Advertisement

Genomic Differences Between “Primary” and “Secondary” Muscle Invasive Bladder Cancer: Implications for Neoadjuvant Chemotherapy

Login to Access Video or Poster Abstract: MP58-02
Sources of Funding: Supported by the Sidney Kimmel Center for Prostate and Urologic Cancers at Memorial Sloan Kettering Cancer Center, Pin Down Bladder Cancer, and the Michael A. and Zena Wiener Research and Therapeutics Program in Bladder Cancer.

Introduction

We recently reported that patients with secondary muscle-invasive bladder cancer (MIBC) have substantially worse outcomes with neoadjuvant chemotherapy compared to primary MIBC (Pietzak, et al. AUA 2016). We subsequently used next-generation sequencing to investigate genetic differences between primary and secondary MIBC specimens.

Methods

We examined MIBC specimens from the Cancer Genome Atlas (TCGA) (n=131), our institutional genomic research database (n=569), and a prospective clinical sequencing protocol (n=214) to identify 342 chemotherapy-naive urothelial MIBC specimens (270 primary and 72 secondary) that underwent whole-exome or targeted exon-capture sequencing. Primary and secondary MIBC specimens were compared for genomic alterations in 341 known cancer genes. Primary MIBC was defined as clinical stage ≥T2 on either initial or re-staging TUR on first bladder tumor diagnosis. Patients with a history of NMIBC (Tis, Ta, or T1 with uninvolved detrusor muscle in specimen) confirmed by a second cystoscopy prior to the eventual diagnosis of clinical stage ≥T2 were considered to have secondary MIBC.

Results

We compared 270 primary and 72 secondary MIBC specimens for differences in genomic alterations in 341 cancer-associated genes. We identified significantly different rates of ERCC2, APC, and FGFR3 alterations. FGFR3-activating mutations (S249C, Y373C) occurred more frequently in secondary MIBC specimens (18% [13/72] vs. 9% [24/270], p=0.03). APC mutations were only seen in primary MIBC specimens (5% [14/270] vs. 0% [0/72], p=0.047). Surprisingly, ERCC2 missense mutations, which are associated with extreme sensitivity to cisplatin chemotherapy, only occurred in primary MIBC specimens (12% [32/270] vs. 0% [0/72], p<0.001). After adjusting for multiple comparisons using the Benjamini-Hochberg false discovery rate method, only ERCC2 mutations remained significant (adjusted p =0.016).

Conclusions

ERCC2 is involved in repair of cisplatin-induced DNA damage, and mutations in ERCC2 are associated with clinical responses to cisplatin. ERCC2 mutations occurred exclusively in primary MIBC patients and may account for their improved outcomes with neoadjuvant chemotherapy compared to secondary MIBC patients.

Funding

Supported by the Sidney Kimmel Center for Prostate and Urologic Cancers at Memorial Sloan Kettering Cancer Center, Pin Down Bladder Cancer, and the Michael A. and Zena Wiener Research and Therapeutics Program in Bladder Cancer.

Authors
Eugene Pietzak
Aditya Bagrodia
Hikmat Al-Ahmadie
Harry Herr
Emily Zabor
David Barron
Qiang Li
Francois Audenet
Samuel Funt
Ahmet Zehir
Maria Arcila
Priscilla Baez
Michael Berger
Nikolaus Schultz
David Solit
Dean Bajorin
Jonathan Rosenberg
Eugene Cha
Bernard Bochner
Gopa Iyer
back to top