EURURO Vol. 72 No. 3

E63 476

Letter to the Editor
Re: Nicholas J. Giacalone, William U. Shipley,
Rebecca H. Clayman, et al. Long-term Outcomes After
Bladder-preserving Tri-modality Therapy for Patients
with Muscle-invasive Bladder Cancer: An Updated
Analysis of the Massachusetts General Hospital
Experience. Eur Urol 2017;71:952–60
Methodological Issues to Avoid Misinterpretation
We read with great interest the paper by Giacalone et al
. The authors evaluated long-term outcomes for patients
with muscle-invasive bladder cancer (MIBC) treated with
trimodal therapy (TMT). The authors showed that on
multivariable analysis, T2 disease, complete response
(CR) to chemoradiation, and the presence of tumor-
associated carcinoma in situ remained in the model as
significant predictors of overall survival (OS) and disease-
specific survival (DSS) . We would like to congratulate
the authors as their results provide insights into long-term
outcomes for TMT among patients with MIBC. However, we
are curious about some methodological and statistical
issues that were not addressed in the study.
The multivariable Cox regression models used were
evaluated directly in the study population and have not
been internally validated This represents apparent
validation, so that the estimated effects for the predictors of
long-term outcomes are optimistic. Investigators assess the
validity of prediction models using internal and external
validation procedures such as bootstrapping and cross-
validation Hence we respectfully suggest that the
authors assess the validity of the prediction models and
report optimism-corrected estimates for the predictors of
long-term outcomes among patients with MIBC treated
with TMT.
The authors mentioned that they used stepwise Cox
regression for model building, but it is unclear which
stepwise selection variant was used to include significant
predictors in model. Stepwise selection comprises several
variants, each of which has pros and cons, and they may
produce different results
A take home message for readers is that prediction
models should be both internally valid and corrected for
optimism.
Conflicts of interest: The authors have nothing to disclose.
References
[1]
.
[2] Gandaglia G, Fossati N, Zaffuto E, et al. Development and internal
validation of a novel model to identify the candidates for extended
pelvic lymph node dissection in prostate cancer. Eur Urol. In press.
.
[3] Karnes RJ, Choeurng V, Ross AE, et al. Validation of a genomic risk
classifier to predict prostate cancer-specific mortality in men with
adverse pathologic features. Eur Urol. In press.
[4]
Saeid Safiria
Erfan Ayubib,c,*
aManagerial Epidemiology Research Center, Department of Public Health,
School of Nursing and Midwifery, Maragheh University of Medical Sciences,
Maragheh, Iran
bDepartment of Epidemiology, School of Public Health, Shahid Beheshti
University of Medical Sciences, Tehran, Iran
c Department of Epidemiology & Biostatistics, School of Public Health, Tehran
University of Medical Sciences, Tehran, Iran
*Corresponding author. Department of Epidemiology, School of Public
Health, Shahid Beheshti University of Medical Sciences, Tehran, Iran.
Tel./Fax: +98 21 22432040.
E-mail address: (E. Ayubi).
June 5, 2017
ava ilable at
journal homepage:
DOI of original article: .
0302-2838/# 2017 European Association of Urology. Published by Elsevier B.V. All rights reserved.