EURURO Vol. 72 No. 6

1025 1030

Re: Adjuvant Chemotherapy vs Observation for Patients
with Adverse Pathologic Features at Radical Cystectomy
Previously Treated With Neoadjuvant Chemotherapy
Seisen T, Jamzadeh A, Loew JJ, et al
JAMA Oncol. In press.
Experts’ summary:
The authors investigated the effect of adjuvant chemotherapy
(AC) versus observation in patients with pT3/T4 and/or pN+
urinary bladder cancer (UBC) previously treated with neoad-
juvant chemotherapy (NAC) and radical cystectomy (RC) . A
retrospective analysis of the National Cancer Data Base (NCDB)
identified 788 patients. Of these, 184 (23.4%) also received AC.
Missing data were assumed to be random and multiple impu-
tation was used to handle them. In an attempt to eliminate
selection bias, the population was weighted using inverse
probability of treatment weighting in adjusted analyses. Mul-
tivariable logistic regression analysis showed a significant
association of younger age, pN+, nonacademic centers, west-
ern facilities, and urban population with administration of AC
after NAC and RC. After median follow-up of 45.7 mo (inter-
quartile range 31.2–67.8), AC administration was associated
with a significant improvement in overall survival (OS; hazard
ratio [HR] 0.78, 95% confidence interval [CI] 0.61–0.99;
p = 0.046). In conclusion, treating patients with NAC and RC
followed by AC administration seems to improve OS by
approximately 5 mo.
Experts’ comments:
The standard of care for patients with nonmetastatic muscle-
invasive UBC is NAC followed by RC in those who can tolerate
cisplatin-based combination chemotherapy (recommenda-
tion grade A). AC can be offered in patients with pT3/T4
and/or pN+ disease if no NAC has been given (recommenda-
tion grade B) . The most recent meta-analysis including
945 patient from nine randomized trials supports administra-
tion of AC, particularly in patients with node-positive disease
. In a retrospective analysis of the NCDB, Galsky et al [4]
assessed the effectiveness of AC in 5653 patients with pT3/T4
and/or pN+ disease. AC receipt was significantly associated
with improved OS using propensity score quintile analysis (HR
0.7, 95% CI 0.64–0.76) and other types of propensity-matched
analyses. This would mean that treating non–organ-confined
UBC with RC and perioperative chemotherapy could be
curative in approximately a quarter of patients. Using a
three-step protocol (NAC then RC followed by AC) could
improve disease control and survival in some patients. By
contrast, using individual patient data in a retrospective study,
Zargar-Soshtari et al [5] compared outcomes for 88 patients
treated with NAC and RC for non –organ-confined UCB; 29
(33%) also received AC. No association of AC with disease
recurrence or cancer-specific mortality was observed in this
limited-size study. The study by Seisen et al represents the
largest investigating the role of AC in patients with pT3/T4
and/or pN+ disease and previously treated with NAC and RC.
The statistical approach of this study is certainly of high
quality, but as with every study, there are some limitations
that must be considered in interpreting the level of evidence.
These are mainly inherent to the retrospective design of the
study and the NCDB. Despite accounting for selection bias via
the use of advanced statistical methods, there are still several
confounders not reported in the NCDB that could have influ-
enced decision-making. First, as addressed by the authors,
information on the chemotherapy agent used and regimens
was not available. Indeed, it is still unknown what number of
cycles should be administered and if the same chemotherapy
agent can be used in NAC and AC settings. Second, younger
patients and those with positive lymph nodes at RC are
selected for aggressive therapy, and this is confirmed in the
multivariable logistic regression. Interestingly, the explorato-
ry analysis showed that the OS benefit of AC decreased sig-
nificantly with age (HR 0.97, 95% CI 0.95–0.99; p = 0.02). Third,
data on cause of death and performance status are not avail-
able in the NCDB. RC is generally associated with high mor-
bidity that depends on age and comorbidity. All these features
could significantly impact OS as well as the likelihood of
receiving AC. Finally, analyses were conducted on a weighted
population. Although this process adjusted for tumor stage, it
could not account for other pathologic features such as histo-
logic variants and downstaging after NAC. The first is probably
under-reported in the NCDB and the latter is missing in the
database. Both features could have biased therapeutic deci-
sion. In conclusion, muscle-invasive UBC is an aggressive
disease with poor outcomes and surgery alone is not sufficient
in most patients. This study supports the treatment of patients
in a multimodal setting, especially if adverse clinicopathologic
features are present.
The indication for AC and the regimen used should be
individually evaluated and based on pathologic features and
response to NAC. One would certainly think that pT3/T4
and/or pN+ disease after cisplatin-based NAC represents a
cisplatin-resistant biology best treated with other thera-
peutic approaches such as checkpoint inhibitors if shown
effective. Moreover, in the age of precision medicine,
resistant tumors should ideally undergo molecular analysis
to identify mechanisms of resistance and possible molecu-
lar hubs for therapy.
Conflicts of interest: The authors have nothing to disclose.
References
[1] Seisen T, Jamzadeh A, Leow JJ, et al. Adjuvant chemotherapy vs
observation for patients with adverse pathologic features at radical
cystectomy previously treated with neoadjuvant chemotherapy.
JAMA Oncol. In press.
[2]
.
[3]
.
[4]
.
[5]
E U R O P E A N U R O L O GY 7 2 ( 2 0 17 ) 10 2 2– 10 2 6 1025