EURURO Vol. 71 No. 5

841 844

number of 36 patients (justified on the basis of frequentist
criteria). Our results were PP = 0 and PP = 1 for PD-L1–
negative and PD-L1–positive patients, respectively. These
figures clearly indicate that no further trial expansion or
enrichment is needed and that treatment efficacy is restricted
to biomarker-positive patients. This is not necessarily true,
however, in other scenarios. For instance, by shifting the
efficacy criterion to ORR > 30% (chosen to represent a 10%
minimum improvement over the 20% baseline rate) and
requiring PP > 0.95 as the stopping criterion, enrichment
with eight patients for the PD-L1–positive stratum would be
required. In addition, other results for different compounds
did not show clear-cut differences between patient sub-
groups .
To conclude, we aimed to provide investigators with
easy-to-use statistical tools for the design and analysis of
biomarker-based, early-phase, noncomparative trials. Sce-
narios that are not clear-cut might arise in future trials on
immune checkpoint inhibitors for which application of
formal yet manageable and efficient methods might better
support the therapeutic development process.
Conflicts of interest: Andrea Necchi has received consultancy and
advisory fees from Roche, Merck Sharp & Dohme, Astra Zeneca, Bayer,
and Janssen, and has received institutional research funding from
Millennium Takeda, Amgen, Merck Sharp & Dohme, Astra Zeneca,
GlaxoSmithKline. The remaining authors have nothing to disclose.
References
[1]
.
[2]
[3]
.
[4]
.
Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy
*Corresponding author. Department of Medical Oncology, Fondazione
IRCCS Istituto Nazionale dei Tumori, Via G. Venezian 1, 20133 Milan,
Italy. Tel. +39 02 23902402; Fax. +39 02 23903150.
E-mail address: (A. Necchi).
September 28, 2016
Mutagenic Factors and Complex Clonal [2_TD$DIFF]Relationship [3_TD$DIFF]of
Multifocal Urothelial Cell Carcinoma
Yiqing Du y, Ruoyan Li b, y, Zhanghua Chen b,y, Xiaofeng Wang Tao Xu Fan Bai b
Synchronous and metachronous development of multifocal
tumors in the urinary tract is a well-known characteristic of
urothelial cell carcinoma (UCC). Several theories have been
proposed to interpret the formation of these genetically
related and unrelated tumors . However, the underly-
ing mechanisms leading to multifocal UCC are unclear. To
shed light on this question, we [5_TD$DIFF]performed exome sequenc-
ing of 13 tumor specimens (including 4 bladder, 6 ureter,
and 3 renal pelvis tumors) from five patients with
multifocal UCC (Fig. 1A, Supplementary Fig. 1, and
Supplementary Table 1).
On average, we obtained 77 exonic coverage (Supple-
mentary Table 2) and identified 2168 (421–4170) somatic
mutations in each tumor (Supplementary Table 1 and
Supplementary Fig. 2). The hypermutational feature may be
related to certain mutagenic factors in the carcinogenesis of
multifocal UCC. To uncover potential mutagenic factors, we
examined the mutational signature of each tumor. Striking-
ly, in all the tumors we found predominant T ! A
transversion, with significant transcriptional strand bias.
The proportion of this transversion was markedly high in the
50 -CpTpG-30 motif (Fig. 1B, Supplementary Figs. 3 and 4).
This remarkable mutational signature was consistent with
the mutagenic features of aristolochic acid (AA), a natural
herb-extracted compound that exists widely in Chinese
traditional medicine [5]. [6_TD$DIFF]The typical AAmutational signature
was also found in a morphologically normal urothelium
specimen (P5NT) surrounding a tumor (Supplementary Figs.
3 and 4). Taken together, although the current study cohort
is relatively small, these findings imply that AA is a potential
mutagenic factor leading to multifocal UCC.
Intriguingly, the mutational signature of sample P5U2
was slightly different to that in a typical AA-affected
tumor such as sample P5U1 (Fig. 1C). By dissecting
mutations according to their clonal fractions (Supple-
mentary Fig. 5), we found a latent mutational signature
other than the AA signature. This C ! T predominant
mutational signature was largely consistent with the
signature of defective DNA mismatch repair ( Fig. 1C)
[6]. Coincidently, we found a frame shift deletion in the
E U R O P E A N U R O L O G Y 7 1 ( 2 0 1 7 ) 8 3 7 – 8 4 3 841