EURURO Vol. 71 No. 4

649 692

1. Introduction
The use of multiparametric magnetic resonance imaging
(MRI) to inform the detection of prostate cancer (PCa) has
grown rapidly in the last few years. Numerous publications
have sought to standardise the conduct and reporting of
prostate MRI . Most recently the European Society of
Uroradiology and the American College of Radiology
published the second version of the Prostate Imaging
Reporting and Data System (PI-RADS) outlining the conduct,
interpretation, and reporting of prostate MRI. These guide-
lines focus on PCa detection, and the questions asked are
‘‘How likely is it that this man has prostate cancer?’’ and
‘‘How can this best be biopsied?’’
The 2014 UK National Institute for Health and Care
Excellence (NICE) PCa guidelines suggest a role for MRI
in the initial and repeat assessment of men on active
surveillance, although no guidance is offered on imaging
criteria for selection or continuation of surveillance. NICE
recommends MRI and/or biopsy for re-evaluation if there is
‘‘concern over prostate-specific antigen (PSA) kinetics or
clinical assessment.’’ The question asked of MRI is then ‘‘Has
there been any significant change?’’ To distinguish between
significant change, measurement error, and natural fluctua-
tions in tumour appearance, we need to understand the
natural history of MRI changes over time in men on active
surveillance in terms of change to MRI lesions and so-called
normal MRI findings. Once these data are established,
radiologic thresholds can be set that indicate significant
actionable, clinical change in disease.
Schoots et al reviewed the evidence for MRI in men on
active surveillance . They found a lack of published data
in the use of MRI in active surveillance follow-up. The
European School of Oncology then convened the Prostate
Cancer Radiological Estimation of Change in Sequential
Evaluation (PRECISE) panel to develop recommendations
for MRI in men on active surveillance for PCa. Formal
consensus methodology, including the use of a face-to-face
meeting, was chosen. This technique is helpful to determine
the level of agreement amongst experts and to identify
areas that require further data before agreement can be
reached. The panel’s objective was to develop recommen-
dations for the reporting of individual MRI studies in men
on active surveillance (the PRECISE report form) and for
researchers reporting the outcomes of cohorts of men
having MRI on active surveillance (the PRECISE checklist).
2. Materials and methods
2.1. Study design
We used the RAND/UCLA Appropriateness Method A core group
(C.M.M., I.G.S., A.K., C.A., and F.G.) developed a draft set of 350 statements
and sent them to all panel members for modification. Statements could
be revised, removed, or added at this stage. A revised set of
394 statements was scored by each panel member on a scale of
agreement from 1 to 9, in which 1 indicated strongest disagreement and
9 indicated strongest agreement. These scores were collated, and a
summary of agreement, uncertainty, or disagreement (derived from the
group median score) was calculated for each statement. Calculations to
determine consensus or lack of consensus for each statement were
performed using the RAND/UCLA classical criteria that take into account
the proportion of panellists scoring within a given category of agreement
(7–9), uncertainty (4–6), or disagreement (1–3). For a statement to have
consensus, a clear majority scoring in that category is needed.
A chair (P.A.) who did not participate in scoring convened a panel
meeting. A graphic representation of the group response was presented
for each statement that included the group median score and the degree
of consensus ). Each statement was discussed. Some statements
were modified or removed; others were added as a result of the
discussions. Following discussion, each statement was rescored anony-
mously by each panel member. Following the meeting, the individual
panellist scores were collated, and the degree of agreement and
consensus was calculated for each statement. The collated scores and
the content of the discussion were used to develop the PRECISE checklist
of reporting criteria for studies of MRI in men on active surveillance and
was scored for agreement on a 9-point scale by each panellist prior to a panel meeting. Each
statement was discussed and rescored at the meeting.
Outcome measurements and statistical analysis: Measures of agreement and consensus
were calculated for each statement. The most important statements, derived from both
group discussion and scores of agreement and consensus, were used to create the Prostate
Cancer Radiological Estimation of Change in Sequential Evaluation (PRECISE) checklist and
case report form.
Results and limitations: Key recommendations include reporting the index lesion size
using absolute values at baseline and at each subsequent MRI. Radiologists should assess
the likelihood of true change over time (ie, change in size or change in lesion characteristics
on one or more sequences) on a 1–5 scale. A checklist of items for reporting a cohort of men
on active surveillance was developed. These items were developed based on expert
consensus in many areas in which data are lacking, and they are expected to develop
and change as evidence is accrued.
Conclusions: The PRECISE recommendations are designed to facilitate the development of a
robust evidence database for documenting changes in prostateMRI findings over time of men
on active surveillance. If used, they will facilitate data collection to distinguish measurement
error and natural variability in MRI appearances from true radiologic progression.
Patient summary: Few published reports are available on how to use and interpret
magnetic resonance imaging for men on active surveillance for prostate cancer. The
PRECISE panel recommends that data should be collected in a standardised manner so
that natural variation in the appearance and measurement of cancer over time can be
distinguished from changes indicating significant tumour progression.
# 2016 European Association of Urology. Published by Elsevier B.V. All rights reserved.
Please visit
to read and
answer questions on-line.
The EU-ACME credits will
then be attributed
automatically.
E U R O P E A N U R O L O G Y 7 1 ( 2 0 1 7 ) 6 4 8 – 6 5 5 649