EURURO Vol. 72 No. 3

367 476

classification, which is characterized by a stromal signature
and immune infiltration, was not found to harbor a
characteristic pattern of genomic alterations. The five Lund
subtypes, however, were characterized by specific (and
anticipated ) mutations and copy number variations. The
more definitive molecular correlation could be interpreted
as an indication that the Lund classification better reflects
true bladder cancer biology. Indeed, this raises the question
of whether genomic alterations could be used to derive a
consensus classification. In the original description, Choi
et al showed that the p53-like subtype has potential
clinical relevance in the context of response to chemother-
apy, also implying important biologic relevance. The
authors suggest that the lack of genomic correlation
indicates that nongenomic factors drive the differentiation
of this subtype from basal and luminal tumors.
Molecular subtyping and genomic sequencing have both
been used independently to address the question of
sensitivity to cisplatin-based chemotherapy in the neoad-
juvant setting. While Choi et al determined that p53-like
tumors were less likely to respond to chemotherapy, we
were unable to validate this finding [5_TD$DIFF]in a larger metadata set
. We observed that patients with basal tumors
experienced the greatest benefit from neoadjuvant chemo-
therapy. Other groups have focused on mutations in single
genes or gene panels. While all of these markers need
further validation before being suitable for clinical imple-
mentation, it is also essential to investigate RNA subtyping
and genomic alterations in the same patients in combina-
tion to determine if this can further augment prediction of
response to neoadjuvant chemotherapy. For example,
survival is poor in patients with claudin-low tumors
receiving neoadjuvant chemotherapy, but some patients
do respond, and it will be important to test whether these
responding patients have specific genomic alterations such
as a mutation in a DNA damage repair gene.
There is no question that RNA-based subtyping has
helped in deconvoluting some of the heterogeneity of
bladder cancer. The differential response to chemotherapy
within one molecular subtype, however, underlines that
there is still molecular, pathologic, and, most importantly,
clinical heterogeneity between tumors in a given subtype.
The integrative analysis of both the transcriptome and the
genome should provide additional insight into this hetero-
geneity.
The analysis by Choi et al advances our understanding
of the bladder cancer subtypes. An updated report from the
TCGA consortium of the entire TCGA patient cohort is
expected in the near future, and is likely to provide
additional insights. Further advances in the field will
include a more formal integrative analysis using computa-
tional tools to predict on the basis of RNA expression which
genomic alterations are driver alterations and which
pathways are therefore worth targeting therapeutically.
In the future, analyses will also have to include the
proteome, which adds an additional layer of complexity
but should reveal further important insights into bladder
cancer biology.
Conflicts of interest: The author has been an advisory board member for
AbbVie, Allergan, Astellas, Bayer, Biocancell, Biosyent, Ferring, Janssen,
Lilly, Merck, Roche, Sanofi, Sitka, and Spectrum; has been a speaker for
AbbVie, Biosyent, Ferring, and Janssen; has received honoraria from
Janssen, Bayer, and Sanofi; has received grant funding from New B
Innovation, iProgen, and GenomeDx; and has been involved in clinical
trials with Astellas, Ferring, Janssen, and Roche.
References
[1]
.
[2]
.
[3]
[4]
.
[5]
.
[6]
.
[7]
[8]
[9]
.
[10] Seiler R, Al Deen Ashab H, Erho N, et al. Impact of molecular
subtypes in muscle-invasive bladder cancer on predicting re-
sponse and survival after neoadjuvant chemotherapy. Eur Urol.
In press.
E U R O P E A N U R O L O G Y 7 2 ( 2 0 1 7 ) 3 6 6 – 3 6 7 367