EURURO Vol. 72 No. 6

953 1030

1. Introduction
Of the estimated 429 000 people to be diagnosed with
bladder cancer in the industrialized world each year,
70–80% will have nonmuscle invasive bladder cancer
(NMIBC) Half of all NMIBC patients will experience
tumor recurrence within 5 yr, and 20–30% will progress to
secondary MIBC Ultimately, as many as 10–15%
of patients presenting with NMIBC will die of bladder
cancer
Previous investigations into NMIBC genetics have been
limited by their inability to comprehensively profile tumors
for multiple cancer-associated genes . More recently,
MIBC were comprehensively investigated by The Cancer
Genome Atlas (TCGA) and other groups using next-
generation sequencing (NGS), leading to the identification
of potential biomarkers and targets for therapeutic inter-
vention However, very few NMIBC tumors have been
examined with NGS methods to date, and these investiga-
tions have been limited by a lack of clinical annotation, the
absence of restaging transurethral resection (TUR) to ensure
appropriate tumor staging, or a failure to differentiate
between primary and recurrent tumors .
In this study, we examined primary treatment-naive
index tumors from a cohort of patients with NMIBC using a
massively parallel, targeted, exon capture-based NGS
platform to define the prevalence of genetic alterations
and their potential clinical implications.
2. Patients and methods
2.1. Patients and samples
Targeted NGS with a 341 or updated 410 cancer-associated gene panel
was performed on formalin-fixed paraffin embedded sections of
treatment-naive index tumors along with matched germline DNA for
105 patients with NMIBC as part of an Institutional Review Board-
approved protocol (Supplementary Fig. 1, Supplementary Table 1,
Supplementary data) . A board-certified genitourinary pathologist
reviewed representative hematoxylin and eosin slides to confirm grade,
stage, and urothelial histology. All patients underwent evaluation and
TUR by a urologic oncologist at Memorial Sloan Kettering Cancer Center
(MSK). All tumors profiled were newly diagnosed and untreated. Patients
who received perioperative mitomycin or other adjuvant perioperative
therapies were excluded. All high-grade T1 (HGT1) tumors had restaging
TUR and confirmation of uninvolved detrusor muscle. Treatment and
management was at the discretion of the treating urologic oncologist.
Patients managed with TUR were followed at MSK with cystoscopy and
urine cytology every 3 mo for the 1st yr, then every 3–6 mo. Recurrence
was defined as histological proven cancer on biopsy or TUR. See
Supplementary data for additional details.
For comparison purposes, we also evaluated the frequency of
genomic alterations seen in 40 pretreatment index tumors in patients
with primary MIBC who were sequenced with MSK-Integrated Mutation
Profiling of Actionable Cancer Targets (MSK-IMPACT) on the same
Institutional Review Board-approved protocol and 98 MIBC specimens
from patients in the TCGA study who were reported to have no prior
history of NMIBC (Supplementary data).
2.2. Statistical analysis
Alterations in oncogenes were deemed significant if they were recurrent
or known functional missense mutations or amplifications (Supplemen-
tary data). Alterations in tumor suppressor and DNA damage repair
genes were deemed significant if truncating mutations (nonsense,
frameshift indels), recurrent missense mutations, or homozygous
deletions were present (Supplementary data). Fisher’s exact tests were
used to analyze categorical associations. Kruskal-Wallis and Wilcoxon
tests were used for continuous variables. Cox regression modeling was
used to determine the association between genomic alterations and
recurrence after bacillus Calmette-Gue´rin (BCG). The Kaplan-Meier
method and log-rank test were used for estimations of recurrence free
survival. A p value of <0.05 was considered statistically significant. All
analyses were conducted using R v.3.3.1.
)
3. Results
3.1. Patient demographics and treatment
To characterize the genomic landscape of NMIBC, we
analyzed 105 tumors across the disease spectrum compris-
ing low-grade Ta (LGTa; n = 23), high-grade Tis (HGTis;
n = 12), high-grade Ta (HGTa; n = 32), and HGT1 (n = 38) for
alterations in 341 cancer-associated genes. Information on
patient demographics and treatments are listed in ,
Supplementary Table 2, and Supplementary Table 3. The
median follow-up for the NMIBC cohort managed by TUR
with or without adjuvant intravesical therapy (n = 100) was
24.4 mo, with recurrences occurring in 46 patients.
Treatment following resection for the 23 low-grade tumors
included surveillance only (48%), intravesical mitomycin
(39%), and intravesical BCG (13%). The 82 high-grade
tumors were treated with intravesical BCG (81%), observa-
tion (12%), or immediate radical cystectomy in five patients
(4 = pT1N0, 1 = pTaN0). Additional details on the NMIBC
cohort and the MIBC samples used for comparison purposes
are available in the Supplementary data, Supplementary
Table 2, and Supplementary Table 3.
associated with increased mutational load, which may have therapeutic implications for
BCG immunotherapy and ongoing trials of systemic checkpoint inhibitors. ARID1A muta-
tions were associated with an increased risk of recurrence after BCG therapy. Whether
ARID1A mutations represent a predictive biomarker of BCG response or are prognostic in
NMIBC patients warrants further investigation.
Patient summary: Analysis of frequently mutated genes in superficial bladder cancer
suggests potential targets for personalized treatment and predictors of treatment response,
and also may help develop noninvasive tumor detection tests.
# 2017 European Association of Urology. Published by Elsevier B.V. All rights reserved.
E U R O P E A N U R O L O G Y 7 2 ( 2 0 1 7 ) 9 5 2 – 9 5 9 953