EURURO Vol. 72 No. 6

958 1030

likely occur early in urothelial carcinogenesis, but it also
supports the development of noninvasive methods to detect
these alterations in urine as a potential screening and/or
surveillance biomarker .
Another issue in NMIBC has been the lack of progress in
treatment outcomes and therapeutic options. For over 40 yr,
BCG has been the most effective intravesical therapy for
high-risk disease, yet pretreatment biomarkers that reliably
predict which patients will benefit from BCG treatment are
still needed . We found that patients whose tumors
harbored an ARID1A mutation had significantly worse
recurrence-free survival after an induction course of BCG.
ARID1A mutations are associated with a poor prognosis in
several cancers and have been shown to be associated with
high-grade bladder cancer , but to our knowledge no
prior study has examined the association between ARID1A
mutations and BCG outcomes. Further work is needed to
clarify whether ARID1A mutations are a predictive bio-
marker for BCG therapy or whether they identify a patient
cohort with an overall worse prognosis. However, if such an
association is confirmed in an independent cohort, not only
could ARID1A mutations serve as a potential biomarker for
BCG response, but drugs that can reverse the epigenetic
consequences of ARID1A inactivation, such as inhibitors of
the EZH2 methyltransferase, may have therapeutic benefit
for such patients
Additionally, our study confirmed the presence of
multiple other potentially actionable targets in NMIBC.
FGFR3 inhibitors are currently being investigated in patients
with persistent or recurrent NMIBC following BCG treatment,
but the utility of these targeted agents has been limited in
part by the systemic toxicity of currently available agents.
With the development of highly selective and less toxic
kinase inhibitors, it would be reasonable to test such agents
as alternatives or adjuncts to BCG in patients whose tumors
harbor anFGFR3 or ERBB2 alteration. Advances in intravesical
delivery systems may also allow targeted agents to be
administered with limited systemic toxicity in the future.
Another important finding fromour investigationwas the
high frequency of DDR gene alterations in high-grade NMIBC,
with ERCC2 missensemutations being themost common. For
several decades, a two-pathway model of low-grade papil-
lary and high-grade invasive bladder cancer has been
postulated, yet there is marked heterogenicity in molecular
profiles and clinical outcomes We found DDR gene
alterations and ERCC2 mutations in particular to be
associated with a larger mutational burden that might
underlie the so-called genomically unstable pathway in
bladder cancer development Impaired DNA repair may
allow the permissive accumulation of multiple molecular
alterations resulting in growth advantage and invasive
capabilities Additionally, the high mutational burden
in NMIBC might have important implications for the use of
systemic checkpoint inhibitor immunotherapies. These
agents have recently demonstrated significant activity in
MIBC and are now being tested in NMIBC patients .
Several studies have found an association between higher
mutational burden and predicted neoantigen burden with
antiprogrammed cell death-1/programmed death-ligand
1 response in patients with metastatic solid tumors,
including bladder cancer . The role that mutational
burden plays in dictating BCG response also warrants further
investigation. While we were unable to find a statistically
significant association on MSK-IMPACT between mutational
burden and BCG response, this relationship will need to be
further explored in a larger cohort and with whole exome
sequencing. The exact mechanism of action for BCG
immunotherapy has remained elusive, so leveraging the
knowledge recently gained from correlative studies of
systemic checkpoint inhibitors may prove fruitful .
There are important limitations to this study. While this
is the largest cohort of NMIBC tumors analyzed to date using
NGS methods, when parsed out by grade and stage the
absolute numbers in each subgroup were relatively small
and longer follow-up will be needed. To facilitate future
larger-scale analyses, we have made all genomic and
clinical data from this study publically available through
the cBioPortal for Cancer Genomics . Furthermore,
intertumor and intratumor heterogenicity may impact our
results. We also found tissue-based sequencing of carcino-
ma in situ specimens to be particularly challenging, with
nearly two-thirds of samples demonstrating inadequate
tumor purity for analysis. To overcome these limitations,
alternative approaches are currently being explored,
including sequencing of cytology specimens, urinary cell-
free DNA, and urinary exosomes. Future advances and
refinements in single-cell sequencing may also overcome
current limitations.
5. Conclusions
NGS of treatment-naive index tumors from patients with
NIMBC identified that the majority of tumors had at least
one potentially actionable alteration that could serve as a
drug target in clinical trials of novel intravesical or systemic
therapy. High rates of DDR gene alterations were identified
in high-grade NMIBC tumors, which might have implica-
tions for BCG immunotherapy and systemic checkpoint
inhibition in NMIBC patients. ARID1A mutations may be
associated with earlier recurrence after BCG and may be a
potential therapeutic target.
Author contributions: Bernard H. Bochner had full access to all the data in
the study and takes responsibility for the integrity of the data and the
accuracy of the data analysis.
Study concept and design: Pietzak, Bagrodia, Cha, Solit, Bochne.
Acquisition of data: Pietzak, Bagrodia, Cha, Li, Baez, Rosenberg, Bajorin,
Dalbagni, Al-Ahmadie, Bochner.
Analysis and interpretation of data: Pietzak, Cha, Iyer, Isharwal, Berger,
Zehir, Schultz, Rosenberg, Bajorin, Dalbagni, Al-Ahmadie, Solit, Bochner.
Drafting of the manuscript: Pietzak, Iyer, Solit, Bochner.
Critical revision of the manuscript for important intellectual content:
Pietzak, Bagrodia, Drill, Iyer, Isharwal, Berger, Zehir, Schultz, Dalbagni,
Al-Ahmadie, Solit, Bochner.
Statistical analysis: Drill, Ostrovnaya.
Obtaining funding: Bochner, Solit.
Administrative, technical, or material support: Baez.
Supervision: Ostrovnaya, Solit, Bochner.
Other: None.
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 9 958