Preston Press - Preston Primary School

Workshop 1: Imaging Primer
Tony Finelli, MD, MSC, FRCS(C)
Silvia D Chang, MD, FRCP(C)
Urologic Oncology/Minimally Invasive Surgery
Department of Surgical Oncology
Princess Margaret Hospital (UHN)
Associate Professor, University of Toronto
Toronto, Ontario
Associate Professor of Radiology
Radiology Residency Program Director
Department of Radiology
University of British Columbia
Head of Abdominal MRI
Department of Radiology
Vancouver General Hospital
Vancouver, British Columbia
S Larry Goldenberg, MD, CM, OBC,
FCAHS, FRCS(C)
Professor and Head
Department of Urologic Sciences
University of British Columbia
Vancouver, British Columbia
Faculty Disclosure Statement
Industry financial relationships for the 2-year period
prior to ICPC
• Ownership/advisory boards: None
• Involvement in research: None
• Honoraria/fees/patent ownership: None
Silvia Chang, MD
Faculty Disclosure Statement
Industry financial relationships for the 2-year period prior to
ICPC
• Ownership/advisory boards:
GlaxoSmithKline, AstraZeneca, Sanofi, Paladin, Amgen, Abbott,
Pfizer
• Involvement in research:
Bayer, Pfizer, Merck
• Honoraria/fees/patent ownership:
GlaxoSmithKline, AstraZeneca, Sanofi, Paladin, Amgen, Pfizer
Tony Finelli, MD
Faculty Disclosure Statement
Industry financial relationships for the 2-year period prior to
ICPC
• GlaxoSmithKline Canada Medical Advisory Board
• Abbott Canada Advisory Board
• Protox Advisory Board
S Larry Goldenberg, MD
Key Learning Objectives
•
Discuss the practical application of multiparametric magnetic
resonance imaging of the prostate.
•
Critique the role of diffusion imaging, spectroscopy and dynamic
enhancement techniques.
•
Debate which technique is best or whether all exams can be
performed at once.
Imaging Primer
ICPC 2012
Silvia D. Chang, MD, FRCPC
Department of Radiology,
University of British Columbia
Vancouver, B.C., Canada
Outline
• Multiparametric MRI:
•
•
•
•
Standard MRI sequences: T1 and T2
MR spectroscopy
MR dynamic contrast imaging
MR diffusion imaging
• With or without a coil?
• Imaging on a 3T system
• Applications and Future directions
MRI: Standard Sequences
T1
T2
MRI: T2 – Weighted Sequences
• High spatial resolution
• Depict prostate
anatomy
• Hyperintense peripheral
zone
• Intermediate signal
transition and central
zone (central gland)
• Localizing and staging
T1-weighted Sequence
Post biopsy hemorrhage
Wait 6 weeks before MRI
Lymphadenopathy
Bone metastases
White S, Hricak H et al. Radiology 195:385, 1995
MRI of Prostate Cancer
Low signal intensity on T2
• Non-specific
• post-biopsy hemorrhage
• post radiation therapy
• hormone therapy
• inflammation
• dystrophic changes
Difficult to detect in the transition zone
MRI - Local Staging
• Extra-capsular
extension
• broad tumor contact
(>12mm)
• smooth or irregular
capsular bulge
• obliteration of
rectoprostatic angle or
asymmetry
• invasion of
neurovascular bundle
• angulation, focal
capsular retraction or
thickening
Yu KK, Hricak H, et al. Radiology 202:697,199
Seminal Vesicle Invasion
•Direct tumor extension into or
around SV
•Tumor extension along
ejaculatory ducts or nonvisualization of ejaculatory
ducts
Axial T2
Local Staging with MRI
•
•
•
Staging accuracy initially 54 - 90%
More recently 75 - 90%
Addition of MR spectroscopy found to
decrease interobserver variability and
improve diagnostic accuracy
Engelbrecht et al Eur Radiol 2002;12:229
Scheidler et al. Radiology 1999;213:473
Yu KK, et al. Radiology 1999;213:481
MRI: Standard Sequences
T1
Post-biopsy hemorrhage
Nodal disease, bone mets
Poor contrast resolution
T2
Detection and staging
Non-specific
Difficult detection in the TZ
MR Spectroscopy of the
Prostate
• 3D proton MR spectroscopic imaging
(MRSI)
• prostate metabollites: choline and
citrate
• prostate carcinoma:
– higher choline, lower citrate levels
– choline + creatine / citrate is increased
Kurhanewicz J. et al. Radiology 198:795,19
MRI and MRS of Prostate Cancer
*
*
Courtesy Dr. Hricak
MRI and MRS of Prostate Cancer
•Metabolic assessment
*
*
•Spectral quality depends on field
homogeneity
•Time consuming acquisition (16
min)
•Time consuming interpretation and
requires special expertise
MRI with Dynamic Contrast Enhanceme
Prostate cancer tends to enhance earlier, faster, to a greater extent and
shows earlier contrast washout compared to normal tissue.
Barentsz. J MRI 1999:10(3):295
Franiel. Invest Radiol 2008;43(7):481
Axial T2
Dynamic Contrast Enhancement
(DCE)
Axial T2
Dynamic Contrast Enhancement
MRI with Dynamic Contrast Enhanceme
Prostate cancer tends to enhance earlier, faster, to a greater extent and
shows earlier contrast washout compared to normal tissue.
Barentsz. J MRI 1999:10(3):295
Franiel. Invest Radiol 2008;43(7):481
Axial T2
Dynamic Contrast Enhancement
(DCE)
Dynamic
Dynamic Contrast
Contrast Enhancement
Enhancement (DC
(DC
• Time-signal intensity parametric maps
(Max. Enhancement, Onset Time, Mean Gradie
• Two-compartment pharmacokinetic modeling
(Ktrans, ve, maximum Gd concentration)
Ktrans (min-1)= volume transfer constant between blood plasma &
extra-vascular extra-cellular space per volume per minute
Ve = volume of extra-vascular extra-cellular space per unit volume
of tissue
0.40
PCa
PZ
Gd-DTPA conc. [mM]
0.35
0.30
0.25
0.20
0.15
0.10
0.05
0.00
0
1
2
3
4
Time [min]
5
6
7
8
T2W
ADC map
anterior
right
left
posterior
Ktrans map
DCE MRI
CG
PZ
PC
a
Max.
Enhancement
Onset Time [min]
Mean Gradient
1.4306 ± 0.1897
0.7333 ± 0.0900
1.0038 ± 0.3108
1.2177 ± 0.2448
0.7414 ± 0.1335
1.0080 ± 0.6590
1.4629 ± 0.2183a
0.6442 ± 0.0890d
2.2970 ± 1.0231a,b
a – PCa significantly different than PZ (p<0.001)
b – PCa significantly different than CG (p<0.001)
d - PCa significantly different than PZ (p<0.05)
Kozlowski P, Chang SD, Jones EC, Berean KW, Chen H,
Goldenberg SL.
J Magn Reson Imaging. 2006;24(1):108-113.
DCE MRI
CG
PZ
PC
a
Ktrans [min-1]
ve
Max. Conc. [mM]
0.5964 ± 0.2884
0.4431 ± 0.0842
0.3144 ± 0.0691
0.5977 ± 0.5599
0.3825 ± 0.1484
0.2738 ± 0.1150
1.2635 ± 0.5437a,b
0.3257 ± 0.1312
0.2572 ± 0.0991
a – PCa significantly different than PZ (p<0.001)
b – PCa significantly different than CG (p<0.001)
Moradi M, Chang SD, Goldenberg
SL et al. JMRI. In press.
MRI with Dynamic Contrast Enhanceme
Prostate cancer tends to enhance earlier, faster, to a greater extent and
shows earlier contrast washout compared to normal tissue.
Localization
Barentsz. J MRI 1999:10(3):295
Franiel. Invest Radiol 2008;43(7):481
•Non-specific
•BPH nodules
•prostatitis
•Lack of consensus for
standardized protocols
•Limited commercialized
tools
•Costs, risks (NSF)
Dynamic Contrast Enhancement
(DCE)
Diffusion Weighted Imaging
• Diffusion Weighted Imaging (DWI):
Random motion of water molecules
(Brownian motion) within extracellular,
intracellular and intravascular spaces.
Diffusion Weighted Imaging
• Restricted diffusion:
• Malignancy (increased number of cells )
• Ischemia (cytotoxic edema)
• Abscess (increased viscosity)
65 yo, PSA 6.9
T2 FSE
b=600
b=0
ADC
49 yo, PSA 6.2
T2 FSE
ADC
Chronic granulomatous prostatitis
85 yo, PSA 23, prev bx neg
T2
T2
b=1000
ADC
59 yo, PSA 54 , bx neg x 2
T2
b=1000
ADC
85 yo, PSA 23, prev bx neg
T2
T2
ADC
Fast, simple, readily available
•Limited spatial resolution
•Susceptible to motion and
magnetic field homogeneity
•Non-specific: inflammation
infection
PROSTATE
PROSTATECANCER
CANCERWITH
WITHEXTRACAPSULAR
EXTRACAPSULAREXTENSION
EXTENSION
55 yo (PSA 2.4) with tumor in the left midgland extending to
involve the neurovascular bundle
b = 600
T2 FSE
ADC
SEMINAL
SEMINAL VESICLE
VESICLE INVASION
INVASION
68yo (PSA 4.3) with left seminal vesicle invasion
T
2
ADC
ADC values (mean ±
SD)
ADC (103mm2/sec)1
b=600
deSouza3
Reinsberg2
b=300,500,80 b=300,500,8
0
00
Tanimoto4
b=1000
PZ
1.99 + 0.21
1.51 + 0.27
1.71 + 0.16
1.72 + 0.35
CG
1.52 + 0.13
1.31 + 0.20
1.46 + 0.14
1.46 + 0.16
PCa
1.21 + 0.25
1.19 + 0.24a
1.03 + 0.18b
1.30 + 0.30
0.93 + 0.11
atumor
> 30% of voxel, btumor > 70% of voxel
1 Kozlowski, Chang,
Goldenberg JMRI
2006;24(1):108-113
1Reinsberg SA. AJR
2007;188:91
ADC values (mean ±
SD)
Gleason Score
ADC (103mm2/sec)
6 (n=17)
1.216 + 0.241
7 (n=16)
1.093 + 0.226
9 (n=5)
0.839 + 0.146a
a – Gleason 9 different than Gleason 6 (p<0.05)
DWI
DCE
DWI+DCE
Sensitivity
59%
60%
80%
Specificity
86%
79%
74%
Prospective, N=31, b=600, ER+pelvic phased array, Bx
Whole
Prostate
Sensitivit
y
Peripheral
Zone
Transition
Zone
T2
T2+D
WI
T2
T2+D
WI
T2
T2+D
WI
54
81
57
87
36
41
Prospective, N=49,b=600,ER +pelvic phased array,
Specificit
91
84
91
wholemount
Haider M.
80
92AJR 2007;188:91
92
y
T2
T2+DWI
T2+DWI+DCE
Sensitivity
73
84
95
Specificity
54
85
74
Tanimoto
Prospective?, N=83, b=1000, 8 channel torso, Px?Bx
A. JMRI 2007;25:146
T2
ADC
T2+ADC
Sensitivity
74,67,67%
75,75,74%
88, 81,78%
Specificity
79,77,57%
86,78,68%
88,89,63%
Accuracy
77,77,61%
82,77,70%
88,86,68%
Retrospective, N=52, b=1000, 1.5T, ER & phased array, step section
Kim HH. Radiology 2009;250(1):145
ADC +
MRS
ADC +DCE
MRS +
DCE
ADC, MRS & DCE
Sensitivity
100%
100%
83%
100%
Specificity
88%
92%
80%
92%
Az
0.94
0.94
0.88
0.92
Prospective, N=20, b=300,500,800 1.5T, ER coil, whole
mount
Riches SF. AJR 2009;193:1583
MRI Coils
Endorectal coil
Pelvic surface coil
MRI of Normal Prostate
Without ER coil
With ER coil
1.5 T vs 3T
• 1.5 T with ER coil = 3T without ER coil
• 3T with ER coil better than 1.5 T with ER coil
Sosna et al. Acad Radiol 2004;11:857
Futterer et al. Investigative Radiology
2004;39(11):671
Bloch et al. Acad Radiol 2004;11:863
3T MR Imaging
• SNR improved x 2
• Improved spatial
resolution
• Improved anatomical
detail
• Improved temporal
resolution
• Increased RF power
deposition x 4
• Increased signal
heterogeneity
• Greater susceptibility
effects
• Changes in relaxation
times ->adjustments
in TR/TE
3T MRI
No ER coil
With ER coil
3T MRI with Contrast
T2
With Contrast
1.5 T
3T
T2
Diffusion
With Contrast
Moradi M, Chang SD,
Goldenberg SL et al. JMRI. In
press
Moradi M, Chang SD, Goldenberg SL
et al. JMRI. In press.
Conclusion
• Multiparametric MRI:
•
•
•
•
Standard MRI sequences: T1 and T2
MR diffusion imaging
MR dynamic contrast imaging
MR spectroscopy
• With or without a coil? Yes, for staging?
• Imaging on a 3T system: Yes
• Future directions: fusion, focal therapy,
monitoring treatment effects.
Thank you
Impact of prostate MRI on disease reclassification among active surveillance candidates ‐ a prospective cohort study
David Margel, Nathan Lawrentschuk , Laurence Klotz, Masoom
Haider, Karen Hersey, Antonio Finelli, Alexandre Zlotta, John Trachtenberg and Neil Fleshner
Division of Urology, Department of Surgical Oncology, Department of Medical Imaging, Princess Margaret Hospital, University Health Network, and Sunnybrook Health Science Centre, Toronto, Ontario,
Canada
Introduction
• Active surveillance‐recommended Tx for low risk prostate cancer (AUA and EUA)
• Fear of under sampling is an impediment to wider‐scale adoption of AS.
• MRI most accurate imaging modality for detection of PCa
AUA+EUA guidelines
Weinreb et al, Radiology 2009
Aim
To report MRI findings among unselected men with low –risk prostate cancer prior to active surveillance
Patients and Methods
• Study design‐ Prospective cohort study
• Population‐ men with low‐grade, low risk PCa
Inclusion criteria
–
–
–
–
–
–
Age between 50 and 80 years
Biopsy‐proven, low‐risk, localized prostate cancer (minimum of 10 cores taken)
Gleason score ≤6 with no Gleason pattern 4
Clinical stage T1c‐T2a
Serum PSA ≤10 ng/mL
Life expectancy more than 5 years
Exclusion criteria
–
Any previous prostate cancer treatment (radiotherapy, chemotherapy, hormonal therapy, oral glucocorticoids, GnRH analogues)
– Any prior prostatic surgery or procedure Patients and Methods • Baseline endorectal‐coil MRI. • 3 groups: – Normal MRI
– cancer on MRI concordant with initial biopsy
– cancer on MRI demonstrating a larger than 1 cm lesion. • All patients second confirmatory biopsy (with in a year). • Primary outcome percentage of patients reclassified not fit for AS • Secondary outcome clinical parameters associated with TDFs.
• Statistical analysis‐
Univariable analysis to assess differences in clinical parameters between groups.
Enrolled
n=60 pts
Normal MRI
38% (23pts)
MRI >1cm
lesion
22% (13pts)
MRI and initial
Bx concordant
40% (24pts)
Excluded
Normal MRI- 2pts
Concordant- 1 pt
MRI >1cm1 pt
Confirmatory biopsy results
Reclassified
3% (2 pts)
Not
Reclassified
32% (21 pts)
No Cancer
17% (10 pts)
Cancer the same
as initial bx
15% (9 pts)
Reclassify
17% (10pts)
Not
Reclassified
3% (2pts)
No Cancer
0% (0 pts)
Cancer the same
as initial bx
3% (2 pts)
Reclassified
10% (6 pts)
No Cancer
12% (7 pts)
Not Reclassified
28% (16 pts)
Cancer the same
as initial bx
17% (10 pts)
Contingency table
MRI lesion Yes
greater than No
1 cm
Patient reclassified on
confirmatory biopsy
Yes
No
10
2
8
36
Sensitivity- 55% (95%CI, 43-67%)
Specificity- 95% (95%CI, 82-99%)
Positive predictive value- 83% (95% CI, 73%-93%)
Negative predictive value- 81% (95% CI, 71%-91%),
MRI Demonstrating a large anterior tumor
missed on initial diagnosis. baseline clinical parameters stratified by MRI prostate cancer findings
MRI with TDF
Cancer on MRI
concordant
with initial bx
No Cancer
detected on
MRI
p
Age in years
(mean±SD )
62.6±7
63.5±6
64±8.2
0.6
PSA ng/ml
(median and range)
5.94 (1.7-10)
4.4 (1.1-9.1)
4.1 (1.1-9.9)
0.07
Prostate Volume cc
(median and range)
37.5 (22-67)
37.5 (19-76)
41 (21-83)
0.09
Number of cores positive
on initial bx
(median and range)
2 (1-3)
2 (1-3)
2 (1-3)
0.5
Percent of cancer in
largest core on initial
biopsy
(mean±SD)
12.1±7.1
15.3±8.2
13.6±5.1
0.3
PSA density
ng/ml/cc
(median and range)
0.15
(0.6-0.05)
0.13
(0.22- 0.03)
0.07
(0.21-0.03)
0.04
Conclusions
• MRI can identify, early and non‐invasively, those candidates for surveillance who harbour occult adverse disease.
• In this series, significant PCa was present in 20% of those eligible for AS.
• MRI has the ability to diagnose and accurately direct a biopsy to these hidden cancers.
• MRI may also be helpful to avoid biopsy‐ when MRI normal only 3% reclassification
• PSA density may be used to help clinicians select the appropriate patients for MRI. 
`