Prostate MRI: Who, when, and how? Report from a UK ,

Clinical Radiology xxx (2013) 1e8
Contents lists available at SciVerse ScienceDirect
Clinical Radiology
journal homepage: www.clinicalradiologyonline.net
Prostate MRI: Who, when, and how? Report from a UK
consensus meeting
A.P.S. Kirkham a, *, P. Haslam b, J.Y. Keanie c, I. McCafferty d, A.R. Padhani e, S. Punwani a,
J. Richenberg f, G. Rottenberg g, A. Sohaib h, P. Thompson f, L.W. Turnbull i, L. Kurban j,
A. Sahdev k, R. Clements l, B.M. Carey m, C. Allen a
a
Department of Imaging, University College Hospital, London, UK
Department of Radiology, Freeman Hospital, Newcastle, UK
c
Department of Radiology, Western General Hospital, Edinburgh, UK
d
Department of Imaging, Queen Elizabeth Hospital, Birmingham, UK
e
Paul Strickland Scanner Center, Mount Vernon Cancer Centre, London, UK
f
Department of Imaging, Royal Sussex County Hospital, Brighton, UK
g
Department of Imaging, Guy’s and St Thomas Hospital, London, UK
h
Department of Imaging, Royal Marsden Hospital, London, UK
i
Hull York Medical School at the University of Hull, Hull, UK
j
Department of Radiology, Aberdeen Royal Infirmary, Aberdeen, UK
k
Department of Imaging, Barts and the London NHS Trust, London, UK
l
Department of Radiology, Royal Gwent Hospital, Newport, UK
m
Department of Radiology, St James Hospital, Leeds, UK
b
art icl e i nformat ion
Article history:
Received 22 January 2013
Received in revised form
16 March 2013
Accepted 20 March 2013
The current pathway for men suspected of having prostate cancer [transrectal biopsy, followed
in some cases by magnetic resonance imaging (MRI) for staging] results in over-diagnosis of
insignificant tumours, and systematically misses disease in the anterior prostate. Multiparametric MRI has the potential to change this pathway, and if performed before biopsy,
might enable the exclusion of significant disease in some men without biopsy, targeted biopsy
in others, and improvements in the performance of active surveillance. For the potential
benefits to be realized, the setting of standards is vital. This article summarizes the outcome of
a meeting of UK radiologists, at which a consensus was achieved on (1) the indications for MRI,
(2) the conduct of the scan, (3) a method and template for reporting, and (4) minimum
standards for radiologists.
! 2013 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
Introduction
The standard pathway for men suspected of having
prostate cancer is resulting in over-diagnosis and over* Guarantor and correspondent: A.P.S. Kirkham, Department of Imaging, University College Hospital, 235 Euston Road, London NW1 2BU,
UK.
E-mail address: [email protected] (A.P.S. Kirkham).
treatment,1 as well as systematically missing significant
tumours, particularly in the anterior and apical parts of the
gland.2 Many of those who routinely request and perform
magnetic resonance imaging (MRI) of the prostate know
that it has the potential to solve many of these problems
and enable effective active surveillance. However, we
recognize that the technique is at a crossroads, and that the
initial experience in many centres is disappointing because
of why and how it is being performed. For MRI to live up to
0009-9260/$ e see front matter ! 2013 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.crad.2013.03.030
Please cite this article in press as: Kirkham APS, et al., Prostate MRI: Who, when, and how? Report from a UK consensus meeting, Clinical
Radiology (2013), http://dx.doi.org/10.1016/j.crad.2013.03.030
2
A.P.S. Kirkham et al. / Clinical Radiology xxx (2013) 1e8
its promise, the setting of standards is vital. This paper is an
attempt to summarize the consensus opinions of a group of
specialist UK radiologists, on the indications for and performance of multiparametric MRI of the prostate. It builds
on two previous European consensus documents in
particular: one from a meeting held in London in 20103 and
the second the European Society of Uroradiology consensus
document on prostate MRI published in 2012.4
Our aim is not an exhaustive summary of the data on
prostate MRI or a duplication of the comprehensive European
Society of Uroradiology (ESUR) guidelines: rather, we aim to
(1) summarize the established and emerging indications for
the technique, together with the performance characteristics
of modern multi-parametric MRI; (2) describe a single protocol that is applicable to the great majority of UK MRI systems; (3) propose a structured, practical approach for
reporting and a standard template; (4) suggest minimum
standards for performing multiparametric MRI in the UK.
Methods
Uroradiologists from nine centres routinely performing
large numbers of prostate MRI or with a research interest in
the field met in April 2012. They addressed a set of predefined topics on prostate MRI under the following categories: indications, conduct, reporting, and minimum
standards. Several other radiologists could not attend but
took part in editing initial and subsequent versions of this
document, or in subsequent meetings to draft the paper,
and are listed as authors. The aim was not to duplicate
previous meetings that have applied a formal consensus
method,3 but to achieve a pragmatic consensus by group
discussion. Unless stated, the statements in this paper are
the unanimous opinion of the group; the exception is the
section on the performance of the MRI examination where,
in several cases, we describe the clinical practice of a
majority.
Findings
Indications
The indications for MRI of the prostate can be broadly
split into two groups: (1) the detection of prostate cancer,
and (2) the local staging of prostate cancer. To some extent
this division is artificial (most examinations will be performed for both indications), but it is useful for a summary
of the evidence.
The detection of prostate cancer
Before discussing the indications for detecting prostate
cancer, it is important to address the issue of significance in
a disease that will affect the majority of the male population, but result in the death of few.5 This is currently a
matter of great debate amongst urologists. It is likely that
Gleason 3 þ 3 cancer very rarely results in death from
prostate cancer,5,6 and that the great majority of tumours
<0.5 cm3 will, similarly, be insignificant.7 If the definition of
significance is tumour in which current treatments make a
significant difference to survival, recent trial data suggest
that the criterion of Gleason 3 þ 3 and <0.5 cm3 for insignificant cancer is too conservative.8
One of the key performance characteristics of prostate
MRI is that it misses the majority of small (<0.1 cm3), lowgrade tumours.9 If the aim is that it will detect the majority
of all cancers, that aim is likely unattainable. Wellperformed, multiparametric MRI can usually (with >80%
sensitivity) detect the following in the whole prostate
(including peripheral and transition zones):
"0.2 cm3 (equivalent to a 7 mm sphere) of Gleason 4 þ 3
or above
"0.5 cm3 (equivalent to a 10 mm sphere) of Gleason 3 þ 4
or above
For Gleason 3 þ 3 tumours, we did not achieve consensus
that tumours >0.5 cm3 will be detected with 80% sensitivity.
Although areas of this size of “focal” or homogeneous Gleason
3 þ 3 disease will usually be detected, those with a more
diffuse pattern and loose stroma on histology often will not.10
These characteristics can only be assumed for multiparametric MRI performed before biopsy, including
diffusion-weighted and dynamic contrast-enhanced sequences, with certain minimum imaging parameters
(described below) and reported by an experienced radiologist. They are in keeping with recent published
results.9,11e14 What do they enable?
(i) Avoiding biopsy: it is likely that a multiparametric MRI
showing no evidence of tumour has a negative predictive value for significant disease similar to or better
than a standard 12 core prostate biopsy15,16 [modelling
studies suggest that 36e47% of lesions >0.5 cm3 may
be missed by a 12 core transrectal approach,17 and
another study showed a detection rate at transrectal
ultrasound (TRUS) of 75% for significant disease18]. If
the performance characteristics for MRI described
above are demonstrable in the center performing the
study, then an MRI performed as the first investigation
in a man suspected of having prostate cancer might in
some cases prevent the need for biopsy. MRI may
detect some tumours missed by biopsy (especially
those lying anteriorly19) and biopsy may detect some
tumours missed at MRI: the combination can be especially powerful in excluding significant tumour, and
may reduce the need for follow-up biopsy when the
prostate-specific antigen (PSA) level remains elevated.
ii) Targeting: if performed before biopsy, the samples
obtained can be targeted to areas of suspicion on the
MRI images. It has recently been shown that performing a mean of 3.8 targeted biopsies based on
multiparametric MRI has a performance for the
detection of significant disease comparable with 12
core untargeted biopsy, and that such a technique may
result considerably reduce the detection of insignificant disease.20 We are not at a stage to recommend
targeted samples only, but there is little doubt that
targeting also produces a more representative sample,
improving the estimation of tumour volume and
Please cite this article in press as: Kirkham APS, et al., Prostate MRI: Who, when, and how? Report from a UK consensus meeting, Clinical
Radiology (2013), http://dx.doi.org/10.1016/j.crad.2013.03.030
A.P.S. Kirkham et al. / Clinical Radiology xxx (2013) 1e8
significantly reducing the number of tumours upgraded at definitive histology.20e22
iii) Previously negative biopsies: in patients with
continuing clinical suspicion of prostate cancer, but a
negative previous biopsy, MRI may detect missed tumours (often anterior) in up to 40%.19,23 This is perhaps
the strongest indication for pre-biopsy MRI; indeed,
we would say that it is a requirement in men with a
persistently raised PSA in whom a second biopsy is
being considered.
iv) Active surveillance: MRI has two potential roles. Firstly,
to confirm that the classification of risk status is accurate and that there is not an unexpected larger or
higher-grade tumour: upgrading of apparently lowgrade tumours found at TRUS biopsy is common at
more definitive histology24 and is more common when
a lesion is conspicuous on MRI.25,26 More speculatively,
if the tumour can be seen on MRI, imaging has the potential to contribute to follow-up. If a tumour is visible,
it can be monitored for change. If it cannot be seen, it is
unlikely (according to the performance criteria above)
that a significant tumour has developed.
Local staging of prostate cancer
Staging aims to determine whether tumour is organconfined, and to detect local and distant spread. Staging of
cancer following a positive biopsy is the only current indication for MRI in the latest edition (2008) of the UK National
Institute for Health and Clinical Excellence (NICE) guidelines,27 which have the following three conclusions: (1)
imaging is not routinely recommended for men in whom no
radical treatment is intended; (2) computed tomography
(CT) is not recommended in those with intermediate or
low-risk disease; and (3) MRI is recommended in those with
high-risk localized (defined “clinically”) and locally
advanced disease being considered for radical treatment.27
The NICE guidelines are 4 years old (new guidance is due
in January 2014), and before recent data on the value of
targeted biopsies.17,22 They do not currently recommend
targeting for patients with intermediate or low-risk disease
(PSA < 20, Gleason # 7), in contrast to the more recent ESUR
guidelines,4 which make the case (as we have done) for MRI
in intermediate and low-risk patients. Ideally, MRI will have
been performed before biopsy (with the benefits of targeting and detection of anterior disease), and thus, before
the risk status is accurately known.
There is little doubt that even T2 sequences alone
improve staging accuracy compared to nomograms,28 and
there are morphological predictors of both seminal vesicle
(T3b)29 and extracapsular (T3a) spread,28 such as degree of
capsular abutment, bulge, irregularity, and fat stranding.
However, although selected groups have achieved moderately high accuracy, overall the detection of small-volume
extracapsular disease remains challenging, with overall
sensitivities of 67e91%, and specificity 67e100%.30,31
Overall, in the opinion of the group, MRI is useful (1)
when biopsies are negative (to exclude anterior disease
missed by a standard biopsy approach19); (2) in active surveillance where underestimation of size and grade is
3
common21,32 and MRI can detect larger or higher-grade
tumours25,26; and (3) to stage disease in men being
considered for radical treatment.4,27 These three scenarios
encompass the great majority of men with a moderately
raised PSA, and lead to the following conclusion: if MRI is
ultimately useful in most men with a suspicion of prostate
cancer, why not perform it before biopsy, when it will be
free of artefact, may enable targeting, and will be immediately available for staging if tumour is found?
The scan
Post-biopsy artefact
It has been known for some time that changes on
T2-weighted and contrast-enhanced sequences may persist
for at least 8 weeks after prostate biopsy,33,34 and that they
can significantly degrade staging performance: in one study
accuracy was 46% if the MRI was performed less than 21
days after biopsy and 83% if performed after.35 The period of
persistence of post-biopsy changes is not known accurately,
but changes on diffusion and T2 sequences can be seen in
some patients for 2e3 months, with artefact on enhanced
sequences lasting considerably longer in many patients.
What are the implications?
i) In the opinion of the group, a staging MRI should be
performed at least 10 weeks after biopsy, and if
possible after 20 weeks. Such a delay is often unacceptable to patients and risks breaching time to
treatment guidelines, so that it may be necessary to
perform MRI sooner. In such cases (MRI examinations
<10 weeks after biopsy), the dynamically enhanced
sequences are often considerably degraded, and a
limited scan using the T1, T2, and diffusion-weighted
parts of the protocol should be considered. The
group consensus was that such scans are “second best”
for the detection and staging of tumour. They might be
termed “limited staging scans”.
ii) In patients in whom the detection of tumour on MRI is
important, the scan should be performed before
biopsy.
A standard imaging protocol
There was a consensus amongst the group on many aspects of the conduct of prostate MRI, which can be generalized to most machines with a field strength of 1.5 T or
greater. They are minimum standards, which with additional techniques (3 T machines and endorectal coils) may
be exceeded.
i) A field strength of 1.5 T is adequate, although optimized images at 3 T are superior.36
ii) Most of the benefits of MRI can be achieved with a
multichannel pelvic phased-array coil. Performance
for both detection of tumour and staging will likely be
improved by the addition of an endorectal coil,37 but
the benefit for routine use does not necessarily
outweigh the costs: patient discomfort, extra time for
Please cite this article in press as: Kirkham APS, et al., Prostate MRI: Who, when, and how? Report from a UK consensus meeting, Clinical
Radiology (2013), http://dx.doi.org/10.1016/j.crad.2013.03.030
4
A.P.S. Kirkham et al. / Clinical Radiology xxx (2013) 1e8
placement, and field inhomogeneity. The majority of
those on the consensus panel did not use endorectal
coils.
iii) For routine clinical use before biopsy, all of the authors
used T2-weighted, diffusion-weighted and dynamically enhanced sequences, and we term this combination “multiparametric”. The relative contribution of
each remains to be determined accurately, but might be
summarized as follows: T2 sequences show anatomy
and detect most tumours, but are not specific, diffusion
sequences are of particular utility in the transition
zone,38 and enhanced sequences add specificity in both
peripheral and transition zones, and sometimes detect
tumours missed with other techniques.39
Minimum standards for resolution and timing have been
well described in the recent ESUR guidelines,4 and are
summarized here with little change (Table 1):
e T2 sequences: #3 mm section thickness. In-plane resolution of 0.7 mm or better. External sphincter, prostate,
and seminal vesicles included. On modern machines it
should be possible to achieve adequate signal-to-noise
ratios using these parameters with scan times for each
sequence of <5 min. Axial and coronal sequences should
always be included. Sagittal sequences are performed by
the majority of the group and are encouraged. The ESUR
guidelines state that axial images should be orthogonal
to the rectum, but the majority of the consensus group
obtain them in the true axial plane, and this is recommended for the purposes of repeatability and comparison between sequences.
e Diffusion-weighted sequences should be acquired with
at least three b-values for calculation of apparent
diffusion coefficient (ADC), with highest b ¼ 1000 s/
mm2. In-plane resolution should be 1.5e2 mm (1.5 mm
is usually attainable on modern machines), and section
thickness of #5 mm. A dedicated long b sequence
(1400 s/mm2 at 1.5 T, 2000 s/mm2 at 3 T) is strongly
recommended, and is possible on most machines. It
usually requires multiple averages for adequate signal
(with a scan time of 4e6 min).
e Contrast-enhanced sequences: pump injection at 3 ml/s,
with a standard dose of contrast medium. The time
taken for each scan should be 15 s or less, and in-plane
resolution 1 mm or better. The ESUR guidelines suggest
a minimum slice thickness of 4 mm, but the majority of
the group use 3 mm. A total scan time of at least 5 min
should be used for the detection of washout. Routine
use of semi-quantitative (inspection of curves) or
quantitative [calculation of pharmacodynamics parameters, such as the transfer coefficient (Ktrans)] are not
necessary in routine practice, but the radiologist should
have access to a workstation that can plot enhancement
curves. The shape of the enhancement curve may be
useful in some cases40; pharmacodynamic parameters
(such as Ktrans) have shown potential in several studies
but have not yet been shown to convincingly improve
diagnostic performance in practice.41,42
Anti-peristaltic drugs [hyoscine butylbromide (Buscopan) or glucagon], although not essential, are recommended as per the ESUR guidelines.4
Post-treatment scans
Although contrast-enhanced sequences are particularly
important for detecting disease after radiotherapy,43 prostatectomy,44 or ablative techniques,45 diffusion-weighted
sequences are also specific for tumour45 and T2-weighted
sequences are always needed for anatomy. There is no
need to modify the sequences used in the standard prostate
MRI protocol.
Additional sequences for staging scans
Nodal involvement is rare in patients with PSA < 20,46 and
the pelvic nodes are involved first in the great majority,47 so
that routine scanning of abdominal nodes on a pre-biopsy
MRI should be reserved for men with PSA > 20. The pelvic
nodes are adequately imaged on the T2 coronal sequence of
the pelvis, and large field of view T1 or T2 sequences are not
necessary in routine practice if the PSA is < 20. The same is
true of dedicated MRI bone imaging [usually a coronal short
tau inversion recovery (STIR) sequence of pelvis and lower
lumbar spine]: bone metastases are rare in men with a
PSA < 10,48 and men with a positive diagnosis of cancer and
PSA > 10 are likely to undergo isotope bone scintigraphy. The
recent ESUR guidelines recommend a full staging MRI bone
examination of the cervical, thoracic, and lumbar spine with
T1, diffusion, and STIR sequences,4 but this is still an area of
controversy, and is not performed by the majority of those on
the consensus panel.
The value of a single “prostate protocol”
Table 1
Minimum standards for multiparametric magnetic resonance imaging using
a pelvic phased-array coil at 1.5 T.
Sequence
Maximum
section
thickness
(mm)
Minimum
in-plane
resolution
(mm)
Scan time (range)
T2 axial, coronal
Diffusion e multi b
Diffusion e long b
Dynamic enhanced
3
5
5
3e4
0.7
2
2
1
3e6 min
4e7 min
4e6 min
<15 s per iteration,
total 5 min
%
%
%
%
0.7
2
2
1
The ESUR guidelines contain an algorithm for deciding
on different protocols for prostate MRI. At the top is TRUS
biopsy, and different protocols are then used at MRI
depending on the biopsy results and intention to treat. All
arms (no tumour, active surveillance, curative intent) end in
an MRI of the prostate: in other words, an MRI is recommended in all patients who undergo biopsy. We have
already described the potential benefits (in particular, the
ability to target and the lack of artefact from haemorrhage)
from performing the MRI before biopsy, and if pre-biopsy
MRI is performed routinely, the MRI must be of a single
protocol that is adequate for excluding significant tumour
Please cite this article in press as: Kirkham APS, et al., Prostate MRI: Who, when, and how? Report from a UK consensus meeting, Clinical
Radiology (2013), http://dx.doi.org/10.1016/j.crad.2013.03.030
A.P.S. Kirkham et al. / Clinical Radiology xxx (2013) 1e8
and staging it if detected. The other great virtue of a single
protocol is simplicity.
Spectroscopy and estimation of grade
A well-conducted, multicentre study published in 2009
showed no benefit of using spectroscopy in addition to T2
sequences for the detection of tumour,49 and none of the
group at the consensus meeting use it in routine clinical
practice. Although much has been published on spectroscopy, a clear benefit has not yet been demonstrated to
justify its drawbacks (the need for an endorectal coil at 1.5 T,
the time for the scan, limited resolution, and the number of
unusable voxels). There is little evidence that it adds
significantly to the performance of the multiparametric
scan we have described.
There is evidence that the degree of metabolic abnormality on spectroscopy correlates with disease grade,50
although there is a similar correlation with ADC value51
and T2 signal intensity.52 In a recent study, spectroscopy
differentiated high from low-grade tumours better than the
ADC value,53 but the group opinion was that the estimation
of tumour grade on MRI remains unreliable, and that this
indication alone cannot justify its use.
Reporting
Grading systems
The use of an ordinal scale to describe the radiologist’s
suspicion of cancer is essential, and as in breast cancer54 a
scale between 1 and 5 should be used: 1 ¼ tumour highly
unlikely; 2 ¼ tumour unlikely; 3 ¼ equivocal; 4 ¼ tumour
likely; and 5 ¼ tumour highly likely.
The recent ESUR guidelines describes a scheme (prostate
imaging, reporting, and data system, “PI-RADS”) for generating scores on each sequence based on semi-objective
criteria, although they fall short of recommending that
they be combined systematically into an overall score,4 and
a recent study has validated the T2-weighted and diffusion
(but not the dynamic contrast enhanced) components.55
The PI-RADS criteria may be useful for the radiologist in
assessing each sequence, but what matters clinically is the
final assessment of the likelihood of disease: the “overall
impression” score.
The unanimous opinion of the group was that MRI of the
prostate should be reported to a level of significance; in
particular, the radiologist’s overall impression of the likelihood of a tumour measuring >0.2 cm3 or Gleason 3 þ 4 or
higher. Small, conspicuous tumours may initially be
thought problematic with such a threshold, but are rarely so
in practice: even if well below 0.2 cm3, conspicuous tumours have a good chance of containing a Gleason 4
component and will be scored as equivocal or greater.10,56 A
score between 1 and 5 should be given for:
(a) sectors of the prostate, divided into anatomical zones.
There is not yet agreement on the optimum number of
zones, though at least 12 are recommended (anterior
and posterior, left and right, at base, mid gland and
apex). Sixteen sectors may be the best balance between
5
detail and reporting burden, and such a scheme was
described in the 2011 consensus.3
(b) Individual lesions, up to a maximum of three. Individual
lesions should be drawn on a diagram of the prostate
and each given a score for likelihood of tumour.
(c) In cases where biopsy has been performed before MRI,
scores of 1e5 should be given for staging parameters, as
follows: (i) likelihood of involvement of bladder neck,
seminal vesicles, rectum, neurovascular bundles, and
external sphincter. (ii) In addition, a score should be
given for the likelihood of extracapsular breach, but
with an indication of whether there is visible extracapsular tumour (“direct” evidence of T3a disease, in
which case the depth of extracapsular tumour should be
given), or it is being inferred from “indirect” signs, such
as bulge, extensive abutment, or loss of capsular definition.28 (iii) Seminal vesicle involvement should also be
scored as “direct” or “indirect” evidence of involvement.
The indirect signs, and the significance that should be
attached to them in terms of a score of 1e5, are summarized in the ESUR guidelines.4
When the MRI is performed before biopsy, an attempt can
be made to estimate the stage of the disease, but it should be
emphasized that this is provisional: the radiologist should
not definitively stage the disease without knowing the histology result, as the grade and size of tumour may affect the
score. A revised report staging the disease should be given
when the histology is known, or during discussion at an
multidisciplinary team (MDT) meeting.
Scans should always be reported with the clinical details,
in particular, the patient’s age and PSA.
Estimation of tumour size
Although there are considerable errors in estimating the
size of tumours at MRI,9 the tumour volume is an important
factor in the estimation of clinical significance7,57 and correlates strongly with grade.58 An estimate by the radiologist of
the volume of suspicious foci is important and can be performed using either a three-axis technique or planimetry; it
should be given for at least the three largest visible lesions.
This is not to say that further lesions should not be reported:
they should be included in the diagram of the position of
tumour, and in the text report. Where an estimation of volume is not possible, maximum tumour diameter correlates
well with histological volume59 and is an alternative.
Diagrams and scoring sheets
Although in most cases the report of a prostate MRI will
include a free-text element summarizing the findings, the
sector scores and position of suspicious foci are much easier
to convey by drawing on a diagram. The elements that
should be included are summarized in Fig 1.
We appreciate that it may be difficult to integrate such
diagrammatic reporting into routine radiology practice and
that in some centres reports will have to remain primarily
text-based. In such cases, as a minimum, scores between 1
and 5 should be given for left and right lobes of the prostate,
and for peripheral and transition zones.
Please cite this article in press as: Kirkham APS, et al., Prostate MRI: Who, when, and how? Report from a UK consensus meeting, Clinical
Radiology (2013), http://dx.doi.org/10.1016/j.crad.2013.03.030
6
A.P.S. Kirkham et al. / Clinical Radiology xxx (2013) 1e8
Figure 1 Components of a proposed scoring sheet. (a) Sixteen-sector diagram of the prostate (for tumour position), (b) 16-sector grid (for sector
scoring), and (c) boxes (to be scored 1e5) for staging information and tumour volume. A practical scoring sheet would likely contain space for
more demographic information and a free-text section.
Minimum standards and training
Like mammography, prostate MRI is challenging and
requires considerable experience to reliably exclude and
detect significant disease. We propose that those reporting
prostate MRI should: (1) report at least 50 multiparametric
scans per year; and (2) regularly attend a meeting at which
the MRI and histology results are discussed. In many cases
this will be part of a National Health Service (NHS) MDT
meeting.
The proposed scoring system simplifies audit, and all
radiologists who regularly report pre-biopsy MRI of the
prostate should be able to audit their performance using the
results of prostate biopsy and histology at radical prostatectomy. In particular, they should be able to answer the
following question: “what do your scores [between 1 and 5]
mean for the likelihood of significant disease?”
The group intends to set up several components that are
hoped will contribute to standardized, high-quality performance of prostate MRI in the UK: (1) training courses; (2)
sets of validated MRI images to enable radiologists to assess
their diagnostic performance, and as a reference for the
image quality that can be achieved when imaging parameters are optimized; and (3) a “sequence bank” of imaging
parameters at 1.5 and 3 T for each of the manufacturers.
Conclusions
MRI of the prostate has the potential to fundamentally
alter our approach to the investigation of men with a raised
PSA, but only if performed well and reported by specialist
radiologists. Most of the benefits can be achieved with a
field strength of 1.5 T, a pelvic phased-array coil, and a
multiparametric approach using T2-weighted, diffusionweighted, and contrast-enhanced sequences. Such machines are available at almost all UK NHS centres where
prostate cancer is treated.
Because MRI is useful in patients with no tumour,
insignificant tumour, or disease requiring radical treatment,
there is a strong case for its routine use before biopsy, which
enables targeting of suspicious lesions and eliminates the
problem of post-biopsy artefact. If it is necessary to perform
staging scans after biopsy, a minimum period of 10 weeks is
recommended to allow this artefact to settle, and scans
performed before 10 weeks should be recognized as of
limited resolution.
As in mammography, a standardized scoring system for
MRI of the prostate is vital, and information conveyed to
urologists can be greatly improved by the use of a diagram
for marking the position of tumour and scoring the likelihood of disease by sector.
References
1. Ilic D, O’Connor D, Green S, et al. Screening for prostate cancer: an
updated Cochrane systematic review. BJU International 2011;107:882e91.
2. Ouzzane A, Puech P, Lemaitre L, et al. Combined multiparametric MRI
and targeted biopsies improve anterior prostate cancer detection,
staging, and grading. Urology 2011;78:1356e62.
3. Dickinson L, Ahmed HU, Allen C, et al. Magnetic resonance imaging for
the detection, localisation, and characterisation of prostate cancer:
recommendations from a European consensus meeting. Eur Urol
2011;59:477e94.
4. Barentsz JO, Richenberg J, Clements R, et al. ESUR prostate MR guidelines 2012. Eur Radiol 2012;22:746e57.
Please cite this article in press as: Kirkham APS, et al., Prostate MRI: Who, when, and how? Report from a UK consensus meeting, Clinical
Radiology (2013), http://dx.doi.org/10.1016/j.crad.2013.03.030
A.P.S. Kirkham et al. / Clinical Radiology xxx (2013) 1e8
5. Parker C, Muston D, Melia J, et al. A model of the natural history of
screen-detected prostate cancer, and the effect of radical treatment on
overall survival. Br J Cancer 2006;94:1361e8.
6. Ahmed HU, Arya M, Freeman A, et al. Do low-grade and low-volume
prostate cancers bear the hallmarks of malignancy? Lancet Oncol
2012;13:e509e17.
7. Wolters T, Roobol MJ, van Leeuwen PJ, et al. A critical analysis of the
tumor volume threshold for clinically insignificant prostate cancer using
a data set of a randomized screening trial. J Urol 2011;185:121e5.
8. Wilt TJ, Brawer MK, Jones KM, et al. Radical prostatectomy versus observation for localized prostate cancer. New Engl J Med 2012;367:203e13.
9. Puech P, Potiron E, Lemaitre L, et al. Dynamic contrast-enhanced-magnetic
resonance imaging evaluation of intraprostatic prostate cancer: correlation
with radical prostatectomy specimens. Urology 2009;74:1094e9.
10. Rosenkrantz AB, Mendrinos S, Babb JS, et al. Prostate cancer foci
detected on multiparametric magnetic resonance imaging are histologically distinct from those not detected. J Urol 2012;187:2032e8.
11. Villers A, Puech P, Mouton D, et al. Dynamic contrast enhanced, pelvic
phased array magnetic resonance imaging of localized prostate cancer
for predicting tumor volume: correlation with radical prostatectomy
findings. J Urol 2006;176:2432e7.
12. Tanimoto A, Nakashima J, Kohno H, et al. Prostate cancer screening: the
clinical value of diffusion-weighted imaging and dynamic MR imaging
in combination with T2-weighted imaging. J Magn Reson Imaging
2007;25:146e52.
13. Villeirs GM, De Meerleer GO, De Visschere PJ, et al. Combined magnetic
resonance imaging and spectroscopy in the assessment of high grade
prostate carcinoma in patients with elevated PSA: a single-institution
experience of 356 patients. Eur J Radiol 2011;77:340e5.
!ge-Lechevallier F, Tonina Senes A, et al. Prostate dynamic
14. Girouin N, Me
contrast-enhanced MRI with simple visual diagnostic criteria: is it
reasonable? Eur Radiol 2007;17:1498e509.
15. Wefer AE, Hricak H, Vigneron DB, et al. Sextant localization of prostate
cancer: comparison of sextant biopsy, magnetic resonance imaging and
magnetic resonance spectroscopic imaging with step section histology. J
Urol 2000;164:400e4.
16. Ahmed HU, Kirkham A, Arya M, et al. Is it time to consider a role for MRI
before prostate biopsy? Nat Rev Clin Oncol 2009;6:197e206.
17. Lecornet E, Ahmed HU, Hu Y, et al. The accuracy of different biopsy
strategies for the detection of clinically important prostate cancer: a
computer simulation. J Urol 2012;188:974e80.
18. Rocco B, de Cobelli O, Leon ME, et al. Sensitivity and detection rate of a
12-core trans-perineal prostate biopsy: preliminary report. Eur Urol
2006;49:827e33.
19. Hambrock T, Somford DM, Hoeks C, et al. Magnetic resonance imaging
guided prostate biopsy in men with repeat negative biopsies and
increased prostate specific antigen. J Urol 2010;183:520e7.
20. Haffner J, Lemaitre L, Puech P, et al. Role of magnetic resonance imaging
before initial biopsy: comparison of magnetic resonance imagingtargeted and systematic biopsy for significant prostate cancer detection. BJU Int 2011;108:E171e8.
21. Hambrock T, Hoeks C, Hulsbergen-van de Kaa C, et al. Prospective
assessment of prostate cancer aggressiveness using 3-T diffusionweighted magnetic resonance imaging-guided biopsies versus a systematic 10-core transrectal ultrasound prostate biopsy cohort. Eur Urol
2012;61:177e84.
22. Moore CM, Robertson NL, Arsanious N, et al. Image-guided prostate
biopsy using magnetic resonance imaging-derived targets: a systematic
review. Eur Urol 2013;63:125e40.
23. Lawrentschuk N, Fleshner N. The role of magnetic resonance imaging in
targeting prostate cancer in patients with previous negative biopsies
and elevated prostate-specific antigen levels. BJU Int 2009;103:730e3.
24. Margel D, Yap SA, Lawrentschuk N, et al. Impact of multiparametric
endorectal coil prostate magnetic resonance imaging on disease
reclassification among active surveillance candidates: a prospective
cohort study. J Urol 2012;187:1247e52.
25. Vargas HA, Akin O, Afaq A, et al. Magnetic resonance imaging for predicting prostate biopsy findings in patients considered for active surveillance of clinically low risk prostate cancer. J Urol 2012;188:1732e8.
26. Borofsky MS, Rosenkrantz AB, Abraham N, et al. Does suspicion of
prostate cancer on integrated T2 and diffusion-weighted MRI predict
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.
38.
39.
40.
41.
42.
43.
44.
45.
46.
47.
48.
7
more adverse pathology on radical prostatectomy? Urology 2013 Feb 7.
http://dx.doi.org/10.1016/j.urology.2012.12.026. pii: S0090-4295(12)
01570-1.
Prostate cancer: diagnosis and treatment. London: National Institute for
Health and Clinical Excellence; 2008. Available at: http://www.nice.org.
uk/cg58 [accessed 4 February 2013].
Wang L, Hricak H, Kattan MW, et al. Prediction of organ-confined
prostate cancer: incremental value of MR imaging and MR spectroscopic imaging to staging nomograms. Radiology 2006;238:597e603.
Wang L, Hricak H, Kattan MW, et al. Prediction of seminal vesicle invasion in prostate cancer: incremental value of adding endorectal MR
imaging to the Kattan nomogram. Radiology 2007;242:182e8.
Hoeks CM, Barentsz JO, Hambrock T, et al. Prostate cancer: multiparametric MR imaging for detection, localization, and staging. Radiology 2011;261:46e66.
Cornud F, Rouanne M, Beuvon F, et al. Endorectal 3D T2-weighted 1mmslice thickness MRI for prostate cancer staging at 1.5 Tesla: should we
reconsider the indirects signs of extracapsular extension according to
the D’Amico tumor risk criteria? Eur J Radiol 2012;81:e591e7.
Duffield AS, Lee TK, Miyamoto H, et al. Radical prostatectomy findings in
patients in whom active surveillance of prostate cancer fails. J Urol
2009;182:2274e8.
Qayyum A, Coakley FV, Lu Y, et al. Organ-confined prostate cancer: effect of prior transrectal biopsy on endorectal MRI and MR spectroscopic
imaging. AJR Am J Roentgenol 2004;183:1079e83.
Tamada T, Sone T, Jo Y, et al. Prostate cancer: relationships between
postbiopsy hemorrhage and tumor detectability at MR diagnosis.
Radiology 2008;248:531e9.
White S, Hricak H, Forstner R, et al. Prostate cancer: effect of postbiopsy
hemorrhage on interpretation of MR images. Radiology 1995;195:385e90.
Cornfeld D, Weinreb J. MR imaging of the prostate: 1.5 T versus 3 T.
Magn Reson Imaging Clin N Am 2007 Aug 1;15(3):433e48.
€ tterer JJ, Hambrock T, et al. Prostate cancer: body-array
Heijmink SWTPJ, Fu
versus endorectal coil MR imaging at 3 Tdcomparison of image quality,
localization, and staging performance. Radiology 2007;244:184e95.
Park BK, Lee HM, Kim CK, et al. Lesion localization in patients with a
previous negative transrectal ultrasound biopsy and persistently
elevated prostate specific antigen level using diffusion-weighted imaging at three Tesla before rebiopsy. Invest Radiol 2008;43:789e93.
Iwazawa J, Mitani T, Sassa S, et al. Prostate cancer detection with
magnetic resonance imaging: is dynamic contrast-enhanced imaging
necessary in addition to diffusion-weighted imaging? Diagn Interv
Radiol 2011;17:243e8.
Puech P, Betrouni N, Makni N, et al. Computer-assisted diagnosis of
prostate cancer using DCE-MRI data: design, implementation and preliminary results. Int J Comput Assist Radiol Surg 2009;4:1e10.
Sung YS, Kwon HJ, Park BW, et al. Prostate cancer detection on dynamic
contrast-enhanced MRI: computer-aided diagnosis versus single perfusion parameter maps. AJR Am J Roentgenol 2011;197:1122e9.
Peng Y, Jiang Y, Yang C, et al. Quantitative analysis of multiparametric
prostate MR images: differentiation between prostate cancer and
normal tissue and correlation with Gleason scoreda computer-aided
diagnosis development study. Radiology 2013 Feb 7.
!re O. MR assessment of recurrent prostate cancer after radiation
Rouvie
therapy. Radiology 2007;242:635e6. author reply 6e7.
Casciani E, Polettini E, Carmenini E, et al. Endorectal and dynamic
contrast-enhanced MRI for detection of local recurrence after radical
prostatectomy. AJR Am J Roentgenol 2008;190:1187e92.
Kim C, Park B, Lee H, et al. MRI techniques for prediction of local tumor
progression after high-intensity focused ultrasonic ablation of prostate
cancer. AJR Am J Roentgenol 2008;190:1180e6.
Makarov DV, Trock BJ, Humphreys EB, et al. Updated nomogram to
predict pathologic stage of prostate cancer given prostate-specific antigen level, clinical stage, and biopsy Gleason score (Partin tables) based
on cases from 2000 to 2005. Urology 2007;69:1095e101.
Briganti A, Suardi N, Capogrosso P, et al. Lymphatic spread of nodal
metastases in high-risk prostate cancer: the ascending pathway from
the pelvis to the retroperitoneum. Prostate 2012;72:186e92.
Haukaas S, Roervik J, Halvorsen OJ, et al. When is bone scintigraphy
necessary in the assessment of newly diagnosed, untreated prostate
cancer? Br J Urol 1997;79:770e6.
Please cite this article in press as: Kirkham APS, et al., Prostate MRI: Who, when, and how? Report from a UK consensus meeting, Clinical
Radiology (2013), http://dx.doi.org/10.1016/j.crad.2013.03.030
8
A.P.S. Kirkham et al. / Clinical Radiology xxx (2013) 1e8
49. Weinreb JC, Blume JD, Coakley FV, et al. Prostate cancer: sextant localization at MR imaging and MR spectroscopic imaging before prostatectomydresults
of
ACRIN
prospective
multi-institutional
clinicopathologic study. Radiology 2009;251:122e33.
50. Kobus T, Hambrock T, Hulsbergen-van de Kaa CA, et al. In vivo assessment of prostate cancer aggressiveness using magnetic resonance
spectroscopic imaging at 3 T with an endorectal coil. Eur Urol
2011;60:1074e80.
51. Mazaheri Y, Shukla-Dave A, Hricak H, et al. Prostate cancer: identification with combined diffusion-weighted MR imaging and 3D 1H MR
spectroscopic imagingdcorrelation with pathologic findings. Radiology
2008;246:480e8.
52. Wang L, Mazaheri Y, Zhang J, et al. Assessment of biologic
aggressiveness of prostate cancer: correlation of MR signal intensity
with Gleason grade after radical prostatectomy. Radiology 2008;
246:168e76.
53. Nagarajan R, Margolis D, Raman S, et al. MR spectroscopic imaging and
diffusion-weighted imaging of prostate cancer with Gleason scores.
J Magn Reson Imaging 2012;36:697e703.
54. Balleyguier C, Ayadi S, Van Nguyen K, et al. BIRADS classification in
mammography. Eur J Radiol 2007;61:192e4.
55. Portalez D, Mozer P, Cornud F, et al. Validation of the European Society
of Urogenital Radiology scoring system for prostate cancer diagnosis on
multiparametric magnetic resonance imaging in a cohort of repeat biopsy patients. Eur Urol 2012;62:986e96.
56. Turkbey B, Mani H, Shah V, et al. Multiparametric 3T prostate magnetic
resonance imaging to detect cancer: histopathological correlation using
prostatectomy specimens processed in customized magnetic resonance
imaging based molds. J Urol 2011;186:1818e24.
57. Bastian P, Carter B, Bjartell A, et al. Insignificant prostate cancer and
active surveillance: from definition to clinical implications. Eur Urol
2009;55:1321e32.
58. Song SY, Kim SR, Ahn G, et al. Pathologic characteristics of prostatic
adenocarcinomas: a mapping analysis of Korean patients. Prostate
Cancer Prostatic Dis 2003;6:143e7.
59. Mizuno R, Nakashima J, Mukai M, et al. Maximum tumor diameter is a
simple and valuable index associated with the local extent of disease in
clinically localized prostate cancer. Int J Urol 2006;13:951e5.
Please cite this article in press as: Kirkham APS, et al., Prostate MRI: Who, when, and how? Report from a UK consensus meeting, Clinical
Radiology (2013), http://dx.doi.org/10.1016/j.crad.2013.03.030
`