EURURO Vol. 73 No. 1

92 148

Platinum Priority – Editorial
Referring to the article published on pp. 81–91 of this issue
Magnetic Resonance Imaging and Detection of Metastases
in Prostate Cancer: Learning Lessons from History
Alex P. Hoyle, Noel W. Clarke
The Christie and Salford Royal Hospitals, Manchester, UK
In February 2017 the inventor of the magnetic resonance
imaging (MRI) scanner, Sir Peter Mansfield, died at the age
of 83 . He was awarded the 2003 Nobel Prize for
medicine, and MR scanning has since become ubiquitous
and indispensable in most fields of cancer, but not quite yet
in urological oncology, particularly for advanced and high-
risk prostate cancer staging. Why have we been so slow to
evaluate its utility properly and to put it to work to best
effect in this disease, especially in the critically important
area of detection, localisation, and quantitation of metasta-
ses?
The review by Woo et al in this issue of European
Urology provides an overview and meta-analysis of the
literature on the utility of MRI in detecting bone metastases,
and compares this in terms of sensitivity and specificity to
the existing urological ‘‘workhorse’’ and traditional 40-yr-
old cornerstone of metastatic prostate cancer staging, the
technetium bone scan . It also highlights the weakness of
the uro-oncological literature in this area and draws our
attention to the paucity of high-quality clinical science in
evaluating and using MRI as a staging tool in the advanced
prostate cancer setting.
The relatively low sensitivity (79%) and specificity (82%)
of bone scintigraphy is well known. In addition, modern
practice dictates that scintigraphy cannot stand alone in the
staging process, as supplementary imaging is required to
complete the assessment of soft-tissue disease status and to
confirm the nature of lesions that are scintigraphically
‘‘equivocal’’. Is it time to move to MRI as the primary
imaging modality? Do the data presented here show or help
clinicians to understand how good MRI is as an alternative,
and if it is better, what type of MRI scanner or scan protocol
should be used?
By definition, the review provides level 2a evidence of
accurate MRI-based detection of bone metastases in
prostate cancer. It is a meta-analysis of ten predominantly
prospective studies and it concludes that overall per-patient
MRI sensitivity and specificity is 96% and 98%, respectively.
However, while the study methodology is fundamentally
sound, the individual studies analysed are generally weak,
with significant heterogeneity and inherent bias that limit
the overall interpretation. Heterogeneity between studies is
recognised by the authors in their submission, and
following meta-regression analysis the number of imaging
planes was the only significant explanatory factor. When
two or more planes were used, the sensitivity and
specificity increased to 99%, but data in relation to this
were only available in a small number of the studies. The
meta-analysis also found no statistically significant differ-
ence between diffusion-weighted imaging (DWI) and
standard MRI sequencing between studies. This seems
surprising at first glance, but the data have limited numbers
for comparison and information relating to standardisation
or co-alignment of protocols is fundamentally weak or
missing. It seems clear that while DWI looks better, it needs
further definitive clarification before safe conclusions can
be drawn.
The authors acknowledge the methodological limita-
tions in their study, including the per-patient analysis of
imaging and the predominant use of a best value
comparator (BCV) over histological confirmation as a
reference standard. This lack of per-lesion analysis hinders
available at
journal homepage:
DOI of original article: .
* Corresponding author. Department of Surgery, Christie Hospital, Wilmslow Road, West Didsbury, Manchester M20 4BX, UK. Tel. +44 161 4463364;
Fax: +44 161 4463365.
E-mail address: (N.W. Clarke).
0302-2838/# 2017 Published by Elsevier B.V. on behalf of European Association of Urology.