EURURO Vol. 72 No. 3

420 476

and Drug Administration (FDA) has lent priority review for
pembrolizumab as second-line treatment of UCC, and the
file has also been submitted to the European Medicine
Agency. Previous phase 2 clinical trials have already
resulted in FDA approval of atezolizumab and nivolumab
for second-line therapy of UCC (Supplementary Tables 1
and 2) . Meanwhile, after obtaining approval in
second-line treatment of UCC, ICIs are currently moving
toward earlier treatment lines and disease stages.
In a recent phase 2 study, cisplatin-ineligible patients
with advanced UCC were treated with first-line atezolizu-
mab (1200 mg intravenously every 3 wk), resulting in an
ORR of 23% (9% complete response rate) and median OS of
15.9 mo (Supplementary Tables 1 and 2) . Similar
results from a phase 2 study with pembrolizumab showed a
comparable ORR in this patient population . Since the
median OS in patients unfit for cisplatin, who receive mostly
carboplatin–gemcitabine as first-line chemotherapy, is
around 9 mo at best , the extrapolated median OS of
more than 12 mo in these studies seem encouraging for
ICIs, having prompted applications for the additional
indication as first-line treatment in the frail cisplatin-unfit
patient population. In addition, several phase 3 studies are
currently evaluating the efficacy of ICIs as first-line
treatment in cisplatin-eligible patients. In several ongoing
first-line RCTs, immune checkpoint blockade (monother-
apy, in combination with platinum-based chemotherapy, or
combination of anti-PD-1 and anti-CTLA-4) is compared
with conventional platinum-based chemotherapy .
Likewise, studies addressing the use of ICIs in the adjuvant
setting are in progress in patients at high risk for disease
progression following radical cystectomy BCG-unre-
sponsive high-risk NMIBC has high recurrence rates and
also a significant risk of progression to muscle-invasive
disease. At present, radical cystectomy is the only available
treatment option for BCG-unresponsive NMIBC . The
use of ICIs as a treatment strategy and potential means to
avoid bladder cancer surgery would be of great potential.
This is further supported by the observation that PD-L1
expression seems to be higher following BCG treatment
and that, like in muscle-invasive disease, a high mutational
load is present in BCG-unresponsive NMIBC . An ongoing
international multicenter phase II clinical trial explores
the efficacy of pembrolizumab in BCG-unresponsive
NMIBC
3.4.2.2. Renal cell cancer. With an almost 6 mo OS benefit ,
nivolumab has been approved by the FDA, thereby replacing
everolimus as second-line treatment of advanced clear cell
RCC. Cost per responder analysis of the CheckMate 025 trial
showed that nivolumab is also cost effective compared with
everolimus, with a monthly cost per responder of
$54 315 for nivolumab compared with $224 711 for ever-
olimus . However, large studies on the efficacy of ICIs for
nonclear cell RCC are still lacking. Furthermore, the place of
nivolumab as second-line treatment of clear cell RCC is
currently shared with cabozantinib, since this tyrosine
kinase inhibitor has been approved more recently as
another second-line treatment option .
In resemblance with UCC, immune checkpoint blockade
is also moving toward earlier treatment lines and disease
stages of RCC, including the neoadjuvant and adjuvant
settings. Current clinical trials mainly focus on several
combination strategies, including the combination
nivolumab–ipilimumab . From a historical perspective,
combining immune checkpoint blockade with antiangio-
genic therapy is a logical step in the treatment of RCC. It has
been shown that bevacizumab increases the migration of
cytotoxic T cells into RCC, thereby potentially enhancing the
local immune response induced by atezolizumab . At
present, several clinical trials have been initiated in which
ICIs are combined with the monoclonal antibody bevaci-
zumab or a tyrosine kinase inhibitor such as axitinib .
3.4.2.3. Prostate cancer. In contrast with UCC and RCC, data
on the efficacy of ICIs in mCRPC are limited. Although the
phase 3 trials did not show benefit for ipilimumab in mCRPC
patients , immune checkpoint blockade may still
play a role in a subset of mCRPC patients. In enzalutamide-
resistant mCRPC patients, early phase 2 studies have shown
efficacy of pembrolizumab when added to enzalutamide
. The survival benefit conferred by sipuleucel-T also
indicates that immunotherapy-boosting T-cell activity can
exert effects in PC patients . In order to enhance T-cell
activity in mCRPC, several combination strategies are
currently under development, including ICIs combined
with anticancer vaccination, PARP inhibition, radium-223,
chemotherapy, or enzalutamide .
3.4.3. Tumor PD-L1 expression as a predictive marker for efficacy
Overall, the results on the value of PD-L1 expression in UCC
and RCC are somewhat conflicting. In the phase 3 trial in
advanced UCC patients, the beneficial effect of pembroli-
zumab over chemotherapy was observed irrespective of PD-
L1 expression, which is underscored by the results of
previous phase 2 trials with either atezolizumab or
nivolumab as second-line treatment in UCC . How-
ever, high PD-L1 expression, defined as a combined positive
score of 10%, was associated with shorter OS in both
chemotherapy- and pembrolizumab-treated UCC patients
. In advanced RCC, high PD-L1 expression ( 1%),
determined as the percentage of PD-L1–positive tumor
cells relative to the total number of tumor cells, was
also associated with an unfavorable outcome in both
nivolumab- and everolimus-treated patients. These findings
suggest that higher PD-L1 expression may be associated
with more aggressive tumor behavior and may be a
prognostic instead of a predictive marker.
Conflicting results on PD-L1 status in different studies
may be related to different targets of the administered
agents (PD-1 and PD-L1) and different methods to deter-
mine PD-L1 expression including differences in assays,
measurements, definition of PD-L1 expression (tumor cells,
tumor-infiltrating immune cells, or combined), semiquan-
titative analyses, and cutoffs. In addition, archival tissue
from primary tumors, often collected years prior to
metastatic disease, was mostly used to determine PD-L1
expression, whereas PD-L1 expression is a dynamic marker
E U R O P E A N U R O L O G Y 7 2 ( 2 0 1 7 ) 4 1 1 – 4 2 3 420