653 692
lesion seen onMRI. It was acknowledged that it is helpful for
the radiologist in the clinical setting to know the location of
positive biopsies, although this information would not be
known in a blinded study.
3.6.
Reporting of magnetic resonance imaging at baseline and
follow-up
Prostate volume on T2-weighted sequences and PSA density
should be reported. Determination of an assessment of the
likelihood of clinically significant disease on a 1–5 scale is
required for each MRI. The use of the term
assessment
was
chosen to include both those groups who use PI-RADS (v.1
or v.2) and those who use a 1–5 scale based on overall
clinical impression without predefined characteristics per
sequence (commonly called a Likert scale). The scale used
should be identified.
The highest likelihood of clinically significant cancer of all
separate lesions should provide the likelihood of clinically
significant cancer for the whole prostate. For men with a
visible lesion, the key metric is the size of the index lesion on
the baseline MRI and at each time point thereafter. The term
index lesion
can be used to denote the largest lesion, or the
one with the highest Gleason grade, or the one of highest
suspicion based on MRI criteria
[6] .It was noted that not all
men with PCa suitable for active surveillance will have a
visible lesion on MRI. It was agreed that size can be
measured using volume (by planimetry or calculated from
three diameters), by biaxial measurement of maximum
diameters on an axial slice, or by a single measurement of
maximum diameter. The panel felt that as yet there was
insufficient evidence to determine which of the methods for
measuring size was optimal for distinguishing between
natural fluctuation in tumour volume, measurement errors
over time, or true disease progression. Some believed that
planimetry volume would be most accurate; others were
concerned that this was too time consuming. For lesions best
seen on functional image sequences (eg, high b-value
images), a single diameter may be more reproducible than
a volume because of the need to use larger voxel sizes in
sequence acquisitions. Comparative data from the same
cohort on the reproducibility of different size measurements
(eg, planimetry volume and biaxial diameter) would be of
great value in exploring this further.
All parameters reported on the baseline MRI should be
reported again on follow-up MRI. In addition, any MRI
report after the baseline MRI report should include an
assessment of the likelihood of significant radiologic
progression from the baseline MRI scan, on a 1–5 scale,
along with a description of the change that has given rise to
that assessment (eg, change in size or change in conspicuity
on one or more sequences).
Table 3shows further details. It
should be noted that there are no robust data on which to
base the threshold for a significant change in size or
conspicuity. The intention is that data collection using the
suggested format will allow such data to be acquired, and
that, in time, thresholds can be set.
3.7.
Clinically significant disease in men on active surveillance
It was agreed that Gleason grading and MCCLs were
important determinants of clinically significant disease in
men on active surveillance, but no cut-off could be agreed
upon. It was agreed that Gleason 4 + 3 or T3a disease or
any involvement of lymph nodes or bone metastases is
clinically significant. Some panellists deemed any Gleason
pattern 4 as significant; others felt that small-volume
secondary pattern 4 disease alone was not necessarily of
clinical significance in all men. PSA and PSA derivatives
such as PSA density and PSA doubling time were deemed of
interest in determining clinically significant disease,
although again no threshold was identified.
It was acknowledged that clinical significance of MRI
lesions is also influenced by patient factors such as age and
comorbidities; a lesion may be deemed significant in a
younger man aged 50 yr but not in an older man with
several comorbidities.
3.8.
Noteworthy areas of uncertainty
There was no agreement on the best way to present change
in lesion size or appearance over time across a cohort of
men. It was acknowledged that some lesions become
nonvisible during follow-up, and there was uncertainty
over how best to deal with this when aggregating results
across a cohort. Concern was expressed that use of
percentage change of lesion volume across a cohort
could yield a large percentage change in small lesions (eg,
a 0.1-cm
3
lesion increasing to a 0.3-cm
3
lesion) and thereby
skew results across the cohort. It was also noted that the
measurement errors of small lesions could be larger than any
change, even if significant in percentage terms.
Table 3 – Assessment of likelihood of radiologic progression on magnetic resonance imaging in men on active surveillance
Likert
Assessment of likelihood of
radiologic progression
Example
1
Resolution of previous features suspicious on MRI
Previously enhancing area no longer enhances
2
Reduction in volume and/or conspicuity of previous
features suspicious on MRI
Reduction in size of previously seen lesion that remains suspicious for clinically
significant disease
3
Stable MRI appearance: no new focal/diffuse lesions
Either no suspicious features or all lesions stable in size and appearance
4
Significant increase in size and/or conspicuity of
features suspicious for prostate cancer
Lesion becomes visible on diffusion-weighted imaging; significant increase in
size of previously seen lesion
5
Definitive radiologic stage progression
Appearance of extracapsular extension, seminal vesicle involvement, lymph node
involvement, or bone metastasis
MRI = magnetic resonance imaging.
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
653