EURURO Vol. 71 No. 5

821 844

The report from the Guideline Office is timely givenmore
recent releases of RCTs that conflict with prior data. Yaxley
et al reported on an RCT comparing open to robot-
assisted radical prostatectomy. They found no difference in
their primary outcome of quality of life (reflecting conti-
nence and sexual function), mirroring prospective RCT data
in colorectal cancer and population-based prospective
studies yet conflicting with a prior MA and surgical
opinions . This conflict probably reflects weaknesses
in prior data (retrospective reports of surgical learning
curves, eg, my higher-volume robotic results are better than
my previous lower-volume open data) and in trial design
(eg, a single surgeon per arm with imbalanced experience;
primary outcomes that reflect open prostatectomy goals
and not the potential benefits of robotic surgery). Surgeons
and patients appear unlikely to adopt the findings of this
RCT. Second, Hamdy et al reported 10-yr outcomes
from the ProtecT RCT. The authors found similar cancer and
overall survival outcomes from surgery and radiotherapy
for localized prostate cancer, conflicting with a prior MA
. This conflict reflects many issues, including the low
event rate in ProtecT and the possible bias towards positive
outcomes in population cohorts. We await longer-term
outcomes from ProtecT to resolve this conflict.
We welcome this thoughtful piece from the Guidelines
Office. It serves to review the limitations of evidence-based
medicine and to understand conflicting sources of evidence.
However, we should not take from this work that RCTs are
redundant in surgery. Rather, there is a need for sound
statistical design and trial conduct for any evaluation.
Realistically, RCTs are the only workable model to overcome
selection bias, which is nearly always strong in clinical
studies of intended positive effects. Medical doctors are
specifically trained to select the best treatment for the right
patient. Often, the most important prognostic factors are
only ever ‘‘recorded’’ in the brain of the attending surgeon
and cannot be accounted for in statistical models. RCTs also
introduce prospective, clearly defined treatment and evalu-
ation protocols, and thus are one of the best ways of raising
clinical standards. Furthermore, a well-designed and well-
conducted trial provides reliable estimates that can be used
to inform patient decision-making. For example, the ProtecT
trial provides high-quality comparative information about
the impact of treatments on patient quality of life. Using this
information, men can make an informed choice about
survival and the impact on issues such as impotence and
incontinence. It is through RCTs that we should establish
why most surgeons prefer robotic surgery and use this
information to drive the generation of better robots.
Thankfully, these lessons are being learned, with more
surgical RCTs running now than ever before. Trial ambition is
also rising, with ongoing prospective surgical RCTs comparing
prostatectomy with focal regional ablation
), prostatectomy in the presence of oligo-
metastases ), and radical
cystectomy with BCG immunotherapy
), among many others. Much of this activity
is in oncology and has been driven by forces beyond urology.
The community dealing with benign urologic conditions
needs to raise its aspirations too, for the benefit of all our
patients. We hope for a future with fewer evidence conflicts
and more robust data for treatment decisions.
Conflicts of interest: The authors have nothing to disclose.
References
[1] Cochrane AL. Effectiveness and efficiency. random reflections on
health services. London, UK: Nuffield Provincial Hospitals Trust;
1972.
[2]
[3]
.
[4]
.
[5] Yaxley JW, Coughlin GD, Chambers SK, et al. Robot-assisted laparo-
scopic prostatectomy versus open radical retropubic prostatectomy:
early outcomes from a randomised controlled phase 3 study. Lancet
2016;388:1057–66.
.
[6]
.
[7]
.
[8]
.
[9]
[10]
.
[11]
[12] Wallis CJD, et al. Surgery versus radiotherapy for clinically-localized
prostate cancer: a systematic review and meta-analysis. Eur Urol
2016;70:21–30. .
E U R O P E A N U R O L O G Y 7 1 ( 2 0 1 7 ) 8 2 0 – 8 2 1 821