EURURO Vol. 72 No. 3

356 476

and more recent studies are characterizing molecular
subtypes that cut across cancer types [14_TD$DIFF] . The potential
clinical significance of the molecular subtypes of these
cancers is being investigated.
3.3. DNA alterations in bladder cancers
The recently completed TCGAMIBC (BLCA) project produced
a comprehensive, open-access catalog of DNA alterations in a
cohort of over 400 MIBCs. The first TCGA ‘‘bladder cancer
study’’ reported the results of a comprehensive integrated
genomic analysis of131 tumors [15_TD$DIFF] , and a recent review
article provided an update on 238 tumors [5_TD$DIFF] . These initial
results were also incorporated into a pan-cancer analysis
that compared the genomic features of 12 different types of
cancers [14_TD$DIFF] . Furthermore, a thorough review of the different
genomic alterations that characterize low-grade papillary
tumors (Ta) and MIBCs was also published recently
Finally, a recent paper reported the comprehensive
transcriptional analysis of a cohort of 460 patients with
NMIBC [16_TD$DIFF] . Therefore, excellent, comprehensive summaries
of the major genomic alterations in the complete spectrum
of bladder cancers can be found elsewhere. The key findings
will now be summarized briefly.
3.3.1. Major drivers of mutagenesis
Cigarette smoking is an established risk factor for bladder
cancer , and chronic exposure to cigarette smoke-like
nitrosamines (ie, BBN) causes bladder cancer in rodents
[17_TD$DIFF] Aromatic compounds in cigarette smoke produce DNA
damage, so it was expected that a history of cigarette
smoking would be associated with specific tobacco-related
DNA mutations in TCGA exome sequencing data. Analyses
of the initial TCGA cohort of 131 tumors failed to identify
such signatures, although tumors from smokers were
enriched with specific DNA methylation patterns
[15_TD$DIFF] . Interestingly, a significant number of bladder cancers
contained mutations in NFE2L2 (NRF2) and TXNIP [5_TD$DIFF]
genes, which encode proteins that inhibit the damaging
effects of the reactive oxygen species that are produced in
response to cigarette smoke carcinogens. Although there
was no obvious relationship between the NFE2L2 and/or
TXNIP alterations and smoking status in bladder cancers,
mutations in these genes were enriched in lung and head
and neck cancers from smokers [18_TD$DIFF] , suggesting poten-
tial causal roles in carcinogenesis and/or tumor progression.
A more recent reanalysis of the original TCGA cohort
identified a novel DNA mutational signature associated
with inactivating mutations in the gene encoding the
nucleotide excision repair protein, ERCC2, and established
that these signatures were enriched in tumors from
smokers [19_TD$DIFF] . Importantly, the signature was much more
strongly associated with ERCC2 inactivation than it was
with smoking, suggesting that the former was the driving
force underlying the signature [19_TD$DIFF] .
The APOBEC family of antiviral enzymes promotes
cytosine deamination and mutagenesis of single-stranded
DNA and mRNAs. Among the APOBEC genes, APOBEC3B
appears to be most commonly overexpressed in solid
tumors, and bladder cancers stand out for expressing some
of the highest levels of APOBEC3B among all solid
malignancies [20_TD$DIFF] Aside from being upregulated by infec-
tion, APOBEC3B activity can also be increased by chemical
carcinogens, which promote APOBEC3B-mediated mutagen-
esis by inducing the formation of the single-strand DNA
intermediates that are formed during DNA damage and
repair [20_TD$DIFF] Analyses of mutational patterns have revealed
that a large proportion of the total mutational burden in
bladder cancer is attributable to APOBEC3B-mediated
mutagenesis [21_TD$DIFF] and the prevalence of APOBEC3B-
associated mutations increased with subclonal evolution in
lung cancers [22_TD$DIFF] . Furthermore, it was recently found that
an APOBEC mutation signature was significantly enriched in
high-risk NMIBCs [16_TD$DIFF] Taken together, accumulating data
suggest that APOBEC-mediated mutations may play a
central causative role in driving bladder cancer genomic
heterogeneity and disease progression.
3.3.2. Major targets of DNA alterations
Histone modifications play central roles in the regulation of
gene expression, and whole exome sequencing studies
revealed that mutations in chromatin-modifying enzymes
were extremely common in bladder cancers [23_TD$DIFF] Among
them, inactivating mutations in the histone H3 lysine 27
(H3K27) demethylase KDM6A (also known as UTX) were
most common and enriched in NMIBCs (32–43%) [24_TD$DIFF] ,
whereas inactivating mutations in the SET family histone
H3 lysine 4 (H3K4) methyltransferase MLL2 were more
common in MIBCs (19%) [5_TD$DIFF] , and mutations in KDM6A and
MLL2 were mutually exclusive [5_TD$DIFF] . Although the biological
consequences of these events have not been defined
experimentally, they would be expected to lead to
decreased RNA polymerase accessibility, gene silencing,
and a less well-differentiated phenotype. In-depth chroma-
tin immunoprecipitation/sequencing (ChIP-seq) studies are
required to directly address this hypothesis.
As introduced above, one of the most striking differences
between NMIBCs and MIBCs is the relative frequency of TP53
gene inactivation and relative levels of genomic instability.
Overall, mutations in TP53 were observed in about 50% of
MIBCs but were less common in NMIBCs (20% of tumors)
[25_TD$DIFF] [26_TD$DIFF] Interestingly, 85% of high-grade NMIBCs (T1G3)
contained p53 pathway alterations [27_TD$DIFF] . Furthermore, am-
plification of TP53’s inhibitor, MDM2, occurred in approxi-
mately 9% of MIBCs [28_TD$DIFF] , indicating that TP53 inactivation
occurred in the majority of muscle-invasive tumors. RB1
inactivation was also much more common in MIBCs as
compared with that in NMIBCs [5_TD$DIFF] and mutations in RB1
tended to be associated with mutations in TP53
[5_TD$DIFF] Interestingly, the same patterns were not observed
with RB1’s upstream inhibitor (CDKN2A), which was deleted
in approximately equal numbers of NMIBCs andMIBCs (50%)
Dysregulation of other genes that promote cell cycle
progression was also common in bladder cancers. Amplifi-
cation of cyclin D1 was reported in approximately 20% of
NMIBCs and MIBCs , and amplification of E2F3 was
observed in high-grade T1 lesions and MIBCs [29_TD$DIFF] An early
study reported that MYC amplification was associated with
E U R O P E A N U R O L O G Y 7 2 ( 2 0 1 7 ) 3 5 4 – 3 6 5 356