurements. This difference increases as we move farther from the rotation axis. The
reason for this difference may be the existence of lines of response (LOR), originated in a certain
source, passing over the localization of other sources. This effect is not visible with 3D OSEM
since in this case LORs are not projected into transaxial planes. These results show that
measuring the PSF with one source of activity moving in the field of view may not be the more
realistic way of measuring the spatial resolution, since in a clinical acquisition we will never
have only one source of activity in the FOV. This choice allows better spatial resolutions results.
However, results obtained like this would probably not be confirmed by clinical tests.
110 — Sunday, October 01, 2006, 8:00 am - 9:30 am, Meeting Room 2
Radiopharmacy/Radiochemistry: Antibodies 1
Comparison of two different Cu-67-labeled chCE7 antibody
variants in nude mice bearing human metastatic ovarian
J. Grünberg, K. Knogler, K. Zimmermann, S. Cohrs, P. A. Schubiger, I.
Novak-Hofer; Paul Scherrer Institute, Villigen, Switzerland.
Aim: Recently we could show, that Cu-67/64 labeled recombinant anti-L1-CAM chCE7F(ab’)2
fragments revealed high tumor accumulation and fast blood clearance, suitable for tumor
imaging, but displayed a higher kidney uptake than the intact monoclonal antibody (J. Grünberg,
I. Novak-Hofer, M. Honer et al, Clin Cancer Res 11, 2005). The objective of our study was to
investigate if point mutations in the heavy chain improve the biodistributions and
pharmacokinetics of chCE7.Methods: An aglycosylated variant of mAb chCE7 (Asn297 was
replaced by Gln) and a format impaired in binding to the FcRn receptor (His310 was replaced by
Ala) were produced in HEK293 cells and derivatized with the bifunctional chelator CPTA for
Cu-67 labeling. Clearance from the blood of the different antibody constructs were evaluated in
I-125 and Cu-67 labeled form. Biodistribution experiments of Cu-67 labeled chCE7 variants
were performed in nude mice with human ovarian cancer metastases two weeks after i.p.
injection of 5x106 SKOV3ip cells.Results: The blood clearance of the H310A construct was
similar in I-125 and Cu-67 labeled form with beta-half life of 10 h vs 11 h, comparable with the
blood clearance of I-125 labeled chCE7F(ab’)2 fragments. When radioiodinated the
aglycosylated variant of chCE7 showed significantly faster blood clearance (beta-half life of 96
h) compared with the parental antibody (beta-half life of 138 h). When Cu-67-labeled both
antibody formats disappeared more rapidly from the blood with beta-half life of 39 h vs 79 h. Cu67-CPTA-chCE7agl revealed high tumor uptake and low liver accumulation in SKOV3ip
metastases bearing nude mice. Eighty-four h post i.v. injection radioactivity present in the tumor
was 49.2±3.95 %ID/g, in the liver 4.44±1.07 %ID/g and blood levels dropped to 8.33±3.28
%ID/g (n=4). By contrast the H310A showed low tumor uptake (6.81±0.77 %ID/g) and a higher
accumulation in the liver (10.7±5.06 %ID/g) after 48 h. Radioactivity left in the blood was
0.26±0.08 %ID/g (n=4).Conclusions: ChCE7 and its aglycosylated counterpart revealed a faster
blood clearance in Cu-67-CPTA labeled form than in radioiodinated form. The aglycosylated Cu67-CPTA labeled variant of the anti-L1-CAM antibody chCE7 represents a very good antibody
format for RIT because of its more rapid clearance from the blood and its excellent tumor
In vitro and in vivo characterisation of 177Lu-huA33
Y. Almqvist1, A. Steffen1, V. Tolmachev1, C. R. Divgi2, A. Sundin1;
Oncology, Radiology and Clinical Immunology, Uppsala, Sweden,
Memorial Sloan-Kettering Cancer Center, New York, United States.
Aim: Colorectal cancer is often detected in a locally advanced stage and is thus usually difficult
to cure. The humanised monoclonal antibody A33 (huA33) has been demonstrated to have
unique targeting specificity for colorectal cancer and normal colon, and has thus been studied for
its therapeutic potential in this disease. The A33 antigen is highly and homogenously expressed
in >95% of all colorectal cancers, both primary tumour and metastases. The aim of this study was
to evaluate the potential of radioimmunotherapy (RIT) with 177Lu-labelled humanised
monoclonal antibody A33 (huA33). Materials and Methods: HuA33 was labelled with the beta
emitting therapeutic nuclide 177Lu using the chelator CHX-A’’-DTPA, and the properties of the
Lu-huA33 conjugate was determined both in vitro, and in vivo in a biodistribution study in
tumour bearing nude mice. Results: The 177Lu-huA33 conjugate bound specifically to colorectal
cancer cells in vitro (with a KD value of 2.3 ± 0.3 nM, determined by a saturation assay) and in
vivo. The tumour uptake of 177Lu-huA33 was very high, peaking at 134 ± 21 %ID/g 72 h post
injection (p.i.). Normal tissue uptake was low; the radioactivity concentration in blood (which
had the second highest radioactivity concentration) was at all time points (8 h to 10 days) lower
than in tumour. The tumour-to-blood ratio increased with time, reaching 70 ± 30 ten days p.i.
Throughout the study, the uptake of 177Lu in bone (known to accumulate free 177Lu) was low, and
the fraction of protein-bound 177Lu in plasma samples was high (95 to 99%). This indicates high
stability of the 177Lu-huA33 conjugate in vivo. Conclusion: The 177Lu-huA33 conjugate shows a
very favourable biodistribution, with an impressively high tumour uptake and high tumour-toorgan ratios, indicating that the conjugate may be suitable for RIT of metastatic colorectal
Radiolabeling of Zevalin with high 90Y activity: optimization
of the procedure and reduction of the radiopharmacist
radiation exposure
S. Papi1, N. Urbano1, G. Tosi2, M. Ferrari2, G. Paganelli1, M. Chinol1;
Nuclear Medicine division, European Institute of Oncology, Milano, Italy,
Health Physics division, European Institute of Oncology, Milano, Italy.
Preparation and evaluation of [ In]-benzyl-DOTA-ZHER2:342 for
imaging of HER2 expression in malignant tumors.
Aim: Zevalin® has been registered for the treatment of NHL. Typically, 1480MBq are labeled
and no more than 1184MBq are administered to the patient. In preparation for a clinical trial with
high activity 90Y-Zevalin, we investigated the parameters influencing the radiopharmaceutical
preparation procedure, in terms of radiochemical purity (RCP) and stability. Due to the amount
of 90Y involved, we also tried to reduce the radiopharmacist radiation exposure by optimizing
some 90Y handling steps and compared the absorbed doses to the fingers with the standard
radiolabeling approach.Materials and Methods: The Zevalin® labeling protocol involves the
transferring of 1480MBq from the source 90Y vial to a reaction vial and labeling at 740MBq/mg
specific activity; after 5 min the reaction is stopped diluting with formulation buffer to 10mL.
After RCP determination, the patient dose is drawn into a 10mL syringe. In the high activity
protocols it is likely to manipulate 5.55-7.40GBq of 90Y (29.6 to 55.5MBq/kg). Therefore we
adopted a 90Y source vial containing the needed amount of activity for the study plus 10%
excess. We then used a shielded syringe with a spinal needle to transfer all the 90YCl3 inside a
larger reaction vial, which was then counted. After addition of the antibody (at 740MBq/mg
specific activity) and waiting 5 min, the reaction was stopped with formulation buffer up to
20mL. After RCP, the activity in excess (the smaller fraction) was drawn out from the reaction
vial, which therefore contained the requested activity. This vial was directly used for patient
administration, with a patented remote injection system. For all the experiments, the finger dose
of the operator was monitored by ThermoLuminescent Dosimeters (TLD) under anti-X
gloves.Results: The RCP of high activity 90Y-Zevalin was similar to the standard activity
procedure (99.1 vs 99.6%). Specific activity was the most critical factor: if increased from 740 to
925 and 1110MBq/mg, the incorporation of 90Y dropped to 94% and 80% respectively. The
stability of radiolabeled conjugate was >98% over a 12h time period. The finger dose,
normalized to 1480MBq, resulted (4.5±1.9)mSv in the Zevalin® labeling, whereas in the high
activity procedure it was (3.2±0.7)mSv.Conclusions: We showed that the labeling of Zevalin
with high 90Y activities was feasible with radiochemical purities and stabilities similar to the
standard preparation. Optimizing some risky steps in the high activity 90Y-Zevalin preparations,
we managed to keep radiation exposure of radiopharmacy personnel comparable to that of
standard Zevalin®.
A. Orlova1, T. Tran1, C. Widström2, V. Tolmachev1; 1Uppsala Univeristy,
Uppsala, Sweden, 2Uppsala Univeristy Hospital, Uppsala, Sweden.
Aim of the study. Imaging of expression of HER2 in breast cancer patients may enable to select
patients, who benefit from treatment with Herceptin, as well as to identify these who may
experience cardiac toxicity by such treatment. Affibody molecule ZHER2:342 is a short (7 kDa)
protein, which binds to HER2 with subnanomolar affinity. The goal of this study was to prepare
indium-111 labelled DOTA-derivative of ZHER2:342 and evaluate if it could be used for imaging of
HER2 overexpression in malignant tumours. Material and methods Isothiocyanate-benzyl-DOTA
was coupled to recombinant ZHER2:342 in alkaline conditions. The conjugate was labelled with
In in 1 M ammonium acetate buffer, pH 5.5 at 60 oC, stability of the label was evaluated. In
vitro specificity of [111In]-benzyl-DOTA-ZHER2:342 binding to HER2 was performed using HER2
expressing breast carcinoma BT474 and ovarian carcinoma SKOV-3 cell lines. Biodistribution of
[111In]-benzyl-DOTA-ZHER2:342 was performed in nude mice bearing HER2-expressing
xenografts. Uptake specificity was controlled by pre-injection of large excess of non-labelled
ZHER2:342. Gamma-camera imaging of SKOV-3 xenografts in mice was performed using a
Siemens e.CAM gamma-camera. Results Efficiency of benzyl-DOTA-ZHER2:342 labelling with
In was 97% after 5 min ay 60 oC. The label was stable in PBS and under challenge with 10000fold excess EDTA. After labelling, [111In]-benzyl-DOTA-ZHER2:342 retained capacity for specific
binding to HER2 expressing cells. In vivo, [111In]-benzyl-DOTA-ZHER2:342 demonstrated quick
clearance from blood and non-specific organs, except kidneys. Four hours pi, tumour uptake was
4.4 ± 1.0 % IA/g, and tumour-to-blood ratio was 23. Gamma camera imaging enabled clear
visualisation of HER2-expressing xenografts. Conclusion [111In]-benzyl-DOTA-ZHER2:342 is a
very promising agent for imaging of HER2 expression in malignant tumours.
determination employing different beta-counting systems
N. Urbano1, S. Papi2, S. Modoni1; 1Nuclear Medicine Unit, Centro di
Riferimento Oncologico della Basilicata, Rionero in Vulture, Italy, 2Nuclear
Medicine Division, European Institute of Oncology, Milano, Italy.
Aim: A clinically meaningful advance in the treatment of relapsed or refractory low-grade,
follicular, or transformed non-Hodgkin’s lymphoma, has been achieved with 90Y-Ibritumomab
tiuxetan (Zevalin®). A mandatory parameter before radiopharmaceutical administration is the
exact determination of radiochemical purity (RCP). We evaluated in terms of sensitivity, radiochromatographic resolution and time consumption, different instruments involved in RCP
measurement.Materials and Methods: After radiolabeling, an aliquot (0.74 MBq /5µL) of 90YIbritumomab tiuxetan was directly spotted on ITLC-SG strip (cm 1 x 10), developed in saline
and then assayed by: I) an auto-radiographic system using high performance storage phosphor
screen; II) a radio- TLC scanner with beta-counting tube. Subsequently, the same radio-ITLC
strip was appropriately cut and measured, after determining accurate isotope geometric factors,
by a Dose Calibrator (III). Moreover, an aliquot of the same radiolabeled compound was suitable
diluted with saline and after ITLC-SG assay (0.037 MBq /5 µL), the strip was cut and the two
radio-ITLC fractions were measured by liquid scintillation beta counter (IV). Another diluted
A novel molecular imaging agent for diagnosis of recurrent
HER2 positive breast cancer. First time in human study using
an Indium-111- or Gallium-68-labeled Affibody molecule
R. P. Baum1, A. Orlova2, V. Tolmachev2, J. Feldwisch2; 1Zentralklinik Bad
Berka, Bad Berka, Germany, 2Affibody AB, Bromma, Sweden.
Aim: A highly specific and sensitive HER2-binding imaging agent could be used to prove in vivo
the HER2 status of lesions not amenable for biopsy and would facilitate treatment planning.
HER2-Scan is a small protein that can be site-specifically labeled with In-111 or Ga-68 using a
DOTA moiety. The protein portion of HER2-Scan is an Affibody® molecule, member of a novel
class of small non-immunoglobulin affinity ligands. Pre-clinical characterization of radiolabeled
HER2-Scan in mice bearing HER2-expressing tumor xenografts showed high specific tumor
targeting with 23 % IA/g at 1 hour post injection (h p.i.), rapid biodistribution kinetics and
clearance and allowed high contrast SPECT imaging as early as 1 h p.i. In this study we evaluate
the use of In-111 and Ga-68 labeled HER2-Scan to specifically detect and stage HER2expressing metastatic lesions in patients with recurrent breast cancer by SPECT or
PET/CT.Methods: The HER2-specific Affibody molecule ZHER2:342 was selected from a
combinatorial library based on a 58 amino acid protein A domain scaffold and a DOTAderivative of ZHER2:342 was made by chemical synthesis in a single synthetic process. Patients
received a microdose (<100 µg) of 123 MBq In-111 or 110 MBq Ga-68 labeled HER2-Scan.
Imaging was done 1 to 3 h p.i. Patients were carefully monitored for adverse effects.Results:
Injection of In-111 or Ga-68 labeled HER2-Scan resulted in high quality SPECT and PET/CT
images enabling the detection of very small lesions (e.g. mediastinal lymph node metastases).
T1/2 (a) in blood was 3.8 min, t1/2 (b) was 1.6 h and t1/2 (c) 18 h. Tumor half-life was 56.9 hrs.
No adverse effects were observed.Conclusions: Molecular imaging using the HER2-specific
Affibody molecule HER2-Scan not only localizes metastatic lesions in vivo, but also adds new
qualitative information not available today by conventional imaging techniques. For the first time
a phage display selected protein was used to determine in vivo the HER2 status of lesions not
amenable for biopsy in patients with breast cancer.
M. Nestor1, M. Anniko1, G. A. M. S. van Dongen2, V. Tolmachev3; 1Unit of
Otolaryngology and Head & Neck Surgery, Uppsala, Sweden, 2Department
of Otolaryngology/Head and Neck Surgery, Amsterdam, The Netherlands,
Unit of Biomedical Radiation Science, Uppsala, Sweden.
201 — Sunday, October 1, 10:00 am - 11:30 am, Friends of Music Hall
Plenary 1: PET/CT as the First Choice Diagnostic Tool
in Staging Cancer
PET/CT in head and neck cancer
Lieselotte Hojgaard (DK)
PET/CT in lymphoma
R. Whal (USA)
PET/CT in lung cancer
Hans C. Steinert (CH)
301 — Sunday, October 1, 11:30 am - 01:00 pm, Friends of Music Hall
CME 2: Neurology: Critical review of diagnostic
methods for assessment of dementia
Role of MRI and CSF markers
H. Hampel (DE)
Indium-111 labeled anti-CD33 monoclonal antibodies (mAbs)
modified with nuclear-localizing sequences (NLS) kill
mitoxantrone-resistant HL-60 acute myelogenous leukemia
(AML) cells
V. Kersemans, B. Cornelissen, J. E. Dick, R. M. Reilly;
Toronto, Toronto, ON, Canada.
University of
Aim: To evaluate the ability of anti-CD33 mAbs, M195 or HuM195 modified with NLS and
labeled with the Auger electron-emitter, 111In to kill wild-type HL-60 and mitoxantrone-resistant
HL-60-MX1 AML cells.Methods: HuM195 and M195 were conjugated with DTPA then
derivatized with 10 ± 3 NLS peptides (CGYGPKKKRKVGG) from SV40 large T-antigen and
labeled to 5 MBq/µg with 111In. The binding affinity (Kd) of HuM195 and M195 was determined
for HL60 and HL60-MX1 cells. Nuclear localization of 111In-NLS-HuM195, 111In-NLS-M195,
In-HuM195 and 111In-M195 was measured by subcellular fractionation. The antiproliferative
effects of 111In-NLS-HuM195, 111In-NLS-M195, 111In-HuM195 and 111In-M195 (2.5-250
kBq/well) were studied on HL60 and HL60-MX1 seeded at 1 × 103 cells/well using the WST-1
assay. Cell viability treated with NLS-HuM195 (20 nM/well), NLS-M195 (20 nM/well), NLS
(200 nM/well) or 111In-acetate (250 kBq/well) was also investigated.Results: Kd values of 0.22 ±
0.08 nM and 0.73 ± 0.17 nM for 123I-HuM195 on HL60 and HL60-MX1, respectively and 4.4 ±
0.78 nM and 3.4 ± 0.63 nM for 123I-M195 on HL60 and HL60-MX1, respectively were obtained.
There was 22.6 ± 3.9 % and 16.1 ± 3.3 % of 111In-NLS-HuM195 in the nucleus of HL60 and
HL60-MX1 cells, respectively. This was 6-fold and 3-fold higher than 111In-HuM195 in HL60
and HL60-MX1 cells, respectively. There were no differences in nuclear uptake between 111InNLS-HuM195 and 111In-NLS-M195. The EC50 for 111In-NLS-HuM195 was 1.9 kBq/well on
HL60 and HL60-MX-1 cell lines. This was 15-25 fold lower than the EC50 of 29.4 and 55.5
kBq/well for 111In-HuM195 on HL60 and HL60-MX1 cells. The EC50 for 111In-NLS-M195 was
14.0 and 15.4 kBq/well for HL60 and HL60-MX1 cells. NLS-HuM195, NLS-M195, NLS and
In-acetate had no effect on cell viability.Conclusions: 111In-NLS-HuM195 and 111In-NLSM195 were routed to the nucleus of AML cells, where the Auger electrons were lethal. The
ability of 111In-NLS-HuM195 and 111In-NLS-M195 to kill HL60-MX-1 cells suggests they may
be effective for treating chemotherapy-resistant AML. This study was supported by a grant from
the Canadian Institutes of Health Research (CIHR) to R.M.R. and a Wyeth-CIHR fellowship to
Combined effect of cetuximab and
and neck tumour cells
At-cMAb U36 in head
Role of FDG-PET
Peter Bartenstein (DE)
Role of amyloid imaging
A. Nordberg (SE)
302 — Sunday, October 01, 2006, 11:30 am - 1:00 pm, Trianti Hall
Cardiovascular: Myocardial Perfusion - Stress Testing
Influence of additional atherogenic risk factors on silent
ischemia detected by sestamibi myocardial perfusion
imaging in essential hypertensive patients
M. Dumont, J. Lefebvre, C. Cote, Y. Lacourciere;
Universitaire de Quebec, Quebec, PQ, Canada.
Centre Hospitalier
Aim: Essential hypertension (EH) and additional atherogenic factors increase the risk of
coronary artery disease (CAD). CAD early detection is important to improve medical
intervention and outcome. We studied the incidence of additional atherogenic risk factors and
their impact on the severity and prevalence of myocardial perfusion imaging (MPI) abnormalities
in asymptomatic EH patients (pts). Material and Methods: We prospectively investigated 1116
pts (579 males, 537 females) (66.3 ± 10 years). All pts underwent stress and rest dipyridamole
Sestamibi MPI. Summed stress scores (SSS) were evaluated by 2 blinded experienced observers.
Results: Overall MPI abnormalities were significantly (p < 0.0001) more prevalent in EH pts
with the addition of diabetes mellitus (DM) (n=462; 41.4%), dyslipidemia (n = 692; 62%),
proteinuria (n = 86; 7.7 %), abnormal resting ECG (n = 487; 43.6 %) and peripheral arterial
disease (PAD) (n = 430; 38.5%). Furthermore, pts with EH and DM as compared with those with
EH alone had greater incidence of high risk MPI (16.4 % vs 10.6%, p < 0.0001). Moreover, the
risk of having abnormal MPI in pts with EH and DM increased by 104% in the presence of
proteinuria (p < 0.0001), 44% (p = 0.001) in the presence of dyslipidemia, 39% (p = 0.008) with
abnormal ECG and by 41% (p = 0.012) in pts with PAD as compared with EH pts with the same
atherogenic risk factors but without DM. Conlusion: In this large-scale high risk population, the
aliquot was mixed with 100µL of 25 mM DTPA for accurate determination of unchelated 90Y,
and assayed by radio-HPLC (V) employing a UV/Visible variable wavelength detector, a radiodetector and a size-exclusion column. Each RCP experiment was performed in triplicate and on
eight different radiolabeled compounds. Results and conclusion: RCP was above 99% with all
instruments used. The auto-radiochromatographic system was very sensitive and allowed
accurate RCP results in only 3-4 min thanks to the ability to measure, at the same time, the three
radio-ITLC employed. Radio-TLC scanner showed high sensitivity too but strips measurement,
separately performed, took about 10 min to complete the whole procedure. Both systems showed
good radio-chromatographic separation. Equal sensitivity but greater practical disadvantage
involved the use of beta counting system, due to the radiosample dilution and the preparation of
radio-ITLC fractions with liquid scintillation cocktail before their measurement. Higher
sensitivity was achieved with radio-HPLC, a very powerful tool for separation of different 90Y
species eventually produced in the radiopharmaceutical, but the required time was much longer.
Dose calibrator was very rapid but less sensitive instrument for the small activities used.
prevalence of silent CAD is significantly greater in EH pts with other atherogenic factors.
Moreover the prevalence and severity in CAD increased significantly in EH pts with the same
atherogenic factors associated with DM. These findings should be clinically relevant in order to
adequately evaluate pts with EH and EH + DM and to initiate therapies aimed at reducing
mortality and morbidity.
Safety and effectiveness of atropine added to submaximal
exercise stress testing with SPECT in patients with known or
suspected coronary artery disease.
M. Spadafora, P. Varrella, R. Sauro, G. Peluso, E. Di Lorenzo, A.
Varricchio, F. Rotondi, R. Golia, A. Martino, G. Stanco, P. Miletto, F.
Manganelli; A.O.R.N Moscati, Avellino, Italy.
Aim. Failure to achieve 85% of age-predicted maximal heart rate (HR) during exercise may
significantly reduce sensitivity and predictive accuracy of SPECT. Unfortunately, several factors
affecting adversely exercise capacity (aging, obesity, peripheral vascular disease) and the
extensive use of ȕ-blockers may lead to a submaximal exercise test. This study was designed to
assess whether atropine administration is safe and effective in patients unable to reach target HR
(> 85% of age-predicted maximal HR) during stress-rest SPECT. Material and Methods. We
recruited patients with known or suspected coronary artery disease undergoing technetium-99m
tetrofosmin or sestamibi SPECT imaging at rest and after standard bycicle exercise. Final
population were constituted by 61 patients who showed signs of fatigue during exercise and had
less of 80% of age-predicted maximal HR. They were randomized to receive intravenously 0.5 to
1 mg of atropine (group A, n=31) or no medication (group B, n=30). All patients were required
to continue exercising for 1 additional minute at the same workload. Results. No significant
differences were observed in age, gender, baseline HR and proportion of patients taking ȕblockers between two groups. HR when patients showed fatigue was 119+10 beats/min (76+5%
of maximal predicted HR) in group A and 118+12 (75+6% of maximal predicted HR) in group B
(p=NS). Maximal HR in group A (135+12 beats/min, 87+8% of maximal predicted HR) was
significantly higher than in group B (124+10 beats/min, 79+7% of maximal predicted HR,
p<0.05). Target HR was achieved in 23 (74%) patients of group A and in 4 (13%) of group B
(p<0.01). Of note, even limiting analysis only to patients receiving ȕ-blockers, addition of
atropine significantly increases the number of patients in whom target HR was achieved (82% in
group A vs 6% in group B, p<0.01). There were no serious complications in either group. Only
one patients in group A experienced dizziness with prompt spontaneuos resolution. Conclusions.
Our study suggests that in patients showing fatigue at submaximal exercise, atropine
administration is a safe and effective way to reach an adequate HR in myocardial perfusion
imaging. Such approach may reduce the number of inconclusive exercise stress tests during
SPECT in patients in whom withdrawal of ȕ-blockers is deemed hazardous. Since myocardial
perfusion imaging has a biological and economic burden, addition of atropine to patients with
submaximal exercise may help to preserve its cost-effectiveness avoiding additional
pharmacologic stress testing.
Safety, tolerability and accuracy of adenosine combined with
maximal exercise protocol in myocardial perfusion imaging
A. Ferrer-Antunes, L. Oliveira, J. Correia, E. Rovira, P. Ferreira, M.
Ferreira, L. Providência, J. Lima; Coimbra Univ. Hosp., Coimbra, Portugal.
INTRODUCTION: Submaximal treadmill stress test (TST) and exercise performed by low
functional capacity patients decrease the sensitivity of myocardial perfusion imaging (MPI) for
detection of coronary artery disease (CAD). On the other hand, the use vasodilator
pharmacological stress (VS) results in loss of important prognostic data obtained in exercise
protocols. Low-level TST has been increasingly applied in combination with VS and results in
reduction of side effects and improvement of image quality. AIM: To evaluate the safety,
tolerability and diagnostic accuracy of a combined protocol with adenosine and maximal TST in
MPI.Materials and Methods: Between 12/2005 and 02/2006 103 consecutive patients (pts)
referred for MPI, with no contraindication for TST or adenosine, were randomly assigned for one
of two groups: Group A, 51 pts, 40 (78%) males, mean age 5 9+/-11 Y, performed combined
protocol of maximal TST (Bruce) + adenosine (4minute protocol); Group B, 52 pts, 37 (71%)
males, mean age 58+/-13Y, performed maximal TST (Bruce). The 2 groups were equivalent
concerning reason for referral, cardiovascular risk factors, history of CAD, cardiovascular
medication and basal hemodynamic variables.Results: The 2 groups had a similar evolution of
heart rate (A 142+/-18; B 144+/'1 71'min, P=ns), exercise duration (A 8.4+/-2.2; B 9.0+/-2.5min,
p=ns), maximal METs achieved (A 10.0+/-2.4; B 10.7+/-2.5, p=ns), symptoms (angina 16% Vs
10%, dizziness 10% Vs 6%, dyspnea 4% Vs 2%, respectively, p=ns), isolated or repetitive
ventricular ectopics 9% Vs 3 7%, respectively, p=ns), supraventricular ectopics (8% Vs 12%,
respectively, p=ns) or heart / liver tracer uptake ratio (2.6+/-2.0 Vs 2.5+/-1.8, respectiveiy, p=ns).
On the other hand, the groups showed different behaviour regarding maximal systolic blood
pressure (A 183+/-26; B 193+/-26, p=0.05) and there was a strong tendency for Group A to have
more positive TST (23% Vs 12%, p=O. 11) and a higher reversibility score (3.5+/-3.9 Vs 2.3+/2.8, p= 0. 16). Using MPI as a standard the sensitivities of stress test were 57% and 36%, and the
specificities were 89% and 97%, respectively. There were no serious adverse events. In Group A
one patient developed a high-degree AV block during exercise due to adenosine overdose,
promptly reverted after dose correction.Conclusions: In this study the combination of adenosine
with maximal treadmill exercise test as stress protocol for MPI was safe and well tolerated, and
offered good quality images. Its diagnostic accuracy didn't seem to be inferior to that obtained
with regular treadmill stress test.
Determination and localization of culprit lesion by perfusion
SPET scintigraphy in patients undergoing percutaneous
coronary intervention
B. Baskot1, S. Rusovic1, S. Obradovic1, A. Ristic-Angelkov1, R. Spaic2;
Medical Military Academy, Belgrade, Serbia and Montenegro, 2Medica
Nucleare, Belgrade, Serbia and Montenegro.
Aim: The coronary angiography provides information on the anatomical state of the coronary
tree and, specifically, on the large epicardial arteries, while perfusion SPET facilitates the
evaluation of the grade of ischemia that a particular stenosis produces. Myocardial perfusion
imaging (MPI) is of considerable use in the procedural indications of partial revascularization in
patients with chronic coronary artery disease (CAD). In these cases the purpose is to detect the
coronary stenosis that provokes the ischemia and is termed the "culprit lesion". The aim of this
study was to evaluate the accuracy of 1-Day DypEx-Rest (dypiridamole exercise low level)
Tc-tetrofosmin SPET for determination culprit lesion in patients undergoing ePCI. Materials
and Methods: Forty-two patients with known CAD were studied. In all of them
angiographically detected significant coronary narrowing (• 75% luminal stenosis). All patients
were submitted to 2 iv. injections of 99mTc-tetrofosmin, one in a peak DypEx (370 MBq) and
the other (740 MBq) at rest 3 h after exercise. Images are taken 15-30 min after injection for
booth studies. In each patient, SPET were analyzed by Stirner program modificated and
standardized myocardial segmentation and nomenclature for topographic imaging of the heart.
Quantification regional tetrofosmin uptake was performed using short-axis myocardial
tomography that was divided on 17-segments for each study. A 5-point scoring system was used
to ass’s difference between uptake degree in stress and rest studies for the same segments (1normal, 2-mild ischemia, 3-moderate ischemia, 4- reversibility and 5- severe reversibility.
Summary score • 3 in the territory of stenoses coronary artery was determinate culprit lesion.
Two of segments with score 5 (index of summary score) in the territory of stenoses coronary
artery was determinate culprit lesion. Results: A total of 126 vascular territories (714 segments)
were analyzed before elective percutaneous coronary intervention (ePCI). Overall sensitivity,
specificity, and accuracy using summary score • 3 were 82, 4%, 93, 3%, 84, 4%, with positive
predictive value 97, 7%. Overall sensitivity, specificity, and accuracy using index summary score
were 86, 3%, 100%, 89, 4%, with positive predictive value 100%. Conclusion: 1- Day DypExRest 99mTc-tetrofosmin tomography significantly improves sensitivity, specificity, and accuracy
for determination culprit lesion in patients undergoing ePCI.
Using adenosine for myocardial perfusion imaging (MPI) in
asthmatic patients
A. Notghi, N. B. Smith, M. Ebanks, S. Mostafa; City Hospital Birmingham,
Birmingham, United Kingdom.
Aim: This study evaluates the safety and practicality of a modified adenosine regime in asthmatic
patients. Introduction: Adenosine is used to inducing vasodilatation for MPI as it is safe, quick
and accurate; enabling a larger patient throughput. Its brochoconstrictor effect is well known;
there is reluctance using it in asthmatic patients especially as severe adverse effects were
reported when Dypridamol was used. However adenosine has a very short half life and thus in
event of brochoconstriction can be stopped with potentially limited adverse effects. There has
been anecdotal use of adenosine in asthmatic patients with no severe adverse effects. Here we
propose a modified regime of adenosine administration for use in these patients and describe our
experience in its use.Methods: Asthmatic patients on multiple inhalers ± oral steroids were
booked into separate “stressing” sessions. Patients were asked to inhale two puffs of solbutamol
inhaler two minutes before the start of the infusion. The adenosine infusion was modified starting
with 7 microgram/Kg adenosine and increasing to the standard 14 microgram/Kg within 1 minute
and maintained for 5 minutes. 99Tcm-tetrofosmin was injected at the 3rd minute. The Patients
blood pressure (BP), pulse (PR), oxygen saturation (finger oxymeter) and ECG were monitored
until the end of the study. All side effects were recorded.Results: Twenty one (10 females, mean
age 65.7) asthmatic patients were given adenosine infusion. One patient developed second degree
heart block (non-asthma related effect), the test was aborted. Twenty patients with asthma
completed the test successfully. They were compared with 20 consecutive non-asthmatic patients
(12 females, 64.8). There was no change in systolic blood pressure during the infusion, the
diastolic BP dropped (asthmatics 83.4 to 78, p<0.05, Non-asthmatics 81.6 to 73.7, p<0.01) and
PR increased (Asthmatics 78.6 to 98.8, p<0.01 and in non-asthmatics 79.3 to 91.6, p<0.001).
There was no significant change in oxygen saturation in asthmatics. There was a significantly
higher number who developed shortness of breath (SOB) in non-asthmatic patients (18/20 nonasthmatic vs. 11/20 asthmatic, p=0.013), only 2 in each group had significant SOB. There was no
significant difference in the number of other side effects.Conclusions:Early results suggest that
modified adenosine infusion with salbutamol pre-medication can be safely used in severe
asthmatic patients with no significant bronchoconstriction.
QRS and ST-segment changes in the determination of
significant myocardial ischemia in vasodilator stressing
E. Moralidis, N. Karavida, G. Arsos, N. Boussios, K. Karakatsanis;
Hippokration Hospital, Thessaloniki, Greece.
Aim: It has been reported that QRS changes during exercise may be predictive of coronary artery
disease (CAD), but limited data exist on their diagnostic efficacy during vasodilator stressing.
This study investigates the diagnostic value of both QRS changes and ST-segment depression
during vasodilator myocardial perfusion imaging in the determination of significant myocardial
ischemia.Methods: 110 consecutive patients (65 male), aged 62±11 years, submitted to routine
vasodilator stressing SPECT myocardial perfusion imaging were recruited. Among them, 29 had
a non-recent myocardial infarction and 26 had been submitted to coronary interventions (15
percutaneous, 11 surgical). Adenosine (n=81) or dipyridamole (n=29) intravenous infusion was
combined with standard low-level bicycle exercise (25W x 6 min). Myocardium was divided into
20 segments and regional Tl-201 uptake at stress and redistribution datasets was graded. The
Sum Stress Score (SSS, number of ischemic segments together with the degree of regional tracer
uptake at stress) and the Sum Difference Score (SDS, reflecting both the area and the degree of
defect reversibility) were calculated. Patients’ 12-lead ECG was interpreted for ST-segment
changes, whereas QRS changes in leads aVF and V5 between rest and stress recordings were
combined in a composite score (Athens score). ROC analysis was used to define a threshold of
abnormality of QRS scoring in the prediction of SSS•9 (the cut-off value providing the best
sensitivity at a desired minimum specificity).Results: ST-segment depression at stress provided a
Prediction of advanced coronary artery disease with exercise
myocardial perfusion imaging: major contribution of lung
thallium-201 uptake measurements.
E. Moralidis1, T. Spyridonidis2, G. Arsos1, C. Anagnostopoulos3;
Hippokration Hospital, Thessaloniki, Greece, 2Hippokration Medical
Center, Larissa, Greece, 3Royal Brompton Hospital, London, United
im: Although myocardial perfusion images offer a high sensitivity for the detection of coronary
artery disease (CAD), the prediction of advanced CAD is suboptimal if based on perfusion
defects alone. This study investigates the relative value of clinical, exercise electrocardiography
and scintigraphic variables in the prediction of advanced CAD.Methods: 398 patients with
suspected or known CAD, but no previous coronary interventions, referred for routine exercise
SPECT myocardial perfusion imaging and submitted to coronary angiography within 3 months
from scan date, were enrolled. Patients were separated into sub-populations of lower- (n=156)
and higher-risk (n=242) CAD. Non-significant coronary lesions (<50% diameter stenosis) and 1vessel disease, other than proximal LAD, was classified as lower-risk CAD. Higher-risk CAD
was considered to be present in cases of a proximal LAD lesion, when multiple vessels showed •
50% luminal narrowing and in LM stem disease. Age, angina as presenting symptom, a history
of myocardial infarction and the Duke treadmill score (DTS) were recorded in all patients.
Summed stress score, transient ventricular dilation (stress to rest left ventricular volume ratio in
SPECT imaging) and lung thallium-201 uptake (L/Hnet, lung to background subtracted
myocardial count density ratio) were also calculated. Logistic regression analysis was used to
assess both the independent contribution of variables in the identification of advanced CAD and
their incremental value.Results: Age (chi-square=16.3, p<0.001), DTS (chi-square=15.0,
p<0.001), SSS (chi-square=39.8, p<0.001) and L/Hnet (chi-square=106.3, p<0.001) were
independent predictors of advanced CAD, whereas angina, a history of myocardial infarction and
transient ventricular dilation were not. DTS offered significant incremental value over clinical
data. SSS added significant predictive information on the model incorporating both clinical data
and the DTS. However, once L/Hnet was entered in the model including all previous variables,
SSS was no longer significant. The model consisting of clinical data, the DTS and L/Hnet offered
the highest information in identifying advanced CAD.Conclusions: In patients with CAD and no
coronary interventions, lung thallium-201 uptake (lung to background subtracted myocardial
count density ratio) offers significant incremental information over clinical data, the Duke
treadmill score and perfusion images in the identification of advanced CAD with exercise
SPECT myocardial perfusion scintigraphy.
Regional washout phenomenon of rest Tc-99m-sestamibi
-comparison between acute myocardial infarction and old
myocardial infarctionH. Yamashina, S. Yamashina, A. Inoue, J. Yamazaki; Toho University
Omori Medical Center, Tokyo, Japan.
Background and Aim: Tc-99m-sestamibi (MIBI) is a myocardial perfusion agent which is
retained in the myocardium for many hours with slow washout. Recently, it has been reported
that in patients with acute myocardial infarction (AMI), myocardial imaging with MIBI shows a
rapid washout phenomenon in infarct area. In this study, we evaluated regional washout rate
(RWR) in patients with AMI and old myocardial infarction (OMI).Materials and Methods: We
studied 30 patients with AMI who received successful reperfusion therapy within 24 hours from
onset, and 10 patients with OMI. Their infarct-related arteries were LAD. Two-phase imaging of
MIBI was performed at rest condition. The SPECT images were divided into 17 segments, and
regional uptake was visually graded using 5 point scoring system. Sum of scores in segments
corresponding with infarct-related area was expressed as total score (TS). Washout score (WS)
was defined as the difference between TS of early image and delayed image. For further
quantitative evaluation, polar map was constructed. Extent score (ES) and severity score (SS)
were calculated for the area which had lower counts than the normal range (mean+/-2SD). Polar
map was divided into coronary-related area and RWR was calculated.Results: In AMI, TS, ES
and SS in delayed image were significantly higher than those in early image (16.1+/-10.1 vs.
9.4+/-9.8 p<0.0001, 42.1+/-21.7 vs. 36.8+/-23.0 p<0.016, 73.6+/-64.1 vs. 49.3+/-58.3
p<0.0001).In constant, there were no significant differences in OMI (24.6+/-3.4 vs. 23.8+/-2.8,
68.5+/-9.5 vs. 67.5+/-8.6, 204.2+/-29.4 vs. 185.5+/-37.9). RWR in infarct area showed no
significant difference between AMI and OMI. However, RWR in infract area was significantly
higher than that in non-infarct area (p<0.0001) in AMI, whereas there was no significant
difference in OMI (Table). RWR in infarct area showed significant correlation with TS
(R=0.656, p<0.0001), ES (R=0.711, p<0.0001) and SS (R=0.603, p<0.0003) in delayed image in
AMI, however, no significant correlation was found in OMI. Significant correlation was
demonstrated between the RWR in infarct area and non-infarct area in both of AMI
(R=0.862,p<0.0001) and OMI (R=0.756,p<0.009).Conclusion: Regarding washout phenomenon,
AMI showed correspondence of visual findings and quantitative evaluation. In OMI, various
increased washout were found in infarct-area despite of showing defect visually. There were no
significant correlations between RWR and each score of SPECT image. There was positive
correlation between RWR in infarct and non-infarct area, and no significant difference between
them, which suggest that RWR in OMI reflected washout in whole left ventricle.
sensitivity of 16% and a specificity of 80% in the prediction of SSS•9. At the corresponding
threshold of abnormality of QRS scoring (” -4), sensitivity (27%) and specificity (80%) values
were not significantly different than those of ST-changes in the detection of SSS•9. In the
identification of SDS•5, ST-segment depression and QRS scoring (abnormal if ” -4) yielded a
sensitivity of 24% and 31%, respectively (ns), and a specificity of 89% and 82%, respectively
(ns).Conclusions: In a mixed population of patients with suspected or known CAD, QRS changes
during vasodilator stressing are not superior to ST-segment depression in the determination of
significant myocardial ischemia, and both are of limited diagnostic value.
Regional washout rate (RWR) in AMI and OMI
RWR in infarct RWR
area (%)
non- difference between infarct and
infarct area (%)
non-infarct area
31.5 +/- 6.7
24.2 +/- 5.8
29.0 +/- 10.4
26.3 +/- 8.0
303 — Sunday, October 01, 2006, 11:30 am - 1:00 pm, Banqueting Hall
Oncology: Central Nervous System Tumours
FET (O-(2-[(18)F]fluoroethyl)-L-tyrosine)- PET for evaluation
of cerebral tumorours lesions suspicous for malignancy
R. Pichler, A. Fischlmayr, J. Pichler, G. Wurm, R. Silye, K. Nußbaumer, F.
T. Aichner; Wagner Jauregg Hospital, Linz, Austria.
Aim: FET-PET has been shown to be useful for evaluation of brain tumors. Our aim was to
investigate this method in brain lesions with suspicous malignancy. Patients and methods: We
evaluated FET-PET imaging in 30 patients (10 females) refered to the Institute of Nuclear
Medicine in our hospital from 11/04 to 3/06. Median age was 47 years [quartiles 29; 60 years],
all presented a cerebral lesion detected by MRI (one patient only had CT). A history of brain
tumor was unknown until then. Most frequently seizures were the reason for admission to the
hospital. 200 MBq 18-F-FET were injected i.v., followed by PET-CT imaging of the brain
(Philips Gemini GXL) 30 minutes later.Results: 16 patients were positive (high uptake of the
lesion) in FET-PET, two undetermined (equal to physiological uptake) and 12 patients had
negative images (defects). Histology was obtained in 21/30 patients, including 14/16 of the FETpositive group (5 glioblastomas, 3 astrocytomas II, 3 oligodendrogliomas II , 2 pilocytic
astrocytomas I, 3 other malignancy). In one patient each with glioblastoma and malignant
ponsglioma surgery was not possible because of reduced general conditions. The FET-negative
group included two astrocytomas II, a single metastasis of lung cancer and some benign tumors
as lipoma or cavernoma. Most cerebral lesions observed in the non-positive groups were at least
considered to be of vascular origin.Conclusions: FET-PET was positive in high-grade brain
tumors, negative images were helpful to exclude that condition and could encourage a wait-andsee strategy. Astrocytomas WHO grade II may or may not have marked FET uptake - negative
FET imaging does not exclude low -grade brain tumor. Caution is recommended when metastasis
of a known primary malignancy is possible.
F-18 FET-PET for preoperative evaluation of brain tumours
showing no Gd-enhancement in the MRI
M. Plotkin1, F. Stockhammer2, H. Amthauer1, U. Thomale2, T. Lehmann2,
C. Woiciechowsky2, T. Denecke1, R. Felix1; 1Department of Radiology,
Nuclear Medicine and Radiooncology, Campus Virchow-Klinikum,
University Clinic Charité, Berlin, Germany, 2Department of Neurosurgery,
Campus Virchow-Klinikum, University Clinic Charité, Berlin, Germany.
Aim: Planning of neurosurgical interventions in patients presenting with non-enhancing brain
masses in MRI is challenging. In such cases, especially in large, inhomogeneous lesions, possible
malignant tumour areas cannot be excluded. FDG-PET was shown to be useful for grading of
brain lesions and defining the optimal biopsy site. However, the sensitivity of FDG-PET for
detection of brain tumours is limited. In this study, we assessed the value of PET using the amino
acid tracer FET for evaluation of brain lesions showing no Gd-enhancement. In addition, the
accuracy of FET-PET and FDG-PET for this issue was compared.Materials and Methods: FETPET was performed in 15 consecutive pat. with suspicious brain lesions (n=11) or known lowgrade gliomas (n=4), in whom MRI demonstrated no Gd-enhancement and who all were
candidates for surgery. 7 of these pat. were also studied by FDG-PET. Data of FET-PET, FDGPET and MRI were fused (Brainlab iPlan V 1,0) and then transferred to the neurosurgical
navigation system (Vector Vision2). In 12 pat. PET-guided biopsies were performed.Results:
FET-PET depicted an increased intratumoral tracer uptake in 11/15 pat. Histology showed
gliomas in 10 of these pat. (WHO II, n=6; WHO III, n=4). In one pat., no biopsy was possible
because of an unfavourable tumour location. Two out of 4 negative findings in FET-PET were
verified by biopsy: one of them corresponded to an astrocytoma WHO II, and another one to an
inflammatory lesion. In one patients, in whom no FET accumulation in a cystic lesion was
detectable; the finding was classified as an old infarction and no biopsy was performed. In
another patient, with known astrocytoma WHO II and negative FET-PET, a “wait-and-see” tactic
was chosen. FDG-PET showed hypermetabolic lesions in 2/7 studied pat. (WHO II, n=1; WHO
III, n=1) and failed to detect gliomas in 5/7 pat. (WHO II, n=4; WHO III, n=1). In 4 of these 5
pat., tumours were correctly detected by FET-PET. The spots with the maximal tracer
accumulation in both patients with positive FET und FDG scans were concordant.Conclusions:
FET-PET is more sensitive than FDG-PET in the detection of brain tumours and represents
therefore a suitable imaging tool for evaluation and biopsy planning of non-Gd-enhancing brain
PET Imaging of brain tumor with N-NH3
Z. Xiangsong; The First Affiliated Hospital, Sun Yat-sen University,,
Guangzhou, China.
Background: Hoop et al. (J Nucl Med. 1976;17:473-479) observed 1 recurrent astrocytoma and
1 glioma showed reduced uptake of 13N-NH3 at the lesions. Schelstraete et al. (Br J Radiol.
1982;55:797-804) observed litter or no extra uptake of 13N-NH3 in 4 primary brain tumors.
Benard et al. (Semi Nucl Med. 2003;33:148-162) thought 13N-13N-NH3 was unsuccessful in
visualizing primary brain tumors. Purpose: The aim of this study was re-evaluated the role of
N-NH3 PET in brain tumor.Methods: 13N- NH3 PET was performed in 47 patients (30male and
17 female patients; age range 18-62 years ) with suspected primary brain tumor or recurrent
tumor detected by MRI. Five minutes after injection of 555~740 MBq of 13N-13N-NH3,
attenuation-corrected brain images were obtained with a dedicated PET scanner. A
histopathological diagnosis was made for 39 of them by open surgery or stereotatic biopsy, and
the others were confirmed by clinical or radiologic follow-up, including 10 cases with
astrocytoma gradeIorII, 15 astrocytomas grade III or IV (7 of them were tumor recurrence), 2
oligodendroglioma gradeI, 1 dysembryoplastic neuroepithelial tumor (DNT), 2 meningiomas, 2
metastatic tumors, 5 radiation necrosis, 7 encephalomyelitic foci, 2 ischemic lesions, 1 reactive
gliosis. PET images were visually inspected, and the uptake ratios of lesion to contralateral
normal brain tissue (T/N) were determined.Results: Twenty-seven of thirty-two brain tumor
exhibited markedly increased uptake of 13N-NH3. 2 oligodendroglioma, 1 DNT, 1 metastatic
tumors and one low-grade astrocytoma in brain stem showed absent 13N-NH3 uptake. The 15
non-neoplastic lesions exhibited absent or less 13N-NH3 uptake. The sensitivity and specificity of
N- NH3 PET for detecting brain tumor were 27 (84%) of 32 and 15 (100%) of 15, respectively.
The sensitivity for detecting astrocytoma was 24 (96%) of 25. There were the significant
differences between the T/N ratios in brain tumors and that in non-neoplastic lesions (T/W ratios:
4.23±1.70, n=32 vs 0.82±0.25, n=15, P<0.01).The significant differences were observed between
high-grade and low-grade gliomas with respect to T/N ratios (T/N ratios: 4.12±1.65, n=15 vs
1.71±0.43, n=13, P<0.01).Conclusions: 13N-NH3 PET can be used for diagnosis of brain tumors,
especially in cerebral astrocytomas. 13N- NH3 PET may enable differentiation between low- and
high-grade gliomas.
Differentiation of recurrent astrocytoma from radiation
necrosis: initial experience with N-NH3 PET
Z. Xiangsong; The First Affiliated Hospital, Sun Yat-sen University,,
Guangzhou, China.
Differentiation of posttherapy radiation necrosis from recurrent brain tumor remains a
challenging diagnostic problem. The combination of the imaging modalities on the basis of
different physiologic mechanisms could improve diagnostic accuracy. The present study assessed
the role of 13N-NH3 PET in differentiating recurrent cerebral astrocytoma from radiation
necrosis.Methods: Twelve patients with suspicion of recurrence based on MR imaging findings
more than 4 months after radiation for cerebral astrocytomas were examined prospectively with
N-NH3 PET. Ten of them also underwent 18F-FDG PET. Three reviewers interpreted the 13NNH3 and FDG PET scans. The qualitative analysis was also graded according to a 5-point
grading system. Inappropriate regional increase in 13N-NH3 and 18F-FDG PET were considered
indicative of tumor recurrence. The final diagnosis based on surgical resection, stereotactic
biopsy or clinicopathologic follow-up was tumor recurrence in 7 cases and radiation necrosis in 5
cases.Results: In all 12 cases the 13N-NH3 PET scans were concordant with the final diagnosis.
Eleven lesions with recurrent tumor showed increased 13N-NH3 uptake (grade = 4-5). All 5
lesions with radiation necrosis showed absent 13N-NH3 uptake (grade = 1). 13N-NH3 PET was
discordant with FDG PET in two contrast-enhanced lesions. The area with reactive gliosis
showed absent 13N-NH3 uptake, however, moderately increased 18F-FDG uptake in one contrastenhanced lesion. One lesion with the recurrent anaplastic astrocytoma showed markedly
increased 13N-NH3 uptake, but hypometabolic 18F-FDG PET abnormality. Conclusions: The
recurrent astrocytomas showed increased 13N-NH3 uptake, and the radiation necrosis showed
absent 13N-NH3 uptake, and 13N-NH3 seem superior to 18F-FDG for this purpose, suggesting that
N-NH3 is a promising tracer for separating radiation necrosis from astrocytoma recurrence.
However, the patient population in this study was small. Thus, more studies are needed to settle
this issue.
Usefulness of N-ammonia PET in the evaluation of brain
lesions that are hypometabolic on F-FDG PET
Z. Xiangsong; The First Affiliated Hospital, Sun Yat-sen University,,
Guangzhou, China.
We investigated the usefulness of 13N-ammonia PET in characterizing brain lesions which show
hypometabolism on 18F-FDG PET. Methods: 13N-ammonia PET was performed in twenty-nine
patients with brain lesions (in 24 for initial diagnosis and in 5 for detection of astrocytoma
recurrence) that showed hypometabolism compared with normal brain tissue on FDG PET. Four
minutes after injection of 555~740 MBq of 13N-ammonia, attenuation-corrected brain images
were obtained with a dedicated PET scanner. The brain lesions comprised 11 astrocytomas, two
meningiomas, two oligodendrogliomas, one metastatic tumor, one dysembryoplastic
neuroepithelial tumor (DNT), and twelve benign lesions (including four cases of radiation
necrosis, one tuberculous granuloma, two ischemic lesions, five encephalitic foci). Results: Ten
of 11 astrocytomas (91%, sensitivity) showed increased 13N-ammonia uptake despite
hypometabolism on FDG PET, and the degree of which exceeded 18F-FDG uptake. The mean
(±SD) uptake ratios of astrocytoma to normal brain on 13N-ammonia and FDG were 1.75±0.32
and 0.48±0.27, respectively. By contrast, all twelve benign lesions showed decreased 13Nammonia uptake (100% specificity). Three non-astrocytoma gliomas and one metastatic tumor
showed decreased 13N-ammonia uptake. Two meningiomas showed intense 13N-ammonia uptake.
Conclusions: 13N-ammonia PET showed high sensitivity and good contrast in the evaluation of
cerebral astrocytomas which showed hypometabolism on 18F-FDG PET. 13N-ammonia PET
could provide additional information when used in combination with FDG PET.
Differential uptake of F-18-fluoroethyl-L-tyrosine and H-3-Lmethionine in focal cortical ischemia
D. Bauer1, G. Stoffels2, D. Pauleit2, G. Reifenberger3, K. Hamacher4, H. H.
Coenen4, K. J. Langen2; 1C. & O. Vogt Institute of Brain Research,
University of Düsseldorf, Germany, 2Institute of Medicine and Brain Imaging
Center West, Research Center Jülich, Germany, 3Institute of
Neuropathology, University of Düsseldorf, Germany, 4Institute of Nuclear
Chemistry and Brain Imaging Center West, Research Center Jülich,
Objectives: C-11-methionine (C-11-MET) is particularly useful in brain tumor diagnosis but
unspecific uptake e.g. in cerebral ischemia has been reported. The F-18-labeled amino acid O-(2[F-18]fluoroethyl)-L-tyrosine (FET) shows a similar clinical potential as MET but is applicable
on a wider clinical scale. The aim of this study was to compare the uptake of FET and H-3-MET
in focal cortical ischemia in rats by dual tracer autoradiography.Methods: Focal cortical ischemia
was induced in 22 Fisher CDF rats using the photothrombosis model (PT). One (n=3), two (n=5),
four (n=3), seven (n=4) days, two (n=1) and six (n=4) weeks after induction of the lesion FET
and H-3-MET were injected intravenously and one hour later the animals were killed. The brains
were frozen in 2-methylbutane at -50°C and cut in coronal sections.The tracer distribution was
detected by dual tracer autoradiography and evaluated by regions of interest (ROIs) placed on
areas with maximal tracer uptake in the lesion and the contralateral brain. Lesion to brain ratios
(L/B) were calculated by dividing lesion and brain value. A L/B ratio > 2.0 was considered as
pathological. The localization of glial fibrillary astrocytic protein (GFAP) and macrophages
(CD68) was analysed immunhistochemically. Results: Variable increased uptake of both tracers
was observed in the PT and its demarcation zone at all stages after PT. The cut-off level of 2.0
was exceeded in 10/22 animals for FET and in 15 of 22 animals for MET. One day after PT the
L/B ratios were 2.0 ± 0.7 for FET vs. 2.1 ± 1.0 for MET (mean ± SD); two days after lesion 2.2 ±
0.6 for FET vs. 2.4 ± 0.2 for MET; four days after lesion 1.8 ± 0.2 for FET vs. 2.6 ± 0.4 for
MET; 7 days after lesion 2.2 ± 0.7 for FET vs. 2.7 ± 1.0 for MET and 6 weeks after lesion 1.3 ±
0.2 for FET vs. 1.9 ± 0.6 for MET. In lesions with differential uptake of FET and MET were
revealed that uptake of FET was associated with GFAP localization while MET uptake was
correlated with CD68 staining.Conclusions: FET like MET may exhibit significant uptake in
infarcted areas or the immediate vincinity which has to be considered in the differential diagnosis
of unkown brain lesions. The discrepancies between FET and MET uptake appear to be caused
by preferential uptake of FET in reactive gliosis and selective uptake of MET in macrophages.
Use of F-18 FET-PET for planning of thermotherapy using
magnetic nanoparticles in patients with recurrent
M. Plotkin1, U. Gneveckow2, P. Wust1, A. Jordan2, K. Meier-Hauff3, T.
Denecke1, H. Amthauer1, R. Felix1; 1Department of Radiology, Nuclear
Medicine and Radiooncology,Campus Virchow-Klinikum, University Clinic
Charité, Berlin, Germany, 2MagForce Nanotechnologies AG, Berlin,
Germany, 3Department of Neurosurgery, Bundeswehrkrankenhaus Berlin,
Berlin, Germany.
Aim: Thermotherapy using magnetic nanoparticles (nano cancer therapy) is a new concept of
local tumour therapy, which is based on controlled heating of intratumoural injected magnetic
nanoparticles. An important prerequisite for the successful therapy is the coverage of the whole
tumour volume with magnetic nanoparticles. The aim of this study was to evaluate the usefulness
of PET with a recently introduced amino acid tracer O-(2-[F-18]fluoroethyl)-L-tyrosine (FET)
for targeting the nanoparticles implantation.Materials and Methods: 11 consecutive pat. (females,
4; males, 7; median age, 44 years; range, 25-75 years) with recurrent glioblastoma, underwent
MRI and FET-PET for planning of the nano cancer therapy. PET studies were performed using
an ECAT-EXACT 47 scanner (Siemens, Erlangen, Germany), starting 10 min. after intravenous
administration of 250 MBq F-18 FET. After coregistering of PET und MRI data, the gross tumor
volumes (GTV) were delineated separately, based on MRI and PET data. Thereafter, the final
GTV were defined, taking the results of both imaging tools in consideration. The resulting data
were transferred to the neurosurgical navigation system (Stealth Station, Medtronic, MN, USA)
and used for stereotactic guidance of the nanoparticles instillation.Results: The MRI-based mean
GTV was 24.3 cm3 (range, 2.5-59.7), and the PET-based mean GTV was 31.9 cm3 (range, 5.277.9). The MRI identified an additional 8.9 ± 4.7 cm3 (mean ± standard deviation), and the FETPET scan an additional 16.5 ± 15.2 cm3 outside of the common GTV (15.4 ± 11.0 cm3). The
mean final GTV accounted to 33.8 cm3 (range, 5.2-77.9). The additional information of FETPET led to an increase in GTV by 22-286% in 8 pat., and to a decrease by 23 and 26% in 2 pat.,
respectively. In one pat., the final GTV was defined on the basis of MRI data only.Conclusions:
FET-PET adds important information on the actual tumour volume in recurrent glioblastomas
and is highly valuable for defining the target volume for the nano cancer therapy. Another
advantage of the FET-PET in planning of the nano cancer therapy is it’s suitability as a baseline
scan for post-treatment evaluation, since MRI cannot be used for this purpose because of
susceptibility artifacts, caused by the magnetic nanoparticles.
PET-CT imaging of recurrent chordoma of the skull base :
Interim analysis of a prospective study comparing FDG-(18F)
and FMISO-(18F)
K. Kerrou1, H. Mammar2, D. Grahek1, F. Gutman1, Y. Petegnief1, V. De
Beco1, F. Montravers1, J. Talbot1; 1Médecine Nucléaire Hôpital Tenon APHP, Paris, France, 2Radiotherapie - Institut Curie, Paris, France.
Table 1: Subjective scores and SUVm ratios
preliminary results provide evidence that a protocol as described above is feasable and can be
used for this clinical evaluation.
Is there already a place for the routine use of iterative
methods in Nuclear Cardiology?
L. F. Metello1, J. M. P. de Lima2, A. I. Ferrer Antunes2, L. Cunha1, A. M.
Abrantes3, B. Ferreira1, A. Nunes1, M. F. Botelho3; 1Nuclear Medicine
Course, High Institute of Health Technologies of Porto, ESTSP-IPP, Porto,
Portugal, 2Nuclear Medicine Department, University Hospital of Coimbra,
Coimbra, Portugal, 3Department of Biophysics and Biomathematics, IBILI,
Medical Faculty, University of Coimbra, Coimbra, Portugal.
Introduction: Filtered Backprojection (FBP) has a very well established place in routine image
processing in the majority of Nuclear Medicine Departments. With the improvement of
computational power actually available, access to the Iterative Methods (IM) should be more and
more considered. We tried to figure what would be the advantage(s) and disadvantage(s) of using
these methods in the actual state of technology. Materials and Methods: After study and establish
what would be best practical compromise considering the wide range of IM available, one
standard methodology had been choosed. 100 consecutive myocardial perfusion imaging (MPI)
examinations had been selected and processed by both methods (FBP and IM), being produced,
each time, the same kind of report. These images were then evaluated by 5 independent experient
physicians, according to three different criteria, but always on the basis of its clinical
value.Conclusions: It had been proved that in the actual state of the art, in this specific issue of
data processing technology, it is possible to take profit of IM without compromising the
praticability of the entire operation and the execution time (the time spent with both methods
being quite similar). Finally, statistical analysis showed notorious advantages of IM over FBP,
demonstrating that the moment for a wide use of IM, especially in this specific field of MPI, is
Optimising the nuclear cardiology service at the brighton &
sussex university hospitals nhs trust
subjective scale Ratio subjective scale Ratio
Subjective analysis correlated well with SUVm ratios for both FDG (r = 0,74; 95% CI = 0,08 0,95; p=0,0345) and FMISO (r = 0,79; 95% CI = 0,26 - 0,95; p=0,0113). Mean SUVm ratio was
higher for FMISO ( 1,59; 95% CI = 1,21 to 1,96) than for FDG (0,41; 95% CI = 0,32 to 0,51)
with significant paired samples t-test p = 0,0005.Conclusions: FMISO-(18F) PET-CT imaging
was more informative than FDG-(18F) in recurrent chordoma of the skull base. These results
could have potential applications for conformal radiotherapy of this highly locally recurrent
304 — Sunday, October 01, 2006, 11:30 am - 1:00 pm, Mitropoulos
Technologists Session 1
Comparison of PET/CT and SPECT/CT for assessment of
myocardial perfusion requires coinjection of 13NNH3 and
M. de Bloeme, R. Milovanovic, G. Radtke, M. Spörri, V. Dürst, T. Berthold,
V. Treyer, P. Kaufmann; University Hospital Zuerich, Zuerich, Switzerland.
Aim: Comparison of 13NNH3 PET versus 99mTc-Tetrofosmin SPECT for the assessment of
coronary artery disease (CAD) requires coinjection of the two tracers. The aim of this study was
to evaluate the feasibility of a protocol with coinjection of the two tracers.Methods: Nine patients
with suspected CAD were included in the present study. A two day protocol was used. The
patients were asked to refrain from caffeine-containing food and beverages for 12 hours. On day
one a stress perfusion with adenosine at standard dose and rate (0.14mg/kg/min over 7 minutes)
was used. After positioning of the patient in the PET-CT (GE Discovery STR) scanner a lowdose CT (80mA; 140kV) for attenuation correction was performed. Thereafter adenosine stress
was started. At end of minute 3 of Adenosine 750Mbq 13NNH3 and 750Mbq 99mTcTetrofosmin were coinjected. Together with the injection dynamic PET acquisition was started
using the following time frames: 9x10”, 6x15”, 3x20”, 2x30”, 1x15’. Two hours later acquisition
on the SPECT/CT camera (GE Infinia) was performed using the following parameters: 3° per
step, 30” per step, low-dose CT (2.5mA;140kV). Images were reconstructed with QPS/QGS
(Xeleris 1.0). All scanning parameters including geometry were carefully stored to be reused on
day two. On day two the same procedure with coinjection was repeated without adenosine stress.
In order to compare the SPECT images on both days a syringe with 2 MBq 99mTc-Tetrofosmin
diluted to 50ml saline was attached to the patients’ chest at the same position near the heart as a
relational reference. The difference in activity between the two days was corrected using a
gamma counter (Packard).Results: Coinjection and image acquisition was successfull in all
patients. Clinically useful images were obtained for all patients on day one and two for either
technique SPECT and PET. The correctionfactor for the SPECT measurements defined with the
syringes was close to 1 (0.97±0.1).Conclusions: Comparison of 13NNH3 PET versus 99mTcTetrofosmin SPECT for the assessment of CAD requires coinjection of the two tracers. Our
A. Vara, C. Garner, L. Jenkins, D. Edwards, J. Berry, K. Miles, S.
Dizdarevic; Brighton & Sussex University Hospital, Brighton, United
Aim: Brighton & Sussex University Hospitals (BSUH) NHS Trust was created in 2001 following
merger of five hospitals. The Nuclear Medicine service is co-ordinated across two sites, The
Royal Sussex County Hospital (RSCH) and Princess Royal Hospital (PRH), and is delivered in
conjunction with the Brighton & Sussex Medical School (BSMS). In 2003, the demand for the
nuclear cardiology service increased with intense pressure. We aimed to optimise the nuclear
cardiology service in order to cope with further service pressures anticipated following future
introduction of the NICE guidance. Methods: Building a new infrastructure was our main focus
to achieve an improvement. We firstly considered adopting new protocols based on current
developments in nuclear cardiology. Secondly, service capacity was increased by the creation of
new staff positions. Thirdly, extra capacity could be created by co-ordinating Nuclear Cardiology
over the two sites. Results: Our first achievement was to change the pharmacological cardiac
stressing method to an Adenosine protocol instead of previously used Dipyridamole. Based on
the two-day protocol, this change has given us the ability to schedule more patients on a single
list with better efficiency. As regards staffing, new appointments were established to support the
nuclear medicine service, including nuclear cardiology. These were:
Two academic posts
Further Consultant
Further Radiographer
Clinical Modality Manager (CMM)
Currently we have one qualified non-medical stress lead i.e. clinical nurse specialist. To further
underpin the stress leads, the CMM extended his role as a stress lead by attending an approved
course. Cardiac stressing is always provided by a 2 staff team; a lead and a nurse assistant. To
assist all the stress leads, nurses from main Radiology are encouraged to participate in cardiac
stressing and to support both sites. This also helps in maintaining the professional developments
of the individuals. Finally, a proposal was made towards a business case for an additional
dedicated cardiac gamma camera. This camera would become the lead instrument for cardiac
imaging, but would be utilised in conjunction with the other dual headed systems across both
sites. Conclusions: The series of initiatives within nuclear cardiology has cumulatively led to
major achievements in service integration and development. We have increased the total number
of patients from approx. 250 to more than 700 per year across the two sites, re-established the
myocardial perfusion imaging at PRH and the integrated the service in the Trust.
Introduction of a flexible Tc99m-Tetrofosmin one-day stress
and rest scanning protocol for myocardial perfusion
E. Müller, M. de Bloeme, M. Spoerri, G. Radtke, T. Berthold, V. Treyer, P.
Kaufmann; University Hospital Zurich, Zurich, Switzerland.
Aim: In our clinic we use a 1-day stress-rest 99mTc-Tetrofosmin protocol to asses myocardial
perfusion. With our normal protocol patients were scanned about 40 minutes after intravenous
injection of 99mTc-Tetrofosmin. To provide more variability as well as to decrease the risk of
high intestine activity close to the heart, we introduced the possibility to scan the patients 5 to 10
minutes after injection.Materials and Methods: Our one-day protocol begins with adenosine or
dobutamine infusion stress test. During maximal stress 300 MBq of 99mTc-Tetrofosmin was
injected intravenously. With the new protocol patients can be scanned 5-10 minutes after tracer
application ( 30s/Frame, 16 Frames/Cycle, 60x3° Steps, 64x64 matrix) on a dual-detector
SPECT/CT camera mounted at right angles and fitted with high-resolution collimators (GE
Chordomas of the skull base are rare locally invasive tumours arising from embryonic
notochordal remnants, associated with high recurrence rates. Their treatment is a combination of
maximal surgical removal and/or radiotherapy with fractionated photon and proton irradiation.
The local control of the tumour is mainly dependent on the quality of radiation, especially dose
uniformity within the GTV. Aim: The aim of this study was to evaluate the potential role of
PET-CT imaging in this indication comparing two PET tracers for characterisation of residual
disease to find the most suitable tracer for conformal radiotherapy, in order to minimise
underdosed areas because of the close proximity of critical structures and to redefine and
possibly escalate dose constraints to more aggressive residual tumour targets.Methods: From
January 2005 to February 2006, 7 patients with recurrent chordoma who were planned for
surgery and/or radiotherapy were included. All patients gave an informed consent to be included
in the prospective phase III study (MIET). All patients had FDG-(18F) PET-CT as reference
imaging. Eight FMISO-(18F) examinations were performed, with one patient having a second
FMISO-(18F) after incomplete surgery and prior to radiotherapy. All tracers were provided by
IASON GMBH, Graz, Austria. Examinations were performed with a GEMINI PET-CT system
(Philips; The Netherlands) post IV injection of 5 MBq/kg body weight, one hour for FDG-(18F)
and two hours for FMISO-(18F). Reconstructions with and without attenuation correction were
performed with a RAMLA 3D iterative algorithm. Images were analysed qualitatively with a 5scale score subjective assessment of uptake on PET only images, at both patient and site levels,
ranging form 1 = no uptake to 5 = intense uptake, and semi-quantitatively computing the ratio of
maximum standardised uptake value (SUVm) of the tumour and cerebellum, with knowledge of
lesions on conventional imaging.Results: Table 1 shows subjective scores and SUVm ratios.
Infinia or GE Millenium Hawkeye VG). This was followed by a low-dose CT (2.5mA, 14 kV,
128x128 matrix) was acquired right afterwards for attenuation correction. Images were
reconstructed and analysed using QPS/QGS tool as implemented Xeleris 1.0 (GE Healthcare).
Images were checked before the second application of the tracer in order to give the opportunity
to repeat stress acquisition. If the quality of the images was good 900 MBq 99mTc-Tetrofosmin
were injected on the camera and the patient was scanned 5 minutes later again with the same
scanning parameters. In cases where a direct scanning of the stress images was not feasible,
direct scanning of the rest condition was still possible.Results: First preliminary observations are
promising, but due to a high workload of our scanners not all patients can be scanned with this
fast protocol. Nevertheless, the flexibility of the new one-day protocol offers the technicians
more room to plan scanner time. Artefacts were seen in both conditions: 5-10 minutes postinjection they were mostly due to high liver uptake in patients with increased liver sizes.
Artefacts seen 40 minutes post-injection were mainly due to high intestine uptake close to the
heart. These artefacts could only be reduced with a second acquisition after another 40 minutes in
which patients were walking around and drank coffee and milk. Further observations are
necessary to evaluate optimal artefact reduction for different patients.Conclusions: This new
open and flexible protocol resulted in a better and ad-hoc changeable workload of the two
SPECT/CT systems as well as in a shorter examination time for some of our patients.
New method of acquisition, by means of (18f)fdg pet/ct, to the
contribution of the stages of carcinoma of the oral cavity
V. Balsamo1, G. Secco2, D. Italiano2, C. Testa2, D. Cagliero2, S. Cusmà3, U.
Ficola1, A. Cistaro2; 11. Nuclear Medicine Department, La Maddalena
Oncologic Hospital, Palermo, Italy, 2Center PET IRMET S.p.A., Torino,
Italy, 32. Radiology Department, La Maddalena Oncologic Hospital,
Palermo, Palermo, Italy.
PET/CT simultaneously supplies functional and anatomic images.The (18F)FDG-PET has often
suffered from the loss of anatomical information,which is particularly critical in the head-neck
region due to the close proximity of multiple anatomical structures and the high frequency of
physiologic (18F)FDG-uptakes in this area.PET/CT has demonstrated to be more accurate than
the PET alone. Aim of this study was to value the utility of a new method(Cistaro) in the staging
of oral cavity carcinoma.Materials and Methods:17 patients were investigated and all were
required to observe the following instructions:1)the exam reservation around mid morning to
avoid the possible (18F)FDG-uptake by the muscles at the tongue basis,that keep the tongue
lowered to the mouth floor during the night time rest;2)observation of silence and the prohibition
to drink 30 minutes before the injection,therefore avoiding (18F)FDG-fixation by the muscles of
the larynx(in particular the vocal-muscles);3)observation of silence and the prohibition to drink
during the period between (18F)FDG-injection and whole-body-scan;4)during this period the
patient must remain seated for the same reasons as reported in number 1. On the tomography
bed,the head is placed inside a helmet.The laser is allined with the orbit-meatal line.CT/PET-scan
of the whole-body is carried out from the orbit-meatal line to the pelvis.Then,with the laser
directed to the open-mouth,a final CT/PET-scan of the head/neck area is carried out,from the
orbit-meatal line to the fossa supraclavicular.Results:Good collaboration on the part of all
patients.There haven’t been any cases of aspecific uptakes about fonoatori muscles and about
tongue muscles.With the open-mouth in 15 of the 17 patients there has been a better evaluation
of the tumor extension.This method has particular advantage with respect to tumors of the
palate,tongue,gum-mucosis,specially of the retromolar-trigon and facilitates the evaluation of
other anatomical structures near the clinically evident tumor more.The method of the open mouth
does not change the lymphonodal staging,that are effectuated with the standard
technique.Conclusions:This work has shown how the PET/CT-scan of the head/neck region with
open-mouth, in this particular type of patient,had the advantage to distinguish with better
accuracy the complicated anatomy of the different organs,intimatly connected inside the oral
cavity and therefore gives potence to the spatial resolution,bettering the anatomical information
achieved by the PET/CT as aposed to the PET scan only.The mid morning execution of the
exam,the relaxing of muscles before the administration and the seated wait permits,in addition,a
reduction of the equivocal physiologic uptakes that are frequent in this area.
Radiotherapy planning with [18F]-DG PET-CT: technical
aspects and preliminary results
J. I. S. Patrina1, D. A. B. Faria1, A. Sevilla1, L. C. Ribeiro1, D. M. Sousa1, C.
A. Nunes1, P. M. R. R. Genésio2, J. V. F. Silva2, J. M. P. Oliveira1, D. C.
HPP Medicina Molecular, Porto, Portugal, 2Clínica de
Radioterapia do Porto, Porto, Portugal.
Nowadays patients undergoing radiation therapy are mainly selected on the basis of tumour
location and volume. Optimization of external beam radiation classicaly relies on CT treatment
planning studies to delineate tumour stuctural volume to be irradiated. In many cases critical
tumour volume obtained with CT scan fails to correctly reveal the most irradiation sensitive areas
of the tumour. In these cases, PET has already demonstrated a major role because it reveals
metabolic activity and is more sensitive to provide adequate definition of the target volumes to be
irradiated. Furthermore this technique provides adequate tools for possible reduction in
irradiation of normal tissues. At the same time it may increase the dose in the tumour. The aim of
this work is to describe the technical aspects of “metabolic radiotherapy planning” with [18F]-DG
PET-CT and the role of nuclear medicine technologists as part of the planning team. The need for
specific hard (fixed external laser beams, flat bed, head holder with radiotherapy mask) and
software tools is emphazised and the PET-CT image acquisition protocol described in detail. We
have, so far, included 10 patients (4 female and 6 male, age range from 30 to 58 years) one with
Malignant Thymoma, 2 Head and Neck tumours, 3 NSCLC and 4 Breast Carcinoma in our
combined molecular imaging/radiotherapy collaborative work. All patients were scanned with a
GE Discovery LS/4 PET-CT Scanner starting 60 to 90 minutes after the administration of 5
MBq/Kg of weight. After patient positioning with CT body external marks, a reference CT
acquisition was always performed using 140 kV, 80 mA and 1.5 pitch. This was followed by a
2D [18F]-DG PET acquisition using our wholebody protocol, in summary 5 to 6 AFOV times 5
minutes per bed position. A further 1 to 2 AFOV late (> 2 hours p.i.) acquisition was undertaken
to better delineate tumoural metabolic volume. Our preliminary experience demonstrates that: 1)
collaborative multidisciplinary teamwork is feasible within our setting and helps to overcome
poorly collaborative patients;2) since we frequently perform additional late segment studies the
radiotherapy planning protocol does not impact significantly on our daily schedule;3) late (>2
hours p.i.) studies have been helpful to improve metabolic tumour contrast and volume.
Inter-observer evaluation of attenuation correction with 137Cs versus Low Dose CT in PET scans where patients have
the arms positioned along the body.
S. Holmboe, C. W. Skøtt, L. S. Jeppson;
Herlev Hospital, Herlev,
Aim: To evaluate inter-observer variation of image quality comparing 137-Cs transmission scan
with LowDoseCT (ldCT) for attenuation correction when the patients have their arms alongside
the body. Background: The study was undertaken on the suspicion that upper extremity bone
might cause problems for the ldCT reconstructed PET images. Materials and methods: Patients,
who were unable to place their arms above the head because of physical deficits or unbearable
pain arising in that position, were included. 29 patients participated, 15 males, 14 females, with a
mean age of 61 years, mean weight 76 kg and a measured blood glucose level below 8 mmol/l.
All patients were examined with both 137-Cs transmission and ldCT for attenuation correction in
the same study. A PET-scan with ldCT takes approximately 20 minutes and an average PETscanning with 137-Cs transmission included lasts another 10 minutes. After 5 hours of fast, an
intravenous injection of 370 MBq 18F-FDG was administered to all patients followed by a 60
minute rest. The PET-scan was preceded by a surview and a ldCT-scan (duration: 25 sec.) using
40 mAs, slice thickness 6.5mm, 5 mm increment. Finally, the PET-scan was performed with 2
min. per bed position and 38 seconds per 137-Cs rotation. The 137-Cs rod source delivers a fan
beam used for attenuation correction. The PET-scan was reconstructed with both ldCT and 137Cs transmission scans. The mean radiation to the patient was 0.1 mSv with 137-Cs transmission
versus 3 mSv when performing ldCT. Two experienced nuclear medicine specialists reviewed
the attenuation corrected PET images independently. They were asked to decide which images
they preferred to use for diagnosis or whether they found the two sets of images to be of equal
quality. Results: The inter-observer study is listed below: Specialist A; ldCT: 11, Equal quality:
13, 137-Cs: 5 Specialist B; ldCT: 15, Equal quality: 6 , 137-Cs: 8 Concordance between the
specialists; ldCT: 11, Equal quality: 6, 137-Cs:0 There was significant concordance between the
observers: Kappa = 0.578 and p<0.001. In most cases when 137-Cs was preferred, it was because
of artefacts from the upper extremities on the ldCT images. Conclusions: PET-scans with ldCT
attenuation correction only caused artefacts in 17% of the patients. PET-scans reconstructed with
ldCT provided better images in 71 % of the patients than with 137-Cs attenuation correction; this
justifies the higher radiation dose. An additional benefit was a higher patient throughput and the
anatomic mapping.
305 — Sunday, October 01, 2006, 11:30 am - 1:00 pm, Scalcotas
Oncology: Therapy Planning & Prediction
Calculation of lesion volume in pulmonary lesions by PET/CT
imaging: a histological comparation between threshold and
mathematical method.
G. Rossi1, S. Fattori1, E. Di Nicola1, A. Poggiu1, S. Ancidei2, S. Barucca2, A.
Berbellini2, F. Capoccetti2, C. Cidda2, F. Ferretti2, S. Sivolella2, E.
Brianzoni2; 1Medical Physic Macerata Hospital, Macerata, Italy, 2Nuclear
Medicine Macerata Hospital, Macerata, Italy.
AIM PET/CT imaging allows to calculate lesion volume: this calculation lead to dosimetric
evaluations and treatment planning on voluming images. The result is strictly correlated with the
technique used to define it, if threshold or mathematical method.Aim of our study is to confirm
or refute imaging threshold, evaluated in our previous work, that better correlates with the true
volume measured after the surgical excision and to find a mathematical methods to define it.
MATERIALS AND METHODS We involved 17 patients that were submitted to surgery after
PET/TC study. We used a standard PET acquisition protocol (scout image, spiral CT, 3D PET
emissive acquisition of 3 min/beds, 5,18 MBq/Kg 18F-FDG, Siemens LSO Biograph). We
reconstructed PET images through software Siemens Syngo, than we selected the region of
interest (ROI), planned Spectrum 10 steps chromatic table to differentiate the imaging threshold
(20%, 30%, 40%, 50%, 60% and 70%) and to calculate volume from ROIs areas multiplied for
the slice thickness (3.4 mm). After we used a mathematical approach to define the imaging
lesions volume applying a Laplacian filter and a program that gives lesion borders as results.
Finally we compared such values with histological volume calculated with classical formula
(sphere or ellipse). RESULTS 13/17 pts showed geometrically regular aspect of the lesions: the
20% threshold is correct for 1/13 case, the 30% threshold for 4/13 cases, the 40% threshold for
3/13, the 50% for 2/13,the 60% for 1/13 and the 70% for 2/13. In 4/17, with not uniform lesions
and infiltrating neighboring anatomical regions, the data point out that the threshold is different
in each case (for 1/4 is 30%, for 1/4 is 40%, for 1/4 is 50% and for 1/4 is 60%) according to
presence or less of necrosis and atelectasia. With the mathematical approach we found good
correlation with anatomical data in 11/17 pts. For the other 6/17 we underestimated volume for
necrosis. CONCLUSIONS We found that 30% threshold is the best choice in cases of classical,
spherical or ellipsoidal morphologies., but really we cannot define a priori a best threshold. With
mathematical method we found a very good correlation in every situation except in case of
necrosis, for which we have to correct histological volume by a percentage we define from
pathological data.
Correlation of [ F]FLT and [ F]FMISO autoradiography with
immunohistochemistry in a series of human squamous cell
carcinoma xenografts
Proliferation and tumor cell hypoxia are tumor characteristics affecting radiosensitivity and
treatment outcome in squamous cell head and neck cancer. Both characteristics may be
determined using PET, with the proliferation marker 3’-[F-18]fluoro-3’-deoxythymidine (FLT)
and hypoxia tracer [F-18]fluoromisonidazole (FMISO). The methods were validated and
compared quantitatively at a microscopic level comparing F-18-autoradiography with
immunohistochemistry of the same sections using image fusion software. For
immunohistochemistry, pimonidazole (PIMO) and bromodeoxyuridine, were used as markers for
hypoxia and proliferation, respectively. The experiments were performed in a series of recently
established xenografts of human head and neck squamous cell carcinomas with a range of
hypoxic and proliferative fractions. Methods. Eight of these xenografted tumor models were
analyzed for proliferation. Mice were injected iv with FLT followed by injection of
bromodeoxyuridine (BrdU). Tumor cell hypoxia was evaluated in 13 xenograft models. Mice
were injected iv with FMISO followed by injection of PIMO. In both groups, the perfusion
marker Hoechst33342 was administered 1 hr later and mice were killed 1 min later. Upon
dissection, tumors were rapidly frozen, sections were cut and analyzed using a storage phosphor
imager (PI) for autoradiography. Subsequently, immunohistochemical (IHC) staining was
performed on the same sections. The PI and IHC images were manually co-registered. Images
were analyzed with in-house developed software. The resampled and co-registered images were
used to create pixel-wise scatterplots. Results. For both FMISO and FLT, correlation between PI
and IHC ranged from good to moderate between the various tumors. Correlation of FMISO with
PIMO was better in tumors with higher grades of hypoxia as compared to those with a lower
hypoxic fraction. Correlation of FLT with BrdU was weak, most likely due to different trapping
mechanisms. Whereas FLT is a measure of thymidine kinase I activity in the cytoplasm, BrdU is
a measure of DNA proliferation in the nucleus. Conclusion. Good to moderate correlations were
obtained when comparing autoradiography and immunohistochemistry imaging modalities,
especially for the hypoxic cell markers. The various histological patterns known for squamous
cell carcinomas were well represented in the PI and IHC images. These data suggest that FMISO
distribution reflected the hypoxia in the tumors, whereas FLT is a reliable tracer for the noninvasive determination of thymidine kinase I activity in the tumor cells.
Contribution of image fusion between FDG PET, MRI and CT
to the definition of target volumes in the radiation therapy of
high grade gliomas
W. Pilloy1, Y. Lassen-Ramshad2, R. Ruchaud2; 1Centre Hospitalier de
Luxembourg, Luxemburg, Luxembourg, 2Centre Francois Baclesse, Esch s/
Alzette, Luxembourg.
Aim : To compare target volumes as defined by dosimetry CT scan alone vs volumes defined by
MRI and FDG PET , in the treatment planning of gliomas Material & Methods : 31 patients were
studied - 19 glioblastomas, 9 astrocytomas, 3 oligodendrocytomas - , of whom 8 only had
macroscopic total resection. Gross Tumour Volumes (GTV) were defined on dosimetry CT ; on
MRI T1c and T2 pre- and post-surgery ; on FDG PET post-surgery. The GTV used for the
treatment is defined as the sum of individual [MRI+PET+CT] GTVs. FDG PET (Philips Gemini)
images were co-registered with MRI images (GE 1.5 Tessla) on a Hermes NUD station. With the
T1c image as reference, areas of grey matter where the uptake was higher than in the
surroundings (residual tumour) or areas in the surgical cavity where uptake was higher than in
white matter , were considered positive for tumour and identified by ROIs. The two images set
and ROIs were co-registred on dosimetry CT images (Philips Pinnacle-Syntegra). NB : CTV
(Clinical Target Volume) is defined as GTV expanded 10mm in every direction. Results : All
tumours could be identified on FDG PET. GTVs defined by CT, MRI, and PET present with
different sizes (respectively 42.3 , 17.5 and 14.3 ml average) and shapes and seldom completely
overlap.As an average, and despite being smaller than CT GTV, more than 50% of MRI GTV ,
and more than 30% of PET GTV are not included in CT GTV. More than 40% of PET GTV is
not included in MRI GTV. As an average, [MRI + PET] CTV is more than 1.5 times smaller than
CT CTV, and in 81% of cases 1.9 times smaller . Conclusion : MRI GTV and PET GTV are
smaller than CT GTV and theoretically identify the most aggressive parts of tumours. Their sum
[MRI + PET] GTV is still smaller than CT GTV , takes into account their limited overlap , and
allows a reduction of the CTV by a factor close to 2 , and a similar decrease in the toxicity
associated to the treatment . It remains to be seen whether this has a consequence on the clinical
Impact of 18f-deoxyglucose positron emission tomography
(FDG-PET) image fusion on radiotherapy target delineation
and treatment outcome in advanced oesophageal carcinoma.
C. Bodet-Milin1, V. Marchand2, A. Oudoux1, B. Bridji3, I. Resche4, F.
Kraeber-Bodere1, M. Mahe5, C. Rousseau4; 1Nuclear Medicine, University
Center, Nantes, France, 2Radiotherapy, Cancer Center, Saint Herblain,
Nantes, France, 3Nuclear Medicine, Cancer Center, Saint Herblain, Nantes,
France, 4Nuclear Medicine, Cancer Center, Saint Herblain-Nantes, France,
Radiotherapy, Cancer Center, Saint Herblain-Nantes, France.
Aim: To evaluate from a retrospective study the impact of fused FDG-PET and dosimetric
computed tomography (DCT) on conformal radiotherapy target delineation in patients with
oesophageal carcinoma and compare toxic effects and survival with previous reference
study1.Materials and Methods: Fifteen patients (14M, 1F), median age 56, with advanced
oesophageal carcinoma, 14 squamous cell and 1 sarcomatoid, stage 2 (n=1), 3 (n=11) or 4 (n=3)
were treated by concomitant radiochemotherapy between 2003 and 2006. Dose of radiotherapy
was 50 (n=9) or 60 Gy (n=6) in 25 or 30 fractions over 5 or 6 weeks, plus cisplatin and
fluorouracil. Each patient underwent DCT and hybrid FDG-PET/CT for simulation treatment in
the same treatment position. FDG-PET were scanned after being injected with 5-7 MBq/kg of
FDG. Target delineation was initially performed on DCT by 2 different investigators and the
FDG-PET data were used as an overlay to DCT data to define target volume.Results: FDG-PET
detected previously unknown distant metastatic disease in 2 patients (1 pleural and 1 pulmonary).
The GTV was decreased by DCT and FDG-PET image fusion in 7 patients (47%) and increased
in 8 patients (53%). Median GTV reduction was 31% [2-61] and median increase 23% [12-56].
Severe acute toxic effects were described in 7 patients (47%) with no significant differences as
described in previous studie1 (44% of severe side effects). Chemotherapy could be administered
as planned in 13/15 patients (86%). One patient died during treatment. Six patients relapsed
(40%), 2 (33%) with local and 4 (77%) with distant recurrences (lung, bone, or liver). The
median time to progression was 9 months [3-12]. Six patients (40%) were in complete remission
at a median time of 11 months [6-24] after diagnosis. Three patients were still in treatment at the
time of the analysis. For 12 patients still alive, median overall survival was 14 months [524].Conclusions: FDG-PET can add information to precisely identify the GTV in patients with
oesophageal carcinoma so as to obtain a better local control of disease. More data and long-term
follow up will be necessary to confirm the interesting facts that no more side effects with better
survival were observed. 1Herskovic A, N Engl J Med, 1992;326: 1593-98.
A preliminary clinical study on the combination imaging of
f-fdg and c-pd153025 pet/ct
L. Yu, X. Liang; Center of PET/CT, Affiliated tumor Hospital of Harbin
Medical University, Harbin, China.
Purpose:Epidermal growth factor receptor (EGFR) is a sort of protein located on the cell surface
or in the sub cells. EGFR is overexpressed and/or mutated in several tumors. PD153035 is the
inhibitor of the epidermal growth factor receptor tyrosine kinase (EGFR-TK) and it can
competitively bind to EGFR-TK and thus we can display the image of EGFR via PD153035.
Nonsmall cell lung cancer is one of the most common pulmonary carcinomas and it is of great
value to promote the therapeutic efficacy of pulmonary carcinomas especially the targeted
therapy effect in clinic. Gefinitib, the inhibitor of the epidermal growth factor receptor tyrosine
kinase (EGFR-TK), has been approved in many countries to cure nonsmall cell lung cancer and
its therapeutic effect has more dependency with EGFR, so the result of 11C-PD153035 imaging
can be an effective director for the targeted therapy of Gefinitib in clinic. Based on the above
reasons, we carried out a clinical study of PET/CT imaging with11C-PD153035.Materials and
Methods: All the 18F- FDG and 11C-PD153035 images of the patients were acquired by the
Center of PET/CT. 11C-PD153035 was synthesized by the method reported by Peter Johnstrom1
but with modification. Discovery ST scanner was used in our study. First perform a whole body
scan in 2D mode at 3 min/table, with18F- FDG injection dose of 0.1 mCi/kg, 50 min after 18FFDG VI; in the 2nd or 3rd day, 11C-PD153035 body or regional scan was performed in 3D mode at
2 min/table, with dose of 0.2-0.3mCi/kg, 8-42 min after VI. Clinical case: 4 patients, 1 man, 3
women, age ranged from 26 to 62.Results: One case only have 11C-PD153025 scan for the clear
clinical diagnosis of lung cancer, after IRESSA treatment, uptake of 11C-PD153025 is low. The
other three cases are all have 18F-FDG and 11C-PD153025 PET/CT scan. All 18F-FDG PET/CT
image are positive. 11C-PD153025 have clear uptake in two cases which pathology are
adenocarcinoma, and in one TB case, 11C-PD153025 only slightly uptakeConclusions:11CPD153025 has value when select lung carcinoma targeted therapy patients and monitoring
therapeutic effect of targeted therapy. Combination of 11C-PD153025 and 18F-FDG is valuable
for diagnosis of chest carcinoma and identification diagnosis.
Imaging tumors of the sympathetic nervous system: whole
body PET/CT with C-11-meta-hydroxyephedrine (11C-HED)
compared with I-123-mIBG SPECT/CT
C. Franzius, K. Hermann, M. Weckesser, K. Kopka, K. U. Jürgens, J.
Vormoor, O. Schober; University Hospital Muenster, Muenster, Germany.
Purpose The C-11-labelled tracer meta-hydroxyephedrine (C-11-HED) is a noradrenalin
analogue that was developed to visualize the sympathetic nervous system with PET. First clinical
studies show a rapid enhancement of C-11-HED in localized tumors of the sympathetic nervous
system. The clinical application of C-11-HED PET as whole body imaging method is now
possible as high end PET scanners allow a rather short examination time. The aim of this study
was to evaluate the clinical implementation of whole body C-11-HED PET/CT for examination
of tumors of the sympathetic nervous system and to compare the results with the standard
diagnostic imaging of these tumors with I-123-mIBG scintigraphy including SPECT/CT. Method
In 19 patients aged from 9 months to 68 yrs. (median 30 yrs.) 24 whole-body C-11-HED PET/CT
examinations (Biograph 16, Siemens/CPS) were performed. Low dose CTs were acquired for
attenuation correction and anatomic correlation. Scans were compared with attenuation corrected
I-123-mIBG SPECT/CT scans (24 h scan, Hawkeye, GE, low-dose CT). Subsequently, in 14
patients with 19 pairs of examinations the following tumors were confirmed histologically: six
neuroblastomas (nine months to 14 yrs.), five pheochromocytomas (33 to 66 yrs.), one
ganglioneuroblastoma (23 yrs.), and two paragangliomas (57 and 59 yrs.). In 5 patients (14 to 68
yrs.) each having one pair of examinations, clinical follow-up and/or histological examination
did not reveal any tumor deriving from the sympathetic nervous system. These C-11-HED
PET/CT examinations were used to describe the normal C-11-HED distribution. Intensity of
pathological tracer accumulation was visually analyzed in both scans, PET and SPECT, using a
4-value scale. In the C-11-HED PET, SUVmax and SUVmean of all lesions were determined.
Results C-11-HED PET/CT detected 80 out of 81 tumor lesions; these were 61 soft tissue lesions
and 19 bone lesions (sensitivity 0.99). I-123-mIBG SPECT detected 75 out of 81 lesions (56 soft
tissue lesions, 19 bone lesions, sensitivity 0.93). The contrast of C-11-HED uptake was higher in
comparison to I-123-mIBG uptake in 32% of the lesions, equal in 48% of the lesions and lower
than I-123-mIBG uptake in 20% of the lesions. Conclusion Whole body imaging using C-11HED PET/CT is feasible in the clinical setting of patients with tumors of the sympathetic nervous
system, even in children. C-11-HED PET/CT detected more tumor lesions than I-123-mIBG
SPECT. However, uptake of C-11-HED in tumor lesions can be higher, equal or lower compared
to I-123-mIBG.
P. Laverman, E. G. C. Troost, M. Philippens, J. Lok, J. Bussink, J. H. A. M.
Kaanders, A. J. van der Kogel, W. J. G. Oyen, O. C. Boerman; Radboud
University Nijmegen Medical Centre, Nijmegen, The Netherlands.
Celecoxib: an effective P-glycoprotein inhibitor?
N. A. R. Vellinga, J. Doorduin, A. van Waarde, R. A. Dierckx, E. F. J. de
Vries; University Medical Center Groningen, Groningen, The Netherlands.
Aim: P-glycoprotein (Pgp) is a transmembrane protein, serving as an ATP-dependent efflux
pump. Pgp is widely expressed in e.g. the intestines, the blood-brain barrier, the blood-testis
barrier and various tumor types. Pgp protects tissues from harmful substances, but also eliminates
drugs. Consequently, these drugs do not reach intracellular concentrations high enough to be
effective. By inhibiting Pgp, the intracellular drug concentration can be increased. Several agents
have been studied as Pgp inhibitors, revealing cyclosporine A as one of the most potent
inhibitors. A major problem of applying cyclosporine A in a clinical setting, is its toxicity at
doses high enough to inhibit Pgp. In recent literature, there are some indications that the antiinflammatory drug celecoxib might be a Pgp inhibitor. In search of a less toxic inhibitor, we
studied celecoxib.Methods: Biodistribution and PET studies with the radiolabeled Pgp substrate
[11C]-verapamil were performed in rats to investigate the effect of Pgp inhibition. The rats were
divided in a control group, a group treated with cyclosporine A (50 mg/kg) and a group treated
with celecoxib (50 mg/kg) as Pgp inhibitor, 30 minutes before tracer injection. Sixty minutes
after tracer injection, the rats were terminated. The brain was isolated and dissected into specific
regions. Samples of several other organs were also taken. The collected tissues were weighed and
analyzed for radioactivity. The results were analyzed with a two-sided Student’s t-test.
Differences were considered significant when P<0.05.Results: Celecoxib increased brain [11C]verapamil uptake 2.1 times (P<0.05) and cyclosporine A increased brain uptake 14.6 times
(P<0.05), as compared to the control group. The PET-scans of controls and celecoxib treated rats
did not show any tracer uptake in the brain (cold spot). In contrast, the brain was clearly
delineated in the PET images after cyclosporine A treatment. Outside the brain, cyclosporine A
could only inhibit Pgp in the testes. Celecoxib did significantly increase [11C]-verapamil-uptake
in the heart, kidney, liver, spleen, submandibular gland, testes, plasma and erythrocytes, as
compared to the control group.Conclusions: Inhibition of Pgp with cyclosporine A can be
visualized with [11C]-verapamil-PET. Celecoxib did increase [11C]-verapamil-uptake in the brain
to some extent, but not enough to be visualized by PET. Celecoxib was able to increase [11C]verapamil uptake in several other organs. This might either be a Pgp-mediated effect or the result
of a celecoxib-mediated effect on tracer clearance. Further research on Pgp inhibition by
celecoxib is required.
Receptor expression in patients with gastrointestinal stromal
tumors (GIST) using a Ga-68-labeled Bombesin analogue and
comparison to F-18-FDG.
A. Dimitrakopoulou-Strauss1, P. Hohenberger2, M. Eisenhut3, H.
Maecke4, U. Haberkorn1, L. G. Strauss1; 1Clinical Cooperation Unit Nuclear
Medicine, German Cancer Research Center, Heidelberg, Germany,
Surgical University Clinic, Klinikum Mannheim, Univ. of Heidelberg,
Mannheim, Germany, 3Department of Radiopharmaceutical Chemistry,
German Cancer Research Center, Heidelberg, Germany, 4Division of
Radiological Chemistry, University of Basel, Heidelberg, Switzerland.
Aim: The purpose of this study was to evaluate the expression of the receptors using Ga-68Bombesin in patients with GIST. We examined the pharmacokinetics of 68Ga-Bombesin
quantitatively and compared the data to those of dynamic FDG studies.Materials and Methods:
Ga-68-Bombesin is a peptide that binds to at least three reseptor types: the GRP (Gastrin
releasing peptide) receptor, the Neuromedin B subtype receptor and the Bombesin receptor
subtype 3 (BSR-3). Experimental data give evidence for a Bombesin/GRP overexpression in
prostate carcinomas, gastrointestinal tumors, breast tumors, bronchial carcinomas and
neuroendocrine tumors. The ongoing study comprises 18 patients (pts) with GIST. All patients
underwent morphological imaging by CT and MRI prior to PET. Dynamic PET scans using Ga68-Bombesin (16 pts) and FDG (18 pts) were obtained on two different days within one week. A
qualitative and quantitative analysis was performed using compartment and non-compartment
analysis based on the fractal dimension (box counting procedure). Multivariate analysis was used
for the evaluation of the kinetic data.Results: Thirteen of 18 pts were positive in FDG imaging,
whereas Ga-68-Bombesin demonstrated an enhanced accumulation in 8 of 16 patients with
metastatic GIST. The highest FDG-uptake was measured in a patient with a recurrent GIST in
the stomach (16.9 SUV). The highest Ga-68-Bombesin uptake was noted in a liver metastasis
from a GIST (5.9 SUV). The mean SUV for Ga-68-Bombesin was 4.7 in comparison to 7.2 SUV
for FDG. Best subset analysis demonstrated that the global SUV (55-60 min p.i.) for FDG was
primarily dependent on k1, followed by k3 and the fractional blood volume VB. In contrast, the
global SUV for Bombesin was primarily dependent of VB, followed by k3 and k1. Multivariate
analysis did not show a significant correlation between the kinetic parameters (k1-k4, VB, SUV)
for FDG and Bombesin.Conclusions: Ga-68-Bombesin and FDG-PET provide different
information regarding the biological properties in GIST. Ga-68-Bombesin may be helpful to gain
additional specific information in patients with GIST.
306 — Sunday, October 01, 2006, 11:30 am - 1:00 pm, Allegro
Neurology/Psychiatry: Parkinson - Miscellaneous
The role of [123I]-FP-CIT SPECT in the diagnosis and clinical
management of Parkinson’s Disease.
L. M. Feggi1, S. Panareo1, C. Sensi2, S. Fabbri3, M. Casali1, E. Fainardi4, C.
Cittanti1, M. Giganti1, N. Prandini1, E. Bagatin1, A. Costanzo1, S.
Mangherini1, F. Preda2, R. Quatrale2; 1Nuclear Medicine Unit, UniversityHospital "S. Anna" Ferrara, Italy, 2Neurology, University-Hospital "S. Anna"
Ferrara, Italy, 3Health Physics, University-Hospital "S. Anna" Ferrara, Italy,
Neuroradiology, University-Hospital "S. Anna" Ferrara, Italy.
Aim: The diagnosis of Parkinson’s Disease (PD) is based on clinical features with imaging
confirmation. Especially early in the course of the disease, the clinical diagnosis can be difficult.
Neuro imaging of presynaptic dopamine transporters has provided a possible diagnostic probe in
the evaluation of PD. In particular [123I]-FP-CIT SPECT (DAT scan) has been successfully used
to detect the loss of dopaminergic nigrostriatal neurons in PD. The purpose of this study was to
evaluate the impact of DAT scan on the diagnosis and clinical management of patients with a
suspicious of PD.Materials and Methods: 136 consecutive patients (69 male, 67 female, mean
age 69 yh) with suspected clinical PD were prospectively studied. All patients underwent to
neurological examination, Magnetic Resonance Imaging (MRI) for the identification of basal
ganglia alterations and DAT scan for the study of nigrostriatal neurons uptake. The radiotracer
uptake was classified in 4 severity degrees: normal (type 0), abnormal putamen omolateral (type
1), abnormal putamen bilateral (type 2) and abnormal putamen and caudate bilateral (type 3). We
verified the correlation between this classification with that based on neurological findings (HY
scale). We also tested the efficacy of DAT scan for the differentiation between PD and other type
of tremors comparing diagnosis before and after DAT scan, to verify the impact of DAT scan in
the management of PD. Finally we compared MRI and DAT scan results.Results: A significant
linear correlation was founded between the classification of patient based on DAT scan degrees
and on HY scale (p < 0.1%). On the other hand the Ȥ2 test demonstrated a significant difference
between diagnosis of PD before and after DAT scan, in particular in doubt cases. Finally, from
the comparison of DAT scan and MRI with Ȥ2 test application, we founded a significant
difference: a 61% positive DAT scan versus 26% positive MRI.Conclusions: Our study shows
the direct relation between DAT scan and neurological findings, in particular when the diagnosis
of PD is clinically difficult to do. On the other hand the difference between positive results on
DAT scan and MRI is related to the absence of morphological alterations on nigrostriatal neurons
detectable on typical PD. The results of DAT scan have changed the diagnosis in the 20% of all
our patients and in the 92% of doubt cases, conditioning the clinical management of patient.
Quantifying DATSCAN images - a comparison of region-ofinterest and fractal analysis
L. Bolt, J. S. Fleming, P. M. Kemp; Southampton University Hospitals
NHS Trust, Southampton, United Kingdom.
Aim: To assess the capability of two different semiquantitative techniques for interpreting
DaTSCANs. Two methods - the region-of-interest (ROI) and the fractal analysis - are assessed
for diagnostic concordance. Materials&Methods: A retrospective study was carried out on 55
scans randomly chosen from the database of patients referred to our Department for DaTSCAN
imaging with symptoms suggestive of Parkinson disease (PD), Parkinsonian syndromes (PS) or
essential tremor (ET). They were classified as 25 “normals” (ET) and 30 “abnormals” (PD and
PS) by the clinical diagnosis reached with the scan support, giving a disease prevalence of 55%
which reflects our clinical practice. The ROI analysis was performed on a two-dimensional (2D)
striatum image, automatically defined by summing all frames with striatal uptake within a 44mm
slab; the fractal analysis was carried out in 2D and 3D. Both analyses were implemented on a
Link Medical workstation using MAPS-10000 software. The ROI analysis measures the specific
binding ratio (SBR), the ratio of specific to non-specific count concentration. A semi-automated
method was used, which limits operator intervention to the striatal ROIs placement. These are
provided as templates of geometrical shape to minimize operator variability, while ensuring
inclusion of partial-volume counts. The specific count concentration is derived from the total
counts within the striatal ROIs. The non-specific count concentration is directly measured in an
automatically defined reference ROI that includes the whole non-striatal cortex. The fractal
analysis measures the fractal dimension (FD) of the striatal uptake, an index related to its
heterogeneity. The method used measures the number of voxels N(k) with counts above an
intensity threshold k. Measurement are recorded for values of k within an appropriate range
where the fractal relationship N(k)=constant*k (-FD) is satisfied. This relationship becomes linear
on a log-log plot, the slope of the regression line gives FD. The threshold k was normalised to the
mean non-specific uptake. The analysis is fully automated.Results: The ROI analysis was able to
separate normal and abnormal groups with a diagnostic concordance of 95% (3 misclassified).
The fractal dimension reached a concordance of 98% in 3D (1 misclassified) and 100% in 2D.
The improvement of the fractal analysis, however, did not reach statistical
significance.Conclusions: Both ROI and fractal analyses provide simple and robust indices that
reach a high diagnostic concordance. The fractal dimension, however, performs marginally better
and has a 100% repeatability. They offer a useful adjunct to the visual interpretation of
In vitro study of
Tc-Annexin V uptake in cellular model of
Parkinson's disease
W. Cao1, J. Y. Wu2, J. S. Huang1, Y. He1, X. L. Lan1, G. X. Cao1, Z. R.
Gao1, R. An1, S. G. Sun1, Y. X. Zhang1; 1Union Hospital, Tongji Medical
College of Huazhong University of Science and Technology, Wuhan,
China, 2Wuhan 1st Hospital, Wuhan, China.
Aim: To evaluate the feasibility of Technetium labeled Annexin V imaging for early diagnosis
of Parkinson’s disease, we prepared 99mTc-Hynic-Annexin V and studied its binding
characteristic to apoptotic cell models of Parkinson’s disease in vitro.Materials and Methods:
Apoptotic cellular models of Parkinson’ s disease were replicated by different concentrations of
1-methyl-4-phenylpyridinium induction in PC12 (rat pheochromocytoma cells) for 48hrs and
SH-SY5Y (human neuroblastoma) cell lines for 72 hrs. The cell apoptosis rates were analyzed by
flow cytometry with fluorescein isothiocyanate- conjugated annexin V (FITC-Annexin V) and
propidium iodide (PI) staining. Annexin V was labeled with 99mTc using the bifunctional
chelating agent hydrazinonicotinamide (HYNIC). The product was purified by Sephadex G-25
column chromatography and analyzed by instant thin layer chromatography (ITLC). The
radiochemical purity was >90%. Cell binding studies (i.e. kinetic, saturation and displacement
competition binding) were carried out with cellular models of Parkinson's disease. Cell uptake
studies were also performed after different levels of MPP+ induction, and the correlation of the
degree of apoptosis (as detected by flow cytometry) and cell uptake of 99mTc-HYNIC-Annexin V
Recover of regional brain activity concentration of FP-CIT
directly from SPECT projections
M. T. De Cristofaro, E. Vanzi, B. Sotgia, S. Ramat, S. Sorbi, A. Pupi, A. R.
Formiconi; University of Florence, Florence, Italy.
The clinical potential of FP-CIT scan in Parkinson disease (PD) is hampered by the low contrast
recovery of conventional SPECT reconstruction algorithms [1], which critically compress
activity concentration measurements, and make FP-CIT SPECT ineffective to monitor PD
progression. In theory, at equilibrium, brain distribution of FP-CIT can be segmented in 5
homogeneous-concentration complementary volumes (HVOI), which are the two striata (ST),
cerebellum (CB), the remaining brain and the background (skull). Thus, the dimension of the
tomographic problem expressed in terms of HVOIs is such (5x5) as to allow for its direct
solution (DS), obtaining regional HVOI activity concentration values (ACV) directly from
projections [2]. This solution is not affected by the ill-conditioning of standard tomographic
reconstruction techniques and offers an optimal contrast recovery capability. SPECT image
conventional reconstruction is required for delineating the HVOIs.Methods: seven Parkinsonian
patients were examined with a conventional FP-CIT SPECT with 120 projections, 60 sec each,
acquired 4h post-injection with a 3 head camera. FP-CIT injected dose was 185 MBq. SPECT
images were reconstructed with the EM algorithm inclusive of collimator spatial response
correction and Chang attenuation correction. The 5 above HVOIs were drawn on these
reconstructed images and their ACV were calculated both with DS and by superimposing the
HVOIs on SPECT images. The DS accuracy was previously assessed with an Alderson phantom
study (presented elsewhere). Binding Potential (BP) was calculated as (ST-CB)/CB from both the
EM images (EM-BP) and the DS (DS-BP).Results: DS-BP has a wider activity range and higher
activity values of EM-BP (Table 1).Conclusions: the use of DS increases the range of BP values
measured with FP-CIT in PD, giving to FP-CIT SPECT the capability of monitoring disease
progression. [1] M. Soret et al. J. Nucl. Med. 44: 1184-93, 2003 [2] E. Vanzi et al. IEEE Trans.
Med. Imag. 23: 363-73, 2004
I. Roca Bielsa, G. Cuberas, M. Velasco, T. Herrero, S. Aguade, F. Porta,
A. Mestre, M. Negre, J. Castell; Hu Vall Hebron, Barcelona, Spain.
Background: Brain SPET detects functional abnormalities in specific language impairment (SLI).
To quantify the abnormal areas, Statistical Parametric Mapping (SPM) is recognized to be the
reference technique, although it requires some resources (technical and human) not available in
many sites. The Neurogam® software allows the individual comparison with a normal database.
Objectives: 1. To describe the brain SPET disturbances in patients with SLI. 2. To compare the
results of SPM and Neurogam in this group of patients.Methods: We have prospectively studied
53 patients with SLI ( 49 males, 4 females, average 5.25 years) in baseline conditions (99mTcHMPAO, reference adult dose 740 MBq, dosage according to EANM Paediatric Committee
schedule, double headed gammacamera, parallel hole collimator LE-HR, 30 sec/image/3º).
According to clinical data, the children have been classified in 3 groups: N=19 mild language
impairment (speech/language delay, normal evolution at 7 years of age, SLD); N= 22
phonologic-syntactic dysphasia (PSD); N= 12 semantic pragmatic syndrom (SPS).Results:
Children with SLD showed a good cortical and subcortical uptake. Brain SPET pattern showed
several hypoactive areas in group 2 (PSD), in both hemispheres and in frontal, temporal and
parietal lobes. Most of the children with SPS had localized, congruent abnormalities, mainly in
both temporal lobes. Using SPM to compare groups 1 and 2, and 1 and 3, we found the following
significant differences (p<0,005):
4 - 6 - 45 - 46 - 48
11 - 20 - 28 - 38 - 48
Neurogam: defining SLD as a reference group, we found significant differences (hypoactivity <2SD in area greater than 50 % pixels) in several areas of frontal and temporal lobes in groups 2
and 3. Comparison SPM-Neurogam: concordant results were found in the fronto-temporoparietal lobes and several Brodmann areas: - PSD: several small areas, with low signifficance in
frontal, temporal and parietal lobes. - SPS: few areas in frontal and temporal lobes, with higher
signifficance. Discussion: Group 1 (children with SLD) can be considered as a reference group.
Comparison of groups 2 and 3 with group 1 showed diffuse and multiple abnormalities in the
PSD, but less involved areas, with higher accuracy and more congruent inside the group in the
SPS. The abnormalities detected in this group of study with primary dysphasia agree with the
published data (parieto-temporal, frontal areas). SPM detects the disturbances with greater
homogeneity in the group. Neurogam detects, on one side, the disturbances present in all patients
and, on the other side, the discordant patients.
Mean (Left and Right) BP estimates for the seven patients with the two methods
Pat. #
Metabolic imaging of pharmacologically-induced tinnitus in
Striatal volume (L+R) (ml) 13.4 12.2 11.0 11.0 15.0 16.4 11.3
1.4 2.3 3.2 4.5 3.7 3.3 5.3
3.7 4.7 6.5 7.5 8.9 9.3 12.0
A. K. Paul, E. Lobarinas, R. R. Salvi, J. C. Luisi, H. A. Nabi; University at
Buffalo, State University of New York, Buffalo, NY, United States.
What is the optimal acquisition interval for [ O]-H2O PET
J. Georgiadis, R. Kortekaas; University Medical Center Groningen,
Groningen, The Netherlands.
Aim It is generally believed that the optimal temporal resolution of [15O]-H2O PET is in the order
of one to two minutes. We aimed to establish the optimal acquisition interval by investigating the
effect of interval timing and length on statistical power. Material and Methods Eight female
volunteers each received eight PET-scans. The radiotracer [15O]-H2O (500 MBq per scan) was
used as an index of regional cerebral blood flow (rCBF). PET-scans were acquired in a dynamic
acquisition mode (7x10s- and 1x50s-frame) and data intervals varying in timing and length were
analyzed using Statistical Parametric Mapping (SPM). For each data interval, a motor task
involving the lower limbs was compared with a non-motor control task. The number of activated
voxels in the dorsal motor cortex and anterior cerebellum was used as a measure for statistical
power. Results Blood flow increases in the anterior cerebellum and the dorsal motor cortex could
already be detected with ten seconds of [15O]-H2O PET-data. Statistical power for the motor
cortex was optimal when only 20 seconds of data were used (interval: 20-40s after tracer
injection). In sharp contrast, statistical power for the cerebellum was optimal when all frames
(interval: 0-120 s after tracer injection) were summed. This regional difference in sensitivity was
condition-independent, as indicated by analysis of the time-activity curves for these
regions.Conclusions: With this new approach we are able to detect rCBF differences on a time
scale well below the presumed optimal time resolution of traditional integrated [15O]-H2O PET.
We also discovered that sensitivity is determined by acquisition interval and location of
activation in interaction, which should be taken into account in future studies with [15O]-H2O
PET in order to optimize statistical power. We hypothesize that this phenomenon is caused by
regional differences in cerebral vascular density.
Brain spet quantification in primary specific language
Aims: The phantom sound of tinnitus is an aversive auditory percept of unknown origin. The
purpose of this study is to investigate the metabolic activities in central auditory structures in vivo
during salicylate-induced tinnitus in rats. Material & Methods. A behavioral paradigm,
schedule induced polydipsia-avoidance conditioning (SIPAC), was used to determine if tinnitus
was present in rats treated with a high dose of salicylate. A dedicated, high resolution animal
positron emission tomography system (microPET Focus 120) was used to image the changes in
brain metabolic activities associated with high-dose salicylate in comparison to the baseline.
Baseline imaging was done first followed by salicaylte administration ((250 g/kg, i.p.) 48 h later.
In both the baseline and salicylate conditions, rats were placed in a quiet environment for 60 min
after the injection of F-18 labeled fluorodeoxyglucose (FDG, ~74 MBq, i.p ). Thereafter,
microPET scans of the rat brain were performed for 60 min in prone position under isofluorane
anesthesia. The radioactivity in central auditory structures, including inferior colliculi (IC) &
temporal (auditory) cortex , was determined and expressed as a fraction of injected FDG dose per
unit mass (%ID/g) based on a calibration factor obtained from imaging a phantom of known
activity.Results: Behavioral assessment with SIPAC showed that salicylate doses from 150-300
mg/kg reliably induced tinnitus in rats. Increase in FDG uptake by the inferior colliculi and
auditory cortices were evident visually. The preliminary quantitative analysis in 3 rats showed an
increase in % ID/g of FDG in the inferior colliculi during tinnitus in comparison to that at
baseline. This suggests increased neuronal activity in central auditory pathway during tinnitus
Conclusion. Our results show metabolic alterations in central auditory pathway during
pharmacologically induced tinnitus in rats. Further studies are ongoing to quantify these changes
and correlate the metabolic changes with pathophysiological, behavioral and neurological
correlates of tinnitus. Supported in part by grant from the American Tinnitus Foundation.
Brain activation during human male ejaculation revisited.
J. Georgiadis, S. Reinders, A. Paans, R. Kortekaas; University Medical
Center Groningen, Groningen, The Netherlands.
Aim In a prior PET-study by our group a broad range of brain regions were reported to be
involved in ejaculation (Holstege et al., 2003). However, recent methodological advances have
indicated that the analysis that was developed may be susceptible to artifact. Here we aim to
evaluate this analysis in order to refine knowledge about ejaculation-related brain structures.
Material and Methods In the original study, eleven subjects each received eight PET-scans while
being sexually aroused and during ejaculation. The radiotracer [15O]-H2O (500 MBq per scan)
was used as an index of regional cerebral blood flow (rCBF). Ejaculation scans were acquired in
was analyzed.Results: 99mTc-HYNIC-Annexin V labeling efficiency and specific activity were
64.56±6.23% and 1.54×105 KBq/mg. Radiochemical purity was 93.6±2.48% and remained above
90% after 4 hours storage at room temperature. The in vitro binding study suggested that the
binding of 99mTc-HYNIC-Annexin V to cellular models of Parkinson's disease was specific,
saturable, and time dependent. Scatchard analysis gave a Kd of 7.2±1.8nmol/L, Bmax values of
179±33 fmol/106cells (PC12) and 220±26 fmol/106cells (SH-SY5Y). MPP+ at different
concentrations can induce cell apoptosis in a dose-dependent manner, the cell uptake tests
suggested that the membrane-bound radiolabeled Annexin V activity linearly correlated to total
fluorescence as observed by FITC-Annexin V flow cytometry (r=0.89, P<0.001).Conclusions:
Results indicate that 99mTc-HYNIC-Annexin V retains its receptor-binding activity and has a
high affinity to cellular models of Parkinson’ disease, and the uptake of radioactivity correlated
well with cell apoptosis rates, thus 99mTc-Annexin V is a potential radiopharmaceutical agent for
imaging of early apoptosis in Parkinson’s disease.
multiple 10s-frames, other scans in one 120s-frame. Results were based on the comparison of
unequal intervals of [15O]-H2O data: 20s of data around the moment of ejaculation versus 120s of
data during stimulation of the erect penis. To validate this analysis we compared unequal data
intervals (20s versus 120s) within the same experimental condition, using the original data set
and another data set. The number of subjects, scans, and experimental conditions were identical
to the original study. In addition, we reanalyzed ejaculation by comparing equal (120s) data
intervals. Results The validation analyses clearly demonstrate that comparing unequal intervals
of [15O]-H2O data induces a typical ‘activation’ pattern, irrespective of the experimental
condition, indicating that many of the brain regions reported in Holstege et al. (2003) were
methodological artifacts and not related to ejaculation. By comparing equal (120s) data intervals
we investigated which brain regions may be related to ejaculation. Ejaculation-related increased
rCBF, as compared to stimulation of the erect penis, was found in the left deep cerebellar nuclei,
the anterior vermis, the pons, and in the left ventrolateral thalamus. These brain regions were also
reported in Holstege et al. (2003). Ejaculation-related decreased rCBF was present throughout
the prefrontal cortex, which is an entirely new finding. Conclusion Revisiting the data analysis
used in Holstege et al. (2003) reveals that the activation pattern previously ascribed to ejaculation
must be divided into task- and non-task-related activations.
Se supplementation on the immunological status.Materials and Methods: Thirty-six consecutive
patients (aged 19-85) with verified HT were included in the present study. In addition to their
treatment, 18 patients (Group I) received 200µg sodiumselenite per day orally for the time span
of 3 months, whereas eighteen patients (Group II) received placebos. All patients had
measurement of thyroid hormones (free thyroxine, free trijodthyronine), thyrotropin (TSH),
autoantibodies (TgAb and TPOAb), Se levels, and intracellular cytokine detection in CD4+ and
CD8+ T cells of peripheral blood mononuclear cells (PBMC) by flow cytometry before and after
Se administration. Eighteen healthy volunteers (Group III) matched in sex and age consisted the
control group.Results: The TPO-autoantibody levels were significantly decreased after Se
administration (524 ± 452 vs. 505 ± 464 U/ml). On the other hand, we found no significant
differences in the CD4+ or CD8+ cytokine pattern (IFNȖ, IL2, IL4, IL5, IL6, IL10, IL13, TNFĮ,
TNFȕ) in HT patients (Group I) before and after Se administration and in HT patients (Group II)
before and after placebo administration. In addition no significant differences between Group I
and Group II before and after Se vs. placebo administration was found.Conclusions: Our data
demonstrate that the peripheral CD4+ and CD8+ T lymphocytes from HT patients show no
significant differences in their cytokine patterns after Se administration. According to our results,
Se administration significantly decreases TPO-autoantibody levels but does not change the
cytokine production patterns in our HT patient’s cohort.
307 — Sunday, October 01, 2006, 11:30 am - 1:00 pm, MC 3
Clinical Science: Endocrinology
Decrease of TPO-autoantibody levels and IFN-g production
by CD8 T cells in patients with Hashimoto's thyroiditis after LThyroxine administration
Improvement of parathyroid lesion visualization in primary
hyperparathyroidism with 99mTc-MIBI pinhole SPECT:
comparison with conventional SPECT and planar
C. Ansquer1, E. Mirallié1, A. Oudoux1, C. Bodet-Milin1, M. Defrise2, A.
Seret3, I. Daumy4, F. Aubron4, F. Kraeber-Bodéré1, T. Carlier1; 1University
hospital, Nantes, France, 2Physics Institute, Liège, Belgium, 3Brussels
University, Brussels, Belgium, 4Ultrasonography center, Nantes, France.
Objectives: The aim of this study was to determine prospectively the interest of 99mTc-MIBI
pinhole SPECT imaging (pSPECT) compared to conventional SPECT (cSPECT) and planar
scintigraphies, for the preoperative localization of parathyroid lesions in primary
hyperparathyroidism.Methods: Twenty seven patients (22 F, 5 M, median age: 55 yrs), all cured
after surgery, underwent neck ultrasound (US) and 3 planar images performed 20 min after
injection of 74 MBq of 99mTcO4- and 15 and 120 min after injection of 925 MBq of 99mTc-MIBI.
A 99mTc-MIBI tomography was performed 30 min and 60 min after injection with respectively a
parallel and a pinhole collimator. pSPECT reconstructions were performed with a dedicated OSEM algorithm. Scintigraphic images were visually analyzed. A diagnostic confidence score (CS)
was assigned to each procedure considering intensity and extra-thyroidal location of suspected
lesions. This CS was defined as follows: 0= negative, 1= doubtful, 2= moderately positive, 3=
positive. The results of these preoperative localization studies were compared to surgical,
pathological and biological follow-up findings.Results: Surgery revealed 30 parathyroid lesions,
5 in posterior and 2 in minor inferior ectopy. The median weight of glands was 0.5g (20 glands ”
0.5g). Sensitivity of US, planar imaging, cSPECT and pSPECT were respectively 53%, 77%,
90% and 93%. The 7 ectopic glands were well localized by cSPECT and pSPECT as well. Five
small glands (”0.5g) were only detected by tomography and one gland (0.3g) only by pSPECT.
Two small hyperplastic glands (0.1 and 0.2g) were not detected by scintigraphies and US. For the
30 pathologic glands, pSPECT showed clearly the highest global CS (71 versus 45 for planar
imaging and 49 for cSPECT). Compared to planar imaging and cSPECT, pSPECT increased CS
for 16 (53%) and 17 (57%) parathyroid lesions respectively, and contributed to markedly reduce
the number of uncertain results.
Number of glands
Variations of CS
pSPECT vs planar imaging pSPECT vs cSPECT
0 to • 2
1 to • 2
2 to 3
No change
3 to 2
Conclusions: In this preliminary study, combination of planar and pSPECT scintigraphies
appears to be at present the optimal preoperative imaging procedure in primary
Selenium administration significantly decreases TPOautoantibody levels but do not change the cytokine
production patterns in patients with Hashimoto's thyroiditis
M. Schuetz, S. Kontur, H. Duan, S. Kommata, K. Wahl, R. Schoen, K.
Kletter, R. Dudczak, M. Willheim, G. Karanikas; Medical University of
Vienna, Vienna, Austria.
Aim: Recently it has been demonstrated that after Selenium (Se) substitution in Hashimoto
Thyroiditis (HT) patients for 3 and 6 months, respectively, there was a significant decrease of
TPO-autoantibody levels. The aim of our study was to evaluate the effect of Se administration on
the lymphocyte cytokine production pattern in HT patients to gain an insight into the influence of
H. Duan, S. Kontur, M. Schuetz, K. Wahl, S. Schwarzmair, A. Antoni, S.
Kommata, K. Kletter, R. Dudczak, M. Willheim, G. Karanikas; Medical
University of Vienna, Vienna, Austria.
Aim: It is well known that T-cell derived cytokines are the key regulators in Hashimoto's
thyroiditis (HT). It has been shown, that after L-Thyroxin substitution in HT patients with
hypothyroidism, there is a significant fall of TPO-autoantibody levels and moreover, the thyroid
volume decreased. The aim of our study is to evaluate the effect of L-Thyroxin administration on
cytokine production pattern, TPO-autoantibody levels and thyroid volume in HT patients with
hypothyroidism.Materials and Methods: Eight consecutive patients (aged 18-70) with verified
HT and hypothyroidism were included in the present study. All patients underwent thyroid
volume determination by sonography and intracellular cytokine detection in CD4+ and CD8+ T
cells of peripheral blood mononuclear cells (PBMC) by flow cytometry before and six months
after thyroid hormone replacement therapy. Eight healthy volunteers matched in sex and age
consisted the control group.Results: The TPO-autoantibody levels were significantly decreased
after thyroid hormone replacement therapy (680 ± 380 vs. 411 ± 379 U/ml). On the other hand,
the thyroid volume showed no significant difference (12 ± 10 vs. 11 ± 9 ml). Most interestingly,
CD8+ T cells had significantly higher percentages of cells producing IFN-Ȗ before thyroid
hormone replacement therapy. We found no significant differences in the CD4+ or CD8+
cytokine pattern regarding IL2, IL4, IL5, IL6, IL10, IL13, TNFĮ, TNFȕ.Conclusions: According
to our data, we were able to demonstrate that thyroid hormone replacement therapy improves the
immunological status, expressed in a decrease of the TPO-autoantibody levels and IFN-Ȗ of the
CD8+ T cells in HT patients.
Thyrotropin concentration in untreated patients with
Hashimoto's thyroiditis correlates with the peak systolic
S. Gaberscek, K. Zaletel, N. Fister, E. Pirnat, S. Hojker; University Medical
Centre, Department for Nuclear Medicine, Ljubljana, Slovenia.
Aim. Data on the thyroid blood flow in homogeneous, untreated groups of patients with different
degrees of hypothyroidism are scarce. Therefore, we decided to evaluate the influence of
thyrotropin (TSH) concentration on the peak systolic velocity (PSV) in a group of untreated
patients with Hashimoto's thyroiditis. Patients and methods. We examined 236 newly diagnosed
patients with Hashimoto's thyroiditis, 223 females, 13 males, aged 16 to 84 (mean, 44.8 ± 14.7).
Concentration of TSH, thyroid peroxidase antibodies (TPOAb) and thyroglobulin antibodies
(TgAb) was measured. All patients were positive for TPOAb and/or TgAb. Colour flow Doppler
sonography was performed using a 7.5 MHz linear transducer. PSV was measured at the level of
intrathyroid arteries with a sampling volume of 2 mm. In each patient PSV was expressed as a
mean of five measurements. Results. Mean TSH concentration in all patients was 20.108 ± 41.06
mU/L, (min. 0.457, max. 295.352 mU/L). According to their TSH concentration, patients were
divided into 6 groups: first group with TSH between 0.457 and 2 mU/L (N = 30) had mean PSV
9.6 ± 2.5 cm/s, second group with TSH between 2.001 and 4 mU/L (N = 47) had mean PSV 10.3
± 2.3 cm/s, third group with TSH between 4.001 and 6 mU/L (N = 45) had mean PSV 11.4 ± 2.4
cm/s, fourth group with TSH between 6.001 and 10 mU/L (N = 39) had mean PSV 12.9 ± 3.2
cm/s, fifth group with TSH between 10.001 and 30 mU/L (N = 41) had mean PSV 14.3 ± 3.7
cm/s and sixth group with TSH higher than 30 mU/L (N = 34) had mean PSV 15.5 ± 5.2 cm/s.
When compared with group one, PSV was significantly higher in group three (p = 0.002), in
group four (p = 0.00002), in group five (p < 0.00001 and in group six (p < 0.000001).
Additionally, in all patients we found significant correlation between TSH concentration and
PSV (R = 0.213, p = 0.001). Conclusion. Colour flow Doppler sonography in a group of newly
diagnosed, untreated patients with Hashimoto's thyroiditis revealed significant positive
correlation between TSH concentration and PSV. Our results are in agreement with some in vitro
data on the stimulative influence of TSH on angiogenesis.
Ablative concept of the radioiodine therapy improves the
post-therapy management of patients with Graves' disease.
K. Zaplatnikov1, I. Seklinski1, M. Welch1, M. Plotkin2, F. Grünwald1;
Nuclear medicine Department, University, Frankfurt/M, Germany,
Aim The widely used ablative radioiodine therapy of Graves’ disease has an advantage of a rapid
and effective elimination of hyperthyroidism. On the other hand, this concept imply a high rate of
therapy-induced hypothyroidism, which borrow a potential risk of ophthalmopathy progression.
The aim of the presented retrospective study was to compare the long-term results of the ablative
and the function-optimised concepts of radioiodine therapy of Graves’ disease. Methods A total
of 120 pat. suffering on Graves’ disease were treated by radioiodine therapy between 1999 and
2005. In 70 of them (8m, 62f) (group. I), a function-optimised concept was applied (organ doses
150-200 Gy). The resting 50 pat. (6m, 44f) (group II), were treated according to the ablative
concept (doses • 250 Gy). The results of the therapy were evaluated by follow-up investigations,
which were performed 3, 6 and 12 months post-therapy and included typically clinical
investigation and thyroid hormone serology. At the time of first presentation, 4/70 pat from the
group 1 and 7/50 pat. from the group 2, had signs of opthalmopathy; the difference was not
significant. All these pat. recieved cortison during the radioiodine therapy. Results In the group
I, a hyperthyroidism/hypothyroidism was observed in 50%/50%, 41%/21% and and 38%/30% of
the pat., 3, 6 and 12 mo after therapy completion, respectively. In the group II, the control
investigations 3, 6 and 12 months post-therapy showed a hyperthyroidism/hypothyroidism in
28%/71%, in 18%/82% and in 10%/90% of the pat., respectively. The pre-existing
ophtalmopathy aggraved in 2/4 pat. from the group I and in 1/7 pat. from the group II. A new
ophtalmopathy developed in 6 pat. from the group I and in no pat. from the group II.
Conclusions The radioidine therapy using ablative dosage concept result in the lower failure
rate. Moreover, the risk to develop an ophtalmopathy is lower in the pat., treated by higher
radioiodine dosages. Consequently, the post-therapy management of pat., treated by ablative
dosage concept is easier, as in case of function-optimised concept.
Evaluation of the effect of diabetes mellitus on stomach,
small bowel and colon by FDG PET imaging
K. M. Alkhawaldeh, S. Dadparvar, G. Bural, M. Houseni, G. El-Haddad, A.
Alavi; Hospital of the University of Pennsylvania, Philadelphia, PA, United
Objectives: The aim of this study was to determine if FDG -PET imaging can detect the possible
effect of diabetes mellitus on the stomach, small intestine and colon by demonstrating any
change in FDG uptake.Methods: Seventy two patients where included in this retrospective study,
(30 males and 42 females, aged 46- 84). Patients were divided into two groups: first one
include36 diabetic patients and second one included 36 age matched non-diabetics. All patients
had FDG- PET studies .Both groups had blood glucose level ranging from 70-150 mg/dl with no
significant difference in blood glucose level between the two groups. Stomach mean SUV (SUV)
was calculated by drawing a region of interest around the stomach wall. Small bowel SUV was
calculated by drawing a region of interest within the mid abdominal cavity .SUV of the colon
was calculated by taking the sum of 3 SUV readings from different regions in the colon. T test
was used to compare the values between the diabetic and non-diabetic groups and P< 0.05 was
considered significantResults: Mean SUV values for stomach, small bowel and colon in diabetic
group were1.98 + 0.60, 1.42+ 0.27 and 5.94 + 1.93 respectively. Those values for non-diabetic
groups were 2.35 + 0.49, 1.58+ 0.22 and 5.34+ 1.22 respectively. Mean SUV for stomach and
small intestine in diabetic group is significantly lower than non-diabetic group (P<0.05).
Although higher colon mean SUV value in diabetic group compared to non-diabetic one, this
difference was not statistically significant (P> 0.05).Conclusions: Our study proves that FDG
PET study can show the effect of diabetes mellitus on the stomach and small bowel by
demonstrating significant lower FDG uptake. This finding can be explained by the diabetic
neuropathy influence on visceral organs that can lead to functional abnormality in smooth
muscles contraction. Although the difference between the diabetic and non diabetic SUV in
colon didn't reach the limits of significance, still we think that the increase in colon activity in
diabetic group can be related to frequency of constipation in those patients .Our study findings
may have future application in follow up of patients with diabetic induced motility and functional
disorders of stomach and small bowel.
antibodies is a risk factor for transition of toxic goiter to
TRAb associated hyperthyroidism following I-131 therapy
W. A. Eichhorn1, A. Koll1, R. Lippold2, P. Bartenstein1; 1Dept. of Nuclear
Medicine, Johannes Gutenberg University, Mainz, Germany, 2Inst. of
Biometrics, Epidemiology and Informatics, Johannes Gutenberg University,
Mainz, Germany.
Aim: The objective of this study was the identification of variables predicting the transition of
autonomous toxic goiter into an autoimmune hyperthyroidism with development of thyrotropin
receptor antibodies (TRAb) as a side effect of I-131 treatment. Methods: In this retrospective
study we analysed 448 patients (149 m/299 f): 104 patients (23%) with a diffuse autonomously
functioning thyroid, 189 (42%) with nodular/diffuse autonomous toxic goiter and 155 (35%) with
solitary or multiple hot nodules. We examined all patients with ultrasound, Tc-99m pertechnetate
scintigraphy and laboratory tests directly before and 6 months after radioiodine therapy, in case
of relapse earlier. The variables gender, age, subtype of autonomous toxic goiter, antithyroid
drug pre-treatment, thyroid function, level of thyrotropin receptor antibodies (TRAb) and antithyroid peroxidase (anti-TPO) were tested by means of uni- and multivariate regression analysis
regarding their prediction for a transition to autoimmune hyperthyroidism. Results: 16 patients
out of 448 (3.6%) developed thyrotropin receptor antibodies (>1.5 IU/l) concomitant with
recurrence of hyperthyroidism after I-131 therapy. Another 12 patients (2.7%) developed also
TRAb, but without autoimmune hyperthyroidism (partly with hypothyroidism needing
substitution). Out of only 12 pat. of the total sample with pretherapeutic borderline TRAb (1-1.5
IU/l) ) 3 pat. (25%) developed posttherapeutic autoimmune hyperthyroidism. In 9 out of 38
patients (24%) with pretherapeutical elevation of anti-TPO we saw a de novo development of
TRAb concomitant with recurrence of hyperthyroidism. The univariate logistic regression
demonstrates that borderline TRAb (p=0.0011), increased anti-TPO (p=0.0001) and level of
TPOAb (p=0.0014) as well as the subtype of autonomous function of the thyroid are significant
variables predicting the transition into an autoimmune hyperthyroidism. Significance persists
also in the multivariate regression analysis. Gender, age, thyroid function and antithyroid drug
pre-treatment are no significant predictors. Conclusions: The presence of autoantibodies in
serum before I-131 therapy - even at low levels - is a predictor of an increased risk of
development of TRAb associated hyperthyroidism following I-131 treatment of autonomous
toxic goiter. In this respect it is remarkable, that the transition incidence of pretherapeutical
elevation of anti-TPO nearly matches that of borderline increased TRAb. In addition patients
with diffuse autonomous toxic goiter have a higher risk of developing Graves` like disease than
with nodular toxic goiter. Patients with pretherapeutically detectable anti-TPO should be treated
as a risk population in a similar way to patients with elevated TRAb and deserve more intense
aftercare following I-131 therapy.
Utility of 18F-FDG-PET (PET) in the diagnosis of local
recurrence or metastasis in follow-up of patients treated for
Differentiated Thyroid Cancer
D. Esteva-Fernández, M. A. Muros, M. Bellón-Guardia, D. Cabello-García,
P. Macías-Mir, J. M. López-Ruíz, J. Jimenez-Alonso, J. M. Llamas-Elvira;
H.U. Virgen de las Nieves, Granada, Spain.
INTRODUCTION: Human thyroglobulin (hTg) determination and 131I whole-body scan (WBS)
are well-stablished methods for the detection of local tumor recurrence and metatases in the
follow-up of patients with differentiated thyroid cancer (DTC). In a certain percentage of patients
in whom hTg determination is positive during follow-up, no lesions with positive radioiodine
contrast are detected. OBJECTIVE: To determine the diagnostic validity of PET in the detection
of local recurrence or distant metastasis in patients diagnosed with differentiated thyroid cancer
(DTC) and presenting elevated serum thyroglobulin and negative 131I whole-body scan (WBS)
during follow-up.Materials and Methods: The study included 50 patients (39 females, 11 males)
with DTC (38 papillary and 12 follicular) referred by the Endocrinology Department and treated
with total thyroidectomy and radioiodine ablation. During follow-up, they presented with
elevated serum thyroglobulin levels and WBS with radioiodine was negative. Replacement
hormone therapy was suspended (4 weeks), and all patients were on an iodine-free diet. 131I
WBS was performed and serum levels of thyroglobulin, anti-thyroglobulin antibodies, FT3, FT4
and TSH were determined. In the week after the WBS, PET (SIEMENS ECAT EXACT 47,
370MBq 18F-FDG iv) was performed using a conventional protocol. PET studies were evaluated
by two nuclear medicine experts. PET findings were confirmed by pathology study or by other
imaging techniques (CT, MRI, US) in patients without surgical treatment.Results: PET was
positive in 36 studies and negative in 14. PET correctly detected the disease in 85% of cases
(sensitivity), with confirmation by other imaging techniques in 20, pathology report in 12, and by
follow-up in 6 patients. Absence of disease was confirmed in 83% of cases (specificity). The
positive predictive value (PPV) was 97%.Conclusions: - This study confirms the utility of PET
in the diagnosis of local recurrence or distant metastases in patients with DTC with elevated
thyroglobulin and negative WBS. - A positive PET result confirmed presence of the disease in
97% of patients (PPV)
308 — Sunday, October 01, 2006, 11:30 am - 1:00 pm, MC 2
2nd ISRTRD - Physics 1: Quantitative Analysis &
Treatment Planning
Tumour dosimetry following radioiodine ablation postthyroidectomy: a dose-response relationship
M. Haq, J. Gear, B. Pratt, G. Cook, C. Harmer, G. Flux; Royal Marsden
Hospital, Sutton, United Kingdom.
Background - In the management of differentiated thyroid cancer (DTC), 131I ablation postthyroidectomy serves to destroy thyroid remnants, facilitate thyroglobulin (Tg) measurement,
reduce recurrence and prolong survival particularly in high-risk patients. The vast majority of
centres administer fixed doses of 1.1-3.7GBq but this often leads to variable results. Optimising
ablation can maximise the initial absorbed dose (Gy), enhance tumour kill and potentially
improve long-term outcome. Aim - To assess the absorbed dose to thyroid remnants following
radioiodine ablation and determine whether ablation success is influenced by initial tumour dose.
Methods - 23 patients (8 M, 15F) with DTC received 3GBq 131I post-thyroidectomy following
appropriate thyroid hormone withdrawal. Patients with incomplete surgery, tumour variants,
Hurthle cell carcinoma and distant metastases were excluded. None of the patients received
recombinant thyroid-stimulating hormone (rhTSH). Following 131I, serial quantitative SPECT
scans were acquired of the neck using a dual-headed gamma camera equipped with a high-energy
general purpose parallel-hole collimator. Processing involved correction for deadtime, scatter and
non-uniform attenuation based on anatomical dimensions of the neck. An iterative algorithm
(OSEM) was used to reconstruct 3D images of tumour uptake. Coregistered images were
incorporated into an in-house dosimetry programme (Royal Marsden Dosimetry Package) to
determine voxel doses based on the MIRD schema. Successful ablation was based on an
undetectable stimulated Tg following rhTSH and a negative neck ultrasound at 6 months.
Results - administration of a fixed ablative dose of 3GBq resulted in a variable absorbed dose of
between 17 to 462 Gy to thyroid remnants. Eighteen patients were successfully ablated in total.
The mean absorbed dose in those successfully ablated (146.8 ± 127.8 Gy) was significantly
different (2 sample t-test, p=0.04) from those who were not (41.4 ± 15.5 Gy). Conclusions ablation with 3GBq 131I following total-thyroidectomy resulted in a wide variation in absorbed
dose to thyroid remnants. Patients successfully ablated received a mean dose of 146.8 Gy
whereas those treated unsuccessfully received a mean dose of only 41.4 Gy. A significant doseresponse relationship was identified. Optimising initial tumour dose may improve ablation
success and potentially lead to better long-term outcome.
Department of Radiology, Nuclear Medicine and Radiooncology, Campus
Virchow-Klinikum, University Clinic Charité, Berlin, Germany.
Lung toxicity in radioiodine therapy of thyroid carcinoma:
dosimetric implications of the 80 mci rule
G. Sgouros, H. Song, P. W. Ladenson, R. L. Wahl; Johns Hopkins
University, School of Medicine, Baltimore, MD, United States.
Based on an extensive data set analyzed by Benua and co-workers (Frontiers in Thyroidology,
1985), a whole-body retention threshold of 2.96 GBq (80 mCi) at 48 hr has been used to limit the
radioactivity of I-131 administered to thyroid cancer patients with pulmonary metastases.
However, lung absorbed doses and dose-rates arising from the 80 mCi limit have not been
thoroughly examined to account for patient-specific differences in lung geometry. This is
important, for example, in children with lung metastases. Methods: The dose-rate constraint
(DRC) was defined as the absorbed dose-rate to adult female lungs when 80 mCi 131I are
distributed in the lungs. The 80-mCi rule was generalized by calculating the activity to yield a
dose-rate equal to DRC using lung-to-lung S-factor-values corresponding to different reference
phantoms. Lung absorbed dose and administered activity were calculated as a function of
effective half-life. Results: A DRC of 45.6 cGy/h was obtained. Applying this to the 15-year-old
phantom, yields an equivalent 48 h activity limit of 2.46 GBq (66.4 mCi). The corresponding
limit for the 10-year old phantom is 1.73 GBq (46.6 mCi) Assuming 100% of whole-body
activity is in the lungs at 48 h and the level in the lungs is constrained by DRC, a relationship
between TE and lung absorbed dose was derived. At TE=33 h, a lung absorbed dose of 59.5 Gy
was predicted. A reduced DRC value of 20 cGy/h gives a lung absorbed dose of 26.1 Gy, also at
TE=33h. Because the female adult phantom was used in the dose-rate calculation, the 80 mCi
equivalent for the adult male is 3.73 GBq (101 mCi). In all cases, approximately 90% of the lung
absorbed dose is delivered by local, electron irradiation. Conclusions: A dose-rate based
formulation of the 80 mCi rule is derived and used to demonstrate application of this rule to
pediatric and adult male patients.. The implications of the 80 mCi rule are also examined. The
assumption of uniform energy deposition in the lungs leads to substantial overestimates of the
absorbed dose. Radiation-induced lung toxicity, expected at normal lung absorbed doses of 25-27
Gy, is probably avoided, because a majority of local electron dose is delivered to tumor tissue
instead of normal lung parenchyma. Use of a dose-rate constraint to adjust treatment for different
clinical circumstances is illustrated by these analyses.
Dosimetry in patients undergoing
with indications for Y-DOTATATE
Lu-DOTATATE therapy
M. Cremonesi1, M. E. Ferrari1, L. Bodei2, M. Bartolomei2, M. Chinol2, R.
Mei2, B. Daou2, G. Tosi1, G. Paganelli2; 1Medical Physics, European
Institute of Oncology, Milano, Italy, 2Nuclear Medicine, European Institute of
Oncology, Milano, Italy.
Background: Receptor radionuclide therapy with 177Lu-DOTATATE represents a recent strategy
for patients with neuroendocrine tumours. Promising results have been obtained with lower
nephrotoxicity compared to 90Y-somatostatin analogues. However, detailed dosimetry
information in patients is limited in the literature. In our Centre a clinical trial is ongoing with
injected activities (IA) ranging from 3.7 to 7.4 GBq/cycle. Aim of this study was to evaluate
dosimetry and to verify if cumulative IA of 22-30 GBq was compatible with acceptable dose to
kidneys. Second aim was to compare absorbed doses with those predicted for 90Y-DOTATATE,
in view of possible combined therapy with 177Lu- and 90Y- peptides.Methods: A personalised
dosimetry was performed in 10 pts (3 males, 7 females) undergoing 177Lu-DOTATATE therapy.
Blood samples, urine collection and whole body images (177Lu windows) were obtained up to 48
h p.i. WB transmission and low dose CT-scans were acquired for attenuation and actual organ
mass corrections. Time-activity curves were derived from images, applying the conjugate view
method. The number of disintegrations was calculated by a compartmental model (SAAMII) for
both 177Lu- and 90Y- derivatives to assess absorbed doses (OLINDA/EXM) in normal organs and
7 measurable lesions ( : 8-50 mm).Results: Blood clearance was rapid (< 1.0±0.4 %IA in the
bloodstream 20 h p.i.) and 70±10 % IA was eliminated within 24 h p.i. in the urine. Median
(range) dose estimates for 177Lu- vs. 90Y- were: 0.62 (0.45-17.74) vs. 2.51 (1.99-6.55), kidneys;
0.18 (0.05-0.34) vs. 0.74 (0.20-158), liver; 0.64 (0.29-2.91) vs. 2.74 (1.09-12.10), spleen; 0.04
(0.02-0.06) vs. 0.13 (0.07-0.27), red marrow; 0.31 (0.22-0.36) vs. 1.35 (0.99-1.66), u.bladder;
0.16 (0.11-0.20) vs. 0.67 (0.37-0.96), testes; 0.04 (0.02-0.08) vs. 0.13 (0.10-0.20), other organs;
0.05 (0.03-0.09) vs. 0.18 (0.13-0.28), total body Gy/GBq. Dosimetry results for 177LuDOTATATE did not suggest a threat for the kidneys with cumulative IA up to 30 GBq, with
median Biologically Effective Dose (BED) of 18 (10-43) Gy. Tumour doses varied in a wide
range: (0.6-56, 177Lu) vs. (2.2-180, 90Y) Gy/GBq. The 90Y-to-177Lu dose ratio was approximately
4 for normal organs, while ranged from 2.1 ( < 2cm) to 4.5 ( > 2cm) for tumours.Conclusions:
Dosimetry results support the tolerability of multiple-cycle therapy with 177Lu-DOTATATE at
cumulative IA even higher than 30 GBq. The 90Y-surrogate is able to deliver 4-fold to normal
organs and 2- to 4-fold doses to lesions but the benefit/risk balance remains to be established for
each patient depending on the clinical needs.
Comparison of
In-DOTA-DPhe -Tyr -octreotide and
InDOTA-lanreotide dosimetry for therapy planning of patients
with somatostatin receptor positive tumours
M. Rodrigues1, B. Ibi2, T. Traub-Weidinger3, S. Li4, I. Virgolini3; 1Inst Nucl
Med, Hietzing Hospital, Vienna, Austria, 2Inst Physics, Hietzing Hospital,
Vienna, Austria, 3Univ Clinic Nucl Med, Innsbruck, Austria, 4Dept Nucl Med,
Univ Hosp, Vienna, Austria.
For radionuclide therapy of somatostatin (SST) receptor (R)-positive tumours, the SST analogues
DOTA-DPhe1-Tyr3-octreotide (DOTA-TOC) and lanreotide (LAN) labelled with 90Y are
frequently used. This therapy is only initiated when a sufficient tumour dose is estimated. The
aim of the present study was to compare directly in patients (pts) with SSTR-positive tumours
organ and tumour absorbed doses (AD) for 90Y-DOTA-TOC and 90Y- DOTA-LAN based on
dosimetric data obtained with the respective 111In-labelled SST analogues.Materials and
Methods: 62 pts (32 M, 30 F; 20-76 a) with several tumour types, SSTR positive, were
investigated with both 111In-DOTA-TOC and 111In-DOTA-LAN scintigraphy for dosimetric
evaluation and therapy planning. Serial whole-body scans were recorded in ap and pa, 15 min
each, at 0.5-1, 2-4, 24 and 48 h after iv bolus injection of 150 MBq (10 µg peptide) 111In-DOTATOC or 111In-DOTA-LAN. For calculation of AD, regions of interest were drawn on each scan
for liver, spleen, kidneys, bladder, tracer accumulations regarded as tumour sites, and
background regions. Tumour volumes were calculated based on CT, MRI and/or sonography.
Dose calculations were performed according to the Medical Internal Radiation Dose (MIRD)
concept.Results: Estimated tumour AD for both radioligands varied among different tumour sites
in the individual pt and showed a wide inter-subject variability. No significant difference
between estimated overall tumour AD for 90Y-DOTA-TOC and 90Y-DOTA-LAN was found. In
16 (26%) pts estimated tumour AD for 90Y-DOTA-TOC were higher than those for 90Y-DOTALAN, whereas the opposite was true in 17 patients (27%). The estimated organ mean AD for
Y-DOTA-TOC and 90Y-DOTA-LAN were, respectively: kidneys 2.31 mGy/MBq, 2.69
mGy/MBq; liver 0.79 mGy/MBq, 0.88 mGy/MBq; spleen 4.84 mGy/MBq, 4.91 mGy/MBq; red
marrow 0.09 mGy/MBq, 0.14 mGy/MBq; urinary bladder 1.11 mGy/MBq, 1.15 mGy/MBq. The
mean effective doses amounted to 0.26 mSv/MBq for 90Y-DOTA-TOC and 0.30 mSv/MBq for
Y-DOTA-LAN.Conclusions: Spleen and kidney receive the highest AD for both radioligands.
Organ and tumour AD for 90Y-DOTA-TOC and 90Y-DOTA-LAN show high inter-pt variation.
Therefore, individual dosimetry is mandatory to decide whether a pt can be recruited or not for
therapy with radiolabelled-DOTA-TOC or DOTA-LAN, and to choose the therapeutic modality
for each pt.
Dosimetry for
Y-DOTATOC therapies in patients with
neuroendocrine tumours
C. Hindorf1, S. Chittenden1, L. Causer2, V. Lewington2, H. R. Maecke3, G.
D. Flux1; 1Joint Dept. of Physics, Royal Marsden Hospital & Inst. of Cancer
Research, Sutton, Surrey, United Kingdom, 2Dept. of Nuclear Medicine,
Royal Marsden Hospital, Sutton, Surrey, United Kingdom, 3Div. of
Radiological Chemistry, University Hospital Basel, Basel, Switzerland.
Aim: The aim of this study was to determine absorbed dose and absorbed dose rate by
quantitative SPECT imaging to tumours in patients recieving therapy with 90Y-DOTATOC. The
interpatient variability of absorbed dose to tumour as well as the question of whether the first
therapy could serve as a guide for future therapies was also investigated.Materials and Methods:
Ten 90Y-DOTATOC therapies were given to five patients diagnosed with refractory stage IV
neuroendocrine tumours (all patients received two therapies). The first and second therapy were
delivered at standard intervals. 90Y-activity was prescribed by surface area (3.7 GBq/m2) and
approximately 100 MBq 111In-DOTATOC was administered concurrently for imaging purposes.
Amino acid co-administration for renal protection was performed. Scintillation camera images
(SPECT) were acquired at various time points after administration of the radiopharmaceutical.
Images were reconstructed using OSEM and either no attenuation correction, a uniform effective
or a uniform linear attenuation correction was applied. Counts per second per voxel from 111In
were converted to 90Y activity using a calibration factor based on phantom measurements and a
voxel S value for 90Y was applied to get the absorbed dose rate. Integration over time then gave
the absorbed dose.Results: For all patients in the maximum tumour voxel the range of the
absorbed dose was 21 - 110 Gy (median: 49 Gy) and the median of the absorbed dose per
injected activity was 7.1 mGy/MBq (range: 3.9 - 20 mGy/MBq). Approximately 25 % of the
total absorbed dose was delivered within the 24 first hours after administration. The ratio of
absorbed dose to the maximum voxel in the tumour during therapy 1 and therapy 2 varied
between 0.8 and 1.4 with a median of 1.0. Applying no attenuation correction underestimated the
activity by 65 - 78 % and applying a uniform linear attenuation correction overestimated the
activity by 9 - 19 %, compared to the uniform effective attenuation correction.Conclusions: The
variation in maximum activity uptake and absorbed dose to the tumour was significantly larger
between patients than variations between subsequent therapies for the same patient. These results
imply that the first therapy could serve as a guide for future therapies.
I-BC8 (anti-CD45) radioimmunotherapy for
patients with high-risk myelodysplastic syndrome and
advanced acute myelogenous leukemkia: radiation dosimetry
for bone marrow and normal organs in treatment planning
and follow-up
D. R. Fisher1, J. M. Pagel2, L. D. Durack3, J. F. Eary3, A. K. Gopal2, F. R.
Appelbaum2, O. W. Press2, J. G. Rajendran3; 1Pacific Northwest National
Laboratory, Richland, WA, United States, 2Fred Hutchinson Cancer
Research Center, Seattle, WA, United States, 3University of Washington,
Seattle, WA, United States.
Objectives: We present methods for calculating internal radiation doses to 108 patients treated
with high-dose 131I-BC8 (anti-CD45) monoclonal antibody for advanced acute myelogenous
leukemia and refractory myelodysplastic syndrome. These are challenging diseases not
successfully cured with standard management techniques, and otherwise fatal. Our objective was
successful radioimmunotherapy by delivering maximal, myeloablative radiation within tolerance
of the critical normal organs, thereby providing maximum therapeutic benefit while avoiding
serious toxicities. Treatment planning prior to myeloablative therapy was based on normal organ
dose assessment using trace-labeled 131I-BC8 antibody and serial imaging. Methods: Tracer
infusions were given using 150 to 370 MBq (4 to 10 mCi) 131I-BC8 antibody (0.5 mg/kg).
Patients underwent serial gamma camera imaging and whole-body counts to determine antibody
biodistributions. Time-activity curves were constructed for the major organs and whole body.
Iliac crest bone marrow biopsies were obtained at 24 h post-infusion and were counted to
determine percent 131I uptake per gram red marrow. Marrow activity was also assessed by planar
imaging, and the time-activity curves were normalized per unit mass from biopsy measurements.
Patient-specific doses were calculated using organ weights from CT volumetric imaging. We
looked at the differences in estimated marrow dose with and without normalizing the biopsy data
to the serial marrow images. Results: Absorbed doses (and means, cGy/mCi) from 131I-BC8
Dosimetry in Myeloablative Y labeled Ibritumomab Tiuxetan
Therapy: possibility of increasing the administered activity
on the base of Biological Effective Dose evaluation
C. Chiesa1, F. Botta2, E. Di Betta2, A. Coliva1, E. Seregni1, F. Elisei3, G.
Aliberti1, S. Bavusi1, L. Devizzi4, A. Guidetti4, A. M. Gianni4, E.
Bombardieri1; 1Nuclear Medicine - Istituto Nazionale Tumori Milano,
Milano, Italy, 2Post graduate Health Physics School, Milano, Italy, 3Post
graduate Nuclear Medicine School - Istituto Nazionale Tumori Milano,
Milano, Italy, 4Medical Oncology C - Istituto Nazionale Tumori Milano,
Milano, Italy.
Aim: In our multicentrical ongoing phase I escalation study, 90Y labeled ibritumomab tiuxetan
(Zevalin) is administered in activities of 0.8 and 1.2 mCi/kg. The radioinduced myelodepression
is overcome by stem cell reinfusion. Since the rate and fractionation of dose delivery in targeted
radiation therapy (TRT) are quite different respect to external radiotherapy (XRT), in TRT the
threshold for deterministic effect to healthy organs other than red marrow is still unknown. The
use of the Linear Quadratic model and of Biological Effective Dose (BED) [Dale Phys Med Biol
41 (1996) 1871] permit to translate known XRT toxicity levels in rate independent BED limits,
which could quantitatively explain the absence/presence of damage in TRT.Methods: We
evaluated 12 patient injected with 0,8 mCi/kg and 2 with 1,2 mCi/kg. Pre treatment planar
dosimetry with diagnostic activity of 111In-Zevalin was performed with conjugate view
technique, background correction (Buijs method), attenuation correction (57Co blank and
transmission scan) and scatter correction (pseudoextrapoation number). Blood samples and WB
scintigram had been collected at least at 0.5 , 48, 120 h . Individual organ mass correction was
applied by ROIs drawing on CT scan. MIRDOSE3.1 was used, but OLINDA S values were used
for red marrow dose. BED was calculated with the formula used by Barone et al, JNM 2005 46:
99S 106S for monoexponential clearance. The parameters Į/ȕ and the halflife for sub-lethal cell
damage T1/2rep were adopted from Barone for kidneys, while for liver, spleen and testes we
adopted Į/ȕ=3 Gy (critical but generally accepted value) and T1/2rep=3 h (less critical value). XRT
absorbed dose threshold D5,5 for toxicity taken from Emami Int J Radiat Oncol Biol Phys [03603016] 1991 May 15;21(1) 109-22 were translated to BEDT according to Dale.Results:
Red marrow
absorbed 1.1±0.4
dose (Gy/GBq)
(0.8 mCi/kg)
Absorbed dose (Gy)
(1.2 mCi/kg)
(0.8 mCi/kg)
(1.2 mCi/kg)
5.8±1.6 3.5±1.3 4.0±2.2
8.0±3.9 9.2±5.6
9.6±2.5 18±7
Absorbed dose (Gy)
Kidneys Liver
Conclusions: 0.8 & 1.2 mCi/kg give average BED at about 40 & 60% of the estimated liver and
kidney toxicity limit. A further increase of injected activity could be possible, provided that stem
cell re-infusion is prepared.
Development and validation of an organ residence time
estimation method for high dose Y-90 ibritumomab tiuxetan
E. C. Frey, B. He, G. Sgouros, I. W. Flinn, R. L. Wahl; Johns Hopkins
University, Baltimore, MD, United States.
Aim: We are currently engaged in a dose escalation trial to find the maximum tolerated dose
(MTD) for a myeloablative Y-90 ibritumomab tiuxetan therapy regimen with autologous stem
cell transplant. In this study, the escalation variable is dose to critical organs. Thus, accurate
estimation of organ residence times is essential. The goal of this work was to develop and
validate a method for estimating organ residence time of the In-111 labeled antibody for use in
this study.Materials and Methods: In the method, anterior/posterior planar scans are acquired at 5
time points plus abdominal and thoracic SPECT/CT acquisitions at 24 hrs post injection. SPECT
images were reconstructed using a quantitative SPECT (Q-SPECT) reconstruction method based
on iterative reconstruction including compensation for attenuation, scatter, and the collimatordetector response. Planar scans were processed using scatter subtraction and geometric mean
methods. Organ volumes-of-interest were defined using the registered CT and SPECT images
and regions of interest were manually defined on the planar scans. Organ time-activity curves
(TACs) were estimated from the planar scans and, for the hybrid method, rescaled to pass
through the organ activity estimated from the Q-SPECT images. The residence times were
estimated by integrating exponential functions fitted to the TACs. The method was evaluated
using physical phantom experiments and Monte Carlo simulation studies by comparing
accuracies of estimated residence times for the liver, the dose-limiting organ in all patients todate. The planar and hybrid methods were also compared using data from 12 patient
studies.Results: The errors in the 24 hr estimates of liver activity were 1.3% and 13% for the QSPECT and planar methods, respectively. The planar method resulted in a 19% underestimation
in the liver compared to 8% for the hybrid method. In the patient studies, the difference between
the planar and hybrid estimates of the residence times was 11% +/- 14%. Of importance, the
variation in the accuracy over the patient population was larger than the average difference.
Large variations in accuracy of dose estimates could result in large variations in the administered
dose compared to the true dose, and thus errors in estimating the MTD.Conclusions: We
conclude that, the hybrid planar/SPECT estimation method provided a substantial improvement
compared to residence time estimates based on planar images. This methodology has been
applied in the ongoing dose escalation study where a 28 Gy dose to the liver has been
administered to three patients without adverse reactions.
Voxel dosimetry of Y-90 micro-spheres in the palliative
treatment of neuroendocrine tumours of the liver
E. McKay; St. George Hospital, Kogarah, Australia.
Background: Y-90 labelled glass micro-spheres administered via the hepatic artery are a recent
development in the treatment of metastatic liver disease. Y-90 is a pure beta emitter so measuring
its in-vivo distribution and consequent dose distribution presents a challenge. Aim: This study
attempts to evaluate mean liver and tumour dose for individuals undergoing palliative treatment
for liver tumours with Y-90 micro-spheres, using bremsstrahlung SPECT imaging.Methods:
Thirty five subjects (23 males, 12 females, mean age 59 years) with neuroendocrine tumours in
the liver (33 carcinoid, 1 gastrinoma, 1 MTC) were treated once or twice (N = 5) with Y-90
micro-spheres (SIRtex), following the manufacturer's guidelines (typical administered activity 2
GBq). Abdominal CT of each subject was obtained prior to therapy, for treatment planning
purposes. Bremsstrahlung imaging was performed subsequently for dose evaluation, using a
protocol validated with physical and software phantom studies. Whole body A/P sweeps and
SPECT imaging were performed using a Philips Axis gamma camera with a high energy
collimator and a single broad energy window (60 - 120 keV). The sweeps were used to exclude
extra-hepatic activity. The SPECT images were reconstructed using OSEM incorporating
attenuation correction and depth dependent resolution, based on a measured point response
function. Reconstructed images were converted to cumulated activity images by assuming a fixed
distribution of micro-spheres (ie, physical decay only). The dose distribution was calculated
using voxel dose kernel convolution in a 3 mm matrix. The CT and SPECT slices were manually
co-registered and regions of interest defining healthy liver and tumour were transferred from the
former to the latter. Dose volume histograms and mean dose measurements were obtained for
total tumour and liver regions for each subject.Results: The mean dose calculated for liver was
43 Gy (range 8 - 78 Gy); for tumour it was 54 Gy (range 7 - 125 Gy). The target/non-target
(TNT) dose ratio averaged 1.3 (range 0.2 - 2.8). These TNT ratios should be considered lower
limits in the subjects for which they are calculated, as the validation study demonstrated that both
registration errors and imperfect resolution recovery in reconstruction tend to reduce this value.
Work is in progress to improve both of these areas.Conclusions: A process for estimating
individual dose distributions for Y-90 SIRT therapy has been developed and is currently being
refined. Presently, the technique shows only modest differentiation between target and non-target
absorbed doses.
Optimisation of energy window settings for scatter correction
in quantitative
In imaging: Comparison of measurements
to Monte Carlo simulations
M. Holstensson1, C. Hindorf2, G. D. Flux3, M. Ljungberg4;
Deparment of Physics, The Royal Marsden NHS Foundation Trust, Sutton,
Surrey, United Kingdom, Joint Deparment of Physics, Institute of Cancer
Research, Sutton, Surrey, United Kingdom, 3Joint Department of Physics,
The Royal Marsden NHS Foundation Trust, Sutton, Surrey, United
Kingdom, 4Medical Radiation Physics, Lund University, Lund, Sweden.
Introduction: Activity quantification in Nuclear Medicine Imaging is highly desirable,
particulary for absorbed dose calculations and biodistribution studies of radiopharmaceuticals.
One of the major problems in quantification in Nuclear Medicine Imaging is that of scattered
radiation which leads to degradation of image quality. Dosimetry for 111In imaging is becoming
increasingly important with the current interest in 90Y radiolabeled antibodies. Aim: The aim of
this study was to optimise the energy window settings for quantitative imaging with 111In using
both experimental measurements and Monte Carlo simulations for a scintillation camera with a
limitation of three energy windows. The improvement on image quality of different scatter
correction methods was investigated. Materials and Methods: Posterior and anterior images of
phantoms filled with activity to mimic clinical data from patients administered with 111In-DOTATyr3-octreotide were acquired using a Philips FORTE scintillation camera. The images were also
simulated by the Monte Carlo code SIMIND. Ten different combinations of window settings
were investigated including the 171 and 245 keV photopeaks and various scatter windows. The
combinations included a Triple Energy Window (TEW) scatter correction for each peak and a
broad scatter window located between the peaks. The images were attenuation corrected using
the geometric mean method. Investigated parameters were spatial resolution, sensitivity, image
contrast and quantitative accuracy. The scatter-to-total ratios and distributions estimated from the
experiments with the window settings were compared with data obtained from Monte Carlo
simulations. Results: We found good agreement between the outcomes of the experimental
were: liver = 1.3 to 5.4 (3.1); spleen = 2.8 to 42 (16); marrow = 1.3 to 29 (7.0); total body = 0.15
to 5.5 (0.54). The mean uptake in red marrow at 24 h was 0.021 %ID/g (range = 0.0036 to
0.088). 131I-BC8 therapy infusions averaged 13.4 GBq (363 mCi) [range = 3.0 to 25.1 GBq, or 81
to 678 mCi]. Conclusions: Radioimmunotherapy with 131I-BC8-antibody delivered substantial
radiation doses to hematopoietic tissues in patients treated for advanced AML and high-risk
MDS. Red marrow dosimetry was best determined using both planar imaging and biopsy.
Dosimetry-based treatment planning enabled delivery of maximally tolerated amounts of 131IBC8 for optimizing therapeutic outcomes. (Supported by NIH P0144991and DOE AC06-76RL01830).
measurements and Monte Carlo simulations both quantitatively and visually. Scatter correction
improved resolution by 5 % to 10 % and image contrast by 3 % to 10 %. Conclusions:
Considering all evaluated parameters the best setting was found to be two main windows centred
at 171 and 245 keV with a broad scatter window in between the photopeaks. Assuming no
scattered events at energies higher than the window at 245 keV the scatter estimation was
extrapolated to the window at 171 keV. This scatter correction improved spatial resolution and
image contrast by 7 % and 10 % respectively and gave relatively good agreement in estimated
and simulated scatter-to-total ratios and distribution. Using both photopeaks approximately
doubled the sensitivity compared to using a single photopeak.
309 — Sunday, October 1, 2006, 11:30 am - 01:00 pm, Lecture Hall 1
Symposium: Physics: Quantitative PET/CT Imaging
CT-based attenuation correction and high-density artifacts
T. Beyer (DE)
Respiratory and cardiac motion, Gating schemes and
K. Schaefers (DE)
New therapy response parameters
W. Nijsen1; 1Dept. of Nuclear Medicine, University Medical Center Utrecht,
Utrecht, The Netherlands, 2Dept. of Radiation, Radionuclides and Reactors,
Delft University of Technology, Delft, The Netherlands.
Aim: The aim of this study was to develop holmium-166 (166Ho) labelled liposomes and
characterise these nanoparticles. Since 166Ho emits both beta particles and gamma rays (Eȕmax =
1.84 MeV (51%), 1,78 (48%), EȖ = 81 keV), these nanoparticles could be deployed as diagnostic
and/or therapeutic particles for oncological indications. Examples of these indications are
glioblastoma (intratumoral administration), liver malignancies (intra-arterial administration) and
lymph node detection/ablation.Materials and Methods: The liposomes were produced by the
conventional thin film hydration technique1. The liposomes consisted of 1,2-dipalmitoyl-snglycero-3-phosphocholine (DPPC), cholesterol (Chol) and 1,2-dipalmitoyl-sn-glycero-3phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000] (PEG-DSPE) in a molar ratio of
1.85: 1: 0.15. The liposomes were labelled with 166Ho by using the amphiphilic molecule
diethylenetriaminepentaacetic acid bisoctadecylamide (DTPA), which was incorporated in the
liposomal bilayer. The holmium labelling efficiency was determined by using a desalting
column. The liposomes were characterised by measuring their size, holmium content and
radiochemical stability after incubation in human serum for 48 h. A proof of principle rabbit
study was conducted in which these liposomes were administered intracutaneously into the hind
leg to assess lymphoid drainage.Results: The liposomes were successfully labelled with 166Ho,
resulting in a labelling efficiency of 95 % and radiochemical stability of >98 %. The liposomes
had a size around 130 nm and a very low polydispersity index of < 0.07. Nuclear imaging in the
rabbits demonstrated lympoid drainage of the liposomes to the popliteal node thirty minutes post
administration, and retention for at least five hours.Conclusions: The labelling efficiency and the
radiochemical stability were very high. Furthermore, the polydispersity index demonstrated that
the size variation of the liposomes was very small. The results from the rabbit study indicate that
these liposomes are suitable for lymph node detection and/or ablation purposes. Additionally,
these liposomes are also apt for loading with for example anticancer agents and/or homing
devices. These possibilities are being further investigated. References: 1. Koning GA, Fretz MM,
Woroniecka U, Storm G, Krijger GC. Targeting liposomes to tumor endothelial cells for neutron
capture therapy. Appl Radiot Isot 2004; 61: 963-7
Wolfgang Weber (USA)
310 — Sunday, October 01, 2006, 11:30 am-1:00 pm, Meeting Room 2
New radiolabeled derivatives of vitamin B12 with abolished
transcobalaminII binding show preferential targeting of
R. Waibel1, D. van Staveren1, S. Mundwiler2, R. Alberto2, R. Schibli1, P. A.
Schubiger1; 1Paul Scherrer Institute/ CRS, Villigen PSI, Switzerland,
University of Zurich, Zurich, Switzerland.
Lipidic Nanocapsules: a new tool for radioimmunotherapy of
E. Jestin1, M. Mougin-Degraef1, A. Faivre-Chauvet1, F. Hindré2, A. K.
Mishra3, P. Thédrez1, C. Saï-Maurel1, J. F. Chatal1, J. Barbet1, J. F. Gestin1;
U601 Inserm, Institute of Cancer Research, Nantes, France, 2U646
Inserm, Angers, France, 3Institute of Nuclear Medicine and Allied Sciences,
Delhi, India.
Aim: Radioimmunotherapy (RIT) is limited in some cases by the low radioactive doses delivered
to tumour cells by antibodies or pretargeted haptens. The aim of our study is to create a new tool
by increasing the dose delivered and decreasing the radiotoxicity with a two step system. Lipidic
nanocapsules (LNC) are proposed as radionuclide vectors (in core or on LNC surface). These
LNC are targeted to cancer cells using a two steps pretargeting method (bispecific antiCarcinoEmbryonic Antigen × anti-hapten antibody) after incorporation of different haptens on
the nanocapsules membrane. Furthermore, to increase residence time in blood,
polyethyleneglycols (PEG2000 and PEG6000) were added in LNC formulations. As PEG presence
could modify bispecific antibody haptens recognition and reduce radiolabelling efficiency,
haptens, bifunctionnal chelating agent (BCA) or Bolton-Hunter reagent (BH) were attached to
PEG extremity to optimize accessibility. DSPE-PEG-haptens, DSPE-PEG-BCA, and DSPEPEG-BH were synthesized to develop these new radiolabelled vector formulations. Materials &
Methods: Lipidic nanocapsules are prepared with a mixture of three components: Labrafac®;
Lipoïd®, the single layer membrane’s constituent; and Solutol®, a hydrophilic surfactant.
Constructs containing haptens or BCA for the radiolabelling were synthesized starting with two
various PEG lengths (PEG2000 and PEG6000). LNC radiolabelling were obtained by incorporation
of constructs DSPE-PEG-BCA or DSPE-PEG-BH in the nanocapsules membrane while in core
radiolabelling was performed using lipophilic molecules. Results: Seven constructs with the two
lengths were synthesized. Three DSPE-PEG-R constructs were prepared (R=
DNP(DinitroPhenyl); DTPA-In or HSGL(HistaminoSuccinylGlycylLysine)). Four others DSPEPEG-BCA constructs were synthesized for 64/67Cu, 111In, 90Y complexation (BCA= DOTA;
DTPA; CHXA’’DTPA; TE3A). One DSPE-PEG-BH construct was synthesized for 125I, 131I or
At radiolabelling. In addition, several lipophilics molecules were developed in order to
radiolabel the LNC core (precursors for 125I, 131I or 211At; N2S2 complexes for 64/67Cu and SSS
complexes for 99mTc and 186/188Re). Stabilities in buffers were performed showing the efficiency
of radiolabelling methods. New LNC formulations in vivo studies showed promising results.
Time residence was really prolonged compare to basics formulations. Conclusion: Several
constructs have been synthesized to formulate new lipidic nanocapsules that can be labelled with
various radionuclides and targeted to tumour cells using bispecific antibodies. A two steps
system is in progress by using an antibody anti-CEA×anti-DTPA-In.
Preliminary studies on holmium-166 loaded liposomes for
diagnostic and therapeutic purposes
S. W. Zielhuis1, W. Bult1, M. A. D. Vente1, A. D. van het Schip1, T. C. de
Wit1, G. C. Krijger2, R. de Roos1, B. A. Zonnenberg1, P. P. van Rijk1, J. F.
Aim. Rapidly growing cells show an increased demand on nutrients and vitamins. The objective
of our work is to use the supply route of vitamin B12 (cobalamin) to visualize hyperproliverative
cells with radiolabeled cobalamin derivatives. Material and Methods. To date, cobalamin has
been radiolabeled with different isotopes of Iodine, Co-57, In-111 and Tc-99m, the latter two
metallic elements by the introduction of pendant metal-binding ligands like DTPA at the corrin
ring system. The results obtained in animals and humans, when using radio-labelled cobalamin
derivatives, showed labelling of tumour tissues, but also a strong accumulation of radioactivity in
healthy tissues, such as kidney and liver. Therefore, imaging and possible radiotherapy are far
from being optimal. In mammals, most of the cellular uptake of cobalamin is regulated by serum
transport vectors (intrinsic factor, haptocorrin, transcobalamin-II) and by cell membrane
receptors. By tailoring a cobalamin derivative with vector-less uptake (abolished transcobalaminII binding), more useful reagents were developed. These conjugates are carrying mono-anionic
ligands with a NNO donor set such as the pama-ligand ([pyridine-2-ylmethyl)-amino]-acetic
acid) and can be efficiently labelled with [Tc-99m(OH2)3(CO)3]+ at yields higher than 95% under
mild conditions (50° C, 30 min, 10-4 M). Results. The biodistribution of the Tc-99m and Re-188
(CO)3 labeled cobalamin derivatives is analyzed in mice having tumor xenografts by
biodistribution studies and by a dedicated small animal CT/SPECT imaging system and
compared with derivatives which still interact with transport proteins. In melanoma tumors, all
cobalamin derivatives accumulate at the tumor site with 6% to 9% of ID/g tissue, but the new
derivatives showed an improvement of the tumor/blood ratio from 1.5/1 to 153/1, tumor/kidney
ratio from 0.33/1 to 6/1 and a tumor/kidney ratio from 0.7/1 to 3/1. Melanoma, bladder
carcinoma, pancreas carcinoma, small cell lung carcinoma and renal carcinoma were positive,
whereas no accumulation could be observed in colon carcinoma, prostate carcinoma and
neuroblastoma. Conclusion. By interfering with the transport mechanism of cobalamin and by
using the very stable tridentate PAMA-ligand for Tc-99m and Re-188 (CO)3 labeling, we have
developed a much improved radiopharmaceutical. A clinical study is initiated.
DNA aptamers as novel radiopharmaceuticals for imaging
and targeted radiotherapy of epithelial tumours
C. Da Pieve1, B. Arash2, A. C. Perkins2, S. Missailidis1; 1The Open
University, Milton Keynes, United Kingdom, 2University of Nottingham,
Nottingham, United Kingdom.
Molecular targeted radiotherapy is an area of increasing interest that has recently seen the
approval of various radiolabelled antibodies for cancer therapy, particularly in the treatment of
leukaemias and lymphomas. However, the use of antibodies as therapeutic conjugates has not
been without significant problems, including the time period required for production, their
immunogenicity and size. We aim to overcome these problems by using DNA aptamers as
recognition molecules to malignant disease. Combinatorial chemistry techniques coupled with
PCR amplification allows the evolution of ligands, through exponential enrichment, which can
bind to almost any target molecule, including extracellular proteins, antibodies, peptides and
small molecules. Our research on the rapid selection of aptamers based on the SELEX
methodology forms the basis for the development of high affinity and specificity molecules,
which bind to surface determinants of tumour cells. The procedure is accomplished within days,
compared with several months required for the selection and purification of antibodies. Aptamers
offer reduced immunogeniciy, greater tumour penetration, rapid uptake and clearance, favouring
their application as effective vehicles for cytotoxic agents or radionuclides. We have previously
reported the generation of high affinity and specificity DNA aptamers against the protein core of
A remote-controlled automatic system for the labeling of
lipiodol with Re-188 in the treatment of hepatocellular
A. Duatti1, L. Uccelli1, M. Pasquali1, A. Boschi1, E. Cazzola1, P. Bedeschi2,
S. Bosi2, M. Giganti1; 1University of Ferrara, Ferrara, Italy, 2Comecer,
Castelbolognese, Italy.
Aim. In the last years, the ȕ-emitting radionuclide rhenium-188 has been employed for the
labeling of lipiodol, which is an effective vector for the selective targeting of hepatocellular
carcinoma (HCC). Recently, we proposed a highly efficient procedure for incorporating a Re-188
lipophilic complex into the hydrophobic lipiodol phase, and this method has been successfully
applied to the treatment of a number of HCC patients. Since the preparation of labeled lipiodol
involves different steps, the radiation exposure of the operator appears significant, particularly
when high starting activities are required for the routine treatment of patients. To overcome this
problem, we attempted to develop an automatic system for the remote-controlled preparation of
Re-188 lipiodol using our labeling method. This work describes the essential features and
advantages of this novel system. Materials and Methods. All required reagents were provided
through a two-vial freeze-dried kit formulation. The preparation involves essentially three basic
steps: (1) formation of the [188ReŁN]2+ core, (2) reaction of this core with diethyldithiocarbamate
to afford the lipophilic complex bis(diethyldithiocarbamato) nitrido rhenium(V) [188ReN-DEDC],
(3) mixing of 188ReN-DEDC with lipiodol with subsequent trapping of the complex into the
hydrophobic phase. Withdrawal of generator-eluted [188ReO4]- and of reactants from the kit vials
was performed automatically and these substances were placed in a reactor vial to accomplish the
production of the final complex 188ReN-DEDC. Preliminary purification of 188ReN-DEDC from
excess of reagents and free ligand was carried out by passing it through a reversed-phase SepPak
cartridge. The radioactive solution was then filtered through a sterile 0.22-µm Millipore filter
before the mixing with lipiodol. All operations were remotely controlled through a dedicated
software package. Results. The radiochemical yield for the complex 188ReN-DEDC, measured
by TLC and HPLC chromatography before purification and incorporation into lipiodol, was
found to fall in the range 96.1 ± 3.0 (n = 8). Extraction of the complex by lipiodol was
quantitative yielding a final radiopharmaceutical with approximately 100% radiochemical purity.
Conclusions. The new remote-controlled system could provide a convenient and efficient tool
for the high-activity production of Re-188 lipiodol in routine treatment of HCC patient, thus
allowing a dramatic decrease of radiation exposure of the operator during the preparation
Preparation and in vitro Studies of
I-UdR-Tf conjugate in
human hepatocellular carcinoma SMMC-7721 cells.
z. Wu;
The PET Center of Union Hospital,Tongji Medical
College,Huazhong University of Science&Technology, Wuhan, China.
[Aim] To explore the feasibility of 125I-UdR-Tf conjugate as a transferrin(Tf) receptor mediated
radiotherapeutic agent against hepatocellular carcinoma SMMC-7721 cells. [Material and
Methods] 2’-deoxyuridine(2’-UdR) labeled with 125I was conjugated to transferrin(Tf).The
bioreactivity,cell uptake,cell retention,cytotoxicity,and intracellular localization of 125I-UdR-Tf
conjugate were tested in tranferrin receptor (TfR)-positive human hepatocellular carcinoma
SMMC-7721 cells,with 125I-UdR,and 125I-Tf as control agent.Transmission electron microscope
was used to investigate the cellular ultrastructral changes.[Results] The specific activity of 125IUdR-Tf conjugate was 3.25×105MBq/mmol.As many as (32±4) 2’-UdR residues were
introduced per mole of transferrin.The cell binding rate of 125I-UdR-Tf conjugate was
14.57%,less than that of 125I-Tf (16.50%),but no significant difference was observed.The peak
uptake time of 125I-UdR-Tf conjugate,125I-Tf and 125I-UdR in cells was 6h,1h and
6h,respectively,while the cell peak uptake rates of 125I-UdR-Tf conjugate,125I-Tf and 125I-UdR
were 24.10%(P<0.05,as compared with 125I-Tf,P<0.01,as compared with 125I-UdR),18.40% and
6.01%,respectively.At cell clearance tests,the cell retention rates of 125I-UdR-Tf conjugate,125I-Tf
and 125I-UdR were 50.61%,15.60% and 38.52%, respectively,after incubation with SMMC-7721
cells for 24h.The 50% lethal doses (LD50) of 125I-UdR-Tf conjugate,125I-Tf and 125I-UdR for
human hepatocellular carcinoma SMMC-7721 cells were (0.49±0.12)KBq/ml(P<0.01,as
compared with 125I-Tf and 125I-UdR),(0.87±0.21)KBq/ml and (0.86±0.15)KBq/ml,respectively.At
the intracellular localization tests,(16.60±1.31)% of the radioactivity of 125I-UdR-Tf conjugate
was bound to the nuclear fraction after incubation with SMMC-7721 cells for 24h.In
contrast,125I-Tf showed (0.21±0.11)% of the radioactivity binding in the nuclear fraction,whereas
I-UdR showed (9.02±0.61)% of radioactivity binding. After exposure to 0.03µCi/ml of 125IUdR-Tf conjugate and 125I-UdR,the radioactivity incorporated into DNA of SMMC-7721 cells at
24h was 2.62×10-7KBq/cell and 7.89×10-8KBq/cell,respectively.The changes of cellular
ultrastructure were viewed at 125I-UdR-Tf conjugate,125I-Tf and 125I-UdR group,especially at 125IUdR-Tf conjugate group.125I-UdR-Tf conjugate had stronger cytotoxicity to human
hepatocellular carcinoma SMMC-7721 cells than 125I-Tf and 125I-UdR.[Conclusions] 125I-UdRTf conjugate can bind to TfR-positive hepatocellular carcinoma SMMC-7721 cells and is
internalized.More radioactivity is retained in the nuclear fraction for 125I-UdR-Tf conjugate than
that for the 125I-Tf and 125I-UdR.125I-UdR-Tf conjugate induces greater cytotoxicity.It is indicated
that 125I-UdR-Tf conjugate is of good prospect to be a radiotherapeutic agent for hepatocellular
carcinoma receptor-mediated targeting therapy.
Synthesis and evaluation of 8-[ I]iodo-L-1,2,3,4-tetrahydro7-hydroxyisoquinoline-3-carboxylic
prostate cancer
S. Samnick, C. Fozing, M. Menger, C. Kirsch; Saarland University Medical
Center, Homburg, Germany.
Aim: To evaluate 8-[123I]iodo-1,2,3,4-tetrahydro-7-hydroxyisoquinoline-3-carboxylic acid
[ITIC(OH)] as an imaging probe for detecting prostate cancer.Materials and Methods: Nocarrier-added ITIC(OH) was synthesized by the IODO-GEN method with 82 ± 7 %
radiochemical yield and > 99% radiochemical purity after HPLC. Thereafter, the tumor affinity
and uptake characteristics of ITIC(OH) were assessed in vitro in human PC-3 and DU-145
prostate tumor cells. In addition, ITIC(OH) was evaluated in experimental models of human
prostate cancer in CD1 nu-nu mice, including measurements of the kinetics of tumor and organ
uptake of the radiotracer by dynamic imaging using a high resolution gamma-camera and
biodistribution by gamma-counting after dissection.Results: The uptake of ITIC(OH) in primary
human prostate tumor cells was rapid. More than 80 % of the total radioactivity accumulation in
cells occurred in the first 2-3 min. The uptake in 106 tumor cells after a 15 min-incubation time
was 35 - 42 % of the total loaded dosis. In addition, ITIC(OH) accumulates rapidly and with high
intensity in human prostate tumor xenografts after intravenous administration. At 60 and 240 min
postinjection, the radioactivity binding in tumors amounted to 13.6 ± 2.1 and 16.2 ± 2.5 % I.D. /
g in the heterotopically implanted tumors compared with 14.8 ± 2.6 and 17.6 ± 3.4 % I.D./g in
the orthotopic tumors. The resulting mean tumor-to-organ ratios at 60 and 240 min were 22.2 ±
4.7 and 48.4 ± 6.6 (tumor/blood), 21.7 ± 4.4 and 26.5 ± 6.6 (tumor/muscle), 13.4 ± 3.1 and 27.6
± 7.5 (tumor/spleen), 2.3 ± 0.7 and 3.9 ± 1.6 (tumor/kidney), 5.2 ± 1.1 and 6.9 ± 2.3
(tumor/intestine), as well as 4.2 ± 1.3 and 13.2 ± 3.5 (tumor/liver). The gamma-camera imaging
allowed a clear visualization of the tumor and a detailed assessment of whole-body tracer
distribution in tumor-bearing mice. Conclusions: The specific and high-level targeting of
ITIC(OH) to human prostate tumor xenografts with a marked tumor-to-background contrast
enabled an accurate detection of the tumors in murine PC-3 and DU-145 human prostate cancer
xenograft models. Therefore, ITIC(OH) appears to be a potential novel radiopharmaceutical for
prostate cancer with SPET.
In vitro and in vivo evaluation of [
I]-2-iodo-D-Tyrosine in
a R1M Rhabdomyosarcoma cell model compared with
M. Bauwens, T. Lahoutte, K. Kersemans, C. Gallez, A. Bossuyt, J.
Mertens; Vrije Universiteit Brussel, 1090 Brussel, Belgium.
Introduction: The LAT1 transport system is the only amino acid transport system currently
known to be capable of transporting D aromatic amino acids. Furthermore, it was shown that the
over-expression of LAT transport systems and the related reversible uptake of amino acid
analogues depended on the proliferation rate of human glioma cells. This urged us to evaluate the
tumour uptake of the D-analogue of [123I]-2-Iodo-L-Tyrosine, which has recently been
introduced into clinical trials. The uptake of 2-Amino-3-(4-hydroxy-2-[123/125I]iodophenyl) - Dpropanoic acid (2-Iodo-D-Tyrosine) was studied in vitro in LAT1 expressing R1M rat
rhabdomyosarcoma cells and in vivo in R1M tumour bearing Wag/Rij rats.Methods: In vitro, the
uptake of [125I]-2-Iodo-L-Tyrosine and [125I]-2-Iodo-D-Tyrosine into R1M cells was determined
in appropriate buffers allowing to study the involved transport systems and determining affinity
values. In vivo the biodistribution in R1M bearing rats (n=4) of [123I]-2-Iodo-L-Tyrosine and
[123I]-2-Iodo-D-Tyrosine was studied by both dynamic planar imaging with a gamma
camera.Results: In in vitro conditions [125I]-2-Iodo-L-Tyrosine and [125I]-2-Iodo-D-Tyrosine
were taken up reversibly in the R1M cells solely by the LAT system with an apparent
accumulation of the D analogue. None of the isomers were incorporated into cell proteins. The
Km value of 2-I-L-Tyr was about 108 µM, while it was 266 µM for 2-I-D-Tyr. DPI showed that
the uptake in the tumours of both [123I]-2-Iodo-D-Tyrosine and [123I]-2-Iodo-L-Tyrosine at 30
minutes p.i. reached mean DUR values of respectively 2.5 ± 0.7 and 1.7 ± 0.6 but with no
significant difference between their tumour / muscle ratio, which was about 4. Both analogues
cleared from the blood to the bladder, with a significantly faster clearance of the D analogue was.
After 30 minutes the activity in the bladder due to the D-form was 6 times higher.Conclusions:
[125I]-2-Iodo-D-Tyrosine in vitro and [123I]-2-Iodo-D-Tyrosine in vivo is reversibly taken up in
R1M tumour cells by the LAT transport system. In the rats the clearance from the blood of [123I]2-Iodo-D-Tyrosine is faster than of the L-analogue resulting in a slightly lower tumour uptake
but a same tumour to background ratio, making [123I]-2-Iodo-D-Tyrosine an interesting
alternative for tumour diagnosis with SPECT.
401 — Sunday, October 1, 02:30 pm - 04:00 pm, Friends of Music Hall
CME 3: Drug Development: Molecular Imaging for
Drug Development
Quality Requirements of Tracers for Drug Development
Alfons Verbruggen (BE)
the MUC1 glycoprotein as a tumour marker on epithelial cancer cells. After ten rounds of
selection and amplification, aptamers were cloned and sequenced, identifying molecules that
bound to MUC1 with a Kd of 0.3nm and high specificity. Post-SELEX modifications at the 2’
position of the sugar at the oligonucleotide ends conferred nuclease resistance and coupling
potential. We now report the coupling and preliminary pre-clinical imaging of these aptamers,
using standard peptide coupling reactions, to commercially available chelators such as MAG3,
MAG2 and DMSA for their radiolabelling with Tc-99 or Re-188, in tumour imaging or targeted
radiotherapy approaches. Additionally, we have synthesised novel ligands for our aptamers that
can utilise their inherent properties and offer improved labelling and imaging properties.
Multiaptamer conjugates as well as coupling to PEG have offered increased molecular weight to
improve the pharmacokinetic properties and therapeutic potential. Use of lanthanides for
labelling offers the possibility to interchange radioligands, such as samarium, with fluorescent
moieties, europium and terbium-for use in diagnostic assays, or gadolinium for MRI and can
allow dual functionalities on the same aptamer molecule.Aptamers have shown great potential
for tumour imaging and targeted radiotherapy in experimental models and are currently under
development as novel targeted radiopharmaceuticals.
Drug Development for the CNS using Imaging
Christer Halldin (SE)
Drug Development for Targets outside the Brain using
Piero A. Salvadori (IT)
402 — Sunday, October 01, 2006, 2:30 pm - 4:00 pm, Trianti Hall
myocardial perfusion defects were seen at inferior wall in both male and female patients on MPS
images. Second mostly observed perfusion defects in women were anterior wall and septum
while in men apex and anterior wall defects were more prominent. The difference between the
mean values of DN in diabetic and nondiabetic groups was statistically significant (5.66±3.14
and 4.51±2.90 respectively. p=0.041). There was also significant difference between mean DN in
patients with normal (•50%) and low (< 50%) EF values, (4.78±3.44 and 7.85±3.24, respectively
p=0.007).Conclusions: A high prevalence of myocardial perfusion defects was observed in our
patients with syndrome X. All of these perfusion abnormalities were reversible. We conclude that
presence of perfusion defects on MPS is an important factor and it is possible to identify
subgroup of patients at high risk of developing cardiac problems over years in cardiac syndrome
Value of myocardial perfusion scintigraphy in pediatric
patients with hyperlipidemia
Cardiovascular: Myocardial Perfusion
P. O. Kiratli, M. Caglar, I. Erdogan, S. Ozkutlu;
Ankara, Turkey.
Purpose: Cardiovascular disease (CVD) represents an immense public health burden with a
mortality of 15 million/year. Studies have shown that hypercholesterolemia is a risk factor for
CVD and there is evidence that atherosclerosis begins in childhood so that CVD risk factors track
from childhood into adulthood. A considerable amount of work in the literature on adults show
how valuable is myocardial perfusion scintigraphy (MPS) on adults, however this information is
lacking among pediatric patients. The purpose of this work is to assess the role of MPS in the
evaluation of CVD in pediatric patients.Materials and Methods: Out of 110 pediatric patients
who were referred to our clinic for MPS with various reasons, 17 patients with the diagnosis of
hypercholesterolemia, either familial or due to eating habits, were gathered. The patients (13
boys and 4 girls) were between 7-17 years of age and all were questioned for their family history,
cigarette smoking, hypertension, and diabetes mellitus apart from their blood lipid values. All
patients had an echocardiography and baseline ECG. All underwent stress gated MPS with Tc99m MIBI and information from their following work-ups were recorded.Results: Twelve
patients had normal gated MPS while 4 patients showed ischemia on MPS and one patient with
normal perfusion showed hypokinesia on gated study. All 5 patients with either abnormality on
MPS underwent angiography and the results were concordant with their MPS. Three patients had
coronary by-pass and two were followed with medical treatment. On the other side 3 patients
with normal MPS who had high clinical suspicion underwent angiography, 2 patients showed
normal results but one had obstruction in LAD on BT angiography and underwent coronary bypass surgery. Discussion and conclusion: In this rather small study group, we observed that gated
MPS enabled the clinicians to choose which patient to have angiography proficiently. The only
patient who was normal on MPS but had an obstruction in one of his vessels could be due the
time interval between the two studies (17 months) and progress in the patient. We believe that
MPS is a relatively noninvasive technique and easily applicable to pediatric population so it
should have a role in the evaluation of patients for hyperlipidemia.
Impaired coronary microvascular function in patients with
chest pain and normal coronary arteries
G. Storto1, A. R. Sorrentino1, T. Pellegrino1, R. Liuzzi1, M. P. Petretta2, M.
Salvatore1, A. Cuocolo1; 1Department of Biomorphological and Functional
Sciences, Institute of Biostructures and Bioimages of the National Council
of Research, University Federico II, Napoli, Italy, 2Department of Internal
Medicine, Cardiovascular and Immunological Sciences, University Federico
II, Napoli, Italy.
Aim: Coronary angiography may reveal normal or near normal epicardial coronary arteries in
many patients with angina. This study was designed to assess the relationship between coronary
microvascular function and myocardial blood flow (MBF) in patients with chest pain, positive
exercise stress test and normal coronary vessels.Materials and Methods: Thirty-five consecutive
patients (25 men; mean age, 59±14 years) with chest pain and normal coronary epicardial vessels
at angiography were studied. All patients had positive exercise electrocardiographic stress test
and 8 patients had reversible perfusion defects at stress SPECT imaging. Twelve age- and sexmatched subjects (9 men; mean age, 59±10 years) with chest pain, negative stress test and normal
coronary angiography represented the control group. These subjects were referred to our
laboratory for functional evaluation and the decision to submit them to coronary angiography
was made by referring physicians. All patients and controls underwent dipyridamole (0.74
mg/kg/body weight) Tc-99m sestamibi imaging. Myocardial blood flow (MBF) was estimated by
measuring first transit counts in pulmonary artery and myocardial counts from SPECT images.
On a separate day, the same acquisition protocol was repeated under resting conditions.
Estimated coronary flow reserve (CFR) was expressed as the ratio MBF stress/MBF rest.
Coronary vascular resistances (CVR) were calculated as the ratio between mean arterial pressure
and MBF.Results: At baseline, heart rate and blood pressure were not different in controls and
patients. Dipyridamole induced an increase (P<0.01) in heart rate, while systolic and diastolic
blood pressure significantly decreased (P<0.01) in both controls and patients. Mean estimated
CFR was 2.40±0.3 in controls and 1.54±0.3 in patients (P<0.0001). Individual data analysis
showed that all the 35 patients had CFR values below the 25th percentile (i.e. 2.1) of controls. In
controls, MBF at baseline was 0.98±0.4 counts/pixel/sec, and it was significantly lower (P<0.02)
as compared to patients (1.30±0.3 counts/pixel/sec). Stress MBF, was not significantly different
between controls (2.34±0.8 counts/pixel/sec) and patients (2.01±0.7 counts/pixel/sec). No
significant relationship between MBF at baseline and CFR was found (n=47, r=0.26, P=NS). At
baseline, CVR values were significantly lower (P<0.01) in patients as compared to controls.
Dipyridamole-induced changes in CVR were significantly greater in controls as compared to
patients (63% vs 35%, P<0.0001). A significant correlation between dipyridamole-induced
changes in CVR and estimated CFR was observed (n=47, r=0.88, P<0.0001).Conclusions:
Coronary microvascular dysfunction may sustain ischemic-related symptoms in patients with
normal coronary arteries by increased baseline MBF and preserved, but insufficient, hyperemic
Tc-99m MIBI myocard perfusion SPECT findings in patients
with cardiac syndrome X
Ö. Ömür1, Z. Burak1, A. Sa÷can2; 1Ege University Medical Faculty,
Department of Nuclear Medicine, øzmir, Turkey, 2Kent Hospital, Department
of Cardiology, øzmir, Turkey.
AIM: The term cardiac syndrome X is used to describe patients with anginal chest pain, a
positive exercise test and a normal coronary angiography. The aim of this study was to determine
the regional distribution of myocardial perfusion abnormalities and defect size using Tc-99m
MIBI myocardial perfusion scintigraphy (MPS) in patients with anginal chest pain, positive
exercise test and angiographically normal coronary arteries.Methods: Eighty-one (81) patients
(40 male, 41 female) who have angina pectoris, positive stress ECG and normal coronary
angiography were included this study. Clinical data of these patients who underwent Tc-99m
MIBI MPS were retrospectively reviewed. All of these patients had basal ECG and rest
echocardiography and they were evaluated for cardiac risk factors. Mean age was 55.2±7.05. For
MPS imaging, two days stress-rest Tc-99m MIBI protocol was used. SPECT data were evaluated
both visually and semi-quantitatively. For the semi-quantitative analysis, 20 segments
myocardial model was used. Number of abnormal segments was defined as the defect number
(DN).Results: There were 65 (80.3%) patients with abnormal MPS and all of these perfusion
abnormalities were reversible defects. Frequency of smoking, hyperlipidemia and diabetes
mellitus was significantly higher in patients with abnormal MPS than normal MPS.
Postmenopausal women proportion was also higher in MPS positive group. Ischemic findings
were detected by exercise stress ECG in all patients. Myocardial hypokinesia was determined on
various segments with rest echocardiography in 17 (21%) patients in whom reversible perfusion
defects were observed in MPS. Seven patients had low LV EF values (<50%). Most of the
Hacettepe University,
Is there any correlation between the severity of anxiety and
myocardial perfusion abnormalities detected by SPECT
K. Ansari-Gilani1, B. Fallahi1, J. Modaresi-Esfeh2, D. Beiki1, A. FardEsfahani1; 1Research Institute For Nuclear Medicine, Tehran University of
Medical Sciences, Tehran, Iran (Islamic Republic of), 2Rasoul-e Akram
Hospital, Iran University of Medical Sciences, Tehran, Iran (Islamic
Republic of).
Aim: It has been shown that type A personality is associated with a higher frequency of ischemic
heart disease (IHD). Anxiety is a main feature of this type of personality. The goal of this study
was to find out if there is any correlation between the severity of anxiety and the severity of
myocardial perfusion abnormality.Materials and Methods: One hundred sixty seven patients with
low to intermediate pre-test probability of coronary artery disease who were referred for
myocardial perfusion imaging were initially assessed with Beck Anxiety Inventory (BAI) and
then underwent a standard rest/exercise 99mTc-sestamibi SPECT. Those with history of
psychiatric illnesses were excluded from the study. Visual interpretation as well as semiquantitative assessment of perfusion score in a five point scaling method for standard 20-segment
model (total perfusion score=100) were used. Spearman's correlation coefficient procedure and
one-way analysis of variances were used for statistical assessment.Results: One hundred sixty
seven patients (71 male, 96 female, 56.6+/- 11.2 years) were entered in the study. When all
patients with normal and abnormal scintigraphy were considered, no significant correlation was
found between perfusion score and the patients' BAI (r= 0.035, p=0.65) . Considering only the
patients with a significant perfusion abnormality (perfusion score less than 90, n=38), there was a
statistically significant correlation between the perfusion score and BAI (r= - 0.598, p=0.013).
Also it was noted that BAI was significantly higher in patients who had perfusion abnormality in
the territory of three main coronary arteries (24.4+/- 5.6) as compared with normal cases
(14.66+/-1.28) and those who had perfusion abnormality in the territory of 1 and 2 coronary
arteries (14.35+/-2.43, 15.22+/-3.05, respectively, P< 0.099). Conclusions: Patients with more
severe anxiety seem to have more extent of perfusion abnormality detected by SPECT imaging.
Also patients with perfusion abnormality in the three-vessel territory have a higher degree of
anxiety as compared with those with perfusion abnormalities in single or double vessel
Relationship between the severity of depression and
myocardial perfusion abnormality using SPECT imaging
K. Ansari-Gilani1, J. Modaresi-Esfeh2, B. Fallahi1, F. Pooyafard3;
Research Institute For Nuclear Medicine, Tehran University of Medical
Sciences, Tehran, Iran (Islamic Republic of), 2Rasoul-e Akram Hospital,
Aim: The previous studies have shown that depression may increase the likelihood of ischemic
heart disease (IHD) even in the absence of other major risk factors. Depression also may worsen
the prognosis of patients who already have IHD. On the other hand chronic IHD can result in
deprressive mood or intensify already existing depression.The goal of this study was to find out
if there is any correlation between the severity of depression and the severity of myocardial
perfusion abnormality.Materials and Methods: One hundred sixty seven patients who were
referred for myocardial perfusion imaging were assessed with Beck Depression Inventory (BDI).
Those with history of psychiatric illnesses were excluded from the study. Patients underwent
rest/exercise 99mTc-sestamibi SPECT. Visual interpretation as well as semi-quantitative
assessment of perfusion score in a five point scaling method for standard 20-segment model
(total perfusion score=100) were used.Results: One hundred sixty seven patients (71 male, 96
female) with mean age of 56.6+/- 11.2 years were entered the study. No significant correlation
was found between perfusion score and the patients BDI (r= -0.013, p=0.865) when all patients
(with normal and abnormal scintigraphy) were considered. Considering only the patients with a
significant perfusion abnormality (perfusion score less than 90, n=38), there was a statistically
significant correlation between the perfusion score and BDI (r= - 0.585, p=0.017). It was also
shown that in patients with normal or near normal perfusion score (more than 95), BDI was
considerably higher in patients with typical (16.17) or atypical (15.98) chest pain as compared
with those who had no chest pain (11.70) (typical chest pain vs. no chest pain, p=0.099; atypical
chest pain vs. no chest pain, p=0.095).Conclusions: It seems that in patients with significant
perfusion defects, the more severe is the abnormality, the higher is the degree of depression. It is
also shown that chest pain perception in patients with normal or near normal myocardial
perfusion score is associated with a higher BDI and this possibly shows a psychological origin
for pain perception in these subgroups of patients.
Quantitative gated myocardial perfusion SPECT in the
assessment of residual ischemia and induced myocardial
damage after Off-pump and On-pump coronary-artery bypass
P. Varrella1, P. Ferrara1, F. Triumbari1, G. Peluso1, V. Rizzo1, F.
Manganelli1, A. Martino1, R. Sauro1, L. Mansi2, C. Lombardi1, P. Miletto1, M.
Spadafora1; 1A.O.R.N. Moscati, Avellino, Italy, 2Sun, Naples, Italy.
Aim. Off-pump coronary revascularization has become a widely used technique during recent
years. Gated-SPECT appears to be the best available measurement tool to evaluate both the
degree of coronary reserve in myocardial reperfused territories, and left ventricular function. The
aim of the study was to compare residual ischemia and induced myocardial damage after Offpump and On-pump coronary-artery bypass surgery (CABG), by gated-SPECT. Material and
methods. We studied 104 consecutive subjects: 55 (mean age 63.9; 47 M and 8 F; 24 without
previous myocardial infarction) and 49 patients (mean age 62.4; 43 M and 6 F; 30 without
previous myocardial infarction) underwent On-pump and Off-pump surgery, respectively. All
patients performed stress-rest gated-SPECT with Tc-99m tetrofosmin or sestamibi, 6 to 9 months
after bypass surgery. The effects of revascularization were evaluated by a quantitative assessment
of both residual ischemia (SDS) and induced scar (SRS). We performed the analysis of SDS in
the entire revascularized myocardium as well as in single territory related to coronary arteries
bypassed. A 5-point scoring system (0, normal, to 4, absence of uptake) in a 17-segment model
was used. In each patient a SDS and SRS values >3 were considered suggestive of ischemia and
scar, respectively. Results. The On-pump group received a mean of 3.4 grafts and the Off-pump
2.1 (P=<0.001). No significant difference in LV ejection fraction between the two groups
(54.59+13.3% versus 56.78+10.2%) was found. In all revascularized myocardial segments, the
SDS in the On-pump versus Off-pump group was 2.49+3.29 and 1.45+2.24 respectively (N.S.).
The percentage of patients with residual ischemia (SDS>3) was 24% in On-pump and 9% in Offpump patients (N.S.). The SDS for left anterior descending, circumflex and right coronary artery
of On-pump versus Off-pump was 0.95+1.98 versus 0.70+1.29, 0.93+1.8 versus 0.92+1.6 and
1.04+1.4 versus 1.08+1.73, respectively (N.S.). In patients without previous myocardial
infarction, no significant differences in SRS between the two groups was observed. However, the
On-pump group had 8 scar (33%) in comparison of 2 necrosis (7%) of the Off-pump group
(P<0.05). Conclusion. Our study suggests that there is no difference between Off-pump and Onpump coronary-artery bypass surgery in completeness of revascularization. However Off-pump
leads to less myocardial damage.
An assessment of common stress myocardial perfusion
scoring systems in the determination of advanced coronary
artery disease
E. Moralidis1, G. Arsos1, T. Spyridonidis2;
Hippokration Hospital,
Thessaloniki, Greece, 2Hippokration Medical Center, Larissa, Greece.
Aim: Limited data exist on the comparative performance of various myocardial perfusion
scoring systems for diagnostic or prognostic purposes. This study investigates the efficacy of two
commonly used approaches for the assessment of myocardial perfusion at stress in the
determination of advanced coronary artery disease (CAD) in patients submitted to coronary
angiography.Methods: 398 patients with suspected or known CAD, but no coronary
interventions, referred for routine exercise SPECT myocardial perfusion imaging and submitted
to coronary angiography within 3 months from scan date, were enrolled. In order to minimize the
bias inherent in assessing patients referred for coronary angiography, the population was
separated into a derivation (n=217) and a validation group (n=181), according to scan date. Nonsignificant coronary lesions (<50% diameter stenosis) and 1-vessel disease, other than proximal
LAD, were classified as limited CAD (n=156). A proximal LAD lesion, multiple vessels with •
50% luminal narrowing and LM stem disease were considered advanced CAD (n=242).
Myocardium was divided into 20 segments and regional Tl-201 uptake at stress was graded. The
sum of ischemic area (SIA, number of ischemic segments) and the sum stress score (SSS,
number of ischemic segments together with the degree of regional tracer uptake) were calculated.
ROC analysis (area under curve (AUC)) was used to compare scoring systems and define
thresholds of abnormality (the cut-off value providing the best sensitivity at a specificity of
•90%).Results: In the derivation group SSS (AUC 0.679, SEE 0.038) was not a better predictor
of advanced CAD than SIA (AUC 0.716, SEE 0.037, p=ns). At the defined thresholds of
abnormality, SSS•25 and SIA•13, sensitivity and specificity in the detection of advanced CAD
were 15% vs 18% and 91% vs 91%, respectively. In the identification of advanced CAD in the
validation cohort, SSS (AUC 0.738, SEE 0.038) had not a better correct ranking probability than
SIA (AUC 0.778, SEE 0.035, p=ns). The previously defined thresholds of abnormality yielded a
sensitivity of 16% vs 14% and a specificity of 90% vs 90% for SSS and SIA, respectively.
Sensitivity and specificity were not significantly different in within and between group
comparisons.Conclusions: In the detection of advanced CAD with myocardial perfusion
scintigraphy, a simple estimation of the area of ischemic myocardium at stress provides similar
diagnostic accuracy to a composite score, incorporating both the area of myocardial ischemia and
the degree of regional perfusion impairment at stress (SSS).
Rest myocardial perfusion imaging as a powerful tool to
evaluate myocardial infarction at Chest Pain Unit.
G. B. Barbirato, P. L. Correa, R. C. M. Felix, J. C. Azevedo, A. Volschan,
M. Viegas, L. Pimenta, H. F. R. Dohman, E. T. Mesquita, F. M.
Casagrande, D. Borges, C. C. Oliveira, D. A. Avila, C. T. Mesquita; PróCardíaco Hospital - Rio de Janeiro - Brazil, Rio de Janeiro, Brazil.
BACKGROUND: Acute rest myocardial perfusion imaging (MPI) have been evaluated
diagnostically in patients with chest pain in the emergency department (ED). We investigated
patients admitted with chest pain or until four hours of symptoms resolution using acute
technetium-99m sestamibi/tetrofosmin rest MPI at Chest Pain Unit and compared to serum
markers. METHODS: Serum markers and acute technetium-99m sestamibi/tetrofosmin rest MPI
were obtained in 108 consecutive patients admitted to the ED with chest pain and nondiagnostic
electrocardiograms. Patients with past myocardial infarction were not excluded (24 patients).
Venous samples were drawn at admission and 6 hours later for cardiac troponin I. Nuclear
physicians performed blinded image interpretation. Myocardial infarction ( MI ) was confirmed
by elevation in troponin I more than three times control. RESULTS: Acute rest MPI results were
abnormal in all 6 patients with MI. Only 1 patient showed normal MPI and elevated serum
markers. An additional 55 patients had positive MPI without MI and 46 patients with both serum
markers and MPI negative. The prevalence of disease was 6,5%. Sensitivity of acute rest MPI for
objective evidence of MI was 85,7% and specificity of 45,5%. The negative predictive value was
97,7%. CONCLUSION: Patients submitted to chest pain protocol with MPI demonstrated an
excellent negative predictive value. Patients with normal scan are a group at lower risk. These
results suggest that acute rest MPI is a useful tool in patients in Chest Pain Unit.
403 — Sunday, October 01, 2006, 2:30 pm - 4:00 pm, Banqueting Hall
Oncology: Lung Cancer
FDG uptake and glucose transporter type 1 expression in
lymph nodes of non-small-cell lung cancer
W. Lee, J. Chung, S. Park, S. Sung, Y. Kim, D. Lee, J. Chung, M. Lee, S.
Kim; Seoul National University College of Medicine, Seongnam, Republic
of Korea.
Aim: FDG uptake in NSCLC is related to glucose transporter type 1 (Glut-1) expression.
However, the direct causal relationship between Glut-1 expression and FDG uptake has not been
clarified. Here, we investigated the correlation between FDG uptake and Glut-1 expression to
determine the role of Glut-1 in FDG uptake by malignant and benign lymph nodes (LNs).
Materials and Methods: Fifty-five curative lung resections in 53 NSCLC patients
(male:female=36:17, age=62.0± 11.8 years) were included in this study. Maximum standardized
uptake values (maxSUV) of LNs in preoperative whole body FDG-PET and Glut-1
immunostaining results were compared. Results: Of 316 pathologically confirmed LNs, 12.3%
(39/316) were malignant, and in malignant LNs, FDG positive LNs were no different from FDG
negative LNs in terms of size (15.0±6.7mm vs 10.0±6.1mm, p>0.05), or in terms of the
proportion of LNs occupied by tumor (60.0±28.8% vs 39.2±38.4%, p>0.05), but had greater
percentages of Glut-1 positive cells in tumors (74.1±31.8% vs 22.7±18.7%, p<0.01), and Glut-1
staining intensities (3.4±0.9 vs 1.8±1.3, p<0.01). FDG negative malignant LNs featured
cytoplasmic Glut-1 expression and adenocarcinoma. Glut-1 staining intensities were found to be
significantly correlated with the maxSUVs of malignant LNs (rho=0.516, p<0.05), but the
percentages of Glut-1 positive cells in tumors were not (r=0.2072, p>0.05). Analysis of FDG
positive benign LNs showed that maxSUV was not correlated with the degree of follicular
hyperplasia, or Glut-1 expression (p>0.05).
Conclusions: Intense Glut-1 immunoreactivity was found to be proportionally related to the
degree of FDG uptake by malignant LNs in NSCLC. However, the finding that Glut-1 expression
in lymphoid follicular hyperplasia showed no correlation with FDG uptake in benign LNs
requires further investigation.
measurements for discriminating benign from malignant
pulmonary lesions
K. J. Nichols, S. S. Tuli, G. G. Tronco, J. N. Rini, M. B. Tomas, C. J.
Palestro; Long Island Jewish Medical Center, New Hyde Park, NY, NY,
United States.
Iran University of Medical Sciences, Tehran, Iran (Islamic Republic of),
Roozbeh Hospital, Tehran University of Medical Sciences, Tehran, Iran
(Islamic Republic of).
Aims: Some investigators have advocated normalizing target counts to background counts for
discriminating benign from malignant lesions. This has the advantage of reducing intra-patient
variability of quantitative FDG-PET, but reproducibility may be compromised because of the
necessity to identify a background region as well as a target region. This investigation was
undertaken to determine the reproducibility of discriminating benign from malignant pulmonary
lesions using target-to-background ratios.Materials and Methods: Early (within 1 hour after
injection) and late (3.3±0.04 hours after injection) FDG-PET scans were performed on 105
patients referred for characterization of “indeterminate” pulmonary lesions identified on CT. For
quantitative analysis, maximum lesion counts were determined at target sites identified by CT
and normalized to background lung counts. Analyses were performed on early and late scans,
and changes between scans. Two observers performed analyses independently of one another,
and without knowledge of pathology results or clinical information other than the CT site of the
suspected target region. One observer repeated measurements without referring to his previous
results or those of the other observer. ROC analysis was used to assess accuracy and generate
thresholds for abnormality. All patients had histopathologic confirmation of final
diagnoses.Results: There were 68 malignant and 37 benign lesions. Accuracy for discriminating
benign from malignant lesions was not different for the two observers for early scans (91±3%
and 87±4%) versus late scans (92±3% and 89±3%, p=NS), but these values were significantly
higher than for changes between scans (76±5% and 66±6%, p<0.0001). For early scans, late
scans and changes between scans, inter-rater agreement for quantitation was kappa = 0.68
(“good”), 0.71 (“good”) and 0.21 (“fair’), respectively, while intra-observer reproducibility was
kappa = 0.73 (“good”), 0.77 (“good”) and 0.37 (“fair”), respectively, (Altman DG, 1991). The
only statistically significant disagreement between observers occurred for quantitative evaluation
of changes between scans (McNemar test p=0.003). Accuracy for discriminating benign from
malignant lesions was highest for late scans, as were measures of inter-observer agreement and
intra-observer reproducibility.Conclusions: For highest accuracy and reproducibility, quantitative
evaluations of pulmonary lesions should be performed for late studies analyzed alone.
FDG-PET-CT in the evaluation of pulmonary nodules: our
experience in 160 patients
F. Giacomuzzi , E. H. Alghriw , F. Bonutti , A. Cecotti , M. Casco , M.
Rocco1, A. Modesto1, A. Follador1, F. Zanconati2, O. Geatti1; 1Azienda
Ospedaliera S. Maria della Misericordia, Udine, Italy, 2Azienda Ospedaliera
Universitaria Ospedali Riuniti, Trieste, Italy.
Aim: incidentally discovered pulmonary nodules usually require histological diagnosis;
nevertheless biopsy is not always feasible for technical or patient-related reasons and sometimes
may yield indeterminate results as well. The purpose of this study was to evaluate if FDG-PETCT can have a role in this differential diagnosis.Materials and Methods: after excluding patients
with a prior history of carcinoma and/or granulomatous disease, we studied with FDG-PET-CT a
series of 160 patients (112 males and 48 females; mean age: 67 years), with newly incidentally
discovered pulmonary nodules, following unsuccessful or indeterminate biopsy or, in a lesser
number of cases, when an attempt at biopsy was considered at too high risk for the patient. This
retrospective study included 131 patients with solitary lung lesions and 29 bearing multiple
nodules. Final diagnosis for patients with positive PET-CT or lesions growth at sequential
radiological examinations was made by histology. The absence of a malignant disease was
assumed for patients with negative PET-CT, if serial CT scans for at least 12 months did not
demonstrate any growth of lesion size.Results: when max-SUV was below 2.5, the lesion was
considered to be a benign process and, based on this criterion, we got 75 true positive and 67 true
negative. The average max SUV was 8.2 ± 4.9 and 1.7 ± 1.3 for benign and malignant lesions
respectively. Overall sensitivity, specificity, accuracy, positive and negative predictive values
were 93% (75/81; 95% CI: 0.87-0.98), 85% (67/79; 95% CI: 0.77-0.93), 89% (142/160; 95% CI:
0.84-0.94), 86% (75/87; 95% CI: 0.79-0.93) and 92% (67/73; 95% CI: 0.85-0.98) respectively.
We had 12 false positive studies, due to a moderate/high FDG-uptake in 8 inflammatory
conditions (5 of which confirmed by histology and 3 by stability of lesion size on CT), in 3
squamous metaplasias and in 1 chondrohamartoma and 6 false negative studies, caused by a
weak FDG uptake in low-grade malignant diseases (2 well-differentiated adenocarcinomas, 1
well-differentiated squamous carcinoma, 1 carcinoid and 2 lymphomas). Cconclusion: FDGPET-CT can be considered a highly accurate “in vivo” metabolic biopsy for the characterization
of pulmonary nodules and may be particularly helpful when a tissue diagnosis cannot be
obtained. In fact a negative study virtually excludes malignancy while a high FDG uptake likely
correlates to malignancy and requires further investigations.
PET / CT in the characterization of solitary lung lesions
S. Kim, M. Allen-Auerbach, M. Dahlbom, J. Goldin, M. Brown, M. E. Phelps,
J. Czernin, C. Schiepers; David Geffen School of Medicine at UCLA, Los
Angeles, CA, United States.
Aim: PET using FDG is a well-established indication for characterization of a solitary pulmonary
nodule. The accuracy of dual modality imaging with PET/CT was investigated, and compared to
the pathological diagnosis.Methods: Studies from Sept 2002 through Sept 2005 were selected.
Sixty patients with a single lung lesion, age 48-84 yr, were available with pathological
confirmation by biopsy or resection. Upper body PET/CT was performed with a REVEAL
system. 55 patients had a non-contrast CT and 5 a CT with i.v. contrast. FDG dose was 7.75
MBq/kg. A shallow-breathing protocol was used for both CT and PET. Lung cancer was
diagnosed in 63%, pulmonary metastasis in 5%, and benign disease in 32%. Blinded
interpretation was performed by 1 radiologist for CT and 2 nuclear medicine physicians for PET
and consensus for PET/CT. Lesions were analyzed by localization, FDG-uptake, texture,
margins, axial dimension, spiculation, and calcification. The data were dichotomized and the
final interpretation scored as benign or malignant.Results: Lesion diameter on CT varied from 730 mm, mean 18 mm. Lesion density varied from -11 to +36 HU, mean +23, compared to a
background of -740 HU. Maximum SUV was 0.5-17 compared to an average background of 0.5
SUV. Comparison of CT vs PET vs PET/CT provided the percentages as shown in the Table. On
PET/CT, there were 6 false positives (all inflammatory lesions), and one false negative
(adenocarcinoma with BAC features of 13 mm). Analysis of lesions less than 10 mm (n=10)
revealed a reduction of PET sensitivity to 50% and CT specificity to 20%. Visual analysis
furnished the same results as semi-quantitative analysis using a cut-off of 2.0 SUV.Conclusions:
PET/CT performance in classifying pulmonary lesions as benign or malignant is excellent. Dual
modality imaging with PET/CT enhances the sensitivity of CT and maintains the specificity of
PET. Visual analysis was sufficient, semi-quantitative analysis with SUV did not change the
interpretation. This technique permits 3D-volume estimates (JNM 2003;8:S285P) and corrections
for partial volume, which may be important for quantitative evaluations in time (wait & watch
regimen), or monitoring therapy.
FDG PET imaging and partial volume correction in the
assessment of pulmonary nodules
K. M. Alkhawaldeh, S. Dadparvar, G. Bural, M. Houseni, G. El-Haddad, A.
Alavi; Hospital of the university of Pennsylvania, Philadelphia, PA, United
FDG PET imaging with SUV>2.5 is well known criterion in the assessment of pulmonary
nodules. Calculated SUV in small lung nodules is thought to be underestimated due to partial
volume effect. The purpose of this study was to evaluate the effect of partial volume correction
of SUVs on diagnostic accuracy of FDG PET imaging.Methods: Two hundred sixty five patients
were included in this study (161men, 104 women, age range: 41-92 years). All patients had
pulmonary nodules detected by computed tomography (CT), and the diagnosis was confirmed by
biopsy or by follow up CT scanning. All patients underwent whole body FDG PET scanning one
hour after intravenous injection of FDG. The maximum standardized uptake values (SUVs) were
calculated for all FDG avid lung nodules. Partial volume correction of calculated SUVs was
applied according to the literature. SUV of > 2.5 was considered as criterion for malignancy with
and without partial volume correction . The results were compared to the biopsy findings and
clinical follow up.Results: Biopsy and follow-up revealed 72 patients with malignant lung
nodules, whereas 193 patients had benign nodules. When considering SUV > 2.5 (without partial
volume correction) as cutoff criterion for malignancy, sensitivity, specificity and accuracy were;
63%, 92% and 85% respectively. After partial volume correction; sensitivity, specificity, and
accuracy were 90%, 80%, and 83% respectively.Conclusions: On FDG PET imaging of
pulmonary nodules, partial volume correction of SUVs have significantly higher sensitivity, with
some loss in specificity, and no significant change in overall accuracy.
The comparison of the diagnostic value of FDG PET/CT vs CT
in the nodal staging of non-small cell lung cancer
Y. M. Chen, G. Huang, X. G. Sun, J. J. Liu, Y. P. Shi, L. R. Wan; Renji
Hospital, Shanghai Jiaotong University, Shanghai, China.
Objective To compare the diagnostic value of PET-CT versus CT in the nodal staging of nonsmall cell lung cancer(NSCLC). Methods FDG PET-CT scanning was performed in 30 NSCLC
patients all of which had surgical records in details and definite post-surgical histopathologic
results before therapy. Each imaging study was interpreted by 4 doctors with a lot of PET-CT
working experience: In CT imaging , Lymph nodes were considered as malignant if they
exceeded 1cm in short-axis diameter; In PET-CT imaging, if the focus of increase FDG uptake in
lymph nodes were above the intensity of the mediastinal blood activity and SUVave>2.5, they
were defined malignant. Nodal stations were localized according to the American Thoracic
Society (ATS) mapping system. PET-CT and CT findings were compared on a station-by-station
basis, which were all correlated with histopathologic results. Results Of the 30 NSCLC patients,
15 had adenocarcinoma, 9 had squamous cell carcinoma, 2 had adeno-squamous mixed
carcinoma, 1 had large cell carcinoma, 1 had pleomorphic carcinoma. Overall 14 patients had
nodal metastases confirmed by histopathologic while 10 patients had mediastinal nodal
metastases. A total of 257 nodals from 115 stations were resected from 30 NSCLC. 50 were
intrapulmonary and hilar nodes(N1),63 were mediastinal nodes(N2),2 were supraclavicular
nodes(N3). Of 115 nodal stations, 32 were malignant and 83 were benign. The sensitivity,
specificity and accuracy of PET-CT and CT in total 115 nodal stations were 68.8%, 95.2%,
87.8% and 56.3%, 88.0%, 75.7%, respectively. Of 65 N2 and N3 stations, 21 were malignant and
44 were benign. The sensitivity, specificity and accuracy of PET-CT and CT in 65 N2 and N3
stations were 90.5%, 97.7, 95.0% and 57%, 81.8%, 60%. In assessing the nodal staging, PET-CT
is correct-staged in 23 cases , over-staged in 2 cases, under-staged in 5 cases while CT is 18, 6,
6,respectively. Conclusion FDG PET-CT has high diagnostic value in assessing nodal staging of
NSCLC which appears to be superior to CT.
Advantage of FDG-PET-CT Retention Index (RI-SUV) over
single point imaging, in the diagnosis of nodal involvement:
our experience in 102 NSCLC patients
F. Giacomuzzi1, E. H. Alghriw1, F. Bonutti1, A. Cecotti1, M. Casco1, M.
Rocco1, A. Modesto1, A. Follador1, F. Zanconati2, O. Geatti1; 1Azienda
Ospedaliera S. Maria della Misericordia, Udine, Italy, 2Azienda Ospedaliera
Universitaria Ospedali Riuniti, Trieste, Italy.
RI-SUV (%)
1-h SUV ( > 2.5)
1-h SUV ( > 3.0)
Accuracy 94%
Sensitivity 94%
Specificity 94%
Role of positron emission tomography (PET) in staging of
patients with non small cell lung carcinoma (NSCLC): our
F. Matteucci1, A. Moretti1, R. Galassi1, M. Monteverde1, G. Giorgetti1, L.
Lazzari Agli2, V. Pilotti3, D. Dell'Amore1, M. Bertocco1, G. Fiorentini1;
Morgagni Pierantoni Hospital, Forlì, Italy, 2Ceccarini Hospital, Riccione,
Italy, 3GVM, Cotignola (Ra), Italy.
Accurate staging of locoregional lymph node directs the management and the prognosis of
patients with NSCLC; the most important decision is between those patients who are candidates
for surgical resection and those who are judged to be unresectable for cure but will benefit from
chemoterapy, radiotherapy or both. CT has been the standard noninvasive staging method for the
mediastinum; enlarged lymph nodes were considered to be metastatic, but size is however a
unspecific criterion. PET has become an important cancer imaging tool, both for diagnosis and
staging, as well as offering prognostic information based on response on therapy. The aim of our
study is to evaluate the accuracy of FDG-PET-CT in mediastinal and hilar lymph nodes staging,
in NSCLC patients. We retrospectively studied 70 consecutive patients (45 males and 25
females; mean age 67.4 years), with proven NSCLC, who were referred for a whole-body PETCT (Biograph Sensation 16 Siemens) imaging in presurgical staging. all patients were fasted and
blood glucose level were evaluated before the injection i.v. of 370 MBq FDG, PET scan were
performed after 60 minutes from the injection with standar tecnique (iteractive reconstruction of
the images with attenuation weighted OSEM, slices of 3.4 mm). Using FDG-PET-CT, a well
definite increase in glucose uptake to a higher level than that in the surroinding tissue at
qualitative analysis was considered to characterize malignancy; on CT findings, lymph nodes
with a short axis diameter greater than 1 cm were considered to be metastatic. The reference
standard were represented by the histopathologic results of lymph node sampling procedures: by
mediastinoscopy (13) or surgical resection (57). Our data show a correct lymph-node stage in 63
pts (90.0%) with FDG-PET-TC; in 3 pts (4.3%), FDG-PET resulted false-negative while 4 pts
(5.7%) were overstaged. On the contrary, with CT we found a correct stage in 45 pts (64%),
while 7 pts (10%) show false-negative results and 18 pts (26%) resulted overstaged. The
sensitivity, specificity, accuracy, positive predictive value and negative predictive values of
FDG-PET-CT were 86%, 92%, 90%, 83% and 94 % respectively, while those with CT were
68%, 64%, 63%, 45%, 81% respectively. Our data show that FDG-PET-CT is useful in the
presurgical management of NSCLC patients and superior than CT in lymph-node staging. This
technique might reduce invasive surgical staging and might become the new standard approach
in presurgical management of patients with NSCLC.
404 — Sunday, October 1, 2006, 02:30 pm — 04:00 pm, Mitropoulos
CTE 2: Information Technology
Progress in IT : possibilities for NM to improve data
J. Rutten (NL)
Aim: purpose of the study was to evaluate the accuracy of RI-SUV in the diagnosis of nodal
involvement, comparing to single point imaging in NSCLC.Materials and Methods: 102
consecutive NSCLC patients (80 m and 32 f; ma: 69 y) referred for preoperative staging, were
enrolled, excluding bronchioalveolar carcinomas and patients who previously underwent any
therapeutic intervention. All mediastinal and hilar lymph nodes were visually analysed in the
early imaging (1h); the nodes showing an increased FDG accumulation with respect to the
adjacent mediastinal structures were considered to be positive and this qualitative evaluation
strictly correlated with a 2.5 SUV threshold. Moreover, to verify and compare accuracies, we
also reanalysed all the data using a 3.0 SUV threshold. A delayed 2h scan was then acquired and
nodal SUV values determined and RI-SUVs calculated according to the following equation: RISUV % = (2h SUVmax - 1h SUVmax) / 1h SUVmax %. In case of an increased 2h-SUV, as
compared to the 1h-SUV (positive RI-SUV), the node was considered to be malignant and final
diagnosis was based on histology (mediastinoscopy: 31/102; surgical resection: 71/102. We also
compared the accuracies yielded by 2.5 or 3 early SUV thresholds and RI-SUV.Results: in the
pathological nodes (49/102) 1h-SUV was 7.3 ± 5.3 and increased to 9.1 ± 7.2 at 2h (+ 20.0% ±
19.2%; P < 0.001); in the benign nodes (53/102) 1h-SUV was 2.1 ± 0.5 and decreased down to
1.7 ± 0.7 at 2h (-20.0% ± 18.9%; P < 0.001). Sensitivity, specificity, accuracy, PPV and NPV
values yielded by RI-SUV were 94%, 94%, 94%, 94% and 94% respectively. We got 3 FP, due
to inflammatory conditions and 3 FN, owing to limited spatial resolution of the equipment.
Single-time-point imaging with a 2.5 SUV-1h threshold, yielded 92%, 92%, 92%, 92% and 92%
respectively, while a 3 SUV-1h threshold improved specificity to 96%, but decreased sensitivity
and accuracy down to 84% and 90% respectively.Conclusions: our preliminary results suggest
RI-SUV can help distinguishing benign from malignant nodes, improving the accuracy of
mediastinal staging in patients with NSCLC.
Image reconstruction and fusion : possibilities and pitfalls
Andrew E. Todd-Pokropek (UK)
405 — Sunday, October 01, 2006, 2:30 pm - 4:00 pm, Scalcotas
Tc-Demogastrin 2 for CCK 2-receptor scintigraphy in
medullary thyroid carcinoma
A. Fröberg1, B. Nock2, T. Maina3, M. de Jong1, W. de Herder4, H. de Wilt5,
A. van der Lugt6, E. de Blois1, M. Verdijsseldonck1, H. Mäcke7, E.
Krenning1; 1Department of Nuclear Medicine, Erasmus MC, Rotterdam,
The Netherlands, 2Radiopharmacy Section, I/R-RP, NCSR "Demokritos",
Athens, Greece, 3Radiopharmacy Section, I/R-RP, NCSR, Athens, Greece,
Department of Internal Medicine, Erasmus MC, Rotterdam, The
Netherlands, 5Department of Surgery, Erasmus MC, Rotterdam, The
Netherlands, 6Department of Radiology, Erasmus MC, Rotterdam, The
Netherlands, 7Radiological Chemistry, University Hospital Basel, Basel,
Aim As 99mTc-N4-Gly-(D)Glu-(Glu)5-Ala-Tyr-Gly-Trp-Met-Asp-Phe-NH2 (99mTc-Demogastrin
2) had shown promising results in preclinical studies, we studied the metabolism and diagnostic
potential of this peptide in patients with evidence of recurrence or metastases of medullary
thyroid cancer after thyroidectomy. Patients and Methods Eleven patients (10 female, 1 male;
age 17- 71 y, median 44 y) were included in the study. Calcitonin levels on the day of
administration ranged from 0.23 - 131 µg/l (median 29 µg/l). Scintigraphy was performed at 3
and/or 6 and/or 24 h post injection (injected in 5 min) of 740 MBq 99mTc-Demogastrin 2
(corresponding to10 µg Demogastrin 2). Blood pressure and heart rate were monitored and
patients were asked for side effects. Results Side effects of the radiopharmaceutical were: raised
heart rate, flushes, some sensation of pressure on the chest, mild sickness, tingling (especially
hands). Heart rate never exceeded 105 bpm, the raise was experienced by only two patients. All
side effects disappeared within 10 min after administration. When comparison at different time
points was possible (n=9), most lesions were visible on the images +/- 6 h pi. Extra 24 h pi
images sometimes provided more certainty in interpretation. In 9 patients pathologic retention of
Tc-Demogastrin 2 was seen. In one patient known liver lesions of +/- 3 mm were equivocally
visible, only on SPECT. In one patient no pathological uptake was seen. Except for the small
liver metastases, all tumour lesions known before scintigraphy were seen on the scans. New
lesions were seen in 7 patients; in brain, neck, liver and bone. In each of these patients at least
one of the new lesions was confirmed by another imaging modality. In 2 patients (new) neck
lesions of 6 and 7 mm respectively were confirmed true positive by ultrasound guided fineneedle aspiration cytology. In the patient with a negative scintigraphy a 5 mm lesion in the neck
was discovered 6 months later by ultrasound. In-vitro autoradiography, performed
postoperatively, demonstrated that this medullary thyroid carcinoma metastasis did not have
CCK 2-receptors. Conclusions CCK 2-receptor imaging with 99mTc-Demogastrin 2 appears to be
a very promising diagnostic tool in patients with evidence of recurrence or metastases of
medullary thyroid cancer. Further studies have to prove its value in patient management.
Comparison of three radiolabelled peptides for CCK 2receptor scintigraphy in medullary thyroid carcinoma
A. Fröberg1, B. Nock2, T. Maina2, S. Good3, J. Erion4, M. Verdijsseldonck1,
W. de Herder5, A. van der Lugt6, A. van Gameren1, M. de Jong1, M. Béhé7,
E. Krenning1; 1Department of Nuclear Medicine, Erasmus MC, Rotterdam,
The Netherlands, 2Radiopharmacy Section, I/R-RP, NCSR "Demokritos",
Athens, Greece, 3Radiological Chemistry, University Hospital Basel, Basel,
Switzerland, 4Biosynthema Inc, Saint Louis, MO, United States,
Department of Internal Medicine, Erasmus MC, Rotterdam, The
Netherlands, 6Department of Radiology, Erasmus MC, Rotterdam, The
Netherlands, 7Department of Nuclear Medicine, Philipps-University
Marburg, Marburg, Germany.
Aim To study and compare the metabolism and diagnostic potential of 99mTc-N4-Gly-(D)Glu(Glu)5-Ala-Tyr-Gly-Trp-Met-Asp-Phe-NH2 (99mTc-Demogastrin 2), 111In-DOTA-(D)Glu-AlaTyr-Gly-Trp-Met-Asp-Phe-NH2 (111In-DOTA-MG11) and 111In-DOTA-(D)Asp-Tyr-Nle-GlyTrp-Asp-Phe-NH2 (111In-DOTA-CCK) in patients with evidence of recurrence or metastases of
medullary thyroid cancer after thyroidectomy. Patients and Methods Six patients (5 female, 1
male; age 17-71 y, median 41 y) were included in the study. Calcitonin levels before the first
administration ranged from 0.23 - 131 µg/L (median 34 µg/L). Scintigraphy was performed at 3
and/or 7 and/or 24 h post injection (injected in 5 min) of 740 MBq 99mTc-Demogastrin 2, 220
MBq 111In-DOTA-MG11 and 220 MBq 111In-DOTA-CCK (all 10 µg). Blood pressure and heart
rate were monitored and patients were asked for side effects. HPLC was performed on serum
taken at 10 min after start of injection. Results Side effects of the radiopharmaceuticals were:
raised heart rate, flushes, some sensation of pressure on the chest, mild sickness, tingling
(especially hands). The analogue causing most side effects was not the same in every patient.
Heart rate never exceeded 105 bpm, only 1 patient experienced the raise. All side effects
disappeared within 10 min after administration. In all patients pathologic retention of 99mTcDemogastrin 2 was seen. Scans showed lesions in brain, neck, liver and bone. 99mTcDemogastrin 2 scintigraphy was easier to interpret in all patients and showed more lesions in 4
patients than either 111In-DOTA-MG11 or 111In-DOTA-CCK. In 4 patients new lesions were seen
(neck, brain, bone and liver) with 99mTc-Demogastrin 2. New liver lesions in 1 patient were
missed by 111In-DOTA-CCK, neck lesions in another by both 111In-DOTA compounds and the
lesion in the brain was only retrospectively seen with the DOTA compounds. In each of these
patients at least one of the new lesions was confirmed by another imaging modality. A neck
lesion only visible on 99mTc-Demogastrin 2 was confirmed true positive by ultrasound guided
fine-needle aspiration cytology (6 mm). HPLC showed slower breakdown of 99mTc-Demogastrin
2 than of the other compounds. This might be one of the explanations of the differences in image
quality. Conclusions CCK 2-receptor imaging appears to be a very promising diagnostic tool in
patients with evidence of recurrence or metastases of medullary thyroid cancer. In this small
group of patients 99mTc-Demogastrin 2 scintigraphy was easier to interpret and showed more
lesions than the other two compounds. Further studies have to prove its value in patient
In vivo detection of early tumor response to tyrosine kinase
inhibitors by microSPECT with 99mTc-MIBI
S. Del Vecchio1, A. Zannetti2, F. Iommelli1, A. Papaccioli2, J. Sommella2, P.
Ferraro1, A. Baiano1, M. Salvatore1; 1University "Federico II", Naples, Italy,
CNR-Istituto di Biostrutture e Bioimmagini, Naples, Italy.
Aim. The aim of our study was to detect early tumor response to tyrosine kinase inhibitors in
animal tumor models by non-invasive SPECT imaging with 99mTc-MIBI. Our previous studies
showed indeed an increase of 99mTc-MIBI uptake in cultured tumor cells upon exposure to
staurosporine. Materials and Methods. A panel of tyrosine kinase inhibitors including gefitinib
(ZD 1839, Iressa, EGFR inhibitor), imatinib (STI571, Gleevec, Bcr-Abl inhibitor), ZD6474
(VEGFR-2 inhibitor) has been selected and tested in cultured tumor cell lines and human
xenografts. MCF-7, MDA-MB231 and T47D (wild type and Bcl-2 transfected) breast cancer
cells, A549 and SKLU-1 lung cancer cells and A431 epidermoid cancer cells were incubated
with increasing concentrations of gefitinib, imatinib and ZD6474 and tested for 99mTc-MIBI
uptake by 60 min incubation. Then control and Bcl-2 overexpressing T47D cells were
subcutaneously injected into opposite flanks of individual nude mice. Imaging studies by
microSPECT and biodistribution analysis were then performed with 99mTc-MIBI before and
after gefitinib treatment (150 mg/Kg/d p.o. for 3 days). Results. Gefitinib, imatinib and ZD6474
treatment caused a 2-fold increase of 99mTc-MIBI uptake in all cancer cell lines tested. As
compared to untreated control, Bcl-2 overexpressing breast cancer cells showed the highest
enhancement of 99mTc-MIBI uptake after treatment with gefitinib. All tumor xenografts could
be detected after 3 days of gefitinib treatment by microSPECT whereas no tracer uptake was
observed in the same tumors at the baseline scan. Consistently, a 2-fold increase of 99mTc-MIBI
uptake determined as % ID/g was also found in treated tumors as compared to untreated controls.
Conclusion. Our findings indicate that early tumor response to tyrosine kinase inhibitors can be
detected by 99mTc-MIBI scan.
In vivo demonstration of a functional relationship between
tumor marker secretion in recurrent thyroid cancer and
scintigraphic uptake of the somatostatin analogue
P. Valsamaki1, A. Gotzamani-Psarrakou2, V. Papantoniou1, S. Tsiouris1, K.
Psarrakos3, D. Kotsias1, I. Tzitzikas4, E. Molyvda-Athanasopoulou3, S.
Gerali1, C. Zerva1; 1Department of Nuclear Medicine, ‘Alexandra’ University
Hospital, Athens, Greece, 2Department of Nuclear Medicine, ‘AHEPA’
University Hospital, Thessaloniki, Greece, 3Department of Medical Physics,
‘Aristotle’ University, Thessaloniki, Greece, 4Department of Radiotherapy,
‘Ahepa’ University Hospital, Thessaloniki, Greece.
Aim: The evaluation of patients with thyroid cancer (TC) and elevated thyroglobulin and/or
calcitonin levels often requires the use of a variety of diagnostic imaging methods. The present
study was designed to assess the possible correlation between the tumor marker levels and 99mTcdepreotide uptake in recurrent metastatic lesions, considering that somatostatin receptors may be
overexpressed in malignant thyroid cells undergoing dedifferentiation and that somatostatin acts
by inhibiting the secretion of many hormones, including calcitonin.Materials and Methods: We
studied 15 patients, aged 19-70 years old (mean: 53.5 ± 17.2), followed-up for TC [8 with
papillary TC (PTC), 1 with Hürthle cell carcinoma and 6 with medullary TC (MTC)]. In all cases
increased serum thyroglobulin and/or calcitonin were indications for further investigation.
Whole-body and tomographic scintigraphy was conducted 2-3 h after the intravenous
administration of 740 MBq 99mTc-depreotide. The findings were compared and verified with a
complete clinical evaluation consisting of biopsy, 131I-WBS for differentiated TC, 123/131I-MIBG
or 99mTc-(V)DMSA for MTC, and other imaging modalities (e.g. neck US, neck and chest CT,
and MRI), performed within 2 weeks maximum. Semiquantitative analysis [lesion-to-background
(L/bg) ratio calculation] was used on the reconstructed tomographic images, with regions of
interest over the lesion and over a body area lacking somatostatin receptor expression,
respectively. Subsequently, non-parametric, Spearman rank correlation analysis was applied
between the calculated L/bg ratios and the serum thyroglobulin as well as calcitonin
values.Results: The L/bg ratios ranged from 2.1 to 5.5 (mean value ± SD: 3.12±1.27) and the
threshold for positivity was 2. Statistical analysis revealed a significant coefficient of correlation
between the L/bg ratios and the thyroglobulin levels (r= 0.899, P<0.001); conversely, there was
only a tendency towards an inverse functional relationship for calcitonin, without a statistically
significant correlation (r= - 0.486, P=0.329).Conclusions: According to our data, scintigraphic
assessment of the somatostatin receptor status with 99mTc-depreotide in TC is associated with the
thyroglobulin levels. There is a tendency towards an inverse correlation between the radiotracer
uptake and the calcitonin levels, although in the present study it does not reach the level of
statistical significance. These results render the modality valuable in terms of TC follow-up and
selection of possible therapeutic intervention by “cold” or “hot” somatostatin-analogue treatment.
Immunohistochemical detection of novel tumour markers in
thyroid cancer lesions
C. F. M. Molthoff1, A. A. M. van der Veldt1, P. J. van Diest2, O. S.
Hoekstra1, M. Boers1, L. Hooft3;
VU University Medical Center,
Amsterdam, The Netherlands, 2Utrecht University Medical Center, Utrecht,
The Netherlands, Academic Medical Center Amsterdam, Amsterdam, The
Aim: Distinguishing malignant thyroid nodules in patients with follicular cytology by fine needle
aspiration (FNA) remains difficult. The large majority of thyroid nodules (> 85%) are
overtreated. This study was undertaken to determine whether biomarkers related to glucose
metabolism and cell proliferation can help to discriminate between benign and malignant thyroid
nodules. Patients and Methods: Surgical specimens of 31 patients with benign (n=12) and
malignant (n=19) thyroid nodules showing follicular/indeterminate cytology were studied by
immunohistochemistry for the expression of galectin-3, hexokinase (HK), glucose transporter-1
(Glut-1), p53, Ki-67 and Cyclin-A. All patients were pre-operatively diagnosed with
indeterminate FNA results (microfollicular cell groups with little or no colloid) and a cold nodule
on a 99mTc scan.Results: Expression of galectin-3 (p= 0.01), HK I (p= 0.05), HK III (p< 0.001),
and Cyclin A (p= < 0.01) differed significantly between benign and malignant thyroid nodules.
Combined staining for galectin-3 and HK III correctly detected 17 of the 19 cancers, and 9 of the
12 benign lesions: sensitivity was 90%, specificity 75%, and the overall accuracy 84%. Other
markers such as glucose transporter-1 gave no additional information.Conclusions: In this series,
we found that combined immunodetection of galectin-3 and HK III on thyroid cells (with
indeterminate cytology) is a sensitive and specific strategy for identifying malignant lesions on
conventional histological samples. The value of the suggested combination of biomarkers
warrants further evaluation in a large prospective study on cytological samples of the
follicular/indeterminate phenotype. The current clinical management to indeterminate thyroid
nodules has a heavy social cost for both patients and health system. The use of our suggested
biomarkers may allow an accurate preoperative diagnosis of thyroid cancer, which can be cost
saving and may avoid serious morbidity such as vocal cord paralysis.
Standardising the protocol and practice of lower extremity
lymphoscintigraphy to improve clinical effectiveness
A. Singh, W. E. Svensson, K. S. Nijran, D. J. Towey, J. Nadarajah, A. ALNahhas; Hammersmith Hospitals NHS Trust, London, United Kingdom.
Aim: The protocol for lymphoscintigraphy is not standardized and differs among nuclear
medicine centres. Differences include the choice of radiotracer, type and site of injection,
variation in types and times of image acquisitions. In this study, we aim to evaluate and compare
certain aspects of qualitative and quantitative assessment of lymphoscintigrams to standardise the
protocol and practice of lymphoscintigraphy with improvement in clinical effectiveness.
Methods and material: Retrospective analysis of lymphoscintigrams was performed for 43
patients (12 males, 31 females; age range 17-83), that is, 86 lower extremities. 99mTc-Nanocoll
was administered as subcutaneous injection in the first interdigital space of each foot. Dynamic
and static images at 30 minutes, and wholebody and static images at 120 minutes were acquired
on Siemens E.CAM-180 dual-detector gamma camera with LEHR collimator. Medical history
and doppler ultrasound were reviewed for 13 patients. Qualitative assessment was performed
using Summed Lymphatic Score (SLS). Quantitative parameters assessed were percentage of
ilio-inguinal node tracer uptake at 120min (NU) and tracer clearance from injection sites (TC).
Statistical analysis was performed to assess the acquired data.Results: SLS was found positive
for lymphoedema in 34/86 (39.54%) and negative in 52/86 (60.46%) extremities. Quantitative
assessment using ilio-inguinal node tracer uptake (NU) confirmed lymphoedema in 15/86
(17.44%), excluded in 36/86 (41.86%), whereas, an equivocal result was obtained in 35/86
(40.69%) extremities. However, tracer clearance from injection site (TC) was positive for
lymphoedema in 20/86 (23.26%) and negative in 66/86 (76.74%) extremities. For 26 lower
extremities, where correlation with medical history and ultrasound was performed, sensitivity,
specificity, PPV, NPV and accuracy for SLS were 71.4%, 40%, 83.3%, 25% and 80.76%; and for
TC were 61.9%, 40%, 81.25%, 20% and 80.76% respectively.Conclusions: These findings
suggest that interpretation of lymphoscintigrams can be standardised using summed lymphatic
score in combination with measure of tracer clearance from injection sites, for establishing the
correct diagnosis of lower extremity lymphoedema. However, it is essential to obtain the
appropriate clinical history. Current protocol can be improved by using strict tracer
administration methodology, implementing changes in image acquisition and replacing current
practice of quantification. This should subsequently lead to an overall improvement in clinical
effectiveness by significant reduction in camera, radiographer and physicist times, with the
appropriate interpretation of lymphoscintigrams.
Tc-Depreotide scintigraphy versus F-FDG-PET in the
diagnosis of suspected recurrent or metastatic disease in
radioiodine negative thyroid cancer patients
M. Rodrigues1, S. Li2, M. Gabriel3, D. Heute3, M. Greifeneder1, I. Virgolini3;
Inst Nucl Med, Hietzing Hosp, Vienna, Austria, 2Dept Nucl Med, Univ
Hosp, Vienna, Austria, 3Univ Clinic Nucl Med, Innsbruck, Austria.
Diagnosis of recurrent (rec) and metastatic (met) thyroid cancer (TC) presents a complex
problem when serum thyroglobulin (Tg) is elevated and radioiodine scans are negative. Papillary
and follicular TC were found recently to express somatostatin receptor (SSTR) 3, 4 and 5. 99mTcdepreotide was initially introduced for the evaluation of SCLC, and found to bind with high
affinity to SSTR 2, 3 and 5. Consequently, the potential role of this tracer in other SSTR
expressing tumours is now under investigation. Aim of this study was to evaluate the diagnostic
value of 99mTc-depreotide in radioiodine negative patients (pts) with suspicion of rec or met TC.
Scintigraphic results were compared to 18F-FDG-PET , considered nowadays as a main
diagnostic tool in this clinical setting.Materials and Methods: 10 consecutive radioiodine
R. Núñez, S. Namwongprom, A. Stachowiak, M. Mar, K. Won, H.
Macapinlac; UT MD Anderson Cancer Center, Houston, TX, United States.
Objective: To determine the value of parathyroid scintigraphy with a new Hybrid
SPECT/Multislice CT scanner for localization of parathyroid adenomas (PA) in patients (pt.)
with primary hyperparathyroidism.Materials and Methods: 10 pt. (mean age 57 years; age range
37 to 73 years; 8 females and 2 males) diagnosed with primary hyperparathyroidism, were
included in this retrospective review. The patients underwent conventional parathyroid imaging
with 25 mCi (925MBq) of 99mTc-MIBI, including early and delayed planar imaging of the neck
and chest, with SPECT and SPECT/CT done between the two planar imaging sessions. The
studies were acquired on a hybrid SPECT/Multislice CT scanner. The planar images were
analyzed first, followed by the SPECT images and subsequently by the fused SPECT/CT images.
For each of the three types of images described above, the degree of certainty regarding the
location and true nature of the finding was classified according to a five point scale. From 5,
being definitively PA and certain of the location, to 1 definitively not PA and very uncertain of
the location. Correlation was done with surgical pathology (available for 7 pt.), and with the
combined findings of dedicated high resolution contrast-enhanced CT of the neck and ultrasound
(in 3 pt.).Results: The 10 patients had 10 PA. In 9 pts SPECT/CT correctly identified the side,
and in 7 of the 9 also the pole of the thyroid lobe where the PA was located. In one patient there
was a PA in the contralateral side of the scintigraphic finding. However, the finding seen in this
same pt. corresponded to a 1 cm thyroid carcinoma, unknown until the time of imaging. From
planar, to SPECT and fused SPECT/CT there was and increase in the certainty of the location
and nature of the findings (see table below). In one pt. the PA was only seen with SPECT and
SPECT/CT, and in another pt. with an equivocal finding by planar imaging, became considerably
more apparent with SPECT and SPECT/CT.Conclusions: 99mTc-MIBI parathyroid imaging with
SPECT/CT improves the confidence in the interpretation of the study, decreasing the number of
equivocal findings and the accuracy in the localization of a PA, specially in comparison with
planar imaging. TABLE. Change in the interpretation of the study in regards to the location and
confidence of findings detected by planar, SPECT and SPECT/CT parathyroid scintigraphy
Planar vs SPECT SPECT vs SPECT/CT Planar vs SPECT/CT
Nature Location Nature
Nature Location
Impact of I-FPCIT SPECT in a Multicenter Clinical Setting: a
Follow-up of 1701 Cases
K. J. Van Laere1, L. Everaert2, B. Brokken3, M. Gonce4, W.
Vandenberghe1, T. Vander Borght5; 1University Hospital Gasthuisberg,
Leuven, Belgium, 2Medical and Pharmaceutical Information, Zemst,
Belgium, 3GE Healthcare Belgium, Diegem, Belgium, 4CHU Citadelle,
Liege, Belgium, 5University Hospital Mont-Godinne, Yvoir, Belgium.
Aim: 123I-FPCIT SPECT has been suggested to be a useful and cost-effective adjunct in
diagnosis of patients with uncertain clinical diagnosis of Parkinson’s disease. This study aimed at
evaluating the current diagnostic and treatment patterns in such patients in a multicenter clinical
setting, based on a complete database of all patients scanned between 1/5/2004 and 31/8/2005
which forms part of an obligatory register for the Belgian National Health Security. Material
and Methods: All national centres that had the technical equipment to perform a 123I-FPCIT
(DatSCAN) SPECT were requested to provide the following data for every patient examined: 1/
result of the SPECT examination (loss vs. no loss of dopaminergic activity or non conclusive). 2/
change in initial diagnosis (when applicable) and 3/ alteration of treatment (when applicable).
Items 2 and 3 were supplied by the referring neurologist. Technical equipement included single-,
dual- and triple-head gammacamera’s. Images were assessed by visual or semiquantitative
reading, according to the current practice at every participating site. Results: In total 1701
questionnaires were received from 57 different centers (range of included scans per center: 3125; number of university based centers = 8). 1461 (86%) of patients had a conclusive 123IFPCIT scan and complete data. Among the patients with an uncertain diagnosis, 59.8 % had loss
of dopaminergic activity (2.4 % of the results were reported as inconclusive). In 51.5 % this was
reflected in a change in initial diagnosis (with 8.3 % incomplete data on diagnosis). Treatment
was reported to be altered in 49 % of the cases (11 % incomplete data on treatment). Patients
with a change of diagnosis were in 83 % also changed in therapy, while in patients without a
change in diagnosis; still 20 % were reported to have a change in therapy. When comparing these
data to an existing pharmaco-economic model, based on the prevalence of true PS in the tested
population, literature-derived prognosis and diagnostic accuracy, as well as clinical opinion
(treatment patterns and resource use) based on an expert-panel for the national situation, there
was a very good correspondence in the predicted changed therapies (predicted 48.5 %).
Conclusion: These results confirm that the routine clinical use of 123I-FPCIT is very influential
in changing diagnosis and resulting therapy, either to rule in or to rule out PS in the majority of
clinically uncertain patients.
Neurology/Psychiatry: Movement Disorders
Assessment of the neuroprotective effect of Creatine in
Parkinsons's disease with FP-CIT SPECT - introducing a twoyear placebo-controlled randomized pilot trial
M. Mustafa, W. Koch, A. Bender, T. Klopstock, K. Tatsch; University of
Munich, Munich, Germany.
406 — Sunday, October 01, 2006, 2:30 pm - 4:00 pm, Allegro
Longitudinal changes of dopamine transporter density in
early-onset parkin disease measured with [ I]FP-CIT SPECT
A. Varrone1, V. Sansone1, M. T. Pellecchia2, M. Amboni2, C. Vitale2, E.
Salvatore2, G. De Michele2, B. Garavaglia3, G. Annesi4, A. Brice5, S.
Pappatà1, P. Barone2, M. Salvatore1; 1IBB-CNR, Dep. of Biomorph.and
Funct. Sci., Univ. "Federico II", Napoli, Italy, 2Dep. of Neurological
Sciences, Univ. "Federico II", Napoli, Italy, 3"C. Besta" Neurol. Institute,
Milano, Italy, 4Inst. of Neurol. Sci., CNR, Cosenza, Italy, 5Inserm U289,
Pitié-Salpetrière Hospital, Paris, France.
Parkin gene mutations were found in approximately 50% of familial and 15% of isolated cases of
early-onset Parkinson’s disease (EOPD). Experimental studies suggested that Parkin is linked to
the expression and function of the dopamine transporter (DAT) and DAT dysfunction is one of
the phenotypic abnormalities of parkin -/- mice. In EOPD, imaging studies demonstrated a more
severe reduction of striatal [18F]F-dopa and DAT density in parkin than nonparkin patients.
Clinical progression is slower in parkin-related disease, but imaging data on the progression of
nigrostriatal dysfunction are limited. Aim. The aim of this study was to evaluate the progression
Aim: 123I-FP-CIT SPECT is a sensitive tool for detection of dopaminergic neuron degeneration
in parkinsonian syndromes and allows to assess progression of Parkinson's disease (PD). Oral
Creatine (Cr) supplementation has been shown to exert neuroprotective effects in several animal
models of neurodegenerative diseases. Aim of this study was to evaluate the effect of Cr on
clinical and SPECT parameters of PD progression.Methods: 60 patients with clinically proven
early stage PD were included in this study. 40 pts were allocated to receive oral Cr, 20 pts
received placebo. Standard dopaminergic medication was continued over the follow-up but
eventually adjusted to dosage at baseline to minimize confounders. 123I-FP-CIT SPECT was
performed at baseline and after two years of therapy following a standardized protocol (injected
dose 185 MBq, triple headed camera, 120 projections, FBP, Chang's attenuation correction) as
was the evaluation by clinical scores (UPDRS and SF-36). Due to various reasons 9 pts were lost
in the therapy arm, 3 pts in the control arm, so final analysis was performed on n=31 pts in the
therapy group vs n=17 pts in the placebo group.Results: Over the 2 year treatment period total
specific striatal 123I-FP-CIT binding decreased from 3.72 ± 1.27 to 3.41 ± 1.03 (decline of 8.33%)
in the placebo group vs. from 3.92 ± 0.83 to 3.61 ± 1.00 (decline of 7.91%) in the Cr therapy
group (n.s.). There were also no significant differences revealed by analysing left/right, ipsi/contralateral or putamen/caudate nucleus parameters separately. Analyses of the clinical
neurological scores showed no significant overall treatment effect, however, there was a better
performance of the Cr therapy group in the “mentation, behaviour, mood“ subscale of the
Added value of parathyroid scintigraphy with a hybrid
spect/multislice ct scanner in patients with primary
of the nigrostriatal deficit in parkin and nonparkin EOPD patients using [123I]FP-CIT SPECT.
Materials and Methods. Eighteen EOPD patients, 8 with parkin gene mutations (6 M/2 F,
age:44±12 yrs, disease duration:11±10 yrs, UPDRS:24±9.2) and 10 without mutations (5 M/5 F,
age:44±5 yrs, disease duration:5±3 yrs, UPDRS:21.2±8.5) underwent [123I]FP-CIT SPECT at
baseline and after 2.1±0.2 yrs. Studies were acquired 4 hr post-injection using a dual-head
camera (E.CAM-Siemens). Images were normalised in the MNI space using a customized
[123I]FP-CIT template and analysed with ROIs (outcome measure:specific-to-nondisplaceable
binding ratio-V3") and SMP2 (uncorrected p<0.01, corrected p<0.05 for cluster extent) using
ANCOVA (sex:nuisance variable) for between-subject and paired t-test for within-subject
comparison. Results. Baseline striatal V3" was lower in parkin (0.99±0.16) than in nonparkin
patients (1.31±0.31, p=0.015), while it was similar at 2 yrs (1.01±0.16 vs. 1.16±0.24, p=0.15).
SPM2 analysis showed that the baseline difference was significant in the left anterior putamen
(x,y,z:-20,12,8) and globus pallidus (x,y,z:-20,0,-4). Within-subject differences from baseline
were significant only in nonparkin patients (p<0.05) in the left anterior putamen (x,y,z:-28,4,-8)
and showed a trend for reduction in the right caudate (x,y,z:20,16,16) and anterior putamen
(x,y,z:28,4,4). Percent change/yr of striatal V3" was lower in parkin (0.4±3.9%/yr) than in
nonparkin patients (-4.3±5.5%/yr, p=0.06). The percent change from baseline showed a
significantly higher DAT loss in nonparkin than parkin patients in the left anterior putamen
(x,y,z:-24,4,-4) by SPM2. Conclusions. In this study we confirm our previous findings that
parkin patients have a more severe striatal DAT loss and nonparkin patients have a more
lateralised DAT loss. Preliminary data showed that the longitudinal striatal DAT loss is slower in
parkin patients, suggesting that parkin-related disease had a more severe but less progressive
nigrostriatal deficit. In addition, the rate of DAT loss even in nonparkin EOPD seems to be
slower than that previously reported in late-onset IPD.
negative pts (7 F, 3 M; 47-78 a) with suspicion of rec or met TC (follicular, 7 pts; papillary, 3
pts) because of elevated Tg were studied. Each pt was investigated with 99mTc-depreotide (740
MBq) scintigraphy and 18F-FDG-PET (370 MBq), performed with a time interval 1-8 weeks. The
presence of loco-regional rec and metastases was confirmed by ultrasonography, CT and/or MRI,
together with cytology or histological examination in selected pts.Results: True positive results
were obtained in 9 pts (90%) with 99mTc-depreotide and in 7 pts (70%) with 18F-FDG-PET.
Tc-depreotide gave better results in terms of detection of rec or met TC compared to 18F-FDGPET in 3 pts, whereas 18F-FDG-PET identified met TC not seen with 99mTc-depreotide in only 1
pt. 99mTc-depreotide detected tumour lesions in 3 pts with negative morphological imaging, and
in 2 pts with negative 111In-DOTA-TOC scans. One false-positive finding was found with both
Tc-depreotide and 18F-FDG-PET on a granuloma. 99mTc-depreotide provided relevant
information for selecting further therapeutic strategy in 6 pts.Conclusions: 99mTc-depreotide
scintigraphy is a valuable procedure for the detection of non-functioning rec or metastases from
differentiated TC. It provides thus a complementary technique in the diagnostic work-up of
radioiodine negative pts with elevated Tg, in addition to 18F-FDG-PET. Furthermore,
determination of the SSTR status with 99mTc-depreotide may be helpful in the selection of SSTR
targeted therapy.
UPDRS and a significantly smaller dose increase of symptomatic dopaminergic therapy in the Cr
group (2.5 fold increase) compared to the control group (7.5 fold increase).Conclusions: Cr
treatment in the given dosage did not reveal a significant neuroprotective effect detectable with
serial FP-CIT SPECT scans. The employed dosage of less than 0.1g per kg bw might have been
too low as suggested in comparison to most rodents treated with higher doses. A trend, however,
may be derived from the significantly smaller increase of symptomatic dopaminergic therapy in
the Cr group going along with better performance in clinical rating scales. The putative
neuroprotective efficacy of Cr for PD, therefore, remains to be proven in higher powered
multicenter studies.
The effect of dopamine agonist drugs on [123I]FP-CIT
imaging in patients with Parkinson’s Disease
D. Volterrani1, R. Ceravolo2, G. Manca1, M. Rossi1, L. Kiferle2, E. Rizza1,
V. Mattone1, F. Pesella1, D. Frosini2, L. Murri2, U. Bonuccelli2, G. Mariani1;
Division of Nuclear Medicine, University of Pisa, Pisa, Italy, 2Dept. of
Neuroscience, University of Pisa, Pisa, Italy.
Objectives: the radioligand [123I]FP-CIT is suitable for the quantification of dopamine (DA)
transporters localized on the membrane of presynaptic neurons. Due to the practical use of
[123I]FP-CIT SPECT, this technique has become a routine diagnostic tool to investigate the
integrity of the nigrostriatal pathway in vivo and also to directly assess the progression of
parkinsonism. It is still debated whether medications (dopaminomimetic, antipsychotic or
antidepressant) affect the striatal DA transporter imaging. Aim of the study was to evaluate with
[123I]FP-CIT SPECT the effect of treatment with dopamine agonist drugs on DA transporters in
patients with Parkinson’s Disease (PD).Methods: [123I]FP-CIT SPECT binding to the DA
transporters was evaluated in 15 patients both without any medication and during treatment with
dopamine agonist drugs (ergot-derivatives and non-ergot-derivatives) after reaching the maximal
individual optimized dose (mean 35 days of treatment). A 2-head gamma camera equipped with
LEHR collimators was used for data acquisition (128 x 128 matrix, pixel size 3 mm, 128
projections, 360° circular orbit) after the i.v. injection of 185 MBq of [123I]FP-CIT. Filtered
back projection with a Butterworth prefilter (0.4 cycles/cm, order 8) was used for reconstruction,
applying uniform attenuation correction (0.12 cm-1). On reoriented transaxials slices binding
ratios were obtained using a template of ROIs which were positioned over the caudate and
putamen nuclei, and the occipital cortex. Striatal, caudate and putamen binding ratios were
calculated.Results: mean striatal [123I]FP-CIT binding ratios were not significantly different on
baseline and during treatment (2.38±0.65 vs 2.32±0.50). No statistically differences were also
observed when left and right caudate and putamen nuclei were individually
evaluated.Conclusions: the results of this study suggest that dopamine agonist medications do not
affect the results of SPECT imaging with [123I]FP-CIT and support the needless of withdrawing
these drugs to measure DA transporter levels. The absence of effect of this treatment on SPECT
measurements might be of particular interest for monitoring disease progression and the effect of
putative neuroprotective drugs.
Effect of levodopa therapy in Parkinson’s disease: an FDGPET study
S. Ramat1, M. De Cristofaro1, C. Polito1, P. Marini1, M. Paganini1, L.
Mosconi2, S. Sorbi1, A. Pupi1; 1University of Florence, Firenze, Italy,
University of New York, New York, NY, United States.
Introduction. Dopaminergic neuronal loss in Parkinson’s disease (PD) induces a demodulation
of the basal ganglia-thalamo-cortical network, resulting in typical motor and non-motor
symptoms. Pharmacological treatment with oral levodopa (LD) improves clinical manifestations.
Twenty de novo PD patients were studied longitudinally using FDG PET to examine changes in
brain glucose metabolism (MRglc) in the cortico-striatal network during short term (3 months)
and long term (2 yrs) LD therapy. We present the results of the first 3-months evaluations.
Methods and results. Twenty untreated early PD patients: 5 F, 15 M; mean age: 65.2 ± 5.5 yrs,
Hoehn-Yahr scale: 1.5 ± 0.5, UPDRS III: 13.0 ± 5.5, without cognitive impairment (MMSE: 28.0
± 2.7) and mood disorders, completed a follow-up clinical and FDG-PET study after 3 months of
LD therapy (300 mg/die). The mean duration of illness was 21 ± 9 months. PET images, sidealigned on the basis of the primarily affected hemisphere (PAH), were analyzed using SPM2
software. For comparison, FDG-PET imaging was also performed in 20 age-matched healthy
subjects. As compared to controls, PD patients showed significant MRglc reductions in the
parahippocampal gyrus (BA 30) and precentral gyrus (BA6), bilaterally. Within the PAH,
significant MRglc reductions were found in the temporal (BA 22/38/39), medial (BA 6) and
superior frontal cortices (BA 8), and in the subthalamic nucleus (STN). Relative MRglc increases
were detected in the globus pallidus and anterior cingulate cortex (BA 24) of the PAH, and in the
medial frontal gyrus (BA 11) of the contralateral hemisphere. Compared to the pre-therapy
phase, a significant motor clinical improvement was found. With respect to the baseline (i.e., pretherapy) PETscan, LD treatment was associated with MRglc increases in both the STN and
cingulate gyrus of the PAH. Areas of lower MRglc were found in the inferior frontal gyrus
(BA9) and inferior temporal gyrus (BA 21) of the contralateral hemisphere and in the
cerebellum. Conclusion. Metabolic changes were found in the cortical-subthalamic-pallidalthalamic-cortical network that may reflect underlying changes in glutamatergic activity, as
hypothesized in PD patients (Obeso JA,Trends in Neuroscience 2004). Moreover, STN MRglc
increases, suggesting an LD-dependent regularization of STN aminoacidic input. Interestingly,
our PD patients showed hypometabolism in the hippocampal-parahippocampal regions as
compared to controls, which was also found in Alzheimer’s disease patients, suggesting a
possible common pathogenesis of these two neurodegenerative diseases (Galpern WR, Ann
Neurol 2006).
Relationship between I ioflupane SPECT and expression of
DAT in peripheral mononuclear cells
D. Prosperi1, G. Capriotti2, A. Tofani1, C. Del Mastro1, T. Antonellis2, A.
Anastasia2, C. Pellicano2, D. Benincasa2, F. E. Pontieri2, F. Scopinaro2;
Hosp. S. Andrea, Rome, Italy, 2University, Rome, Italy.
The early clinical symptoms of Parkinson's disease (PD) may be difficult to perceive and are
frequently overlooked. SPECT with 123I-ioflupane (DATSPECT) allows the identification of the
reduction of dopamine transporter (DAT) in nigrostriatal terminals in PD. Recent data indicate
that DAT is expressed on peripheral blood mononuclear cells (PBMC) (1). In this study, we
investigated the correlation between central and peripheral DAT in PD patients, with the aim to
verify whether DAT characterization in PBMC might have a potential role as diagnostic tool for
PD.Methods: 26 patients were submitted to DATSPECT for differential diagnosis between PD
and essential tremor. They were not affected by hypertension neither alcoholism, and were free
from antiparkinsonian drugs. On the morning of the scan, a venous blood sample was drawn
from each subject. Measurement of DAT immunoreactivity in PBMC (APBMCOD) was carried
out as described previously (1), by means of computer-assisted densitometric analysis of the
cells, using a monoclonal anti-DAT antibody. DATSPECT were evaluated a)visually and b)with
semi-quantitative analysis as reported by others (2). ROIs were drawn on Striatum (S), Putamen
(Put) and occipital cortex (CO) and results expressed as Put/CO, S/CO Put/Put and so on.
DATSPECT were classified as normal, mild positive and positive. Positive scans were defined as
at least full involvement of one Put, that means Put /CO ratio lower than 1.50 or asymmetric
involvement of more than one ganglion. Mild positive scan was defined as a) partial involvement
of only one Put and b) put/CO ratio ranging 1.8 -1.50 (2).Results: 6 pts showed negative, 5
mildly positive and 15 positive DATSPECT. Final diagnosis was PD for positive and mild
positive Pts, and essential tremor for all Pts with negative DATSPECT. Pts with negative scan
showed 0.08±0.008 APBMCOD whereas pts with positive scan showed 0.051±0.006 ( P< 0.01
vs. negative). Pts with mildly positive scan showed 0.053±0.007 APBMCOD (P < 0.01vs.
negative, NS vs mild positive ). One patient with final diagnosis of essential tremor showed
0.064 OD -2 sd under normal average- and normal SPECT.Conclusions: Our results suggest that
measurement of DAT in PBMC may represent a non-invasive and cheap screening test for the
identification of dopamine damage in PD. However, such method doesn't appear able to
substitute DATSPECT for differential diagnosis of PD. 1) J Neural Transm. 2001;108:803 2)
Nuc Med Comm. 2005,26: 421
Difference of motor activation in Parkinsonism: a PET study.
P. Payoux1, C. Brefel-Courbon1, J. Azulay2, F. Durif3, F. Tison4, O. Blin2, J.
Esquerré1, O. Rascol1; 1University Hospital, Toulouse, France, 2University
Hospital, Marseille, France, 3University Hospital, Clermont Ferrand, France,
University Hospital, Bordeaux, France.
Introduction : Despite a resemblance of clinical signs physiopathological mechanisms
underlying MPI, MSA-C and MSA-P are poorly understood. Few functional imaging studies
have reported a striatal and frontal hypometabolism at rest in MSA patients but none study has
been performed during motor task. Aim : This prospective study investigated the motor
activation induced by a right hand joystick movement in 3 groups of patients with Parkinson's
disease (PD), cerebellar form of multiple system atrophy (MSA-C) and parkinsonian form of
multiple system atrophy (MSA-P) and one control group. Pathological motor activation was
compared with normal motor activation. Materials and methods : 9 PD patients and 18 MSA
patients (9 MSA-C and 9 MSA-P patients) were studied OFF medication. rCBF measurements
(H215O PET Scan) were performed at rest and during an auditory-cued right hand movement
using a joystick. All scans were replicated. Order of rest/movement conditions was randomized.
Statistical parametric mapping (SPM2) was used to analyze motor vs. rest. Significance was set
at Z score > 3.2 (p<0.01). As control group, ten age matched volunteers were included. Results.
Compare to Controls, during right hand movement MPI patients activate cerebellum but there's
no activation in SMA. In contrast there's no motor activation in cerebellum in MSA patient, but
we observe an important activation in SMA. In addition there's no statistical difference between
MSA-C and MSA-P motor activation. Such results suggest that Parkinsonism despite a same
clinical presentation have two different network failure. Striato-mesio frontal for MPI and
cerebellar for MSA.
Brain functional activity in patients with gait disturbancies: a
rCBF SPECT study
D. Volterrani1, M. C. Carboncini2, M. Della Porta1, G. Barsotti2, G. Manca1,
M. Tolaini2, P. Lazzeri1, E. Borsò1, B. Rossi2, G. Mariani1; 1Division of
Nuclear Medicine, University of Pisa, Pisa, Italy, 2Neurorehabilitation Unit,
University of Pisa, Pisa, Italy.
Patients with deep subcortical white matter vascular lesions can present various degrees of gait
disturbancies characterized by decrease in stride step, shuffling, freezing and postural instability,
the so called vascular pseudoparkinsonism or gait apraxia. The clinical features of the gait in
these patients are similar to those with Parkinson’s Disease, suggesting a possible common
pathophysiologic hypotesis: a failure of the thalamocortical projection to facilitate the motorrelated cortical areas (supplementary motor area, cingulated and primary motor areas). Aim of
the study was to evaluate the SPECT rCBF changes during gait on treadmill in patients with
vascular pseudoparkinsonism.Methods: twelve consecutive patients with subcortical white matter
lesions documented by MRI were enrolled in the study. Seven patients (age 77±9 yrs) had gait
apraxia, whereas five (age 78±3 yrs) presented a normal gait (Tinetti and Nevitt evaluation
scale). All patients underwent an rCBF SPECT study with 99mTc-Bicisate (ECD, Neurolite
Bristol Meyers Squibb) in two different conditions on separate days, respectively at rest and
during walk. In the rest condition tracer was administered with the patients laying supine in a
quiet room. In the walk condition all patients walked for 8 minutes on a treadmill (adapting speed
to each patient) and tracer was injected 4 minutes after starting walking. Imaging was performed
about 1 hr after the injection using a dual-head gamma camera (Optima NX, GE Medical
Systems) equipped with low-energy-high resolution collimators. Data were acquired using the
following parameters: 128x128 matrix (pixel size 3mm), 128 projections (rotation 360°), 40s per
projection. Filtered back-projection with a Butterworth filter (cut-off 0.55 cycles/cm, order 8)
and attenuation correction were applied for reconstruction. Statistical Parametric Mapping
(SPM2) was used for SPECT data analysis. SPM analysis evaluated the differences of the effect
407 — Sunday, October 01, 2006, 2:30 pm - 4:00 pm, MC 3
Clinical Science: Nephrology & Urology
Agreement between Tc-99m-DTPA blood sample method and
prediction equation methods in determination of renal
function in Japanese
K. Itoh; Keiyukai Sapporo Hospital, Sapporo, Japan.
Background: Since National Kidney Foundation has established the recommendation for the
estimation of renal function in chronic kidney disease, comparative studies on estimation of renal
function by clearance and prediction methods have been reported in many countries. The purpose
of this study was to assess agreement between plasma clearance and predicted renal function in
Japanese. Materials and Methods: There were 339 studies in 300 inpatients (183 men and 117
women; age range 20 to 94 year-old; median age 68 y). They included various renal diseases
with varying degrees of the renal function (range of serum creatinine level from 0.33 to 5.66
mg/dL; median 0.95 mg/dL). Patients ware hydrated with 5 mL/kg water 20 min prior to the
examination. Tc-99m-DTPA was labeled in the hospital with a commercially available freezedried kit. Renal scintigraphy with administered dose of 37 to 200 MBq was preformed for 20
min. The glomerular filtration rate (GFRm) was determined by 2 plasma clearance methods:1)
single plasma sample at 180 min-post-injection with Christensen-Groth equation in case of serum
creatinine (Scr) < 1.2 mg/dL, 2) 2-plasma sample with correction of Brochner-Mortensen’s
equation at 120 and 240 min after injection in cases of Scr >/=2.0 mg/dL. These data were
compared to GFR which was predicted with Cockcroft-Gault (CG) and Modification of Diet in
Renal Disease (MDRD) equations. Results: Predicted GFR (GFRe) correlated well with GFRm
(r= 0.8265 in CG and r=0.8239 in MDRD). In agreement analysis, both prediction equations
overestimated GFRm in average by 8.85 ml/min/1.73 m2 in CG and by 29.87 ml/min/1.73 m2 in
MDRD. The accuracy (% absolute difference = abs(GFRe- GFRm)/GFRm x100%) of 50% was
83% in CG and 54 % in MDRD. Conclusion: Both equations which are recommended for
estimation of GFR correlated well with the clearance method. CG seems to be more suitable than
MDRD in Japanese. However, agreement between GFRm and both of GFRe is moderately poor.
Prediction methods are very convenient and useful in bed-side calculation of renal function, but
cannot replace a clearance method as the determination of accurate renal function.
Comparison of the methods in the determination of
glomerular filtration rate: Two plasma sample, single plasma
sample, gamma camera-Gates, and creatinine clearence
F. Aydın1, A. K. Cengiz1, F. Güngör1, E. Mahsereci2, M. Tuncer2, M.
Cenkçi3; 1Akdeniz University, School of Medicine, Department of Nuclear
Medicine, Antalya, Turkey, 2Akdeniz University, School of Medicine,
Department of Internal Medicine, Antalya, Turkey, 3Akdeniz University,
Faculty of Arts and Sciences, Department of Mathematics, Antalya, Turkey.
Aim: Glomerular filtration rate (GFR) is an important index for renal function. A large number
of methods for measuring GFR have been developed for a more accurate and/or more simplified
technique. In this study it was aimed to compare GFR values measured by single plasma sample
methods (SPSMs), gamma camera-Gates method, creatinine clearence [CrCl (calculated from the
serum and 24 h urine creatinine level)], formulas for predicted CrCl [“Cockgroft-Gault” formula
(CGf), and “Modification of Renal Disease” formula (MDRDf)] with two plasma sample method
(TPSM) considering as the reference method. Materials and Methods: A total of 127 subjects
(potential kidney donors, 52 male, 75 female, age between: 18-74, mean age: 42.8±12.5). Tc99m DTPA renal scintigraphy was performed on all subjects. Blood samples were taken at 120,
180, 240 min after injection of Tc-99m DTPA. The methods of Constable (CM), Dakubu (DM),
Groth and Aasted (GAM), Morgan (MM), Russel (RM), and Christensen and Groth (CGM) were
used as the SPSMs (180 min sample). TPSM-GFR was measured using by slope-intercept
method (120, 240 min samples). The appropriate formulas were used for MDRDf and CGf. GFR
was calculated from the renal uptake between 2 and 3 minute after the injection of Tc-99m
DTPA in gamma camera-Gates method. Pearson’s correlation and Bland-Altman analysis were
used to compare the results. Results: There were strong statistically significant correlation
between TPSM and all SPSMs. Among them, GAM showed the best correlation and the lowest
“standard error of estimation” (see) (r=0.96; p<0.0001, see=3.7 ml/min/1.73m2). The correlation
coefficient and see values were calculated as r= 0.49, see=20 ml/min/1.73m2 ; r=0.27, see=34
ml/min/1.73m2 ; r=0.43, see=18 ml/min/1.73m2 ; and r= 0.48, see=20 ml/min/1.73m2 for gamma
camera-Gates, CrCl, MDRDf, and CGf, respectively. According to Bland-Altman analysis, all
SPSMs showed good agreement (95% limits of agreement) with TPSM. The mean absolute
difference ± 2SD value between TPSM and GAM was found as 1.5 ± 7.1 ml/min/1.73 m2.
Gamma camera-Gates, CrCl, MDRDf, and CGf did not show good agreement with TPSM.
Conclusion: This study indicated that gamma camera-Gates, CrCl, MDRDf, and CGf are not
accurate in the measurement of GFR. Since all SPSMs showed high correlation and good
agreement with TPSM, they can be routinely used for determination of GFR.
of treadmill on rCBF between the two groups of patients (statistical design: 2 groups, 2
conditions).Results: significant rCBF increase (p <0.001) induced by the treadmill (gait-induced
activation) was found in the medial and superior frontal gyrus bilaterally, and in the anterior
lobes of the cerebellum of asymptomatic patients when compared to those with gait
apraxia.Conclusions: these prelimiray results indicate that gait disturbance in patients with
subcortical white matter lesions may be associated with reduced activity in the medial motor area
and cerebellar hemispheres during walk. These findings are in agreement with those obtained by
other authors in patients with Parkinson’s Disease suggesting that the mechanisms underlying
gait disturbances are probably the same.
Validation of the Abbreviated form of the MDRD Equation in
patients referred for Glomerular Filtration rate (GFR)
measurement by 51Cr-EDTA
G. Arsos1, E. Moralidis1, N. Karavida1, N. Boussios1, C. Chatzikyrkoy2, A.
Papagianni2, C. Karakatsanis1; 1Dept of Nuclear Medicine, Aristotle
University Medical School, Hippokration Hospital, Thessaloniki, Greece,
Dept of Nephrology, Aristotle University Medical School, Hippokration
Hospital, Thessaloniki, Greece.
Aim : A multivariate linear model for GFR estimation (GFRes) is expressed by the Abbreviated
form of the MDRD Equation, which incorporates serum creatinine (SCr) level, age, sex and race.
Chronic kidney disease (CKD) staging according to estimated GFR level has been recently
proposed by the K/DOQI guidelines (2002). The aim of the present study was to validate
aMDRD GFRes and the resultant CKD staging against 51Cr-EDTA GFR measurements (GFRm)
in a mixed non-pediatric population referred for radionuclide GFR measurement. Patients and
methods : One hundred fifty two individuals (84 women), aged 48.7±16.4 yrs were enrolled in
the study (59 with chronic kidney disease [CKD], 42 kidney donor candidates, 37 renal
transplant recipients, 10 renal transplant donors, and 4 with surgically corrected congenital
hydronephrosis). All patients were submitted to slope-intercept, two-sample 51Cr-EDTA GFR
measurement as well as to full biochemical evaluation including SCr measurement. GFRes were
obtained using the aMDRD equation. The association between GFRes and GFRm was assessed
by linear regression. The GFR limits proposed by K/DOQI 2002 guidelines (•90, 60-89, 30-59,
15-29 and <15 ml/min/1.73 m2) were applied for CKD staging. The kappa statistics was used to
assess the agreement between GFRes and GFRm in CDK staging. The scatterplot of the
differences GFRes - GFRm vs their average (GFRes + GFRm)/2 was constructed to assess trends
and biases. Differences among continuous variables were evaluated by ANOVA. Results : The
correlation between GFRes and GFRm was good (r=0.87, p<0.0001). However, their agreement
in CKD staging was only fair (ț = 0.404). The scatterplot of differences vs averages of GFRes
and GFRm showed wide, heteroscedastic scattering, rendering impossible formal Bland-Altman
analysis throughout the entire range of average GFR values. The mean GFRes - GFRm
differences over various average GFR increments (<60, 60-90 and •90 ml/min/1.73 m2), were
8.7, -0.9 and -18.9 ml/min/1.73 m2 respectively (ANOVA, p<0.05). Conclusion : GFR estimates
obtained by the application of the abbreviated form of the MDRD equation are well correlated
with GFR measurements by 51Cr-EDTA. However, GFRes fail to correctly classify patients with
CKD as they frequently over- and underestimate low (<60 ml/min/1.73 m2) and high (>90
ml/min/1.73 m2) GFR values respectively. Such a misclassification may be of clinical importance
for both CDK patients and healthy kidney transplant donor candidates.
Is mean transit time superior to Tmax for estimating renal
J. Kuyvenhoven1, A. Scholtens1, A. Piepsz2, H. Ham2; 1Ziekenhuis
Gelderse Vallei, Ede, The Netherlands, 2Ghent University Hospital, Ghent,
Aim: Mean transit time (MTT) obtained by deconvolution is not influenced by the level of
overall renal function, contrarily to empirical renogram parameters such as Tmax. In clinical
practice however, the superiority of MTT to Tmax has not been demonstrated. This work aims at
determining, for a wide range of MTT values and using different input functions, the relationship
between MTT and Tmax.Materials and Methods: In a simulated acquisition (duration of
acquisition 1200 sec, 480 frames, time per frame 2.5 sec), 10 input functions (plasma
disappearance curves of 99mTc-MAG3 with renal clearances ranging 33 - 405 ml/min) were
convolved with 522 retention functions, characterized by minimal transit time ranging 150 - 720
sec, MTT ranging 160 - 1400 sec and maximal transit time (MaxTT) ranging 170 - 2080 sec.
Since in clinical practice the retention function is necessarily truncated once MaxTT is longer
than the duration of the acquisition, the original value of MTT was corrected into MTTc,
calculated as the surface under the retention function up to 1200 sec. MTTc was compared to
Tmax for different levels of renal function as expressed by renal clearance.Results: A total of 5220
renograms resulted with Tmax ranging 163 - 1200 sec. For MTTc ranging 150 - 249, 250 - 349,
350 - 449, 450 - 599, 600 - 899 and • 900 sec, mean ± sd of Tmax equalled respectively 223 ± 36,
335 ± 47, 449 ± 61, 591 ± 88, 854 ± 164, and 1116 ± 106 sec for low clearances (< 100 ml/min, n
= 1566), 219 ± 32, 322 ± 37, 423 ± 39, 546 ± 51, 754 ± 92, and 1003 ± 97 sec for moderate
clearances (100 - 300 ml/min, n = 2088), and 215 ± 29, 310 ± 32, 403 ± 34, 516 ± 47, 696 ± 82,
and 903 ± 102 sec for normal clearances (> 300 ml/min, n = 1566). Only minimal changes of
Tmax due to changes in renal clearance were observed.Conclusions: Although theoretically MTT
is a better reflection of renal transit than Tmax, in clinical practice Tmax can be used as a parameter
for cortical as well as whole kidney transit whatever the level of overall renal function. This is
obvious in case of rather short transit time (e.g. cortical transit in renovascular hypertension,
obstructive uropathy). In case of prolonged transit both Tmax and MTT cannot be quantified.
Comparison of methods for determination of glomerular
filtration rate: low and high dose tc-99m-DTPA renography,
predicted creatinine clearance method and plasma sample
M. Eftekhari, M. Assadi, M. Hozhabrosadati, M. Saghari, A. Fard-Esfahani,
B. Fallahi Sichani, D. Beiki, D. Beiki; Research Institute for Nuclear
Medicine, Tehran University of Medical Sciences, Shariati Hospital, Tehran,
Iran (Islamic Republic of).
Aim: The gamma camera uptake method with Tc-99m-DTPA is simple and less time consuming
for determination of the glomerular filtration rate (GFR) but its diagnostic accuracy is debated.
Gate’s method (Low dose; LD), high dose method (HD), predicted clearance method and plasma
clearance method with Tc-99m-DTPA are compared in the present study. We also performed
GFR measurement and diuretic renogram at the same time.Materials and Methods: Tc-99m
DTPA renography was performed on 36 patients aged 18-72 years with a wide range of renal
function (Serum creatinine 1.36 ± .50). The GFR was determined simultaneously by 4 methods:
1-Gamma camera uptake method with low dose Tc-99m DTPA(gate’s,LD), 2- gamma camera
uptake method with high dose Tc-99m DTPA(gate’s ;HD), 3- predicted creatinine clearance
method(Cockcroft-Gualt, CG) and 4- plasma sample clearance(PSC) using mono exponential
curve. The PSC was chosen as a reference.Results: the regression equation of the LD, HD and
CG against PSC was y=5.7 +1.019 x; (r = .83; p value<0.001), y=3.13 +0.928 x; (r = .91; p
value<0.001) and y=0.688 +0.773 x (r =0.72; p value<0.001) respectively.Conclusions: the three
methods were highly in agreement with the PSC but the high dose GFR had less error in
estimation of GFR. Furthermore, GFR measurement as well as diuretic renography could be
performed at the same time when the high dose method is used. Due to low cost and negligible
radiation burden, this method might be preferred for routine nuclear medicine practice.
discerned again on the same positions. If necessary, static images were repeated after 24 hours.
Results: Patients were divided into 3 groups according to clinical findings and complaints. 22
patients with edema in abdominal (13/22) and genital region (9/22), 3 patients with decreased of
taking dialysis fluid, 6 patients with non-specific complaints were respectively grouped as 1, 2
and 3. In Group 1, a leakage into inguinal canal was detected in 6/9 (% 66,7) patients with edema
in genital region and pericatheter external leaks or abdominal wall hernias were detected in 10/13
(% 76,9) patients with abdominal edema. 2/3 patients had pericatheter external leaks and 1/3
patient had pleural leak in Group 2. Normal scintigraphic findings were observed in all group 3
patients (6/6) with non-specific complaints. Conclusions: Our study showed that the scintigraphic
technique was a productive diagnostic method for patients performing CAPD therapy due to
CRF. Moreover, the importance of objective findings and subjective symptoms could be
identified by scintigraphic method in the patient with suspected leakage.
408 — Sunday, October 01, 2006, 2:30 pm - 4:00 pm, MC 2
Reproducibility of MAG3 dynamic renography: population
2nd ISRTRD - Physics 2: Biological and Long-Term
Effects/Animal and in-vitro Studies
L. Lezaic, J. Fettich, M. Grmek, M. Milcinski; University medical centre
Ljubljana, Ljubljana, Slovenia.
AIM: MAG3 dynamic renography is performed throughout the age specter for a number of
indications, some of which require longitudinal follow - up of the patient. Result of the
investigation, especially one that demonstrates significant change on follow - up, frequently
influences patient management. Therefore, highest possible reproducibility of the investigation is
desired. MATERIAL, METHODS: We have evaluated 40 pediatric (age 0 - 7ys) and 50 adult (20
- 84ys) renograms for relative function (expressed in %) and drainage (evaluated as good, partial
or poor) for both kidneys (L, R). All renograms were evaluated twice by three observers in
random sequence for inter - (between observers) and intraobserver (within each observer)
reproducibility by repeatability parameter through analysis of variance (relative function),
intraclass correlation coefficient [ICC] (relative function, drainage) and weighted kappa test
(drainage).Results: Pediatric renograms, interobserver reproducibility, relative function: ICC
0.96, repeatability +/- 9.0%; drainage: wkappa L 0.82, R 0.83. Pediatric renograms, intraobserver
reproducibility, relative function: ICC 0.99/0.99/0.97, repeatability +/- 6.8/4.7/9.0%; drainage:
wkappa L 0.79/1.0/0.95, R 0.87/0.86/0.91. Adult renograms, interobserver reproducibility,
relative function: ICC 0.99, repeatability +/- 4.2%; drainage: wkappa L 0.91, R 1.0. Adult
renograms, intraobserver reproducibility, relative function: ICC 0.99/0.99/0.99, repeatability +/1.9/3.5/2.36%; drainage: wkappa L 0.93/0.98/1.0, R 1.0/0.97/0.92.Conclusions: For both
evaluated parameters and both comparisons (inter - and intraobserver), reproducibility of adult
renograms is better than that of the pediatric population.
Use of ordered-subsets expectation maximization (OSEM) for
lesion detection and localization in Tc-99m DMSA SPECT
D. Bedel, G. Durmus-Altun, A. Ergulen, S. Berkarda; Trakya University,
Edirne, Turkey.
AIM: Tc-99m DMSA scintigraphy is an accurate method for evaluation of regional cortical
impairment during acute pyelonephritis and later on, for detection of permanent scarring.
Iterative reconstruction algorithms produce accurate images without streak artifacts as in altered
backprojection. They allow improved incorporation of important corrections for image degrading
effects. Aim of the study was to evaluate the role of OSEM reconstruction algorithm on DMSA
SPECT and to comparing lesion detection and localization on 360 and 180 degree SPECT
acquisition.Methods: Thirty consecutive patients enrolled in the study group (3 male, 27 female,
mean age 22±16 yrs, and ranges: 6-55 yrs). Standard static imaging, posterior 180 (SPECT180)
and 360 degree (SPECT360) renal SPECT images were obtained simultaneously by using a dualhead camera. Visual evaluation was performed on the same frames in both acquisition
methods.Results: Ten of 30 patients were without cortical defect according the all three
accusations (planar, SPECT180 and 360). There were significant differences for lesion detection
and localization on all three imaging techniques (x2: 163, p: 0.0001). On the planar images,
cortical defects were obtained in 16 patients, however, no sign of renal disease was obtained in
14 patients (47%). 18 of patients (60%) and 17 of patients (57%) had normal result on the 180
degree and 360 degree acquisitions, respectively. For the lesion detection and localization,
SPECT180 was significantly correlation with SPECT360 (r: 0.87, p: 0.001). Image quality of
SPECT images is superior in SPECT360 than SPECT180.Conclusions: Contrast to general
acceptance, Tc-99m DMSA SPECT imaging has not overestimated number of the detectable
renal cortical defect on both SPECT180 and SPECT360 when the OSEM algorithms are used for
SPECT reconstruction. SPECT360 could be a preferential acquisition technique due to best
image quality, however, SPECT180 has an advantage on imagining time.
The usefulness of scintigraphic method to evaluate objective
findings or subjective symptoms in patients undergoing
continuous ambulatory peritoneal dialysis
N. O. Kucuk, S. S. Gultekin; Ankara University, Ankara, Turkey.
Aim: Continuous ambulatory peritoneal dialysis (CAPD) frequently uses as an alternative to
hemodialysis in the treatment of patients with chronic renal failure (CRF). The application may
cause some clinic and anatomical problems in the long period. The aim of this study was to
evaluate the usefulness of scintigraphic method in patients with objective findings or subjective
symptoms. Methods: 31 patients (gender; 15 F, 16 M, age; 27-66 years, mean average; 43 + 10
years) with CRF were included in the study. 25/31 had objective findings (edema in abdominal
or genital area, the decrease dialysis fluid more than 500 cc. last a few days), 6/31 had subjective
symptoms (abdominal distention or pain). 2 mCi (74 MBq) 99mTc sulphur colloid was injected
via catheter to peritoneal space in combination with 2 L of peritoneal dialysis fluid. Dynamic
anterior images were obtained for 10 min at 60 sec intervals from abdominal region and
following static images were obtained for 5 minutes on anterior and lateral positions from thorax,
abdomen and pelvic regions. Before and after emptying the dialysate, static images were
The use of BED and EUD concepts in heterogeneous
radioactivity distributions for Targeted Radionuclide Therapy
A. Malaroda1, E. Kalogianni2, G. Flux1; 1Institute of Cancer Research,
London, United Kingdom, 2The Royal Free Hospital, London, United
AIM: In Targeted Radionuclide Therapy (TRT), it is extremely important to irradiate all cancer
cells avoiding normal tissue. However, there is evidence that non-uniform activity distribution
within tumours might cause TRT to fail. The aim of this study was to investigate the effects of
the temporal and the spatial behaviour of the radioactivity in TRT, focusing on heterogeneous
radiopharmaceutical distribution on a multi-cellular scale.Methods: Various activity distributions
at the multi-cellular level from three radionuclides (32P, 90Y and 131I) were simulated in cubic
matrices of 1, 3 and 4 mm per side. The in-house software DOVE was used to calculate dose-rate
maps from these activity distributions, and Survival Fractions were calculated taking into account
an uptake and a clearance phase in the Linear Quadratic model which included the dose-rate
effect via the Lea-Catchside factor. The effect from non-uniform activity distributions was
analysed in terms of Dose Volume Histograms (DVH), Biologically Effective Dose (BED) and
the Effective Uniform Dose (EUD).Results: The fraction of the absorbed dose that is “wasted”
without producing a biological effect to the treatment reaches 60% in the highly non-uniform
distributions. For 32P and 90Y, the loss of therapeutic effectiveness was shown to be less than for
I. However, 90Y due to its shorter physical half-life, presented lower mean BED values in
almost every geometry, compared to 32P and 131I, and thus was less effective. 131I, amongst all
geometries, appeared to be more effective in more homogeneous activity distributions and in the
1mm VOI, whilst it was the least effective radionuclide in the more heterogeneous activity
distributions. 32P, presented the highest values of EUD, compared to 90Y and 131I.Conclusions:
The EUD is a unique value that facilitates comparisons between different activity distributions in
terms of treatment outcome. This study showed that as the degree of the heterogeneity in the dose
distributions increases the therapy effectiveness worsens. Non-uniform absorbed dose
distributions can create a situation in which a fraction of cells are under-irradiated, while another
fraction of cells is “over-killed”.
Differences in pharmacokinetics between therapeutic and
trace amounts of
Lu-DOTA-Tyr -octreotate in a small cell
lung cancer animal model
A. Schmitt1, P. Bernhardt1, O. Nilsson2, H. Ahlman3, L. Kölby3, E. ForssellAronsson1; 1Radiation Physics, Göteborg, Sweden, 2Pathology, Göteborg,
Sweden, 3Surgery, Göteborg, Sweden.
We have previously shown that 177Lu-DOTA-Tyr3-octreotate (Emean=150 keV, T½=6.7 days) is a
potential radiopharmaceutical for treatment of somatostatin receptor positive small cell lung
cancer (SCLC). The aim of this study was to compare the pharmacokinetics between therapeutic
and trace amounts of 177Lu-DOTA-Tyr3-octreotate in the same SCLC animal model.Methods:
Female nude mice (n=20), bearing tumours from the human SCLC cell line NCI-H69, were
subcutaneously injected with three fractions of 30 MBq (1 µg) 177Lu-DOTA-Tyr3-octreotate with
24 h between the fractions. Activity concentration (%IA/g) in tumour, liver, spleen, blood and
the kidneys was determined at days 3, 7, 14 and 22 (first injection at day 0), and dosimetric
estimations were done according to MIRD formalism with the cumulated activity obtained
through integration of the time-activity curves. Morphological analyses of the tumour samples
were done, and the results were compared with those from tumours from untreated controls
(n=7).Results: The tumour activity concentration decreased from 5.0±0.3 %IA/g at day 3 to
2.8±0.4 %IA/g at day 7, p<0.05, but thereafter the activity concentration increased again to
3.7±0.4 %IA/g at day 14. The mean relative tumour volume in the groups showed large
differences at time of measurement, due to massive therapeutic effect of 177Lu-DOTA-Tyr3octreotate. At day 22 only one of five tumours was visible. The activity concentration in normal
tissues decreased during the time studied. Tumour-to-normal-tissue activity concentration ratios
increased with time for all normal tissues. The absorbed dose to tumour per injected amount of
activity was 0.78 Gy/MBq. This corresponding value was 0.29 Gy/MBq after 3 MBq (trace
amounts) given i.v. (Schmitt et al., Cancer Biother Radiopharm 2003,18:593-599). However,
also the absorbed dose to kidney became higher after therapeutic amounts than after trace
amounts (0.25 Gy/MBq compared to 0.11 Gy/MBq), but for spleen, liver and blood no difference
could be seen.Conclusions: The pharmacokinetics in the SCLC model after the therapeutic
amount 3*30 MBq, given s.c. with 24 hours between the fractions, resulted in a 2.7-fold higher
absorbed dose to tumour per injected amount of activity compared to after the trace amount 3
Tumour targeted therapy with
J. Dahle , J. Borrebæk , K. B. Melhus1, Ø. S. Bruland3, R. H. Larsen2;
Institute for Cancer Research, The Norwegian Radium Hospital, Oslo,
Norway, 2Algeta ASA, Oslo, Norway, 3The Norwegian Radium Hospital,
Oslo, Norway.
The novel alpha-particle emitting radionuclide 227Th was conjugated to the CD20-specific
monoclonal antibody rituximab and used to treat nude Balb/c mice bearing growing macroscopic
(32-256 mm3) Raji human B-cell lymphoma xenografts. Thorium-227 has a relatively long halflife of 18.7 days and can be produced in practically unlimited amounts from 227Ac (T1/2 = 22
years). The T1/2 of 227Th is compatible with the in vivo administration of 227Th-labeled antibody
conjugates and selective tumor targeting before a significant amount of 223Ra is generated. The
relatively long half-life of 223Ra, in turn, would assure that this nuclide would be largely excreted
or trapped in skeletal hydroxyapatite before decay would occur. Thus, a therapeutic window may
exist, permitting therapy with 227Th without unacceptable toxicity from the generated
Ra.Methods: Therapeutic levels of 227Th-DOTA-p-benzyl-rituximab as well as Zevalin (90Ytiuxetan-ibritumomab) was injected intravenously into tumor-bearing mice. Survival and tumor
growth delay was measured and compared with controls.Results: Median survival and median
growth delay after treatment with 400 kBq/kg 227Th-DOTA-p-benzyl-rituximab was 74 days and
43 days, respectively, while the median survival after treatment with 30 MBq/kg Zevalin was 23
days and 2 days, respectively. No toxic effects were observed after either treatment. These
studies suggest that 227Th is a very effective nuclide for radioimmunotherapy, and its relatively
long half-life may make it a promising alpha emitter for systemic tumor therapy.
Increased therapeutic efficacy through combination of Lu177-DOTATOC and chemotherapy in neuroendocrine
tumours in vivo
M. Gotthardt1, D. Librizzi2, D. Wolf2, G. Lalyko2, T. M. Behr2, M. Behe2;
Radboud University Nijmegen Medical Center, Nijmegen, The
Netherlands, 2Department of Nuclear Medicine, Philipps-University,
Marburg, Germany.
Aim: in oncologic therapy, the combination of therapeutic modalities is used to increase
tumoricidal effects. In a nude mouse tumour model, we compared the efficacy of monotherapy to
a combination of Lu-177-DOTA-Tyr3-Octreotide (DOTATOC) and chemotherapy.Materials and
Methods: maximal tolerated doses (MTD) were determined for Lu-177-DOTATOC, doxorubicin
(DX), and cisplatinum(CS). The MTD was reached if one or more animals in a treated group
died as a result of therapy if all animals survived in the group receiving a 10% lower dose level.
BALBc nu/nu nude mice were used in our experiments with 10 animals per group. For tumourinduction, 10 000 000 pancreas-tumour cells (AR42J) were subcutaneously injected to create
somatostatin receptor-positive tumours. Two weeks later, mono- and combination-therapy were
applied. Tumour size was determined in the follow-up in comparison to an un-treated control
group. Kidney toxicity was evaluated by histological examination.Results: the MTDs were
determined to be 84 MBq for DOTATOC, 300µg for DX, and 260µg for CS per 25g mouse. The
combined therapy with either DOTATOC and DX or DOTATOC and CS lead to a strong
inhibition of tumour growth in comparison to either mono-therapy. Though only 50% MTD of
each agent was applied, the combination of DOTATOC and CS was more effective than the
combination of 75% MTD each of DOTATOC and DX. After 4 weeks, the experiments were
stopped due to the tumour growth in the control groups. The average tumour sizes at the end of
the experiments were DX 302µl, DOTATOC 633µl, combination 160µl, control 3356µl, and CS
645µl, DOTATOC 800µl; combination 6µl, control 3425µl. The differences shown were
statistically significant (ANOVA, all p<0.05). If combined at 50% or 75% MTD each, the
substances did only show slight nephrotoxicity as determined by histological
examination.Conclusions: our in vivo data show that the combination of radiopeptide therapy and
chemotherapy offers an effective means to increase the therapeutic efficacy in the treatment of
neuroendocrine tumours. Clinical trials have to be performed to evaluate this therapy regimen in
man. Care needs to be taken in respect to nephrotoxicity.
growing ovarian cancer in nude mice with 211At-MX35
(Fab´)2: therapeutic efficacy and myelotoxicity.
J. Elgqvist; Dept of Radiation Physics, Göteborg, Sweden.
The aim of this study was to investigate the therapeutic efficacy and myelotoxicity during
fractionated radioimmunotherapy of ovarian cancer in mice. The study was performed using the
monoclonal antibody (mAb) MX35 F(ab´)2 labeled with 211At. Methods: Animals were
intraperitoneally inoculated with ~1×107 NIH:OVCAR-3 cells. Four weeks later the mice were
given the first treatment. Six groups of animals were intraperitoneally injected with ~800,
3×~267, ~400, 3×~133, ~50, or 3×~17 kBq 211At-MX35 F(ab´)2 (n=6×18). The second and third
injections for groups 2, 4, and 6 were given 4 and 8 d after the first, respectively. As controls,
animals were treated with MX35 F(ab´)2 (n=12). Eight weeks after the last injection the animals
were sacrificed and the presence of macro- and microscopic tumors and ascites was determined.
Blood counts were determined for each mouse in groups 1 and 2 before, and 3, 7, 11, 15, and 23
d after the first injection. The mean absorbed dose in the bone marrow was based on the ratio
between the 211At activity concentration in bone and blood (the bone-to-blood ratio, BBLR) as
well as in bone marrow and blood (bone-marrow-to-blood ratio, BMBLR), and the cumulated
activity and absorbed fraction of the Į-particles emitted by 211At in the bone marrow. Results:
The tumor-free fractions of animals were 56% and 41% when treated with ~800 kBq or 3×~267
kBq 211At-MX35 F(ab´)2, respectively, 39% and 28% when treated with ~400 kBq or 3×~133
kBq 211At-MX35 F(ab´)2, respectively, and 17% and 22% when treated with ~50 kBq or 3×~17
kBq 211At-MX35 F(ab´)2, respectively. The suppression of WBC counts was decreased (46%19%, compared to the baseline WBC counts) and delayed (from 4 d to 11 d after the first
injection) during the fractionated treatment compared to the single-dose treatment. The %IA/g
for blood, bone, and bone marrow all peaked 6 h after injection at: 13.80 ± 1.34, 4.00 ± 0.69, and
8.28 ± 1.38, respectively. The BBLR and BMBLR were 0.20 ± 0.04 and 0.58 ± 0.01, respectively.
The mean absorbed dose in bone marrow was ~0.4 Gy after intraperitoneally injecting ~800 kBq
At-MX35 F(ab´). Conclusion: No advantage was observed in the therapeutic efficacy of using
a fractionated compared to a single administration, with the same total amount of administered
activity. Alleviation of the myelotoxicity was observed during the fractionated regimen in terms
of decreased suppression and delayed nadir of the WBC counts. No thrombocytopenia was
observed during either regimen.
Microdosimetric analysis of cell survival after alpha-particle
N. Chouin1, K. Bernardeau1, M. Cherel1, A. Faivre-Chauvet1, C.
Apostolidis2, A. Morgenstern2, A. Lisbona3, J. Barbet1, F. Davodeau1, M.
Inserm U601, Nantes, France, 2Transuranium Institute,
Karlsruhe, Germany, 3CRLCC Nantes-Atlantique Renee Gauducheau, St.
Herblain, France.
The aim of our study was to compare the microdosimetric cell sensitivity, z0, according to
absorbed dose delivery (particle hits from free antibodies in the solution or from antibodies
bound to cell surface). We carried out in vitro targeting of the tumour specific MHC/peptide
complex, present on one cell line (T2) by 7D4 and 8A11 antibodies coupled to 213Bi. The great
interest of this model, for dosimetry, is the possibility to modulate the number of antigenic sites
on the cell surface within the experimental set up. It is therefore possible to vary the proportion
of surface and volume alpha decays and so to evaluate their respective impact on cell survival.
Cell survival was measured by [3H]thymidine incorporation on more than 500 experimental
points. Microdosimetry calculations were performed, for which many parameters were
determined. Cell packing configuration and a representative distribution of radii of the cell
population were obtained by microscopy. The number of bound labelled antibodies was
calculated from the number of antigenic sites (evaluated using Scatchard plot analysis) and
antibody affinity. Our microdosimetric model is based on analytical calculations. It is derived
from Stinchcomb method. Our study consisted in the calculation of the cell sensitivity, z0, for 4
different experimental configurations: when the irradiation was totally non specific, when less
than 10% of the absorbed dose came from the surface decays, when more than 10% or 50% of
the absorbed dose came from the surface decays. The cell survival plotted as a function of
absorbed dose is a one-phase exponential decay with a plateau around 15% of cell survival. For
doses ranging from 0.2 Gy to 0.4 Gy, survival data points were scattered around the experimental
curve . Similar values of z0 were obtained for the 4 categories:
Percentage of absorbed dose that comes from surface z0
Standard error of the
< 10%
> 10%
> 50%
A large uncertainty is associated to these values. It may have different origins: - the
measurement of many biological parameters (number of antigenic sites, antibody affinity) is
difficult and always approximated - cell survival can be inherently variable. However, the
microscopic cell survival parameter, z0, seems to remain constant, for different levels of
irradiation specificity. This work highlights the need for more experimental studies, in which as
many biological parameters as possible are kept constant and carefully assessed in order to
decrease the potential sources of variation. It should be then possible to fully express the
potential of a complete microdosimetric approach.
Radionuclide therapy of micro-xenografts using
A´´-DTPA-ABD-(ZHER2:342)2 Affibody molecule
V. Tolmachev1, A. Orlova1, R. Pehrson2, J. Galli2, B. Baastrup2, K.
Andersson1, M. Sandström3, D. Rosik2, J. Carlsson1, H. Lundqvist1, A.
Wennborg2, F. Y. Nilsson1; 1Uppsala Univeristy, Uppsala, Sweden,
Affibody AB, Bromma, Sweden, 3Uppsala Univeristy Hospital, Uppsala,
Aim An anti-HER2 Affibody® molecule, ZHER2:342, targets HER2-expressing xenografts with
high selectivity and contrast. However, its small size (about 7 kDa) determines renal excretion,
which causes high accumulation in kidneys and complicates radionuclide therapy using
radiometal labeled ZHER2:342. We propose a reversible binding to serum albumin as a method to
reduce renal excretion and improve radionuclide therapy using ZHER2:342. Material and methods
The dimeric form (ZHER2:342)2 was fused with albumin binding domain (ABD) derived from the
albumin-binding motif of streptococcal protein G. Recombinant ABD-(ZHER2:342)2 was
conjugated with isothiocyanate derivative of CHX-A´´-DTPA and labeled with the low-energy
beta-emitter 177Lu. Cellular binding and processing of 177Lu-CHX-A´´-DTPA-ABD-(ZHER2:342)2
was evaluated in living HER2-expressing SKOV-3 ovarian carcinoma cells. Biodistribution
experiments in normal and tumor-bearing mice were performed to evaluate the effect of fusion
with ABD and capacity to target HER2-expressing xenografts. Experimental radionuclide
therapy of HER2-expressing SKOV-3 micro-xenografts, which should resemble
MBq. The absorbed dose to tumour after 3*30 MBq was approximately 70 Gy, which resulted in
complete remission in 50 % of treated animals (Schmitt et al., Eur J Nucl Med Mol Imaging
2005, 32(S1):S47). Tumour-to-normal-tissue activity concentration ratios increased with time for
all studied normal tissues, which is favorable in a therapeutic situation.
micrometastases, was performed. The time to establishment of xenograft tumors after SKOV-3
inoculation was used as an end-point for therapeutic effect. Results CHX-A´´-DTPA-ABD(ZHER2:342)2 had a picomolar affinity to HER2 and low nanomolar affinity to human and murine
albumin. Radiolabeled 177Lu-CHX-A´´-DTPA-ABD-(ZHER2:342)2 demonstrated specific binding
to HER2-expressing cells and excellent cellular retention. In vivo, fusion with ABD enabled a
25-fold reduction of renal uptake in comparison with non-fused dimer molecule 177Lu-CHX-A´´DTPA- (ZHER2:342)2. Biodistribution in tumor-bearing mice demonstrated high and specific uptake
of conjugate in HER2-expressing xenografts. Treatment of SKOV-3 micro-xenografts with
Lu-CHX-A´´-DTPA-ABD-(ZHER2:342)2 completely prevented formation of tumors, while
tumors were established in control animals treated with PBS (median tumor free-survival 43
days) or the same radioactivity of a non-specific Affibody molecule 177Lu-CHX-A´´-DTPAABD-(Zabeta)2 (median tumor free-survival 43 days). Conclusion Fusion with ABD enables
modification of in vivo biodistribution of small tumor-targeting proteins and peptides. 177LuCHX-A´´-DTPA-ABD-(ZHER2:342)2 is a promising candidate for a treatment of micrometastases
of HER2-expressing malignant tumors.
Amifostine can reduce toxicity of targeted radionuclide
therapy without affecting anti-tumour-effect
F. Forrer, E. Rolleman, B. Bernard, R. Valkema, E. P. Krenning, M. De
Jong; Erasmus MC, Rotterdam, The Netherlands.
Aim: Amifostine is a radical scavenger and acts as a radio-protector. Higher doses in healthy
tissues are achieved by higher activating enzyme concentrations in non-tumor tissue. Promising
first results demonstrated effectiveness in kidney protection during peptide receptor radionuclide
therapy (PRRT) with [Lu-177-DOTA0,Tyr3]octreotate. Now we investigated the effect of
amifostine on tumour growth and kidney function during PRRT to exclude negative effects on
anti-tumour-activity.Methods: 144 tumor bearing rats (CA 20948 pancreatic tumor) were treated
with 0, 278 or 555MBq [Lu-177-DOTA0,Tyr3]octreotate. Some rats were co-injected with
amifostine, lysine or a combination of both. SPECT/CT with a dedicated multipinhole animal
SPECT/CT NanoSPECT (Bioscan Inc., Washington D.C.) was acquired 5 days after treatment.
Approx 1 month after treatment SPECT/CT with In-111-Octreoscan and approx. 3 months after
treatment with Tc-99m-DMSA was acquired. Kidneys were analysed histologically.Results:
High quality SPECT/CT images were acquired with all radionuclides (Lu-177, In-111, Tc-99m).
Images could be quantified absolutely to determine tumor uptake or kidney-uptake respectively.
Approximation of organ and tumor size was feasible on CT. A significant, dose-dependent
reduction of renal DMSA uptake compared to the control rats was found after [Lu-177DOTA0,Tyr3]octreotate-treatment. The amifostine groups showed a significantly higher DMSA
uptake than the corresponding groups without kidney protection. The results correlated very well
with histological findings, blood- and urine-parameters. No negative influence of amifostine on
tumor-uptake nor on tumor-response after [Lu-177-DOTA0,Tyr3]octreotate treatment was
found.Conclusions: Amifostin is a most promising agent to protect healthy tissue from radiation
damage during peptide receptor radionuclide therapy. The NanoSPECT/CT is an accurate tool to
follow physiological processes in the same animal over time with different tracers. It allows a
precise determination of size (CT) and absorbed activity in vivo (SPECT). Amifostine is a
promising agent to reduce toxicity in radionuclide therapies.
Reduced iodide transport (stunning) and DNA synthesis in
cultured porcine thyrocytes exposed to low absorbed doses
from I
C. Johanson , M. Nordén , M. Nilsson , E. Forssell-Aronsson ; Dept of
Radiation Physics, Göteborg, Sweden, 2Dept of anatomy and cell biology,
Göteborg, Sweden, 3Dept of antomy and cell biology, Göteborg, Sweden.
Thyroid stunning is defined as a reduced therapeutic uptake of 131I in the thyroid gland (or
thyroid tumor) compared to the uptake of a previously administered diagnostic amount of 131I.
The phenomenon is clinically important, as it can potentially lead to under-treatment of thyroid
disease. Previous clinical and experimental studies indicate that thyroid stunning is dosedependent. However, is yet unknown at what lowest absorbed dose from 131I it is induced. The
aim of this study was to study in vitro effects of low absorbed doses of 131I on iodide transport
and DNA synthesis on normal thyroid cells concerning DNA-synthesis. Methods: Porcine
thyrocytes were grown on microporous filters in bicameral chambers as a confluent and polarized
monolayer, corresponding to the thyroid follicular epithelium. Cells were continuously irradiated
by 131I in the culture medium for 48 h to absorbed doses of 0.0015-9 Gy. Three days after
irradiation the transepithelial, basal to apical, iodide transport was evaluated using 125I-. The
effect of 131I on DNA synthesis was estimated by measuring [3H]thymidine incorporation of both
subconfluent and confluent cells. Total DNA content was measured to quantify cell proliferation.
All irradiated cell cultures were compared to non-irradiated control cells.Results: Reduction of
125 I transport was seen for absorbed doses •0.15 Gy. An absorbed dose of 1.5 Gy reduced the
125 I -transport with around 50% compared to non-irradiated control cells. Cell proliferation was
reduced already at absorbed doses of 0.01 Gy, while 0.15-0.3 Gy did not affect the proliferation.
Absorbed doses • 1 Gy showed a continuously reduced proliferation with increased absorbed
dose. A 50% reduction was obtained at approximately 4 Gy. Total DNA measurements showed a
reduction of the cell number at 8 Gy.Conclusions: The lowest absorbed dose that statistically
significantly reduced the 125I--transport was 0.15 Gy. Reduced cell proliferation was obtained at
0.01-0.1 Gy and at absorbed doses •1 Gy, showing a hypersensitivity region at 0.01-0.1 Gy.
Total DNA measurements, however, indicated that no decrease in cell number was obtained at
corresponding absorbed doses.
Multimodality imaging of radioactive holmium-loaded
polymer microspheres for treatment of liver tumours
J. F. W. Nijsen1, J. Seppenwoolde2, T. C. de Wit1, M. A. D. Vente1, G. C.
Krijger3, S. W. Zielhuis1, P. R. Seevinck2, M. G. G. Hobbelink1, R. de Roos1,
C. J. G. Bakker2, B. A. Zonnenberg1, P. P. van Rijk1, A. D. van het Schip1;
Dept. of Nuclear Medicine, University Medical Center, Utrecht, The
Netherlands, 2Image Sciences Institute, University Medical Center, Utrecht,
The Netherlands, 3Dept. of Radiation, Radionuclides & Reactors, Faculty of
Applied Sciences, Delft, The Netherlands.
Introduction For treatment of unresectable liver tumours, internal radiation therapy is a
promising option. For this therapy, radioactive poly (L-lactic acid) microspheres (Ho-PLLA-Ms)
are administered into the hepatic artery. These microspheres will lodge in and around the tumour
and irradiate the surrounding tissue. The microspheres emit both ȕ-radiation, allowing internal
radionuclide therapy and Ȗ-rays suitable for imaging with a gamma-camera (half life 26.8h).
Since holmium is highly paramagnetic, the microspheres can be visualized directly with MRI as
well. In this study the benefit of multimodality imaging and image guided therapy was explored
in rabbits and pigs. Materials and Methods Radioactive Ho-PLLA-Ms were administered into
the hepatic artery of tumour bearing rabbits and normal pigs. In-vivo and ex-vivo images of the
distribution of the radioactive microspheres were performed by nuclear (planar and SPECT), MR
and CT imaging. Fusion of images was performed. To evaluate unintend redistribution and
toxicity of the microspheres the animals were monitored for several weeks. After termination the
organs were examined. Results and DiscussionCombining systems like SPECT and CT, and
PET and CT is one approach to use the complementary imaging properties of different
techniques. A second approach is the production of diagnostic and therapeutic agents that can be
co-imaged by different imaging systems. This study showed that the element holmium is
equipped with the necessary properties to allow for co-imaging. In pigs and rabbits small tracer
amounts were used to assess biodistribution prior to the administration of the therapeutic amount
of radioactive microspheres. Administration of Ho-PLLA-Ms through catheterisation in pigs was
performed under MR or X-ray guidance. MR imaging resulted in imaging of the Ho-PLLA-Ms
as well as the anatomic structures. This can be very helpful in the interpretation of the
biodistribution. However, detection with MRI of microspheres in the lungs and stomach is not
feasible. Therefore nuclear imaging is more practical. Fusion of the images resulted in a superior
distribution analysis. Since holmium is paramagnetic the accumulation of decayed radioactive
holmium can be investigated by MRI weeks after administration. During neutron irradiation of
holmium-165 also a minute amount of metastable holmium-166 is formed (half life 1200y). In
this study the distribution of the microspheres in organs was quantitatively analysed by
measuring 166mHo activity. The imaging information will give extra knowledge in the
understanding of the treatment results. We conclude that radioactive holmium is a very useful
isotope for diagnosis and therapy in oncological indications.
409 — Sunday, October 01, 2006, 2:30 pm - 4:00 pm, Lecture Hall 1
Physics: Instrumentation 1
Investigation of the luminescence properties of the LYSO:
Ce, LSO: Ce and GSO: Ce single crystal scintillators at 140
keV for applications in SPECT
I. Valais; 1)Department of Medical Instruments Technology,Technological
Educational Institute (TEI) of Athens, Ag. Spyridonos, Aigaleo, 122 10
Athens,2)Department of Medical Physics,Medical School,University of
Patras, 265 00 Patras, Athens, Greece.
Introduction:The aim of the present study was to determine the light emission efficiency of
(Lu,Y)2SiO5:Ce (LYSO: Ce), Lu2SiO5 (LSO: Ce) and Gd2SiO5: Ce (GSO: Ce) single crystal
scintillators under low energy -ray excitation (140 keV) for single photon tomographic imaging
(SPECT). All crystals are fast emitting (40-60 ns), non-hygroscopic scintillators exhibiting high
radiation absorption efficiency in the energy range used in medical imaging
applications.Materials and Methods: Crystals with dimensions 10x10x10 mm3 were excited by
140 keV -rays emitted by a Tc-99m source. The light flux emitted by the excited scintillators was
measured by a light integration sphere (Oriel 70451) coupled to a photomultiplier (EMI 9798B).
The variation of the absolute luminescence efficiency (emitted light energy flux per unit of
incident -ray exposure rate) with -ray flux was determined by these measurements. In addition
the optical spectrum of LYSO: Ce, LSO: Ce and GSO: Ce was measured, under -ray excitation
using an optical spectrometer (Ocean Optics Inc., S2000). Optical spectrum data were used to
determine the spectral compatibility of the three scintillators to various optical photon detectors
(photodiodes, photocathodes) incorporated in detectors of nuclear medicine imaging systems.
Finally intrinsic parameters such as the quantum detection efficiency and the intrinsic conversion
efficiency were calculated using theoretical models and tabulated data. Results: LSO: Ce was
found to exhibit higher absolute efficiency while LYSO: Ce was found with increased intrinsic
efficiency. Their emission spectrum (390 to over 470 nm), peaking at 430 nm for LYSO: Ce and
LSO: Ce and at 440 nm for GSO: Ce was found compatible (55% - 95%) to most currently
employed optical photon detectors (amorphous silicon photodiode, crystalline silicon photodiode,
photocathodes, charge coupled devices etc). All crystals were found of increased performance in
the energy range employed in SPECT imaging.
Monte Carlo study of the Detection Efficiency of various
scintillators for use in positron emission tomography (PET)
D. Nikolopoulos;
Department of Medical Instruments Technology,
Technological Educational Institution of Athens, Ag. Spyridonos, Aigaleo,
122 10, Athens, Greece.
Introduction:Five crystal scintillators designed to be used in positron emission imaging (PET)
scanners, were evaluated in terms of radiation detection efficiency by applying Monte-Carlo
Methods.Materials and Methods: The detection efficiency of every scintillator was studied as a
function of the crystal thickness. The investigated scintillators were the Lu2SiO5 (LSO), the
LuAlO3 (LuAP), the (Lu,Y)2SiO5:Ce (LYSO), the Gd2SiO5 (GSO) and the YAlO3 (YAP). The
crystals were modeled as blocks of 45x45 mm2 entrance area and of varying thickness ranging
from 5 to 30 mm. Scintillator exposure was simulated as originating from monoenergetic 511
Monte Carlo Simulation Study of Several Camera Designs for
the PET Component of a Dedicated Breast PET/CT Scanner
S. L. Bowen1, S. R. Cherry1, J. M. Boone2, R. D. Badawi2; 1Department of
Biomedical Engineering, UC Davis, Davis, CA, United States, 2Department
of Radiology, UC Davis Medical Center, Sacramento, CA, United States.
Aim: To explore the performance of a range of detector geometries for the PET component of a
dual modality dedicated breast PET/CT system using Monte Carlo simulations and an
anthropomorphic model.Materials and Methods: The open source software GATE was used. A
cylindrical system consisting of 22 LSO block detectors (9x9 arrays of 3x3 mm2 crystals with
variable length and 3.3 mm pitch) per ring was simulated (inner diameter 206 mm).
Combinations of 20 or 30 mm long crystals, and 4 or 6 modules axially were considered (the
existing prototype consists of two planar detectors with 4x4 blocks of 3x3x203 mm crystals) .
The anthropomorphic model consisted of the NURBS NCAT phantom with the original breast
substituted by a pendant model [Lamare et al, conf. rec. IEEE MIC 2005]. System parameters
were set as follows: paralysable block dead time - 256 ns; non-paralysable coincidence dead time
- 256 ns; coincidence window - 7.50 ns (all achievable but not yet implemented on our
prototype). Trigger threshold was set to 120 keV and the energy window was 350-650 keV.
Singles, trues, randoms, scatter and NEC (1R formulation) were determined as a function of total
activity.Results: NEC rates increased significantly in comparison to the planar dual head
configuration. For the full-ring system with 6 axial blocks and 30 mm crystals, NEC increased
~4-fold at clinically viable activities; this did not change significantly for the 4 axial block
configuration. This suggests that the lengthened axial FOV provides little benefit for a mediansized breast. Increasing the crystal thickness from 20 mm to 30 mm increased NEC by
approximately 30%. Singles-to-trues ratios were approximately equal to 300 for all cases. Scatter
fraction for the fixed ring system (~0.29) was 20% greater on average than the dual head scanner
regardless of the configuration.Conclusions: For our breast PET/CT scanner, a significant
improvement in NEC over the current dual-headed geometry is possible by utilizing a full-ring
configuration, at the cost of increased scatter fraction. Future Work: We are currently
examining the effects of different breast sizes on count-rates and aim to explore the impact of
depth-of-interaction resolution on reconstructed images in the coming months. We also hope to
simulate data for a variable-diameter split-ring and five-sided box geometry.
The engineering and preliminary results of a transformable
ultra-high resolution PET camera
W. Wong, H. Li, H. Baghaei, Y. Wang, Y. Zhang, S. Kim, R. A. Rameriz, J.
Liu, S. Liu; University of Texas M.D. Anderson Cancer Center, Houston,
TX, United States.
We have developed an ultrahigh resolution transformable PET camera that can be transfigured
into different organ-specific PET systems to (a) test the clinical and research usefulness of
ultrahigh resolution PET in body imaging and cancer staging, (b) test the usefulness of dedicated
brain PET and breast-caner PET, and (c) test the utility of a flexible transformable PET design
concept for molecular imaging research. This PET has the following attributes: (1) for
wholebody cancer staging, an intrinsic image resolution of 2.6 mm with similar coincidencedetection sensitivity as a typical commercial PET, (2) a dedicated full-ring breast PET that has an
intrinsic resolution of 2.4 mm (2.5-3.0 mm practical), for detecting early small breast tumors; this
mode has a 10X or higher detection sensitivity than a clinical PET. It samples 143 slices (1.4-mm
apart) for pendulous breast up to 21-cm long with the option of including the axilla region, (3) a
dedicated high-resolution brain PET with 3 times the detection sensitivity of a regular PET, with
a large 21-cm axial field of view for concurrent imaging of the brain and carotid arteries (to
concurrently measure arterial input function without the invasive arterial puncture), (5) the
production cost may be lower or equal to that of the standard PET despite its ultra-high
resolution and using multiple times more detectors (38106 versus 10,000-18,000). This system
has 38016 BGO detectors (2.7 x 2.7 x 18 mm3) using only 924 photomultipliers (round) by
applying the low-cost high-resolution PMT-quadrant-sharing detector design. To compensate for
the slower scintillation timing of BGO detectors, we applied the "HYPER" pileup-up event
recovery front-end electronics. The detection system was engineered to provide minimal dead
space (at the base of the breast) when imaging the breast in the breast mode. It has a ring
diameter of 83-cm and a 13-cm axial FOV in the wholebody mode. The system has been put
together and is being tested for its imaging characteristics. The overall engineering and the
preliminary imaging performance of this system will be presented. This flexible PET may
provide a more flexible and useful alternative to the current commercial clinical PET (designed
mainly for cancer staging) for the target of molecular imaging research.
keV photons impinging at the centre of each scintillator block. The study was performed via a
general photon transport Monte Carlo code previously generated and validated by the reporting
team. This code is based on random event generation via which tracking of all photon energy
histories is achieved. Taking into account the three basic interaction processes governing photon
transport at 140 keV, the energy transferred, absorbed, redistributed and reabsorbed in every
crystal was calculated. Moreover, the secondary energy redistribution due to K- or Lcharacteristic fluorescence radiation was also recorded and investigated. To be specialised for
certain scintillator media the code was fed with the interaction cross-sectional data and the nonrelativistic form factors and scatter factors as downloaded from the NIST Reference Database.
The code can be easily manipulated at source level so as to evaluate a wide variety of physical
properties. Various parameters were calculated such as the fraction of the incident photon energy
absorbed, transmitted or redistributed as fluorescence radiation, the scatter to primary ratio, the
photon and energy distribution within each scintillator block etc. Results:As being most
significant, the fraction of the incident photon energy absorbed was found to increase with
increasing crystal thickness tending to form a plateau above the 30 mm thickness. For LSO,
LuAP, GSO and YAP scintillators respectively, this fraction had the value of 38.9%, 35.8%,
31.2% and 13.3% at the 10 mm thickness and 71.4%, 68.1%, 63.9% and 35.5% at the 50 mm
thickness. Approximately (52-54) %, (53-55) %, (57-59) % and (89-90) % of this fraction, was
due to scattered and reabsorbed radiation. For all scintillators, a negligible fraction (<0.1%) of
the absorbed energy escapes the crystal as fluorescence radiation.
Performance investigation of a dual-modality SPECT/optical
small animal imager
J. Peter, W. Semmler;
German Cancer Research Center, Heidelberg,
Objective: Intended for simultaneous detection of (near-infrared) fluorescent molecular markers
and radiolabeled pharmaceutical distributions in vivo we have build a tomographic dual-modality
SPECT/optical small animal imaging system. Key objective of this investigation was to perform
a series of phantom acquisition experiments to identify the imaging performances of the subsystems. Methods: For SPECT imaging a compact gamma detector (Thomas Jefferson National
Accelerator Facility, Newport News, VA, USA) is implemented. It consists of a 2x2 array of
Hamamatsu H8500 position sensitive photomultiplier tubes which are attached to a pixellated
scintillator array (St. Gobain Crystals and Detectors, Newbury, OH, USA). The scintillation
crystal consists of a 66x66 array of opto-decoupled 1.3x1.3x6 mm3 NaI(Tl) crystal elements
yielding a total detector area of 9.9x9.9 cm2. A purposely designed roof-shaped pinhole (PH)
collimator made of tungsten alloy (Densimet, Plansee AG, Austria) has been built in our
laboratory. The focal length of the collimator is 7.5 cm. Optical photons are directed through a
system of reflective surfaces to compose an optical projection that is aligned with the angular
field of view of the SPECT camera. For optical imaging a high resolution ORCA AG cooled
CCD camera (Hamamatsu Photonics, K.K., Japan) is used. Results: In order to assess spatial
resolution properties of the SPECT sub-system planar projections were acquired of capillary line
sources filled with various tracers. In extracts, for a 1 mm diameter knife-edged-shaped PH the
intrinsic spatial detector resolution was measured to be 1.2 mm FWHM, and improves to 0.57
mm FWHM for a 0.5 mm diameter PH, both measured for Tc99m at 3.5 mm line source to PH
distance. Energy resolution of the SPECT sub-system, measured for Tc99m, is 12 % FWHM.
Spatial resolution of the optical sub-system is largely constrained by scattering and attenuation of
light within the imaged object. Discussion: Prime advantage of the dual-modality imager is the
ability to directly detect and compare the individual sub-tracer/marker kinetics as both photon
types can be measured simultaneously from the same projection angle at identical imaging
geometries and animal positioning. The performance characteristics of the individual subsystems is nearly unaffected by the dual-modality construction principle except for a slightly
higher minimal object-to-PH distance of the SPECT imager which is caused by the mirror
assembly. Sensitivity of the optical system is degraded somewhat by the imperfect mirror
reflectance (about 82 % per mirror).
The nanospect/ct: a high-sensitivity small-animal spect/ct
with submillimeter spatial resolution
N. U. Schramm1, C. Lackas1, J. W. Hoppin1, F. Forrer2, M. de Jong2;
Research Center Juelich, Juelich, Germany, 2Erasmus MC, Rotterdam,
The Netherlands.
Objectives: Multi-Pinhole SPECT has become a proven modality in small-animal molecular
imaging. Although the spatial-resolution capabilities of SPECT are greater than those of PET, the
latter is generally considered the gold-standard nuclear imaging modality due to highsensitivities. In this work, we present a high-throughput SPECT system that achieves
submillimeter reconstructed resolutions while simultaneously approaching the sensitivity of PET.
This increase in sensitivity combined with existing advantages of SPECT, e.g. tracer chemistry,
cost and dual isotope capabilities, improves the standing of SPECT as a molecular
imager.Methods: The NanoSPECT/CT consists of four high-resolution SPECT detectors and a
high-performance x-ray CT mounted on a high-precision gantry. Each of the SPECT detectors is
outfitted with an interchangeable 9-pinhole aperture for a total of 36 pinholes surrounding the
field of view (FOV). Pinhole diameter and FOV are chosen in accordance with the prescribed
application, e.g., mouse or rat imaging. The axial FOV is extended using helical scanning (userselectable range from 20 to 290mm). Additionally, helical orbits provide an increase in the
angular sampling. All told, this increase in sensitivity and sampling greatly improves image
quality both for detection and estimation (quantification) as compared to standard SPECT
acquisition techniques. The x-ray CT unit provides accurate anatomical information to aid the
interpretation of the SPECT data.Results: We will present a detailed description of the
NanoSPECT/CT along with numerous phantom studies and small-animal scans performed with
an array of I-125, Tc-99m, I-123, Lu-177 and In-111 tracers. The results will address resolution,
sensitivity, imaging times, injected dose and quantification results as well as multi-isotope,
dynamic and dual-modality (SPECT/CT) imaging.Conclusions: We have developed a highperformance dual-modality SPECT/CT system for imaging small rodents. The system provides
submillimeter resolution (down to 0.5mm), high sensitivity (up to 2600cps/MBq), and fast
acquisition times (less than 1 min) with quantification and dynamic scanning capabilities.
The MADPET-II project: Performance evaluation and first
images from a dual layer LSO-APD small animal PET
V. Spanoudaki, I. Torres-Espallardo, S. Ziegler; Klinikum rechts der Isar,
Nuclear Medicine Clinic, Technical University of Munich, Munich, Germany.
Aim: The use of avalanche photodiodes for the readout of scintillation crystals offers the
possibility of using compact detector modules in positron emission tomography. This study aims
to prove the feasibility of acquiring accurate information from a small animal PET scanner with
two radial LSO-APD detector layers, one-to-one coupling, and individual electronic
readout.Methods: MADPET-II features individual readout of LSO scintillation crystals
(2x2x6mm3 front layer and 2x2x8 mm3 back layer) by avalanche photodiode (APDs) arrays
(single APD active area: 1.6x1.6 mm2). The complete system will consist of 1152 LSO-APD
channels arranged in a dual layer detector ring in order to reduce parallax effects and thus to
maintain uniform spatial resolution along the field of view. At the same time, events that have
suffered Compton scatter within one detector block can be identified and used in order to
increase the system’s sensitivity. Singles events are detected and stored in list-mode format, each
one characterized by it’s own energy and timing information. Coincidence sorting is done postacquisition offering maximum flexibility in applying variable energy and timing window
settings. Data are reconstructed using a MLEM algorithm and a Monte Carlo based system
matrix. Images of 0.3 mm diameter FDG-18 line sources were reconstructed to measure the
spatial resolution.Results: A mean energy resolution of 22% and a system-wide time resolution
of 10.5 nsec have been measured for our 384 channel prototype. The maximum system count rate
without any data loss is 10.000 counts per second per channel. The intrinsic axial spatial
resolution and the reconstructed transverse spatial resolution were measured to be 1.1 mm and
1.3 mm (FWHM), respectively which is in good accordance with the expected values from
simulation studies. Structure phantoms showed homogeneous resolution across the whole fieldof-view.Conclusions: Reconstructed phantom images verify the feasibility of acquiring singles
list-mode data from a dual layer LSO-APD detector system and sorting coincidences postacquisition in software. The scanner exhibits satisfactory performance in terms of time and
spatial resolution, count rate and sensitivity performance for high resolution PET imaging.
501 — Sunday, October 1, 04:30 pm - 06:00 pm, Friends of Music Hall
Clinical applications in Oncology
18FDOPA striatal uptake in Macaca fascicularis monkeys:
Methodological development for a Mosaic microPET scanner
Orazio Schillaci (IT)
J. M. Martí-Climent1, M. Collantes2, I. Peñuelas2, L. Álvarez-Erviti3, F. J.
Blesa3, M. Ecay2, J. Arbizu1, J. Obeso3, J. A. Richter1; 1Nuclear Medicine
Service, University Hospital , University of Navarra, Pamplona, Spain,
MicroPET Research Unit, CIMA-CUN, University of Navarra, Pamplona,
Spain, 3Movement Disorders Laboratory, Neurosciences, CIMA, University
of Navarra, Pamplona, Spain.
Due to the experimental character of the small animal PET scanners, the analysis software
usually shows limitations for detailed quantitative evaluation compared to those used in clinical
applications in humans. Aim: The objective was to develop and validate the methodology (PET
scanner calibration, acquisition, reconstruction and corrections) that would allow the
quantification of 18FDOPA decarboxilation constant (Ki) in the striatum of Macaca fascicularis
monkeys using a small animal PET scanner.Materials and Methods: The resolution, sensitivity
and uniformity of a microPET Mosaic (Philips) PET scanner were evaluated to asses the
feasibility for brain monkey studies. Scanner calibration was done using a Fluor-18 cylinder
phantom (diameter: 6 cm). Five healthy animals were studied. One hour before the study,
carbidopa (50 mg) was orally administered to block peripheral 18FDOPA decarboxylation.
Anesthesia was induced with ketamine and midazolan before the transmission scan, followed by
a list mode acquisition (100 minutes) that started at the injection of 18FDOPA (77±20 MBq).
Data were sorted in 23 frames (10x90”, 9x300” y 4x600”). During the reconstruction using the
3D-Ramla algorithm the following corrections were applied: death time, attenuation, scatter and
decay. The 18FDOPA Ki influx constant (min-1) was calculated pixel by pixel with Patlak
graphical analysis, generating parametric images with a 2 mm pixel size, with the PMOD
software. The occipital cortex was used as a reference of non-specific uptake. ROIs over the left
and right striatum were drawn on the dynamic average image and transferred to the Ki
image.Results: Resolution (2,0 mm), sensitivity (1,2 %) and uniformity resulted to be adequate
for the 18FDOPA uptake quantification in monkey basal ganglia. List mode acquisition allowed a
proper synchronization with 18FDOPA injection. Death time correction factors (< 1,53) were
obtained from a look up table generated during phantom calibration. The reconstructed dynamic
file was exported in Dicom format, but raw data for each image frame needed a further
normalisation to the frame duration and a transformation to activity concentration by means of a
calibration factor, obtained with the cylinder phantom. Mean (± SD) Ki values for both striatal
regions were 8,94 ± 2,11 and 8,41 ±1,12 10-3 min-1.Conclusions: A proper calibration of the
Mosaic small animal PET scanner and a detailed methodology allowed the quantification of the
monkey striatal uptake by means of a dynamic study with 18F-FDOPA. Measured Ki values,
similar to those previously published in monkeys using PET scanners, confirms the overall
experimental procedure.
502 — Sunday, October 01, 2006, 4:30 pm - 6:00 pm, Trianti Hall
410 — Sunday, October 1, 2006, 02:30 pm - 04:00 pm, Lecture Hall 2
The radiopharmaceutical basics of Re-188
CME 4: Fusion Imaging: Is there a clinical role for
Premises and perspectives of SPECT- CTQuality
Luigi Mansi (IT)
Clinical applications in non-oncological diseases
Torsten Kuwert (DE)
Cardiovascular: Gated SPECT Imaging
Evaluation of a decision support system for interpretation of
gated SPECT
M. Lomsky1, P. Gjertsson1, L. Johansson1, J. Richter2, M. Ohlsson3, A. van
Aswegen4, R. Underwood5, L. Edenbrandt1; 1Sahlgrenska University
Hospital, Gothenburg, Sweden, 2WeAidU in Europe AB, Lund, Sweden,
Department of Theoretical Physics, Lund, Sweden, 4Medical Physics
Department, Bloemfontein, South Africa, 5Imperial College, London, United
Purpose: We have recently presented an automated decision support system for interpreting
ECG-gated myocardial perfusion scintigraphy (MPS). The purpose of this study was to evaluate
the system based on artificial neural networks in a separate hospital from where it was trained
and to compare it with a quantification software package.Methods: A completely automated
method based on computerised image processing and artificial neural networks was trained for
the interpretation regarding myocardial ischemia and infarction using 418 MPS from one
hospital. Features from each examination describing perfusion, regional and global function were
used as inputs to different artificial neural networks. After the training session, the system was
evaluated using 535 MPS from another hospital. The patients were studied using a one day
Tc-tetrofosmin protocol at the training site and using a two day 99mTc- sestamibi protocol at
the test site. These test images were also evaluated using computer derived summed rest score
(SSS) and summed difference score (SDS) and extent of rest defects and reversibility (Emory
Cardiac Toolbox, CEqual quantitation). The images were interpreted by one of three experienced
clinicians at the training hospital and one experienced clinician at the test hospital and these
interpretations regarding the presence or absence of myocardial infarction were used as the
standard.Results: The decision support system had a sensitivity of 87% and a specificity of 97%
regarding the diagnosis of myocardial infarction. The corresponding specificity for summed rest
score was 63% and for extent of rest defect 87%, measured at the same level of sensitivity. The
specificities for myocardial ischaemia were 88% for the extent of reversible defect measure, 54%
for the summed difference score and 89% for the decision support system, all measured at a
sensitivity of 84%.Conclusions: A decision support system based on artificial neural networks
presents interpretations more similar to experienced clinicians compared to conventional
automated quantification programs. This study shows the feasibility of disseminating the
expertise of experienced clinicians to less experienced physicians by the use of artificial neural
Furn F. Knapp, Jr. (USA)
Algorithm specific normal values (classified by MRI) for left
ventricular volumes and ejection fraction from gated
Bone pain palliation with Re-188 radiopharmaceuticals
Holger Palmedo (DE)
W. M. Schaefer, C. S. A. Lipke, H. Kuehl, T. Krohn, H. Kaiser, U. Buell;
University Hospital Aachen, Aachen, Germany.
Intratumoral and intracavitary
Hans Bender (DE)
Liver cancer therapy with Re-188 radiopharmaceuticals
Joachim Kropp (DE)
Aim Gated myocardial perfusion SPECT allows calculation of end-diastolic and end-systolic
volumes (EDV, ESV) as well as of ejection fraction (LVEF). Despite good agreement with the
current gold standard cardiac MRI (cMRI) the gated SPECT algorithms QGS, 4D-MSPECT and
ECTB lead to systematic differing results (1). Aim was therefore to establish algorithm specific
gated SPECT normal values for EDV, ESV and LVEF based on “normal patients“ with regular
heart functions classified by reference values from cMRI. Materials & Methods Seventy
patients (54 male, 16 female, mean age 59.7±12.4 years, range 33-82 years) with suspected or
known coronary artery disease (CAD) were examined with gated 99mTc-MIBI SPECT (8
gates/cardiac cycle) 60 minutes after tracer injection at rest. EDV, ESV and LVEF were
calculated from gated 99mTc-MIBI SPECT by means of ECTB, 4D-MSPECT and QGS. Directly
before or after gated SPECT imaging, cMRI (20 gates/cardiac cycle) was performed and EDV,
ESV and LVEF were calculated using the modified Simpson's rule. Based on the cMRI results of
the Off-spring cohort of the Framingham heart study (2) our cMRI findings were categorized as
Comparison of gated SPECT, echocardiography and cardiac
magnetic resonance imaging for the assessment of left
ventricular functions
H. Demir1, Y. Z. Tan1, G. Kozdag2, Y. Akoz Anik3, D. Ural2, C. Balci3, S.
Isgoren1, F. Berk1, A. Demirci3; 1Kocaeli University School of Medicine
Department of Nuclear Medicine, Kocaeli, Turkey, 2Kocaeli University
School of Medicine Department of Cardiology, Kocaeli, Turkey, 3Kocaeli
University School of Medicine Department of Radiology, Kocaeli, Turkey.
Aim: Left ventricular ejection fraction (LVEF), end-diastolic volume (EDV) and end-systolic
volume (ESV) can be determined non-invasively by two-dimensional echocardiography
(ECHO), gated single photon emission computed tomography (GSPECT) and cardiac magnetic
resonance imaging (CMRI). This study was designed to analyse the concordance between LVEF,
EDV, ESV values and left ventricular regional wall motion scores (RWMS) derived from ECHO,
GSPECT and CMRI.Materials and Methods: ECHO, GSPECT and CMRI were performed in a
group of 21 patients with suspected coronary artery diseases. LVEF, EDV, ESV values were
calculated. For regional wall motion analysis a 16-segment model with a five-point scoring
system was used to enable direct comparison of the same areas.Results: The mean LVEF
measured with GSPECT, ECHO and CMRI were 55.9±17.8%, 55.7±16.4% and 56.4±15.7%,
respectively. The mean EDV measured with GSPECT, ECHO and CMRI were 109.2±42.4 ml.,
127.5±42.2 ml. and 91.1±38.0 ml, respectively. The mean ESV measured with GSPECT, ECHO
and CMRI were 54.2±41.2 ml., 59.9±37.6 ml. and 41.8±26.9 ml., respectively. The results of
linear regression analysis showed good correlations between LVEF, EDV and ESV values
derived from GSPECT, ECHO and CMRI (r:0.91, r:0.92, r:0.97 for LVEF; r:0.71, r:0.68, r:0.73
for EDV; and r:0.86, r:0.91, r:0.91 for ESV, P<0.01, respectively). As shown by means of BlandAltman analysis LVEF, EDV or ESV were not under- or overestimated by GSPECT compared
with ECHO. There was no systematic under- or overestimation of LVEF and EDV with GSPECT
compared with CMRI. ESV was underestimated by CMRI compared with GSPECT. LVEF were
not under- or overestimated by CMRI compared with ECHO. However, EDV and ESV were
underestimated by CMRI compared with ECHO. For regional wall motion analysis the overall
agreements were 87%, 82% and 80% between ECHO and CMRI (k:0.41), between GSPECT and
ECHO (k:0.30) and between GSPECT and CMRI (k:0.16).Conclusions: This study showed a
good overall correlation between LVEF, EDV and ESV values derived from GSPECT, ECHO
and CMRI over a wide range of values. On the other hand, overall concordance of RWMS
between these imaging modalities was fair to poor. Estimates of wall motions from one technique
cannot be used interchangeably with those derived from other technique.
Comparison of 4D-MSPECT and QGS in quantification of left
ventricular volumes and ejection fraction from 16- and
rebinned 8-frame gated
Tc-Tetrofosmin SPECT
W. M. Schaefer, T. Krohn, K. C. Koch, H. Kaiser, U. Buell; University
Hospital Aachen, Aachen, Germany.
Aim Gated myocardial perfusion SPECT allows estimation of left ventricular enddiastolic/systolic volumes (EDV, ESV) and ejection fraction (LVEF). Currently gated SPECT is
acquired mostly using 8-frames per cardiac cycle. This relatively low temporal resolution results
in blurred end-diastolic and end-systolic phase images, leading to underestimated EDVs and
overestimated ESVs, which in turn result in underestimated LVEFs. Using 16 frames per cardiac
cycle, however, decreases the signal-to-noise-ratio significantly. Aim of this study was therefore
to compare the results from 16-frame and then rebinned 8-frame gated SPECT data in a large
patient cohort using the QGS and 4D-MSPECT algorithms. Materials & Methods 120 patients
were examined using gated SPECT (16-frames/cardiac cycle) on a Siemens Multispect 3 (triplehead gamma camera; Siemens Gammasonics Inc., Hoffman Estates, IL) 60 minutes after
intravenous administration of about 450 MBq (two-day protocol) or about 750 MBq (one-day
protocol) 99mTc-Tetrofosmin at rest. Short axis slices (16-frames) were summed framewise (1+2,
3+4, etc.) yielding 8-frame data sets. EDV, ESV and LVEF were calculated for both data sets
using 4D-MSPECT and QGS. Results QGS succeeded in 119/120 cases. In the remaining case,
an extensive defect caused the cardiac apex to be incorrectly delineated in both the 16-frame and
the 8-frame run. 4D-MSPECT succeeded in 117/120 cases, with the remaining 3 showing an
extreme divergence of the time-volume curves between 16-frame and 8-frame runs. For the
remaining 116 patients, higher EDV (+0.7%/+3.5%) and LVEF (+2.5%/+4.8%) and lower ESV
(-3.2%/-1.4%) (4D-MSPECT/QGS) were found for 16-frame runs. Bland-Altman limits were
smaller for QGS than 4D-MSPECT (EDV 32/12ml, ESV 21/10ml, LVEF 17/7% (4DMSPECT/QGS)). Conclusion Contour finding by QGS was stabler than by 4D-MSPECT in
lower count data sets. Both algorithms met the expectations of higher EDV-/LVEF- and lower
ESV-values. These effects are readily explained by the better temporal resolution of the 16-frame
data, which gives more precise, and hence less blurred images, of the final phases. Nevertheless,
due to better count statistics and only small and predictable effects, we favor 8-frames for clinical
routine. If 16-frame acquisition still must be done, the data should be rebinned to 8-frame format
with a second evaluation to check the results of the 16-frame data, particularly with use of the
4D-MSPECT algorithm.
Evaluation of exercise-induced myocardial stunning by
means of immediate post exercise Tc-99m sestamibi gated
A. Manrique, M. J. Ouvrier, M. Bernard, J. Roset, C. Ribeiro, P. Vera;
Centre Henri Becquerel and Laboratoire Quant.IF., Rouen, France.
Repeated episodes of exercise-induced myocardial stunning may lead to chronic ventricular
dysfunction in patients with myocardial ischemia. The aim of this study was to assess the
parameters that may influence the occurrence of post exercise myocardial stunning in patients
undergoing myocardial perfusion imaging.Methods: Six hundred consecutive patients with
known or suspected coronary artery disease were prospectively studied (438 men, mean age 61 ±
11 y/o). All underwent a stress and rest Tc-99m sestamibi gated SPECT (one day protocol).
Symptom limited exercise stress was performed on a treadmill (Bruce protocol). Tc-99m
sestamibi (3.7 MBq/kg) was injected at peak exercise and acquisitions were performed within 15
minutes on a dual camera (16-frame gating, 60 sec/projection). Rest imaging was obtained 3 hrs
later, after a rest injection (11.1 MBq/kg). Summed stress (sss), rest (srs) and difference (sds)
perfusion scores were calculated using the automated QPS software. LV function (EDV, ESV,
and EF) was assessed using the QGS software on post stress and rest imaging. Myocardial
stunning was defined as the association of significant reversible perfusion defect (sss • 4 and sds
> 0) and a decrease in post stress > 5% compared to rest EF.Results: Ischemia was found in 221 /
600 patients (37%). Among these patients, 61/221 (27%) showed myocardial stunning (G1) and
160/221 (73%) did not (G2). Compared to G2, G1 patients had decreased rest ESV (G1: 48±25
ml vs G2: 65±545 ml, p= 0.03), increased rest EF (G1: 58±11% vs G2: 52±14%, p= 0.005) and
decreased post-stress EF (G1: 49±10% vs G2: 53±14%, p < 0.02). Perfusion scores, stress and
rest EDV and stress ESV were not significantly different between G1 and G2 [respectively: sss:
13±8 vs 12.6±9 (ns), srs: 5.8±7 vs 6±8 (ns), sds: 7.2±4 vs 6.4±4 (ns), stress EDV: 104±41 ml vs
115±60 ml (ns), rest EDV: 108±37 ml vs 121±63 ml (ns), stress ESV: 56±30 ml vs 61±51 ml
(ns)]. Logistic regression showed that rest EF > 45% was an independent predictor of myocardial
stunning in patients with ischemia (RR: 4, p < 0.03), whereas rest EDV, rest ESV and extent of
myocardial ischemia were not.Conclusions: In patients with myocardial ischemia, exerciseinduced myocardial stunning was associated with preserved rest ventricular function. A rest EF >
45% was the only independent predictor of exercise-induced myocardial stunning in this clinical
Myocardial stunning during gated SPECT indicates presence
of severe coronary artery disease
G. Biswas, S. Mohanadi, H. Salman, A. R. Shukur; Chest hospital, Kuwait
city, Kuwait.
Aim:Transient LV dysfunction is known to occur following stress in patients with coronary
artery disease[CAD]. Our aim was to show that post stress LV dysfunction[myocardial stunning]
seen with gated SPECT is a powerful indicator of severe coronary artery disease.Materials and
Methods:We evaluated 62 patients [38males,24 females, mean age 47 yrs] referred with chest
pain for inducible myocardial ischaemia. All patients underwent two day stress/rest Tc-99mtetrofosmin myocardial perfusion gated SPECT[GSPECT] with dipyridamole[n=35] or physical
exercise[n=23] or dobutamine[n=4].On the basis of a 20 LV segments model stress/rest GSPECT
analysis evaluated perfusion as stress/rest summed scores[SSS,SRS] and difference scores[SDS]
and regional functional indexes RWM and RWT.LV function was also obtained as post stress
and basal LVEF.SDS[0-1] was considered normal,SDS[2-6] was considered mild to moderate
ischaemia and SDS>6 was considered to be severe ischaemia. Results17patients showed
presence of severe ischaemia [SDS>6] with post stress reduction of LVEF[mean 7.3±3.8] and
RWM and RWT abnormalities in ischaemic myocardial segments compared to basal values. All
patients in this group showed presence of extensive double or triple vessel CAD on
angiography.25 patients showed mild to moderate ischaemia[SDS 2-6] with mild decrease in post
stress LVEF[mean 2.8±1.7] and no RWM and RWT abnormalities compared to basal values.18
patients in this group underwent angiography and all of them had mild to moderate and less
extensive CAD than the previous group.20 patients were considered normal[SDS 0-1] and
showed post stress rise of LVEF[mean 5.6±2.5] compared to basal values.Conclusion: Our study
shows that myocardial stunning during gated SPECT is induced by severe post stress myocardial
ischaemia and is a powerful indicator of extensive coronary artery disease.
Diagnostic value of ecg-gated tc-99m tetrofosmin spect wall
thickening abnormalities in regions with normal or
equivocally reduced resting perfusion
O. Partos1, K. Zupán2, G. Fontos1, B. Kári3; 1Hungarian Inst.of Cardiology,
Budapest, Hungary, 2Semmelweis University, Dept. of Cardiology,
Budapest, Hungary, 3Semmelweis University, Dept. of Diagnostic
Radiology and Oncotherapy, Budapest, Hungary.
Aim. Ischemic heart disease results in regional functional and perfusion abnormalities of the left
ventricle. These can be evaluated by ECG-gated myocardial SPECT simultaneously. The
objective of this study was to assess the importance of regional systolic wall thickening
abnormalities in myocardial regions with normal or nearly normal resting perfusion (where
transmural necrosis can be excluded). Materials and Methods. Clinical and imaging data of 79
consecutive patients, referred for both ECG-Gated Tc-99m Tetrofosmin SPECT and coronary
angiography (CA) were retrospectively studied. Patients with a history of bypass surgery and
with intraventricular conduction disorders were not included. SPECT acquisition was performed
at rest in 8 frames. Segmental wall thickening and perfusion defects were evaluated in a 17segment model with Emory Cardiac Toolbox™. Regional score values were calculated for the
"normal" or "abnormal". For cMRI-"normal"-patients the mean values of the gated SPECT
results (LVEF, and to the body-surface-area (BSA) normalized EDV (EDV_BSA) and ESV
(ESV_BSA) values) were calculated. Results In the cMRI LVEF-normal-range (male&female)
of 61-79% fell 36 from the 70 gated SPECT patients (8 female). Following data resulted: LVEF:
QGS 58.8±5.5 %; 4D-MSPECT 65.6 ± 6.7%; ECTB 68.4±5.8 % (normal ranges (mean±2SD):
QGS 48-70 %, 4D-MSPECT 52-79 % und ECTB 57-80%). For the EDV_BSA und ESV_BSA
normal values only male patients were included (too few females) (EDV_BSA 39 pat.,
ESV_BSA 34 pat.): QGS EDV_BSA 55.5±10.2 ml/m2, ESV_BSA 22.3±5.7 ml/m2; 4DMSPECT EDV_BSA 59.2±12.9 ml/m2, ESV_BSA 19.8±5.8 ml/m2; ECTB EDV_BSA
60.3±12.1 ml/m2, ESV_BSA 18.2±5.5 ml/m2 (resulting normal ranges: QGS EDV_BSA 35-76
ml/m2, ESV_BSA 11-34 ml/m2; 4D-MSPECT EDV_BSA 33-85 ml/m2, ESV_BSA 8-31 ml/m2
and ECTB EDV_BSA 36-85 ml/m2, ESV_BSA 7-29 ml/m2). Conclusions The gated SPECT
normal ranges for EDV_BSA, ESV_BSA and LVEF vary algorithm specific thus distinctly (e.g.
LVEF max. ca. 10%), that these differences must be taken into account when judging gated
SPECT results. As a consequence the use of normal values proves reliable only within one
algorithm. References: (1) Schaefer et al. J Nucl Med 2005;46:1256-63. (2) Salton et al. J Am
Coll Cardioll 2002;39:1055-60.
major coronary territories and were compared to the CA results. Results. In the 181 analyzed
vascular territories with normal or equivocally reduced resting perfusion (perfusion defect score
< 4), the semiquantitative wall thickening score was predictive of the stenosis of the
corresponding coronary artery (ROC area under the curve: 0.750 ± 0.05). The association was
comparable in normally perfused and in equivocally hypoperfused regions, and was also
observable in patients with and in those without previous myocardial infarction. These regional
wall thickening abnormalities may therefore be caused by non-transmural (subendocardial)
necrosis and / or prolonged myocardial stunning. Conclusion. Decreased regional wall
thickening in the presence of normal or nearly normal resting perfusion draws attention to the
involvement of the supplying coronary artery.
[Purpose] The aim of this study is to evaluate the correlation between colorectal occult cancer
risk and the risk factors such as high degree of incidental colonic fluorodeoxyglucose (FDG)
uptakes, serum CEA elevations, and positive fecal occult blood test (FOBT) findings, in patients
without colorectal carcinoma (CRC) who underwent whole-body FDG positron emission
tomography (PET) for other purposes. [Method] Cases of 300 patients without known history of
CRC with incidental colonic uptake on FDG-PET were reviewed retrospectively. In 171 patients
among these, colonoscopy was performed and histpathological diagnoses were confirmed (29
carcinomas, 142 benign lesions or negative findings), and FOBT and serum CEA findings were
also obtained. FDG-PET findings is qualitatively evaluated by two or more radiologist experts
and classified into 3 groups, strongly-positive/moderately-positive/equivocal. We constructed a
multivariate logistic regression model, using these pathological diagnosis, FDG-PET findings
and serum CEA levels and FOBT findings (performed 2 times consecutively), and estimated the
CRC risks. [Result] The odds ratios of PET strongly-positive/moderately-positive finding, serum
CEA elevation, FOBT 1-time positive/2-times positive, were 7.24, 4.42, 1.54, 2.09, 2.37,
respectively. For a patient with PET strongly-positive and negative CEA/FOBT findings, the risk
of CRC is estimated at 43.8%. For a patient with PET strongly-positive and either CEA or FOBT
positive, and positive CEA/FOBT findings, estimated CRC risk is 77.4%, 79.0%, and 94.3%,
respectively. For a patient with PET moderately-positive, if CEA/FOBT are negative, estimated
CRC risk is no more than 6.2%. On the other hand, for a patient with PET moderately-positive
and CEA positive, FOBT 1 time/2 times positive, CEA/FOBT positive cases, estimated CRC risk
is 22.4%, 26.4%, 39.9%, and 74.5%, respectively. For a patient with PET equivocal findings and
CEA positive, FOBT 1-time/2-times positive, estimated CRC risk is 1.3%, 1.6%, 2.3%,
respectively. [Conclusion] In general, the patients with incidental colonic FDG uptake are
recommended for further evaluation with colonoscopy. However, actual CRC risk may differ
according to not only FDG-PET findings, but also serum CEA status and FOBT. Therefore,
examinations of serum CEA and FOBT are useful as complementary methods for interpretation
of abnormal colonic uptake on FDG-PET. Particularly, for a patient with PET moderatelypositive and CEA/FOBT negative, or with equivocal PET findings, CRC risk is estimated to be
low, and it may yield little benefit to enforce colonoscopy on these low CRC risk patients.
Comparison of SPECT and delayed-enhanced MRI in the
diagnosis of myocardial sarcoidosis
F. Rouzet1, D. Lasalmonie1, J. Laissy1, L. Sarda-Mantel1, R. Lebtahi1, J.
Serfaty1, B. Crestani1, A. Tazi2, P. Merlet1, D. Le Guludec1; 1Hopital Bichat,
Paris, France, 2Hopital Saint-Louis, Paris, France.
Aim: Early diagnosis of myocardial sarcoidosis (MS) is essential because aggressive treatment
improves the prognosis. Since endomyocardial biopsy, which is the gold standard, has a poor
sensitivity and may be harmful, other non-invasive techniques have been assessed: single photon
emission computed tomography (SPECT) using perfusion tracers, and more recently magnetic
resonance imaging (MRI) using gadolinium infusion. The aim of this study was to compare
evidences of MS using 2 non-invasive methods: SPECT with sestamibi as a tracer and contrastenhanced MRI.Materials and Methods: Twenty-five consecutive patients (mean age: 47±9 years,
males: 48%) with histologically proven extra-cardiac involvement of sarcoidosis were included.
On rest ECG, conduction abnormalities were present in 7 pts and isolated repolarization
abnormalities in 2. Echocardiography showed left ventricular enlargement and systolic
dysfunction in 2 pts. Gated-SPECT was acquired in a 2 days protocol. All pts underwent a first
acquisition at rest, after intravenous injection of 900 MBq Tc-99m sestamibi. In case of
abnormality, a second acquisition was performed after Dipyridamole infusion. MRI included
cine-SSFP sequences, and delayed-enhanced acquisitions 10 min after gadolinium injection in
short axis, vertical and horizontal long axis. Global abnormalities were analyzed on both SPECT
and MRI and were defined as enlargement of ventricles and/or global dysfunction. Segmental
abnormalities were defined on SPECT as areas of low tracer uptake and on MRI as areas of late
enhancement. Results: Nine of the 25 pts had neither global nor segmental abnormality on both
SPECT and MRI. Four pts had global and segmental abnormalities evidenced by both methods.
Additionally, segmental abnormalities alone were present on both methods in 1 patient, only on
SPECT in 2 pts, and only on MRI in 9 pts. The total number and location of abnormal segments
on SPECT and MRI are detailed in table. Out of 7 pts with segmental abnormalities on SPECT at
rest, 2 showed an increase of the tracer uptake after Dipyridamole infusion (not performed in 2
others with a history of asthma), suggesting myocardial scar in the 3 remaining. Conclusions: In
this study, SPECT and MRI yield was similar regarding ventricular global function and size.
Conversely, the discrepancies regarding segmental abnormalities, including their location,
support that pathophysiological mechanisms evidenced by both methods are different (perfusion
vs inflammation).
SPECT (n=13)
MRI (n=22)
both SPECT and MRI
503 — Sunday, October 01, 2006, 4:30 pm - 6:00 pm, Banqueting Hall
Oncology: Colorectal Cancer
Accuracy of tumor detection based on quantitative PET data
obtained in colorectal tumors.
L. G. Strauss1, S. Klippel2, K. Schoenleben2, U. Haberkorn1, A.
Dimitrakopoulou-Strauss1; 1German Cancer Research Center, Heidelberg,
Germany, 2Klinikum Ludwigshafen, Ludwigshafen, Germany.
Aim: Assessment of the contribution of quantitative PET data to the detection of colorectal
tumors.Materials and Methods: A modified whole body protocol, combining a dynamic study
with additional static images was used. Dynamic PET examinations (60 min) with F-18Deoxyglucose (FDG) were acquired in 22 patients with colorectal tumors from the tumor region.
All PET studies were performed within two days prior to surgery. Volumes-of-interest (VOI)
were used to obtain time-dependent tracer data for the tumor region as well as for the normal
colon. The dynamic data were evaluated by applying a two-tissue compartment model and a noncompartment model to the data. SUV (55-60 min) and FDG influx were calculated.Results: All
tumors were confirmed by histology. Discriminant analysis was applied to the data, using the
histopathological data for reference. The use of the SUV alone resulted in a misclassification rate
of 22 %, mainly due to false negative results in tumors with a low FDG accumulation. The noncompartment model (calculation of the fractal dimension (FD)) was the best single parameter
with a misclassification rate of 14 %. The two-tissue compartment model provided the
compartment parameters k1-k4 as well as the fractional blood volume (VB) in the target area.
We obtained a misclassification rate of 8 % if only the compartment data were used to
differentiate tumor and normal colon. Best results were obtained using the model parameters and
FD with a misclassification rate of 6 % (one false negative, one false positive).Conclusions: Best
results are obtained if quantitative dynamic data are used to detect a colorectal tumor. The use of
only the SUV limits the sensitivity to less than 80 %.
Assessment of colorectal occult cancer risk in patients with
incidental focal colonic fluorodeoxyglucose uptake:
correlation with serum tumor marker and fecal occult blood
test, multivariate logistic regression analysis.
K. Shibata, K. Uno, J. Wu, W. Ka, Y. Matsuo, T. Suzuki; Nishidai Clinic
Diagnostic Imaging Center, Tokyo, Japan.
How FDG-PET impacts the management of patients with
colorectal cancer ? Preliminary results of a follow-up study
on 341 patients.
J. Esnault, H. Kolesnikov-Gauthier, A. Adenis, C. Fournier, P. Carpentier;
Centre Oscar Lambret, Lille, France.
Background/ Aim / Purpose: FDG-PET is widely used throughout western world as a
diagnosis, staging and prognosis tool in a large spectrum of malignant diseases. It has been found
particularly accurate in the staging of colorectal cancer. On the basis of a first retrospective
analysis, we launched a prospective cohort study in order to evaluate how FDG-PET impacts the
management of patients with colorectal cancer. Methods: From 1/2003 to 7/2005, 341
consecutive patients (mean age of 63 yr) with colorectal cancer underwent 423 PET scans. A first
set of 183 patients with 236 PET (group A) were retrospectively included in a specific database
on the basis of our medical charts. A second set of 158 patients with 187 PET (group B) were
prospectively included from 3/2004 to 7/2005. A questionnaire was sent to the referring
physician of each individual pt, before and after the PET exam, to assess whether and how PET
altered their own decision-making process. Physicians were also questioned about their own
perception of disease extent and intended treatment before PET, and staging and choice of
treatment after PET. PET findings were validated either by histological proof (in case of
surgery), or by conventional imaging or by follow up (mean: 13 months). Results: These
preliminary results are based on 206/341 patients and 261/423 PET (group A: 62%; group B:
60%). Indications of PET were as follows: staging before surgery (126/261, 49%), suspicion of
recurrence (43/261, 17%), elevated tumour markers with normal imaging (16%), other reasons
(18%). With PET findings, the tumour stage changed in 180/261 (69%) of the cases with up- or
down-staging in 46% and 18% of the cases, respectively. These changes in terms of staging
induced changes in terms of clinical therapeutic management in 101/261 (39%) of the cases.
Most of the time the therapeutic change was to move from a surgical treatment to a medical one
(chemotherapy or best supportive care or radiation therapy) (53/101 of the changes). Conversely,
in 17/101 of the changes, the referring physician moved from a medical treatment to a surgical
treatment with curative intent. Finally when PET imaging was assessed for an isolated increased
of tumour markers, hot spots were found in 30/42 (71%) cases which induced therapeutic
changes in 20 cases (surgery: 6/20; medical treatment: 14/20). Conclusions: These preliminary
results show that FDG-PET imaging has a major impact on the therapeutic management of
patients with colorectal cancer.
Usefulness of FDG PET/TC in patients with suspect of
relapse of colon rectal cancer.
A. Moretti1, F. Matteucci1, R. Galassi1, S. Fanti2, M. Bertocco1, G.
Fiorentini1, M. Farsad2; 1Morgagni Pierantoni Hospital, Forlì, Italy, 2S.
Orsola Malpighi Hospital, Bologna, Italy.
PURPOSE: aim of this study was to evaluate the impact of 18F-fluoro-2-deoxy-D-glucose FDG
PET/CT versus CT alone in restaging patients with suspect, clinical, biochemical or with other
conventional diagnostic studies, of colon rectal cancer relapse. Clinical efficacy of positron
emission tomography was evaluated with regard of the impact in patient management.Materials
The role of PET/CT in restaging of colorectal cancer
V. Valotassiou, J. Malamitsi, R. Efthimiadou, M. Pagou, A. Rousakis, J.
Andreou; Department of PET/CT D.T.C.A. Hygeia - Harvard Medical
International, Athens, Greece.
Aim: The aim of the study was a) to determine sensitivity and specificity rates of PET/CT scan
in restaging of colorectal cancer as compared with high dose contrast CT and b) to assess the
impact of PET/CT scan on treatment strategy in patients with recurrent colorectal cancer.
Materials-methods: Thirty-one patients (16 men, 15 women, mean age: 63.2 years) with
histologically proven colorectal cancer underwent restaging on a PET/CT LSO tomograph. Two
nuclear medicine physicians and two radiologists evaluated PET, CT and fused PET/CT images.
Disease stage was determined by using the TNM and the American Joint Committee on Cancer
staging systems. Histology and a more than 6 months follow-up served as the reference
standards.Results: The total number of lesions was for PET alone 63, for CT alone 57 and for
PET/CT fused images 79. PET/CT sensitivity and specificity were 95% and 96% respectively, vs
65% and 90% for CT values. PET scan was true positive in 3/5 patients with elevated tumor
markers (CEA and CA 19-9) and a negative CT study. PET was false negative in 3 patients with
micronodular pulmonary lesions that proved to be metastatic. On PET/CT, 14/31 patients were
correctly upstaged, 4/31 patients were correctly downstaged and 13/31 patients were diagnosed
correctly as being on the same stage. On the basis of PET/CT results, a change of management
was induced in 11/31 patients (35%).Conclusions: PET/CT scan provides important information
on restaging of colorectal cancer and therefore plays a significant role in the management of
patients with suspected recurrent colorectal cancer.
FDG-PET for prediction of survival of patients with metastatic
colorectal carcinoma
L. de Geus-Oei, B. Wiering, P. F. M. Krabbe, T. J. M. Ruers, C. J. A. Punt,
W. J. G. Oyen; University Medical Center Nijmegen, Nijmegen, The
Aim: The current study focuses on the prognostic value of pretreatment metabolic activity in
metastatic lesions as measured with [18F]fluorodeoxyglucose positron emission tomography
(FDG-PET), as an indicator of survival in patients with colorectal cancer.Methods: In a
prospective series of 152 patients with metastatic colorectal cancer, of whom 67 were treated
with resection of metastases and 85 with chemotherapy, standardized uptake values (SUV) as a
semiquantitative parameter of glucose metabolism determined with FDG-PET, were calculated
prior to treatment. Survival probabilities were estimated by the Cox proportional regression
analysis. For Kaplan-Meier analysis SUV was stratified by the median value. Survival
differences between groups were assessed using the log-rank test.Results: SUV in metastases
proved to be a significant predictor for overall survival (hazard ratio 1.17, 95% confidence
interval 1.06-1.30, p = 0.002), independent of the subsequent treatment. According to the median
value of the patient population a low uptake (SUV < 4.26) and high uptake group (SUV > 4.26)
was defined. The median survival and the 2- and 3-year survival rates were 32 months, 59% and
45%, respectively, in the low-uptake group and 19 months, 37% and 28%, respectively, in the
high-uptake group (p = 0.017).Conclusions: A significant survival benefit was observed in
patients with low FDG uptake in metastases of colorectal cancer. SUV calculations provide
relevant prognostic information prior to any surgical or medical treatment, and may therefore be
considered an important guide in decision-making and stratifying patients for prospective studies
when different therapeutic options are to be compared.
Neoadjuvant treatment consisted of external beam radiotherapy (45 Gy) to the pelvis and 5fluoruracile-based chemotherapy. Both FDG PET/CT and CMA were obtained 1-2 weeks before
preoperative radiochemotherapy and repeated 4-6 weeks after completion of RCT. PET/CT
evaluation included semi-quantitative measures of tumour uptake (SUV max value and mean
value) before and after treatment. PET/CT and CMA data were compared with histopathological
findings on surgical specimens. Pathological assessment of response was made using a tumor
regression grade score (TRG) Results. Histopathological evaluation of the resected tumour
revealed complete tumour regression (TRG1) in 7 patients, rare malignant cells (TRG2) in 8,
malignant disease with predominant fibrosis (TRG3) in 2, and persisting cancer outgrowing
fibrosis (TRG4) in other 2 cases. PET/CT correctly identified 11 out of 15 patients with
pathologic response TRG1 and TRG2, showing a decrease of SUVmax, as well as SUVmean,
>70% in the rectal wall, whereas all 4 patients with persisting malignant disease (TRG3 and
TRG4) showed a decrease of SUV <70%. In contrast, among 15 patients with tumour regression
TRG1 and TRG2, CMA assessed complete response (T0 disease) in 4 patients and persisting
malignant disease (T3 an T4 disease) in 8. Conclusion. In our small series of patients FDG
PET/CT improved the assessment of rectal cancer response to preoperative RCT in comparison
with CMA, since PET/CT estimated more accurately the extent of the pathological response.
FDG PET has a great potential in the evaluation of rectal cancer response to preoperative RCT to
select patients for conservative surgical treatment.
Planning transarterial radioembolisation of colorectal liver
metastases with Y90-microspheres: diagnostic imaging
T. Denecke, E. Lopez Hänninen, R. Rühl, M. Plotkin, L. Stelter, J. Ricke, R.
Felix, H. Amthauer; Charité - Universitätsmedizin Berlin, Berlin, Germany.
Aim: We assessed the value of a variety of radiological and nuclear medicine exminations for
patient selection and risk minimization in transarterial radioembolisation (RE).Materials and
Methods: 20 patients (m, 14; f, 6; age, 58(±17) years) with known irresectable colorectal liver
metastases were prospectively enrolled in this preliminary analysis. They underwent the
following examinations for staging: 16-slice-CT (thorax/abdomen including CT-arteriography
(CTA), n=20), MRI (liver, hepatocyte specific contrast agent, n=16), and FDG-PET (stand alone
PET, n=8; PET-CT, n=7). If RE with Y90-microspheres (Sirtex, Australia) was indicated, a
digital subtraction angiography (DSA) with transarterial hepatic perfusion scintigraphy (planar
and SPECT-CT) with Tc99m-MAA was performed (n=12). The results were analysed regarding
their impact on the treatment strategy.Results: CT revealed contraindications for RE in 4/20
patients (portal vein thrombosis, n=1; extrahepatic progression, n=3). MRI revealed peritoneal
carcinosis in 1 patient. In another 3 patients, locally ablative treatment was chosen instead of RE,
based on the MRI and PET findings. FDG-PET showed bone metastases in 1 patient (not
visualized on CT). In 1 patient, FDG-PET excluded a recurrent rectal carcinoma (suggested on
CT). SPECT-CT revealed unexpected gastrointestinal MAA accumulation in 3/12 patients. The
responsible non-target arteries were not visualized in initial DSA, but were successfully avoided
during a second application of MAA. CTA failed to visualize the small aberrant gastrointestinal
vessels responsible for extrahepatic MAA accumulation (n=3). Planar scintigraphy revealed an
increased hepatopulmonary shunt volume, relevant for dosimetry, in 5 cases. Treatment was
successfully delivered in 12 patients without pulmonary or gastrointestinal
morbidity.Conclusions: CT is a useful first line diagnostic tool in RE planning. Additional PET
and MRI improves the evaluation process. CTA is not capable to visualize small non-target
arteries. The latter are best seen on SPECT-CT.
504 — Sunday, October 1, 2006, 04:30 pm — 06:00 pm, Mitropoulos
CTE 3: Education & E-learning
E-learning in practice
Martha Webb Pickett (USA)
VirRAD a virtual learning resource
Stephen J. Mather (UK)
505 — Sunday, October 01, 2006, 4:30 pm - 6:00 pm, Scalcotas
Assessment of rectal cancer response to neoadjuvant
radiochemotherapy using FDG PET/CT: correlation with
histopathology and conventional imaging
Oncology: Neuroendocrine Cancer Scintigraphy
E. Borsatti1, A. De Paoli2, V. Canzonieri3, C. Rossi4, T. Baresic1, R. Ruffo1,
G. Boz2, R. Sigon5, M. Cimitan1; 1Nuclear Medicine CRO, Aviano, Italy,
Radiotherapy CRO, Aviano, Italy, 3Pathology CRO, Aviano, Italy, 4Surgery
CRO, Aviano, Italy, 5Surgery CRO, Aviano, Italy.
Comparison of in-111 octreotide scintigraphy with tc-99m
hynic-toc/tate scintigraphy in the same patient group for
diagnosis of somatostatin receptor expressing tumors
Background. Clinical and morphological imaging techniques (EUS, MRI) have revealed a lack
of accuracy in predicting response of rectal cancer induced by preoperative chemoradiation
(RCT), because of their limited specificity in differentiating post-therapeutic scar from vital
tissue. In contrast, FDG-PET seems to accurately estimate the extent of pathologic response of
rectal cancer to preoperative treatment. Aim. In this study we evaluate the usefulness of FDG
PET/CT in adjunct to conventional clinical and diagnostic morphological (EUS,MRI) assessment
(CMA) to improve the response prediction of rectal cancer after RCT. Materials and methods.
19 consecutive patients with proven locally advanced rectal cancer were enrolled in this study.
S. Sager, L. Kabasakal, S. Yımaz, M. Ocak, Y. Sanlı, H. Maecke, C. Onsel,
I. Uslu; Istanbul University, Cerrahpasa Medical Faculty, Istanbul, Turkey.
Objectives: In-111 labelled Octreotide scintigraphy has been shown to be a valuable tool for the
diagnosis of somatostatin expressing tumours and this method has been widely used for imaging
of this tumors. However, In-111 has several drawbacks such as high radiation burden, high cost
and limited availability. To overcome these drawbacks octreotide (TOC) and octreotate (TATE)
has been recently labelled with Tc-99m, using hynic as a conjugate. The aim of the present study
was to compare the diagnostic value of In-111 Octreotide scintigraphy and Tc-99m Hynic
and Methods: 120 patients (69 man, 51 female, mean age 65.54 aa) with history of surgery and
chemo/radiotherapy for colon rectal cancer were enrolled in this study with suspect of local
recurrence or relapse of disease documented with elevation of biotumoral markers (34 pts) and/or
conventional imaging dubious (86 pts). All patients were far from therapy and were studied with
PET/CT (Biograph Sensation 16 Siemens) and standard CT in a short time (2-3 weeks) between
the two examinations. All patients were fasted and blood glucose level were evaluated before the
injection i.v. of 370 MBq FDG, PET scan were performed after 60 minutes from the injection
with standar tecnique (iteractive reconstruction of the images with attenuation weighted OSEM,
slices of 3.4 mm). Clinical correlation was confirmed histopathologically at time of operation or
by follow up of at least of 1.6 years.Results: the overall sensitivity, specificity and accuracy of
FDG PET/CT in detection of lesions were higher (94.87, 95.24 and 95% respectively) than for
CT (79.71, 35.29 and 65.05%). FDG PET/CT altered clinical management in a beneficial manner
in 70 percent of cases (84/120) when compared with evaluation by computed tomography plus
other conventional diagnostic studies.Conclusions: FDG PET/CT is more sensitive than CT for
the detection of metastatic or recurrent colorectal cancer and may improve clinical management
in almost three-quarter of cases.
Toc/Tate scintigraphy in the same patient group.Methods: The study was composed of 24
patients, 14 of them were female and 28 of them were male. The ages were ranged from 30 to 74
years with a mean of 48.83±13.46. Among 24 patients there were 11 Neuroendocrine Tumors, 7
Carsinoids, 2 Medullary Thyroid Cancers, 2 Ectopic Cushing Disease, 1 Pheocromasitoma and 1
Lung Cancer. In-111 Octreotide was performed 4 hour and 24 hour after injection. Scintigraphy
with Tc-99m Hynic Toc/Tate was performed 1 and 4 hour after injection. . In 18 patients Tc-99m
hynic/TOC and in 6 patients Tc-99m hynic/TATE were studied. SPECT studies of areas of
interest were permormed 4 hour after injection for both tracers as well as at 24 hour after
injection for In-111 Octreotide. The time interval between the studies using each tracer ranged
from 2 to 18 days. None of the patient received any treatment between two studies.Results: The
diagnostic results of both studies were similar. Both studies were positive in 15 patients (62.5 %)
and negative in 9 patients (37.5 %). However the number of the lesions is higher with Tc-99m
Hynic Toc/Tate. There were 92 lesions with Tc-99m Hynic Toc/Tate and 64 lesions with In-111.
The image quality was higher and less liver and bowel uptake was observed with Tc-99m Hynic
Toc/Tate images as compared to those with In-111 images. SPECT images were extremely
helpful in detecting and localizing lesions in both studies.Conclusions: Technetium labelled
somatostatin receptor analogs octreotide and octreotate are excellent alternative to In-111
labelled octreotide for imaging somatostatin receptor positive tumors. Radiolabelling is easy,
they are readily available for routine use and they give excellent images for clinical diagnosis.
54 (-)ve and in group C 26 (+)ve and 105 (-)ve. No patient in group B had a pre-intervention
Octreoscan®. Three of 12 patients (25%) in group B with negative initial scans developed
positive scans at a later period (up to 6 years later).Conclusions: 1) The Octreoscan® provided
significant positive diagnostic information’s in 20% of patients referred for possible
neuroendocrine tumors (group C). 2) The overall sensitivity of detection of known SSTR
expressing tumors (group A) in the study was only moderate 44%. 3) It is hypothesized that
patients in theses two groups with (-)ve scans would not benefit from restaging follow-up
Octreoscan®. 4) It is suggested though not proven that patients from group B with an initial or
sometime late (+)ve follow-up Octreoscan® would have had a (+)ve pre-intervention scan if been
performed. 5) A negative initial scan in group B could not exclude follow-up Octreoscan®, since
a preoperative scan was not available to detect possible SSTR in the primary tumor.
AIM: To prospectively determine whether 111In-Pentetreotide SPECT/CT (Octreo-SPECT/CT) is
more accurate than 111In-Pentetreotide SPECT(Octreo-SPECT) for the detection and localization
of neuroendocrine tumors (NETs). Materials and Methods: Sixty consecutive patients with
known or suspected NET underwent 111In-Pentetreotide scintigraphy using an integrated
SPECT/low-energy-CT system. Planar images were acquired 6h and 24h after tracer injection
while abdominal and thoracic SPECT/CT were performed respectively at 6h and 24h. SPECT
and fused SPECT/CT images were interpreted separately and a lesion-by-lesion analysis was
performed with regard to classification (probability of NET graded on a five-point scale) and
localization of each abnormal focal tracer uptake. A patient-by-patient analysis for likehood of
NET in each patient was also performed. Reference for confirming the presence or absence of
NET was either histophatology or clinical/imaging follow-up data. The value of SPECT/CT
imaging was assessed by using receiving operating characteristic analysis (ROC) and the
McNemar test.Results: A final diagnosis of NET was achieved in 29/60 patients and a total of
116 areas (90 NET and 26 benign/physiological) with abnormal tracer uptake were included in
the final lesion-by-lesion analysis. ROC analysis showed that Octreo-SPECT/CT performed
significantly better than Octreo-SPECT for the detection of NET on both the patient and lesionbased analysis, improving especially evaluation of abnormal tracer uptake in the abdomen.
Moreover, Octreo-SPECT/CT allowed an accurate localization of 93.3% of the lesions
significantly higher than SPECT alone (44.8%) with greatest improvement for nodal and skeletal
lesions.Conclusions: Octreo-SPECT/CT imaging appears to be more accurate than simple
Octreo-SPECT for the detection and localization of NETs with major benefit for the evaluation
of abdominal abnormalities.
The use of a new somatostatin analogue ( Tc-EDDA/HYNICoctreotate) in pre- and intraoperative visualisation of
pancreatic neuroendocrine tumours.
A. Hubalewska-Dydejczyk1, K. Fröss-Baron1, P. SzybiĔski2, R.
Mikoáajczak3, D. Pach1, A. Sowa-Staszczak1, A. Gilis-Januszewska1, B.
Huszno1, J. Kulig2; 1Nuclear Medicine Unit, Chair and Department of
Endocrinology Medical College, Jagiellonian University, Krakow, Poland,
Gastrointestinal and General Surgery Department, Medical College,
Jagiellonian University, Krakow, Poland, 3Radioisotope Centre Polatom,
Otwock-Swierk, Poland.
Introduction: Pancreatic NETs are relatively rare endocrinopathies with malignant potential.
They are usually recognized because of specific clinical symptoms and some of them cause
serious difficulties in imaging diagnostics. Diagnostic procedures like USG, EUS, contrast
enhanced CT, MRI, MRI angiography can detect the primary lesion and metastases with varying
from 9% to 76 % sensitivity. SRS become a routine investigation and one the most important tool
in localizing and staging pancreatic NETs. A new somatostatin analogue 99mTc marked octreotate - should significantly improve sensitivity of the diagnostics due to better affinity to
SSR2 and the higher count rate in comparison to 111In -Octreoscan used until recently. Aim:
Assessment of the usefulness of 99mTc-EDDA/HYNIC-octreotate in pre- and intraoperative
visualization of pancreatic NET.Materials and Methods: 99mTc-EDDA/HYNIC-octreotate
(740MBq) scintigraphy was performed in 44 patients (mean age 50,3±17,3y). Group A: 23 pts
with suspected pancreatic NET on the basis of clinical symptoms, biochemical tests, biopsy
results and/or the presence of the focal lesion in CT/MRI. Group B consisted of 11 pts with
postoperatively confirmed NET (glucagonoma-4, gastrinoma-3, insulinoma-1, somatostatinoma1, non-functioning-3) with suspected local recurrence and metastases. Target/non-target counts
ratio was assessed for focal lesions. The patients with positive results of scintigraphy were
qualified for the operation (some of them with RGS and IOUSG). The subjects with disseminated
tumours were referred for radionuclide therapy with 90Y-DOTA-octreotate.Results: Group A: the
result of scintigraphy was positive in 8 pts. Histopathology revealed: insulinoma in 5 pts (3 surgery was successful because RGS had been used), non-secreting NET in 1pt, and gastrinoma
in 1, glucagonoma-1. CT and MRI results were negative in 4 of these pts. Group B: SRS
visualized metastatic lesions in 5 pts (gastrinoma-1, glucagonoma- 3, non-functioning-1) and
focal lesion suggesting local recurrence in one patient with diagnosed malignant insulinoma- in
hist.-pat. a postoperative cyst with inflammatory reaction. The mean target/non-target counts
ratio was the following: 2,8±0,9 for pancreatic NET and 4,1±0,8 for liver meta. SRS and
CT/MRI sensitivity was 100% and 57% respectively in our group of patients.Conclusions:
Tc-EDDA/HYNIC-octreotate SRS seems to be a sensitive method in pre- and intraoperative
visualization of pancreatic NETs but the further studies in bigger group of patients are necessary.
The positive SRS result, with accompanying clinical symptoms and biochemical tests suggesting
pancreatic NET, should be the indication for operation, even with negative results of other
imaging methods.
The role of positive and negative “baseline” Somatostatin
Receptor Scintigram (SRS) in follow-up studies
J. Koutsikos, V. Valotassiou, S. Tsiouris, E. Mainta, I. E. Datseris, E.
Papadaki, M. Souvatzoglou, V. Makripoulias, M. Passadi, C. Zerva, A.
Leondi; Department of Nuclear Medicine, Alexandra University Hospital,
Athens, Greece.
Aim: SRS with In-111- DTPA- octreotide (Octreoscan®) can offer additional diagnostic
information in the pre- and postoperative management of patients with various tumors of
neuroendocrine and other origin. The aim of this retrospective study was to evaluate the
usefulness of SRS in follow-up protocols and the role of a baseline study in the decision making
for such studies to be included in such protocols.Materials and Methods: We studied 250
patients, in order to evaluate pathological conditions that express SSTR (50 in pituitary gland, 32
with thyroid cancer, 109 with GEP tumors, 27 in APUDomas, 11 with lung cancer and 21 in
miscellaneous tumors that express SSTR). SRS was performed 4 hr after injection of 5 mCi
Octreoscan®. In all patients there were available conventional radiologic examinations (CT, MRI,
U/S) and in 119 patients also pathologic findings (FNA and/or biopsy). In 34 patients (group A)
staging baseline Octreoscan® was performed (pre-intervention scan) in cases of histologically
established neuroendocrine tumor. In another 85 patients (group B) Octreoscan® was performed
for the detection of residual and/or secondary disease (post- intervention scan). In 131 patients
(group C) diagnostic baseline Octreoscan® was performed in suspected neuroendocrine tumors
due to clinical symptoms and imaging and following various other imaging and laboratory
findings.Results: In group A we interpreted 15 (+)ve scans and 19 (-)ve, in group B 31 (+)ve and
Octreo-SPECT/CT imaging for accurate detection
localization of suspected neuroendocrine tumours.
M. Perri, P. Erba, D. Volterrani, E. Lazzeri, G. Boni, G. Manca, G. Mariani,
G. Mariani; Nuclear Medicine, Pisa, Italy.
Comparison of 111In-Octreotide and 99mTc-Depreotide in the
detection and follow up of neuroendocrine tumors
M. T. Siabanopoulou1, T. Kalathas1, E. Raptou1, V. Hatzipavlidou1, J.
Giavroglou1, M. Kotzassarlidou2, N. Salem1; 1Department of Nuclear
Medicine, "Theagenion" Cancer Institute, Thessaloniki, Greece,
Department of Medical Physics, "Theagenion" Cancer Institute,
Thessaloniki, Greece.
In vivo somatostatin receptor (SST-R) scintigraphy using Octreoscan is a valuable method for the
visualization of human endocrine tumors and their metastases. Since neuroendocrine tumors are
known to express various somatostatin receptor subtypes, new alternative SST radioligands have
been synthesized for diagnostic or therapeutic use in vivo. Improved image quality with lower
radiation doses and faster tumor visualization (1-d protocol) can be expected with the recently
developed radiotracer 99mTc-Depreotide (Neospect). Aim: The aim of this study is to evaluate
the diagnostic ability of 99m-Tc Depreotide compared with that of 111-In Octreotide, to detect
and localize endocrine tumors. Final diagnoses, in patients considered to have a GEP tumor or
metastases, were based on the results of the recent conventional radiological imaging methods
(CT, MRI). Methods: 21 patients (13 men, 8 women, age range: 35-78y, mean age: 56,5y) with
histological proven neuroendocrine tumors were prospectively included (8 patients with
carcinoid tumors, 1 patient with medullary tumor, 12 patients with GEP or other functioning
endocrine tumors). 111-In Octreotide planar images (whole body, head, chest, abdomen) were
obtained 4h and 24h after IV injection of 148MBq (4mCi), while SPECT (thorax-upper
abdomen) was performed 24h post injection. Similar, planar images and SPECT were obtained
3h-4h after the IV injection of 740MBq (20mCi) 99mTc-depreotide. All scintigrarphy image
findings were evaluated qualitatively as mild and moderate to marked uptake. The tumor-tobackground ratios in the lesions were compared. Results: 111-In Octreotide detected tumor sites
in 13 patients whereas 99mTc-depreotide detected lesions in 6 patients. Focusing on liver
metastases, in 7 patients were detected by 111-In Octreotide scintigrarphy, while in only 2
patients 99m-Tc depreotide showed abnormal uptake. In that case, the ratio of tumor uptake to
normal-tissue uptake was significantly higher for 111-In Octreotide scintigraphy. Conclusions:
In patients with neuroendocrine tumors, 111-In Octreotide scintigrarphy appeared to be more
sensitive especially for liver metastases, while the detection rate of 99mTc-depreotide was higher
in pulmonary lesions.
The Utility of Somatostatin receptor imaging in detection of
of breast neoplasm and Correlates with Expression of SSTR
F. Wang; Nanjing Medical University ,Nanjing first hospital, Nanjing Jiangsu
province ,China, China.
Aim:o evaluate the value of 99mTc-Sandostatin somatostatin receptor scintigraphy in the detection
of primary breast tumor and axillary lymph-node invovement in comparison with 99mTc-MIBI
scintimammography, to correlate with the expression of somatostatin receptors.Materials and
Methods: 42 female patients(range,30-71y mean age :57y) with palpable breast lesion were
included in this study. 99mTc- Sandostatin scintigraphy were undertaken in all patients, 99mTc-
506 — Sunday, October 1, 2006, 04:30 pm — 06:00 pm, Allegro
Symposium: Monoamine Transporters as Targets for
PET and SPECT Imaging…
Andrea Varrone (IT)
Jan Booij (NL)
Kirk A.Frey (USA)
First experiences with ga-68-dota-lanreotide pet in tumor
T. Traub-Weidinger, E. von Guggenberg, G. Dobrozemsky, D. Kendler, W.
Eisterer, R. Bale, D. Heute, M. Gabriel, C. Decristoforo, I. Virgolini; Medical
University of Innsbruck, Innsbruck, Austria.
Aim: Ga-68 radiolabeled somatostatin receptor (SSTR) peptide positron emission tomography
(PET) has recently been evaluated in patients with SSTR positive tumors. First promising results
in lung and thyroid tumor patients with In-111-DOTA-lanreotide (DOTALAN) scintigraphy
have been described. We report our first experiences with Ga-68 labeled DOTALAN. Methods:
Seven patients (2 patients with NSCLC, 2 patients with SCLC, 2 patients with radioiodine
negative metastatic thyroid cancer and 1 patient with medullary thyroid cancer) were
investigated. After intravenous injection of 75-150 MBq Ga-68-DOTALAN patients were
imaged with a dedicated PET system (GE Advance). Dynamic studies over the known tumor site
within the first 40 minutes were performed in 3 patients, and 2 whole body scans 20 minutes and
50 minutes after tracer injection in 2 patients. Whole body acquisitions 90 min after injection
were done in all 7 patients. Images creation was performed by means of itertative reconstruction
utilizing additional transmission scans for attenuation correction. Vital parameters were recorded
during the PET study and up to 24 hours after tracer injection. Blood and urinary sampling was
done until 4 hours after tracer injection in 6 patients. PET results were compared to CT and/or
MRI. Results: After an initial rapid flow from blood down to 25.6 ± 4.5% of the incjected dose
(ID) after 40 minutes and 20.7 ± 4.1% of the ID after 1 hour, the radioactivity was excreted into
the urine and amounted to 35.6% of the ID within the first 4 hours. No acute side effects were
recorded. In all 3 patients with dynamic studies the tumor sites were visualized during the first
minutes by Ga-68-DOTALAN PET. In the whole body acquisitions the primary tumor/local
recurrence was found in 6/6 patients, lung metastases in 3/4 patients, thoracal lymph node
metastases in 4/4 patients, adrenal gland metastasis in 1/1 patient and bone metastases in 2/2
patients. The CT/PET comparison showed concordant results in 3/7 patients and partially
concordant results in 4/7 patients. Especially micronodular metastases (<4mm) of the lung and
the cerebrum were not visualized by Ga-68-DOTALAN PET. The maximal SUVof lund and
bone tumor lesions ranged from 6 to 8. Conclusion: Ga-68-DOTALAN visualized the majority
of CT known tumor lesions without any acute side effect. Further studies have to be done to
investigate the clinical value, organ and tumor dosimetry and the best imaging protocol of this
new PET tracer.
18-F-DOPA PET for staging neuroendocrine tumors
K. P. Koopmans, E. G. E. de Vries, I. P. Kema, P. H. Elsinga, O. Neels, W.
J. Sluiter, A. N. A. van der Horst - Schrivers, P. L. Jager; University
Medical Center Groningen, Groningen, The Netherlands.
Introduction: The aim of the study was to assess the value of whole body 18-fluoro-L-DOPA
PET for staging and restaging of patients with neuroendocrine tumors (NET), in comparison with
the generally used gold standard of abdominal CT in combination with whole body somatostatin
receptor scintigraphy (SRS).Methods: All patients had histologically verified NETs. Whole body
PET from the neck to the upper legs was performed after injection of 100 - 200 MBq DOPA and
oral pretreatment with carbidopa. CT and SRS were performed within a short interval using
standard methods. Endpoints were the number of positive patients, positive body regions and
individual lesions. Image interpretation of PET was blinded from other studies, but for precise
localization PET-CT image fusion was carried out. Newly detected lesions on PET were
validated using conventional methods when feasible.Results: In this ongoing study 60 patients
(M:F=31:29) were entered. DOPA PET was positive in 59 patients (sensitivity 98%), and
detected a mean of 12 lesions per patient (lpp), CT was positive in 52 patients (sensitivity 87%, 7
lpp) and SRS in 54 patients (sensitivity 90%, 6 lpp). DOPA PET detected 146 of 157 positive
regions (93% sensitivity), versus 77 for CT (49%) and 99 for SRS (63%). On a lesional analysis
DOPA PET detected 722/790 positive lesions (the number of lesions were truncated at 10 per
region) versus 425 for CT and 366 for SRS (sens. 91%/54%/46%). The largest numbers of
lesions were detected in the liver and abdomen (DOPA PET sens 90% and 94%, CT 71% and
45%, SRS 53% and 39%). In 27/56 patients software fusion of DOPA PET and CT led to a better
localization of lesions. Discussion: DOPA PET was clearly superior to CT and SRS for staging
of neuroendocrine tumors. Software based image fusion improved localization of tumor lesions.
507 — Sunday, October 01, 2006, 4:30 pm - 6:00 pm, MC 3
Clinical Science: Pediatrics
Practical GFR measurements in kidney transplanted children
using either a scintillation probe or the Schwartz formula comparison with findings based on the Sapirstein method as
a gold standard
G. Berding1, L. Pape2, S. Geisler1, G. Offner2, A. Luehr1, W. H. Knapp1, J.
H. H. Ehrich2; 1Department of Nuclear Medicine, University School of
Medicine, Hannover, Germany, 2Department of Pediatric Nephrology,
University School of Medicine, Hannover, Germany.
Aim: The purpose of this study was (1) to evaluate the accuracy of two different procedures
(Oberhausen vs. Schwartz) to measure the glomerular filtration rate (GFR) in children after
kidney transplantation and (2) to assess graft function using Sapirstein as the gold
standard.Materials and Methods: 73 children (mean age 14±4, body height 153±18 cm, weight
49±18 kg) that had received a kidney transplant (42 living donor and 31 cadaveric renal
transplantations) were included in this study. After i.v. injection of 80 kBq Cr-51-EDTA per kg
body weight time activity curves were registered for 1 hour from the right shoulder using a
collimated scintillation probe. Additionally blood samples were obtained at 5, 10, 15, 25, 60 and
120-min post injection. The GFR was calculated according to Oberhausen (including the 60-min
blood measurement) and according to Schwartz based on plasma creatinine (GFR = body height
* k / creatinine) using a k-value of 38. The degree of impairment in GFR was classified as low
(60-90), intermediate (30-60) or high (< 30 ml/min/1.73qm).Results: No significant differences
in mean values of GFR obtained according to Oberhausen (48±21 ml/min/1,73qm), Schwartz
(50±18) and Sapirstein (49±22) were observed (t-test). However, measurements obtained
according to Oberhausen showed a higher degree of correlation to the gold standard compared to
values calculated using the Schwarz formula (r-square: 0.66 vs. 0.54). The tables of contingency
revealed more correspondence with repect to the assessed degree of impairment of kidney
function between Oberhausen method and gold standard (46 of 73 cases) than the Schwartz
method and Sapirstein (40 of 73 cases). The Oberhausen method tended to underestimate the
degree of impairment less frequently than Schwartz method (12 vs. 19 cases).Conclusions: In the
mean all three methods demonstrate an intermediate degree of impaired kidney function in a
group of kidney transplanted children. Nevertheless, values obtained with the Oberhausen
method based on measurements with a scintillation probe showed more analogy to the results of
the Sapirstein gold standard than values obtained with the Schwartz formula. Therefore, the
Oberhausen method seems to be superior for simple individual assessment of kidney function in
children after kidney transplantation.
Validation of a new software package to standardize direct
radionuclide cystography
R. Malveiro1, A. I. Santos2, P. Almeida1; 1University of Lisbon, Faculty of
Sciences, Biophysics and Biomedical Engineering Institute, Lisbon and
Nuclear Medicine Department-Hospital Garcia de Orta, Almada, Portugal,
Nuclear Medicine Department-Hospital Garcia de Orta, Almada, Portugal.
Aim: In order to allow a more complete and standard reporting of direct radionuclide
cystography (DRC), a new software, in Aladdin programmability, has been designed.Materials
and Methods: This new software package allows the user to define the standard areas (both
kidneys, ureters and bladder), perform motion correction, quantify volumes (Post Voiding
Residual Bladder Volume (PRBV), Maximum Volume of Bladder Filling (MVBF), Maximum
Right Refluxing Volume (MRRV), Maximum Left Refluxing Volume (MLRV)), display
sequential images showing reflux, generate time activity curves from bladder and kidneys ROIs
and present compressed images, representing bladder and each kidney separately. In order to
validate the programme, it was tested in a total of 83 patients, corresponding to 38 males and 45
females, with a mean age of 3,9 years. Of these, 43 did not have Vesicoureteral Reflux (VUR),
18 had bilateral VUR, 13 had left VUR and 9 had right VUR. The following types of bladder
fillings were taken into account: single instillation (SI), multiple instillations (MI) and
MIBI scintimammography were also performed 1 day later,The region of interest was drawn
around each lesion, The tumor uptake of 99mTc-sandostatin were measured and expressed as the
ratio of T/Ns .The final clinical diagnosis was confirmed by histopathological study. mRNA
expression of SSTR and protein level of SSTR were measured by means of RT-PCR and
immunohistochemical staining respectively.Results:31 patients were confirmed invasive ductal
breast carcinoma, 11 patients had benign lesions. Out of 32 patients with malignant lesin ,17
patients were verified as axlliary lymph nodes involvement finally. The sensitivity, specificity,
accuracy of 99mTc-sandostatin in detection of primary lesion were 90.5%, 73%, 76.6%
respectively. The positive predictive value was 72.7% , negative predictive value were 95.6%.
the sensitivity , specificity and accuracy of 99mTc-MIBI scintimammography were
91.4%,78.3%,85.7%,respectively , positive predictive was 81.8%,and negative predictive value
was 96.8% . In the detection of axlliary lymph node metastasis, out of 17 patients ,only 7 patients
were visualized by 99mTc-sandostatin imaging, 12 patients were founded by 99mTc-MIBI. The
significant higher mRNA expression of SSTR were found in breast cancer,good correlation was
shown between the ratio of T/NT and the expression of SSTR2 mRNA (r=0.728, P<0.01). The
level of SSTR2 and SSTR5 are higher among five subtypes in primary lesion.
Immunohistochemical staining found all malignant lesion were positive for SSTR2 and
SSTR5,the protein level of SSTR2 and SSTR5 were significant higher than benign
tissues.Conclusions: 99mTc-Sandostatin scintigraphy are equally sensitive for the detection of
primary lesion of breast cancer,but had limited value in detection of axilliary node invovement in
comparison with 99mTc-MIBI scintimammography. Somatostatin receptor scintigraphy revealed
the uptake of 99mTc-Sandostatin in primary tumor correlated with expression of SSTR, it may
promote somatostatin mediated radiotherapy in breasr cancer dramatically.
instillations with initial resident bladder volume (IRBV). The validation of this method was made
defining as gold standard methods refered on EANM Paediatric Guidelines for DRC. Four
quantitative parameters have been correlated: PRBV, MVBF, MRRV and MLRV, through
Spearman Rank Correlation test, with values obtained with those calculated following methods
published on the aforementioned guidelines (based on formulas using voided volume of urine),
except PRBV that was correlated with those calculated following both methods published on
aforementioned guidelines: PRBV1 and PRBV2, using voided volume of urine and volume of
fluid infused, respectively.Results: We found statistically significant correlations between values
calculated by our program and the values obtained using the EANM guidelines: correlation
coefficient was higher than 97% for all the parameters except MVBF, with
r=0.9216.Conclusions: The results obtained for parameters quantification suggest that the method
developped, easily available through dedicated software, may be useful for the routine
quantification of the aforementioned parameters. Is its also valid in all variants of bladder
fillings, even in those with multiple instillations and/or when complete emptying of the bladder,
previous to the exam, was not achieved. This programme seems to be a strong tool to process the
DRC and the comprehensive presentation include the major aspects to be analyzed in a DRC.
Evaluation of clinical impact of FDG-PET on pediatric
patients with lymphoma
I. Borrego Dorado, E. Quiroga Cantero*, C. Márquez Vega*, G. Ramírez
Villar*, R. Vázquez Albertino, L. Caballero Gullón, G. Pineda Cuevas*;
Hospitals Virgen del Rocío., Seville, Spain.
Aims: Evaluation of the clinical impact of FDG-PET in pediatric patients with
lymphoma.Materials and Methods: We have retrospectively studied 22 pediatric patients (11 M /
11 F) aged 5 to 14 years (mean age: 10.7 years) with 65 consecutive FDG-PET scans, and
diagnosis of lymphoma (15 LH and 7 LNH). Whole body FDG-PET (ECAT HR+) were done
after an intravenous injection of 18F-FDG (6 MBq/Kg weight), in normoglucemia conditions,
muscular relaxant, hydration and diuretic. Images were evaluated by 2 expert nuclear physicians
and were analysed qualitatively and semiquantitatively (SUV). We have compared results of
FDG-PET with conventional diagnosis methods (CT, MRI, ultrasonography and bone marrow
PAAF). The results had histopathological confirmation and clinical evolution (follow-up period
> 12 months).Results: Indications for FDG-PET were: initial staging (n: 18), evaluation of
response to induction chemotherapy (n: 12 LH), evaluation of response to therapy (n: 26), restaging for malignant or equivocal lesion (n: 4), and control follow-up (n: 5). PET improved
initial staging in 4/13 patients with LH (30.7 %), changing stage level in 1 of them and confirmed
3 equivocal lesions of CT. FDG-PET led to a management change in 7 / 15 patients with LH (46.
6 %), avoided radiotherapy in 6 of them, and detected an unknown recurrence. In the group of 7
patients with LNH, PET-FDG led to a management change in 2 patients (28.5 %). Clinical
impact of FDG-PET in our patients was 40.9 % (9 / 22). In all 38 cases studied for therapy or
induction chemotherapy response, SUV levels showed good correlation with clinical
evaluation.Conclusions: FDG-PET has elevated clinical impact in pediatric patients with
lymphoma, and must be incorporated to the diagnosis protocols.
Paediatric non-Hodgkin Lymphoma: Evaluation of FDG-PET
for initial staging and therapy monitoring
C. Furth1, T. Voelker2, J. Ruf1, T. Denecke1, D. Misch1, B. Stoever3, G.
Henze2, H. Amthauer1; 1Klinik für Strahlenheilkunde, Campus VirchowKlinikum, Charité-Universitätsmedizin Berlin, Berlin, Germany, 2Klinik für
Pädiatrie m.S. Hämatologie und Onkologie, Campus Virchow-Klinikum,
Kinderradiologie, Klinik für Strahlenheilkunde, Campus Virchow-Klinikum,
Charité-Universitätsmedizin Berlin, Berlin, Germany.
Aim: In this prospective analysis, the use of FDG-PET for initial staging and therapy monitoring
was evaluated in paediatric patients with non-Hodgkin lymphoma (NHL). Imaging findings were
compared to the established conventional imaging modalities (CIM). Material and methods: 20
patients with paediatric NHL (6 f, 14 m, range 2.8-17.6 yrs; mean 11.9) were examined using
whole-body FDG-PET initially (n=15), during therapy (n=12) and after completion of therapy
(n=12), respectively. PET findings were compared to CIM performed according to the protocol
of the German NHL-BFM 95/04 studies. The results were evaluated for their impact on disease
classification and therapy decision (St. Jude, REAL) in correspondence to a clinical follow-up of
at least 12 months. Results: Concerning initial staging, all lymphoma manifestations detected by
CIM were also detected by FDG-PET (27 nodal, 16 extranodal). Furthermore, an additional
nodal lesion was detected by PET in three patients and bone marrow involvement in one patient.
This resulted in an upstaging followed by an intensified polychemotherapy in two patients. 12
patients showed residual masses on CIM during therapy, FDG-PET indicated viable residual
tumour in 4 cases. One of these 4 patients showed a progression of disease during follow-up, the
three other PET positive patients and the 8 PET negative patients did not. After completion of
initial therapy, PET revealed in 2 out of 12 patients persistent positive tumour metabolism in the
primary lesions and/or detected new manifestations. One of these patients died shortly after
restaging due to disease progression. The other patient had disease relapse. PET indicated
complete metabolic response in ten patients after completion of therapy. These patients are up to
now without recurrence. Conclusions: FDG-PET is shown to be very valuable for initial staging
of paediatric NHL. In cases of residual masses detected by CIM, FDG-PET can differentiate
between viable or non-viable lymphoma tissue.
Childhood and adolescent thyroid carcinoma in comparison
with adult patients
Z. Bence-Zigman1, J. Ille2, M. Dumic2, D. Dodig3, T. Zigman4, T. Zarkovic4;
Clinical Department for Nuclear Medicine and Radiation Protection,
Clinical Hospital Center Zagreb, Zagreb, Croatia, 2Pediatric Clinic, Clinical
Hospital Center Zagreb, Zagreb, Croatia, 3Clinical Department fo Nuclear
Medicine and Radiation Protection, Clinical Hospital Center Zagreb,
Zagreb, Croatia, 4Clinical Hospital "Sisters of Mercy", Zagreb, Croatia.
AIM: Evaluation of presentation, therapy and outcome of the thyroid carcinoma (TC) in children
and adolescent in comparison with adult patients (pts).Materials and Methods: 958 pts treated for
TC in our department were divided in 4 groups in view of age: 1. Children in pre-puberty period:
9 pts (3-12 years of age) 2.Teenagers/adolescent: 29 pts (12-21 yrs of age) 3.Young and middle
aged: 536 pts (22-50 yrs of age) 4.Elderly group: 384 pts (51-80 yrs of age). Mean follow up in
the groups was 10, 12, 7 and 5 yrs, respectively. Female/male ratio was 1:1, 1:4, 1:5 and 1:5,
respectively. Total thyroidectomy was performed in all children and adolescent, and in 98% of
adult pts. Lymph node (LN) dissection was performed in pts with ultrasonically guide fine needle
biopsy (UGFNB) diagnosed metastases. Radioiodine ablation of thyroid remnant was performed
in 85% of children and adolescent, and in 94% of adult pts. Pts with 131I positive distant
metastases (DM) were treated with additional 131I therapies.Results: papillary carcinoma was
found in the groups: 100%, 90%, 82 % and 69%, follicular carcinoma in 0%, 0%, 12% and 19%,
Hürthle cell carcinoma in 0%, 0%, 2% and 3%, medullary carcinoma in 0%, 10%, 4%, and 8% of
pts, respectively, and anaplastic carcinoma in 1% of the forth group, only. Primary TC < 1 cm
was found in the groups: 0%, 8%, 32% and 33%, respectively. Extrathyroid tumour infiltration
was found in 56%, 7%, 9% and 19%, multicentric foci in 56%, 30%, 24% and 25%, LN
metastases in 67%, 40%, 36% and 29%, and DM in 33%, 7%, 5% and 9% of pts, respectively.
During the follow-up period two pts died from the first group, nobody from the second group,
0,7% of pts died from the third group and 5% from the fourth group.Conclusions: Children and
adolescent had advanced disease at the time of initial diagnosis, especially pts younger than 12
yrs of age, what suggest different biological behaviour but also later diagnosis. We suggest more
often use of ultrasound examination of the thyroid and the neck in children and adolescent
population, UGFNB, total thyroidectomy and LN dissection if TC and LN metastases was
diagnosed, and radioiodine ablation of thyroid remnant. Despite an overall good prognosis, it
cannot be overemphasized that the survival rates of children with TC as a whole are lower than
those of young and middle aged pts.
Optimised PET/CT protocols with diagnostic contrast
enhanced multi-slice CT and with low-dose CT in paediatric
patients: analysis of more than 350 examinations
C. Franzius, J. Vormoor, M. Weckesser, K. U. Jürgens, O. Schober;
University Hospital Muenster, Muenster, Germany.
Purpose Examinations of children with PET-CT demand a particular preparation of the patients
and an adaptation of acquisition protocols to their young age and low body weight. The aim of
this study was to implement optimised PET/CT protocols for children und to analyse special
features and problems with the occurring in pediatric examinations. Methods 355 children (mean
age 13 yrs.) were examined with PET-CT (Biograph 16, Siemens/CPS) according to special
paediatric protocols for PET/CT with diagnostic contrast enhanced CT and for low-dose CT.
Special features related to the scanning procedure were documented. These features were the
need of sedation or anaesthesia and the cooperation of the children during the scan. The
incidence of scanning artifacts due to breathing or motion was evaluated prospectively for a 11
months period of time (n=175) on a 3-value scale (no, little, distinct) and was compared with the
incidence in a group of 1290 adults (range 18 - 85 y, mean age 57 y) within the same period of
time. Results The age distribution in the whole pediatric group was the following: 0-5 yrs., n=47;
6-10 yrs., n=68; 11-15 yrs., n=142; 16-18 yrs., n=98. 186 children had diagnostic CT scans of at
least one part of the body, 189 children only had low dose CT for attenuation correction. Out of
37 children younger than 3 1/2 years, 10 children were examined during anaesthesia (27 %), 21
children were sedated (57 %), 1 baby slept well after milk feeding, and 2 children (both 3-yrs.old) were compliant without any sedation/anaesthesia. 2 children could not be examined due to
insufficient sedation (not included in n=355; 2/357; 0,6%). In the artifact analysis, 5 artifacts
were detected the pediatric group (3 %). These were 2 artifacts due to breathing (1 little, 1
distinct) and 3 due to motion (3 distinct). Out of 1290 PET-CT examinations in adults, 60 had
artifacts (5 %): 59 breathing artifacts (45 little, 14 distinct) and 1 distinct motion artifact.
Conclusion Whole body PET-CT imaging is feasible in pediatric patients in all age groups.
Special PET/CT protocols with optimisation of examination parameters should used to lower
radiation exposure and to meliorate children's compliance. Children at the age of 4 or older can
usually be examined without sedation or anaesthesia. Artifacts due to motion or breathing do not
occur more often than in adults.
Relationship between dopadecarboxylase
secretion in hyperinsulinism of infancy
M. S. Ribeiro1, P. DeLonlay2, S. Bourgeois1, T. Delzescaux1, N. Boddaert2,
C. Nihoul-Fekete2, F. Jaubert2, F. Brunelle2, A. Syrota1;
Hospitalier Frederic Joliot, Orsay, France, 2Hopital Necker-Enfants
Malades, AP-HP, Paris, France.
Aim: Congenital hyperinsulinism of infancy (HI) is characterized by profound hypoglycaemia
caused by inappropriate insulin secretion. Focal and diffuse forms of HI share a similar clinical
presentation, but their treatment is radically different. HI can revert in children with focal lesions
by selective surgical resection contrary to the diffuse form which requires subtotal
pancreatectomy when resistant to medicamental treatment. The aim of this study is to assess the
use of positron emission tomography (PET) with [18F]-fluoro-L-DOPA in the differential
diagnosis between focal and diffuse HI. For operated patients, we evaluated the relationship
between the uptake of [18F]fluoro-L-DOPA and the pancreatic concentration of
dopadecarboxylase (AADC).Materials and Methods: 42 patients (28 boys, 14 girls, age: 1 month
to 5 years old) with HI were studied with [18F]fluoro-L-DOPA using an ECAT HR+. All
patients fasted for at least 6 hours prior the PET study. In 41 cases the medication was
discontinued for 72 hours. The examination was performed under sedation and data were
acquired 30 to 60 min after the injection of 4 MBq/kg of [18F]fluoro-L-DOPA. Standardized
uptake values (SUV) were calculated for the head, body and tail of the pancreas. For operated
patients, immunohistochemical detection of AADC was performed with anti-AADC
antibodies.Results: An abnormal focal uptake accumulation of [18F]fluoro-L-DOPA was
Radionuclide scintigraphy in the evaluation of gastroesophageal reflux in preterm infants
A. Bhattacharya, C. Morigeri, K. Mukhopadhyay, A. Narang, B. R. Mittal;
Pgimer, Chandigarh, India.
Objective: Gastroesophageal reflux (GER) is common in both full-term and preterm infants.
Most of them are asymptomatic at birth, but may subsequently develop respiratory complications
like recurrent apneas, aspiration pneumonia and lung collapse. The reported incidence in
symptomatic patients varies from 22-85%, mainly due to the different methods and criteria used
for diagnosis. Studies on asymptomatic preterm infants are limited and literature is lacking
regarding the incidence of GER in this group. We used radionuclide scintigraphy to evaluate the
incidence of GER in both groups of preterm neonates.Methods: The study included 106 preterm
infants of less than 34 weeks gestation, of which 52 were symptomatic and 54 asymptomatic for
GER. Feeding was avoided for 2 hours preceding the study. Tc99m sulphur colloid was
administered in a dose of 0.05 mCi in 1 ml, followed by 10 ml of milk, through an orogastric
tube. The tube was removed and sequential anterior images acquired for 20 minutes, with the
baby supine in the incubator, under a gamma camera. Reflux was graded as low (into lower half
of esophagus) or high (into upper esophagus or oropharynx), and reflux episodes during the study
were counted.Results: The incidence of GER in the symptomatic group was 71.2% and in
asymptomatic babies 61.1%. High-grade reflux was more common (71.4%) than low-grade
(28.6%) in both groups. The incidence of high-grade GER in symptomatic babies was 67.6% and
in asymptomatic babies 75.8%. There was no statistically significant difference in grades of
refluxes between symptomatic and asymptomatic babies. Mean number of reflux episodes in
symptomatic babies was 4.41 ± 2.4 in 20 minutes and in asymptomatic babies 4.97 ± 2.2.
Symptoms like vomiting, regurgitation, apnea, dips in saturation, unexplained bradycardia and
recurrent lung collapse were more common in the symptomatic group with positive scintigraphy
and high grade GER.Conclusions: GER is common in preterm infants of less than 34 weeks
gestation. However, the occurrence of GER was not significantly different in symptomatic
preterm infants when compared to asymptomatic preterm infants. Radionuclide scintigraphy is a
simple, quick and non-invasive investigation for assessment of GER in these babies, with no
adverse effects. The initial study may be used as a baseline study for follow-up of preterm infants
positive for GER in the neonatal period.
509 — Sunday, October 01, 2006, 4:30 pm - 6:00 pm, Lecture Hall 1
Oncology: Radiation Protection
Dosimetry in a nuclear medicine department: an instrument
to evaluate radiation protection
E. De Geest, M. Bricoult, J. Van Cauteren; AV Controlatom, Vilvoorde,
Aim Dosimetric follow up of nuclear workers in a nuclear medicine department can be used as
an instrument to evaluate radiation protection. The aim of this study is to evaluate the
effectiveness of different radiation protection measures and point out the most critical factors in
ALARA policy. Materials & methods The occupational exposure of radiation workers is
monitored using personal dosimeters and is filed both at the health physics and the occupational
medicine department. The dosimetry results of 80% of the occupationally exposed workers in
Belgium and of about 20 nuclear medicine departments will be examined. The follow up over
several years (1998-2000 and 2003-2005) will be discussed and changes explained. Results The
dose distribution of the occupationally exposed persons in all sectors shows the highest doses in
the medical sector. If we zoom in for the nuclear medicine departments, differences between
them can be up to a factor of 5. The availability of radiological protection equipment such as
syringe holders, lead aprons , the design of the department, the activities administered to the
patient, the number of personnel per patient, training of the technologist seem to be important
parameters to evaluate the differences in doses. Nevertheless, doses exceeding the legal dose
limit (Belgian law and European Directives: 20 mSv/y) are rare. Conclusion The follow up of
individual dosimetry remains even nowadays important and can be used as an effective
instrument to steer radiation protection measures and to reduce individual doses.
Impact of FDG-PET in the occupational whole-body dose
received by several professional groups in a Nuclear
Medicine Department
J. A. M. Santos, I. Ruiz, A. L. Bastos, F. Ponte; Instituto Português de
Oncologia do Porto, EPE, Porto, Portugal.
Aim: The purpose of this work is to evaluate the impact of a PET unit in the personal radiation
whole-body doses of several professional groups in a Nuclear Medicine Department.Materials
and Methods: We used the personal whole-body dose values received by the workers over a
period covering the starting of a PET unit and the subsequent year, relating the observed values
with the initial increasing number of PET exams and the FDG handling techniques. The groups
of studied workers were physicians, physicists, technologists, nurses, administrative, and
auxiliary personnel. The assignment of tasks, which implies a repetitive daily exposition to F-18,
is as follows: dose calibration: radiopharmacist (1) and physicians (3); injection: nurse (1); exam:
technologists (4). The physicist, administrative and auxiliary personnel, as well as one non-PET
technologist, has no significant contact with the daily FDG-PET procedures. All the
measurements considered in this work were executed by the same personal dosimeter provider
(NRPB, UK).Results: The technologists are the professional group where we can observe a more
evidenced increase on the average whole-body dose (from 0.15 mSv/month to 0.3 mSv/month)
following the increasing number of PET exams before reaching a constant number around 80
exams/month over a period of 1 y. The average dose received by the nurse, who has the higher
exposition from all the workers (0.7 mSv/month), does not show any relevant variation during
the same period.Conclusions: The overall dose variation does not seem to be significant as a
result of the FDG-PET implementation. The higher dose received by the nurse (~ 7.7 mSv/y),
simply reflects the large amount of exams (PET and non-PET), as the increase in dose received
by the technologists can be related with the time it is needed to handle the patient at the exam and
the higher exposition rate constant of F-18 when compared with other isotopes used in Nuclear
Medicine (namely Tc-99m). This is evidenced by the comparison between the non-PET
technologist, which receives significantly lower doses than the PET technologists. In conclusion,
we can say that the higher exposition factor due to FDG-PET in the department arises from the
professionals (technologists and nurse) contact with the injected patients rather from the FDG
dose preparation.
T. Sera1, L. Csernay1, H. Gavaller2, L. Boda3, G. Szanto3, L. Kardos1, L.
Euromedic Diagnostics Szeged Ltd, Szeged, Hungary,
Department of Cardiology, University of Szeged, Szeged, Hungary,
Cardiology Unit, County Hospital, Deszk, Hungary, 4Department of Nuclear
Medicine, University of Szeged, Szeged, Hungary.
Introduction and Aim: Our nuclear medicine department has recently been reorganised.
Myocardial stress SPECT investigations are performed three days a week, on each day by a
different team, comprising a cardiologist and a technician. The aim of this study was to
investigate the radiation exposure of the cardiologist teams participating in the nuclear medicine
work only in these 99m-Tc-MIBI stress studies.Materials and Methods: To determine their
exposure, PC-attachable, calibrated electronic (energy-compensated silicon pin diode detector)
personal dosimeters were used. The cardiologist and the technician wore the dosimeters on their
chests. The data were processed with the aid of a computer program. The personal doses were
monitored for each team throughout 7 weeks, during 10 patient stress studies/day, each patient
receiving 700 MBq 99m-Tc-MIBI. The working methods of the teams were analysed in
connection with their radiation exposure.Results: The daily effective doses (mean±SD) of the
cardiologists from the three teams were: 9.25±1.72, 16.16±4.02 and 15.07±3.3 microSv, while
the daily maximum dose rates were {7,5-12}, {13,6-26,7} and {10,6-20,9} microSv/h,
respectively. The corresponding values for the technicians were: 9.74±1.05, 12.67±2.20 and
10.67±3.0 microSv, and {9.5-14}, {10.5-18.5}, {4.8-15.9} microSv/h, respectively. The average
equivalent dose of the natural background (mean±SD) for 10 hours, repeated for 10 days, was
1.25+0.2 microSv; this was used for the background correction of the above data. The personal
doses and the maximum dose rates of the cardiologists and technicians in the three teams
exhibited considerable differences, but were well below the dose limits for occupational
exposure. The low doses for team 1 reflected their 10 years of experience in nuclear cardiology.
The higher values for teams 2 and 3 could be due to less optimal work organisation: overlap
between consecutive studies, an improper distance from the patients, and questioning patients
after radiopharmaceutical administration, all causing needless exposure from already-injected
patients.Conclusions: Direct-reading dosimeters are especially useful in monitoring personnel
relatively inexperienced with radiation work on a daily basis.
Establishing restriction period after Iodine-131 therapy in
hyperthyroidism: are dose rate measurements needed?
A. Prata, S. Carmona, A. I. Santos; Serviço de Medicina Nuclear - Hospital
Garcia de Orta, Almada, Portugal.
Aim: To evaluate if the estimate Iodine-131 residual activity at 24h is accurate enough to
establishing the restriction period, based on the “Radiation Protection 97” document, published
by EURATOM. Populations and Methods: We have evaluated 121 patients (19 males and 102
females, <age>=54,8±14years), submitted to I-131 therapy for hyperthyroidism, between
February 2002 and March 2006. The I-131 therapeutic activity (TA) - 333±111MBq (74555MBq) was calculated based on a standard gamma camera method, in which the percentage of
iodine thyroid uptake at 24 hours (TU) is obtained following I-131 administration (740Bq). The
restriction period (RP) was established based on the estimated 24 hours residual activity (RA),
calculated by RA=TAxTU, and extended for a few days, for precaution reasons. After
completing the RP, all patients came to our department to measure the dose rate at 1 meter
distance from the patient (DR). Patients were divided in three groups, according to the primary
thyroid disease: multinodular goiter (MNG) (n=36); toxic adenoma (TAd) (n=28) and diffuse
goiter (DG) (n=57).Results: After completing the RP, DR was >3µSv.h-1 in 46 (38%) patients.
The average RP extension was 3,2 ± 1,8 days and <DR> was 3,07±2,16µSv.h-1. Comparing
patients with DR within the expected values with those with higher values, we tried to find if
there were differences in terms of the TA administered, TU or RA, but these were not
statistically significant. The same comparison was made in terms of primary thyroid disease, and
differences were, again, not statistically significant.Conclusions: A considerable number of
patients needed further restrictions days than those recommended by EURATOM, based on our
24 hours residual activity estimation. As so, we consider recommendable to measure the dose
rate at 1 meter distance from the patient before stopping restrictions, especially in those patients
who have contact with small children or pregnant women.
observed in 14 patients whereas a diffuse uptake of the radiotracer was observed over the
pancreatic area of the other patients. The mean±SD of SUV were 2.8±1.1 and 2.1±0.7
respectively for the pancreas hot spot in focal and pancreatic area in diffuse HI (p<0.02). All
focal forms and 7 patients with diffuse radiotracer accumulation unresponsive to the treatment
were submitted to surgery. One patient was submitted to 2 PET studies: the diffuse uptake found
before an inhibitor AADC administration disappeared completely after its administration. The
histological results confirmed PET findings in 19/21 operated patients.Conclusions: The results
obtained validate the use of [18F]fluoro-L-DOPA PET as a reliable test to differentiate focal
from diffuse HI. The PET results are supported by the immunohistochemical analysis performed
after partial (focal) or subtotal (diffuse resistant) pancreatectomy.
Adsorption of
I in urine by using anion exchange resin
P. Unak, S. Teksoz, E. I. Medine; Ege University, Izmir, Turkey.
Aim: The aim of this study is the adsorption of iodine-131 which comes from the waste of the
patients’ urine through anion exchange resin and investigate relevant parameters such as
temperature, flow rate, particle size etc., and then determine the optimum conditions.
Introduction: Iodine is one of the trace elements in human body and it is essential for growth
and development of body. Unless sufficient amount iodine is taken in daily diet, the level of
thyroid hormones will decrease. It is known as Iodine Deficiency (ID) and is a problem for
almost all the countries in the world. For the last 15 years, there has been an increase in the
number of patients with deficiency of thyroid hormone. Those patients have been treated with
methods involving the use of 131I at nuclear medicine centers. Of all the radioactive iodine
isotopes, iodine-131 has been used successfully in the treatment of patients with thyroid diseases
for 50 years. 80% of iodine given to the patient is excreted from body in urine after
treatment.Materials and Methods: In this study Amberlite IRA 410 strong basic anion exchange
resin have been used, and also adsorption and elution parameters of iodine-131 in the patients’
urine have been examined through the resin.Results: The results prove that the Amberlite IRA410 anion exchange resin is very effective in adsorbing 131I and is not eluted by
water.Conclusions: By this way, the collection of discharged patients’ urine with radioactive
iodine-131 through the resin in a small volume can be prevented from being hazardous waste for
environmental and public health, and it can be concluded that it is- as a new system- proposed an
alteration to the tank system by means of concentration. Key Words: Iodine-131, anion
exchange resin, urine, waste management
from the dose coefficients published by the International Atomic Energy Agency (IAEA Safety
Report no.115) assuming M lung clearance class and a prolonged intake over a 7hours work-day.
The 50-year CED from each day’s intake ranges from (1.78±0.03) to (3.66±0.02) microSv
depending on different working activities, while the external dose daily received by technologist,
ranges from (22±0.7) to (30.5±1.0) microSv. Results are summarised in the next table.
Conclusion This study calculated the total effective dose, including the external dose and the 50year Committed Effective Dose from inhaled fluorine-18 fluorodeoxyglucose for a technologist
working in a mobile PET unit. A number of situations have been modelled according to the
different physician’s activities showing that the maximum dose that he may receive is (32±1)
µSv/day. The major contributions to technologist’s dose comes from external exposure.
Table1: Radiation exposure for a technologist performing 18F-FDG scans in a mobile PET unit.
Patients treated with Sr-89-Chloride, Re-186-HEDP or Sm153-EDTMP cause radiation exposure of beta-particles to
non-patients! A reconsideration of general ideas on radiation
M. G. E. H. Lam , A. Hoekstra , P. P. van Rijk , J. M. H. de Klerk , B. A.
Zonnenberg1; 1University Medical Center Utrecht, Utrecht, The
Netherlands, 2Meander Medisch Centrum, Amersfoort, The Netherlands.
Aim: After treatment with beta-emitting radiopharmaceuticals as 89SrCl2, 186Re-HEDP and
Sm-EDTMP, the radiation exposure to non-patients is generally thought to be caused by so
called ‘bremsstrahlung’ and gamma-radiation, with negligible contribution of beta-radiation. The
latter however, may be contradicting reality. The aim of this prospective study was the
investigation of radiation safety, based on exposure measurements of beta-particle and photon
radiation after treatment with 89SrCl2, 186Re-HEDP and 153Sm-EDTMP. Methods: 35 patients
with hormone refractory prostate carcinoma with painful osseous metastases were treated with
150 MBq 89SrCl2 (10 patients), 1295 MBq 186Re-HEDP (13 patients) or 37 MBq/kg 153SmEDTMP (12 patients). External exposure at a distance of 30 cm and 1 meter from the patient was
measured in time (0-72 hours p.i.). To identify and quantify the contribution of bremsstrahlung,
beta-particles and gamma-rays, a calibrated survey meter (PDM1; Nuclear Enterprises Ltd) was
used, with a shutter (open/closed) to block beta-radiation. For each radiopharmaceutical a curve
of best fit was drawn on the measurement-points. An estimation was made of the total radiation
exposure to non-patients, defined as the area under the fitted curve. Estimations of residence
times and distances of relatives in relation to the patient, were used to calculate skin doses and
effective doses with and without instructions (sleeping apart, keeping distance etc).Results: The
total calculated mean radiation dose (based on the AUC of the fitted line) at 1 meter distance
from the patient was 170 µSv for 89SrCl2, 242 µSv for 186Re-HEDP and 247 µSv for 153SmEDTMP. The radiation dose (and percentage of total dose) for beta-radiation only was ±170 µSv
(>99%) for 89SrCl2, 211 µSv (87%) for 186Re-HEDP and 68 µSv (28%) for 153Sm-EDTMP. The
total skin dose with instructions was 43 µGy for 89SrCl2, 61 µGy for 186Re-HEDP and 62 µGy for
Sm-EDTMP, and 238 µGy, 339 µGy and 346 µGy respectively, without instruction. The
effective doses were 0.4 µSv for 89SrCl2, 0.6 µSv for 186Re-HEDP and 0.6 µSv for 153Sm-EDTMP
with instruction, and 2.4 µSv, 3.4 µSv and 3.5 µSv respectively, without instruction.Conclusions:
Patients treated with 89SrCl2, 186Re-HEDP or 153Sm-EDTMP emit a spectrum of radiation,
including non-negligible beta-particles. Beta-particles in superficial tissue expose the
environment. This is supported by an increase of exposure with higher beta-energy. It has major
consequences for the use of all beta-emitting radiopharmaceuticals.
Injection and diagnostic activities
Escorting patient to the scanner room after
the tracer uptake period and positioning him
escorting patient out of the mobile unit
yr Effective
3.66 ± 0.02
24.4 ± 0.8
28.0 ± 0.8
2.92 ± 0.03
22.0 ± 0.7
25.0 ± 0.7
1.78 ± 0.03
30.5 ± 1.0
32.2 ± 0.9
External dosimetry and personnel monitoring for medical
workers operating 18F FDG PET/CT scans
K. Dalianis1, J. Malamitsi1, L. Gogou1, M. Pagou1, K. Gogos1, R.
Efthimiadou1, J. Andreou1, E. Georgiou2; 1Department of PET/CT D.T.C.A.
Hygeia - Harvard Medical International, Athens, Greece, 2Department of
Medical Physics University Medical School, Athens, Greece.
Aim: Positron emission tomography is considered one of the most relevant diagnostic techniques
providing functional and quantitave information of the organ of interest. Additional radiation
safety precautions must be taken considering the high energy annihilation gamma rays (511
KeV) emitted by PET radiopharmaceuticals and the long patient-uptake time. Therefore we
measured the radiation dose of the staff in our PET/CT Department at the Diagnostic and
Therapeutic Center of Athens HYGEIA Harvard Medical International. As for the time being
only 2-deoxy-2-[18F]fluoro-D-glucose PET studies are performed, radiation dose measurements
concern those derived from dispensing of the radiopharmaceutical as well as from the injected
patients. Our aim is to develop more effective protective measures against radionuclide
expposure.Methods: To estimate the effective dose from external exposure all 7 members of the
staff (2 nurses, 2 medical physicists, 2 technologists, 1 secretary) had TLD badges worn at the
upper pocket of their overall, TLD rings on the right wrist and digital dosimeters in their upper
side pocket. In addition isodose curves were measured with thermoluminescence detectors for
distances of 20,50,70 and 100 cm away from the patients.Results: In the period of June 2004March 2006 a total of 1243 PET/CT studies were performed in which patients received 250-420
MBq of 18F-FDG. The collective effective and finger doses received by all 7 members of the
staff were the following: Nurse 1 received 3,77 mSv as a whole body dose and 1,2 mSv as a hand
dose and Nurse 2 received 3.92 mSv whole body dose and 1,3 mSv hand dose respectively. For
medical physicist 1 doses were: 2,12 mSv whole body dose and 1,2 mSv hand dose, for medical
physicist 2 1,78 mSv whole body dose and 1,1 mSv hand dose and for the technologist 1 and 2
the whole body doses were 1,23 mSv and 1,47 mSv respectively. Lastly, the secretary received
0,34 mSv as a whole body annual dose.Conclusions: The personnel dose results are significantly
less than the recommended annual dose by International Commission for Radiological
Protection. However a greater effort should be made to reduce the doses further in line with the
ALARA principle.
510 — Sunday, October 01, 2006, 4:30 pm - 6:00 pm, Lecture Hall 2
Radiation exposure for a technologist performing 18F-FDG
scans in a mobile PET unit
Radiopharmacy/Radiochemistry: Radiopharmacology
- Miscellaneous
M. De Marco1, S. Maggi1, G. Biscontini2, G. Ascoli2; 1Ospedali Riuniti
Ancona - Medical Physics dept., Ancona, Italy, 2Ospedali Riuniti Ancona Nuclear Medicine dept., Ancona, Italy.
Aim The use of mobile PET truck is becoming more widespread throughout Europe and the
world. These equipments usually house a control room adjacent the scan room and the injection
room on the opposite side. The routinely handling of unsealed radioactive isotopes in this small
room without a dedicated fume exhaust cabinet, may result in the production of airborne
radioactivity. The aim of our study was to quantify the occupational radiation exposure that
technologists working in a mobile PET unit are committed to receive over time, evaluating the
independent contributions coming from internal contamination and external irradiation.
Materials & methods Internal exposure: each technologist was issued with a Personal Air
Sampler (PAS) that collects 5 µm AMAD particles on a 20mm-diameter paper filter. Flow rate
through the filter was set at 20l/min (1.2m3/h) corresponding to the standard breathing rate. PAS
was worn for 20 minutes, then filter’s radioactive contamination has been characterised with a
gamma spectroscopy that provided the estimates of personal daily intake. External radiation
exposure: personal dose equivalent Hp(10) has been quantified using both thermoluminescent
and electronic dosimeters. Occupational exposure has been evaluated twice a week for three
months based upon an operating efficiency of 9 procedures/day and an average radiotracer
administered of 300 MBq/patient. Results Committed Effective Dose (CED) has been calculated
Renal uptake of radiolabeled antibody fragments mediated by
S. Borkowski1, K. Thelle2, E. I. Christensen2, J. E. Cyr1, S. K. Moestrup3,
C. Jacobsen3, J. T. Nielsen4, D. Berndorff1, L. M. Dinkelborg1; 1Research
Laboratories Schering AG, Berlin, Germany, 2Department of Cell Biology,
Institute of Anatomy, University of Aarhus, Aarhus, Denmark, 3Department
of Medical Biochemistry, University of Aarhus, Aarhus, Denmark,
Department of Clinical Physiology and Nuclear Medicine, University
Hospital, University of Aarhus, Aarhus, Denmark.
Aim: Renal uptake of radiolabeled antibody fragments may compromise their use in diagnostic
imaging as well as in radiotherapy. We therefore investigated the role of the endocytic receptor
megalin in mediating uptake of selected tumor-targeting antibody fragments both in vitro in yolk
sac cells and in vivo in receptor-associated-protein (RAP)-deficient mice which are characterized
by a reduced megalin expression.Materials and Methods: The single-chain antibody fragment
D-amino acid peptides improve tumor targeting using
bispecific antibody-based pretargeting of renal cell
J. J. E. M. van Eerd1, F. van Schaijk1, E. Oosterwijk1, B. McBride2, D.
Goldenberg2, F. Corstens1, O. C. Boerman1; 1University Medical Center
Nijmegen, Nijmegen, The Netherlands, 2Immunomedics Inc, Morris Plains,
NJ, United States.
Objectives: Previous studies have shown that pretargeting using bispecific antibodies (bsAb) and
radiolabeled peptides, results in effective targeting of renal cell carcinomas (RCC). For tumor
visualization and radioimmunotherapy, high tumor uptake and retention of the radiolabeled
peptide in the tumor is required. To further improve the tumor residence time of radiolabeled
peptide, new peptidase-resitant bivalent peptides were synthesized. These peptides consisted of 4
D-amino acids (D-a.a. peptide) and were conjugated with appropriate chelators to allow
radiolabeling. Here we investigated the tumor targeting characteristics of the radiolabeled D-a.a.
peptides in nude mice with SK-RC-52 RCC tumors. Methods: The uptake and retention in the
tumor of 111In-, 99mTc- and 131I-labeled bivalent peptides (L-a.a. peptide and D-a.a. peptide) were
studied in female BALB/c athymic mice with subcutaneous SK-RC-52 RCC tumors. Tumors
were pretargeted with 15 µg bsAb G250xDTIn-1 and 72 h later, mice were injected intravenously
with one the radiolabeled peptides. Imaging experiments were performed with the 99mTc-labeled
peptides. Results: The maximum uptake and retention of D-a.a. peptides in the tumor were
significantly higher compared with those of the radiolabeled L-amino acid peptides . Uptake of
the 99mTc-labeled D-a.a.peptide was 10.5 %ID/g and of the 131I-D.a.a. was 22.2 %ID/g while
uptake of the L-a.a. labeled peptides was 4.0 %ID/g and 5.12 %ID/g respectively. Uptake of the
D- and L-a.a. peptides in the normal tissues were similar. Gamma-camera images showed
improved visualization of the tumors after injection of the radiolabeled D-amino acid peptides.
Conclusion: Uptake and retention in the tumor of the radiolabeled peptides after pretargeting
with a bs-mAb could be significantly improved using D-amino acid peptides. Accordingly, the
visualization and radiation dose to the tumor can be enhanced substantially, using these new
peptidase-resistant radiolabeled peptides.
A novel biotin derivative carrying two DOTA groups of
potential interest to increase the tumor dose
S. Papi, M. Miranda Cona, N. Urbano, C. Grana, G. Paganelli, M. Chinol;
Nuclear Medicine division, European Institute of Oncology, Milano, Italy.
Aim: Avidin-biotin pretargeting has shown its potential in radionuclide tumor therapy. We
designed a novel biotin analogue carrying two DOTA groups per molecule (BDB: BisDOTABiotin); this molecule can potentially be labeled with increased specific activity if compared to a
mono-DOTA analogue previously developed thus to deliver higher doses to the tumor. In this
work we investigated the labeling of BDB with 111In, 90Y and 177Lu, in-vitro stability and avidin
binding affinity.Materials and Methods: The optimal radiolabeling pH was studied incubating
BDB, MCl3 (M= 111In, 90Y, 177Lu) and sodium acetate 1M buffer at 95ºC for 30 min in pH range
3.5-7.0, using a specific activity of 10.6 MBq/nmol. The radiochemical purity (RCP) was
determined by RadioITLC in saline, adding an aliquot of the radiolabeled solution to a molar
excess of Avidin and DTPA. The maximum achievable specific activity (SA) (RCP>99%) was
then evaluated using the best pH. Stability of radiolabeled conjugates was studied by RadioITLC over 48h incubation at 37ºC in saline (with and without ascorbic acid, AA) or in human
serum:saline 1:1. The affinity of the cold-metal labeled BDB towards avidin was studied using
HABA colorimetric test.Results: The best radiolabeling pH was found to be 4.5-5.0
(RCP>99.8%) for all isotopes.The maximum SA obtained was 53 MBq/nmol for 90Y (RCP=
99.0%) and 105.7 MBq/nmol for 177Lu (RCP= 99.3%). For 111In, it was not possible to exceed
10.6 MBq/nmol, probably due to 111Cd. For 90Y, the max SA implies a metal:BDB molar ratio of
1:34, whereas for 177Lu the ratio is 1:7 (1:1.2 considering total lutetium). In comparison the
mono-DOTA biotin (studied in a previous work) showed max SA of 5.3 MBq/nmol for 90Y
(RCP= 99.5%) and 130 MBq/nmol for 177Lu (RCP= 99.4%), with a metal:Biotin molar ratio of
1:340 and 1:5.6 (1:1) respectively. 90Y- and 177Lu-BDB were stable up to 24h in AA (98.1% and
97.9%) and in serum (98.7% and 99%) when labeled up to their max SA. The affinity for avidin
at 1:4 molar ratio was 63%, lower than that found for the mono-DOTA derivative (87%),
probably due to the steric hindrance of the two DOTA groups.Conclusions: This new BisDOTAbiotin derivative may be potentially useful to increase the radiation dose to the tumor. Although
the avidin affinity needs further structural improvements, the optimal RCP achieved at higher
specific activity and its stability over time is promising especially for 90Y-based therapeutic
AP39 (~50 kDa) and the small immunoprotein L19-SIP (~80 kDa) both targeting the oncofetal
matrix protein ED-B fibronectin were tested for binding affinity to the megalin and cubilin
receptor by surface plasmon resonance (SPR) analysis. Cellular uptake, lysosomal degradation
and specific inhibition of I-125-labeled antibody fragments were studied in megalin-expressing
BN16 yolk sac cells. AP39 either labeled with Tc-99m or with In-111 (via mx-DTPA) was
further examined for kidney uptake and biodistribution in RAP-deficient mice and compared
with the corresponding control strain. Urinary metabolites of the mice were analysed by SEHPLC.Results: SPR analysis showed no significant binding of the antibody fragments neither to
megalin nor to cubilin in a concentration range of 10-1000 nM. In contrast, cellular uptake of the
iodinated proteins in yolk sac cells was high (AP39 = 79 ± 8; L19-SIP = 232 ± 22 fmol/mg cell
protein) and megalin-mediated as demonstrated by inhibition studies. Lysosomal degradation of
both proteins could be inhibited almost completely by addition of the protease leupeptin resulting
in cellular accumulation of radioactivity. Kidney uptake of Tc-99m- and In-111-mx-DTPAlabeled AP39 which was 4.03% ID/g ± 0.98 and 67.02% ID/g ± 5.27, respectively in normal
mice was strongly reduced by 46% and 87% in RAP-deficient mice compared to the control
strain. HPLC analysis showed 50-54% of the intact labeled protein in the urine of RAP-deficient
mice.Conclusions: Megalin mediates renal uptake of radiolabeled antibody fragments
independent of their low binding affinity to the receptor and the isotope used for labeling.
Different molecular structures of antibody fragments result in different levels of cellular uptake
transported by megalin. Kidney retention of radioactivity is consequently due to the speed of
intracellular metabolism.
In vitro 131I uptake of FRTL-5 cells as an experimental model
for the stunning phenomenon
B. Meller, W. Deisting, E. Gaspar, B. E. Wenzel, M. Baehre; University of
Luebeck, Luebeck, Germany.
Aim: In radioiodine therapy the “stunning phenomenon” in the thyroid gland is defined as a
reduction of radioiodine uptake after diagnostic application of 131I. Up to now, only few in vitro
investigations of stunning have been carried out. Consequently, mechanisms of thyroid stunning
are not yet understood. The TSH-depending FRTL-5 monolayer cell culture is a well defined cell
system for simulating thyrocyte function in vitro. In the current study, we established an in vitro
model to investigate stunning after application of 131I using FRTL-5 cells.Methods: TSHstimulated FRTL-5 cells were incubated during their period of growth with 131I in the presence of
low concentrations of stable iodid 30 µg/l (0.2 µM 127I-) Controls were additionally incubated
with perchlorate or 9.4 µM 127I-. Time-dependent 131I uptake and the viability of FRTL-5 cells
were determined 2-144 h after radioiodine application. Uptake measurements were repeated in
TSH-deprived cultures. Data were corrected for number of viable cells, half life and 131I
concentrations. NIS mRNA was isolated (RNeasy, Qiagen) and levels were quantified by real
time PCR. Statistical analyses were performed using nonparametric tests.Results: Independent of
TSH stimulation FRTL-5 monolayer cell cultures showed a specific maximum uptake of 131I 2448 h after application. Significantly decreased 131I uptake values were observed independently of
TSH levels after 72-144 h. Incubation with 131I reduced iodine uptake significantly by an average
of 90 % over the investigation period. The observed uptake reduction was comparable to the
competitive decrease achieved by coincubation with high concentrations of 127I-. Decreasing
radioiodine uptake was correlated with decreasing mRNA levels of NIS.Conclusions: Since
decreased radioiodine uptake was independent of TSH stimulation and cell proliferation, 131I
appears to influence iodine biokinetics in FRTL-5 cells significantly. Due to our investigations,
FRTL-5 cells were shown to be a valuable tool for further investigations of the stunning
phenomenon on molecular level.
accumulation of tumour targeting radiofolates
C. Mueller1, R. Schibli2, A. P. Schubiger2, E. P. Krenning1, M. de Jong1;
Erasmus MC, Rotterdam, The Netherlands, 2ETH, Zurich, Switzerland.
Aim. Folate-based radiotracers have the potential to be used as imaging agents of folate-receptor
(FR)-positive tumours. However, the tumour (tu) uptake of (radio)folates is relatively low
compared to uptake in FR-positive kidneys. The development of a method to increase the tumour
uptake while protecting the kidneys is crucial, in particular with regard to potential use of
radiofolates for therapeutic purposes. The aim of this study was to investigate the impact of the
clinically used antifolate pemetrexed (PMX) on the biodistribution of the recently synthesized
Tc-PAMA-folate. Materials & Methods. In vivo studies were performed with nude mice
bearing FR-positive KB-tumours. The synthesis of the 99mTc-radiofolate was performed using the
“Isolink” kit. PMX was intravenously injected at different time points prior to the intravenous
administration of the 99mTc-folate radiotracer. Biodistribution of radioactivity was determined 4
p.i. of 99mTc-PAMA-folate. Multimodality imaging studies were performed in anesthetized mice
using the small-animal SPECT/CT system ‘X-SPECT’ (Gamma Medica). Results. PMX was
injected 2h, 1h, 30min or 15min prior to the administration of the folate radiotracer. Optimal
tumour-to-non-tumour ratios (tu/blood; tu/kidney; tu/liver) were found when PMX was injected 1
h prior to the radiotracer. The tumour uptake of the radiotracer was not affected (2.21 ± 0.34 %, 4
h p.i. vs. control: 2.33 ± 0.36 %), but a very impressive reduction of the uptake in the critical
renal tissue (1.14 ± 0.18 %, 4h p.i. vs. control: 18.48 % ± 0.72 %) was found, resulting in > 10fold increased tu/kidney ratio (1.99 ± 0.51; 4h p.i. vs. control: 0.13 ± 0.02; 4h p.i.). The tu/blood
ratio (82.1 ± 4.4, 4h p.i. vs. control: 58.0 ± 12.2) and tu/liver ratio of radioactivity (2.38 ± 0.73,
4h p.i. vs. control: 2.53 ± 2.13; 4h p.i.) were not affected significantly by PMX administration.
Small animal SPECT/CT scans corroborated the superior tumour-to-background ratios of
radioactivity if mice were pre-injected with PMX. Conclusion. Combination of 99mTc-PAMAfolate with PMX significantly reduced the undesired accumulation of radiotracer in renal tissue,
with consistently high tumour uptake. This effect increased the target-to-non-target ratio of
radioactivity by a factor of > 10 fold. These findings are of high interest with respect to a
potential folate-mediated radionuclide therapy.
Bimodal imaging of an orthotopic pancreatic neuroendocrine
tumor-model by 3T MRI and Ga-68 animal-PET in the mouse
L. H. F. Stelter1, A. Scholz2, A. Rexin2, J. Pinkernelle1, T. Denecke1, R.
Felix1, B. Wiedenmann2, R. Michel1, H. Amthauer1; 1Department of
Universitätsmedizin-Berlin, Campus Virchow-Klinikum, Berlin, Germany,
Universitätsmedizin-Berlin, Campus Virchow-Klinikum, Berlin, Germany.
Introduction Detection and monitoring tumor growth and metastatic spread in orthotopic animal
models can be of great use and benefit for clinical practice. We analyzed and optimized specific
functional and morphological imaging of SSTR-positive NETs in a novel orthotopic mousemodel. Materials and Methods Neuroendocrine differentiated and SSTR-positive rat pancreatic
tumorcells (AR42J) were injected in the head of the pancreas of nude mice. After establishing
tumor-management and growth 5 MBq of the positron-emitting radioisotope Ga-68-DOTATOC
and Ga-68-DOTANOC were injected intravenously. The animals were examined in a dedicated
animal PET-scanner starting with a dynamic series followed by an emission scan. Additionally
post-contrast T2-weighted imaging was performed using a clinical 3 T MRI-scanner. All imaging
results were verified with the macro- and microscopic dissection findings. To evaluate the
biodistribution we measured activity distribution 1 hour after injection in tumor and organs.
Results All animals showed a beginning radionuclide distribution in the region of the primary
tumor after 10 minutes. Definite anatomical correlation was performed using digital image fusion