Genomic Differences Between "Primary" and "Secondary" Muscle-invasive Bladder Cancer as a Basis for Disparate Outcomes to Cisplatin-based Neoadjuvant Chemotherapy

Abstract

Background: Cisplatin-based neoadjuvant chemotherapy (NAC) followed by radical cystectomy (RC) is the standard of care for patients with muscle-invasive bladder cancer (MIBC). It is unknown whether this treatment strategy is appropriate for patients who progress to MIBC after treatment for prior noninvasive disease (secondary MIBC).

Objective: To determine whether clinical and genomic differences exist between primary and secondary MIBC treated with NAC and RC.

Design, setting, and participants: Clinicopathologic outcomes were compared between 245 patients with clinical T2-4aN0M0-stage primary MIBC and 43 with secondary MIBC treated with NAC and RC at Memorial Sloan Kettering Cancer Center (MSKCC) from 2001 to 2015. Genomic differences were assessed in a retrospective cohort of 385 prechemotherapy specimens sequenced by whole-exome or targeted exon capture by the Cancer Genome Atlas or at MSKCC. Findings were confirmed in an independent validation cohort of 94 MIBC patients undergoing prospective targeted exon sequencing at MSKCC.

Outcome measurements and statistical analysis: Pathologic response rates, recurrence-free survival (RFS), bladder cancer-specific survival (CSS), and overall survival (OS) were measured. Differences in somatic genomic alteration rates were compared using Fisher's exact test and the Benjamini-Hochberg false discovery rate method.

Results and limitations: Patients with secondary MIBC had lower pathologic response rates following NAC than those with primary MIBC (univariable: 26% vs 45%, multivariable: odds ratio=0.4 [95% confidence interval=0.18-0.84] p=0.02) and significantly worse RFS, CSS, and OS. Patients with secondary MIBC treated with NAC had worse CSS compared with cystectomy alone (p=0.002). In a separate genomic analysis, we detected significantly more likely deleterious somatic ERCC2 missense mutations in primary MIBC tumors in both the discovery (10.9% [36/330] vs 1.8% [1/55], p=0.04) and the validation (15.7% [12/70] vs 0% [0/24], p=0.03) cohort.

Conclusions: Patients with secondary MIBC treated with NAC had worse clinical outcomes than similarly treated patients with primary MIBC. ERCC2 mutations predicted to result in increased cisplatin sensitivity were enriched in primary versus secondary MIBC. Prospective validation is still needed, but given the lack of clinical benefit with cisplatin-based NAC in patients with secondary MIBC, upfront RC or enrollment in clinical trials should be considered.

Patient summary: A retrospective cohort study of patients with "primary" and "secondary" muscle-invasive bladder cancer (MIBC) treated with chemotherapy before surgical removal of the bladder identified lower response rates and shorter survival in patients with secondary MIBC. Tumor genetic sequencing of separate discovery and validation cohorts revealed that chemotherapy-sensitizing DNA damage repair gene mutations occur predominantly in primary MIBC tumors and may underlie the greater sensitivity of primary MIBC to chemotherapy. Prospective validation is still needed, but patients with secondary MIBC may derive greater benefit from upfront surgery or enrollment in clinical trials rather than from standard chemotherapy.

Keywords: Bladder cancer; ERCC2; Muscle-invasive bladder cancer; Neoadjuvant chemotherapy; Non–muscle-invasive bladder cancer; Radical cystectomy.

Fig. 1 –

Recurrence-free survival (RFS) and overall survival (OS) in patients with primary and secondary MIBC. (A) Comparison of RFS in patients with primary versus secondary MIBC treated with neoadjuvant chemotherapy followed by radical cystectomy. (B) Comparison of OS in patients with primary versus secondary MIBC treated with neoadjuvant chemotherapy followed by radical cystectomy. MIBC = muscle-invasive bladder cancer.

Fig. 2 –

Genomic analyses of MIBC tumors. (A) Alteration frequency of genes previously associated with chemotherapy response in bladder cancer in primary versus secondary MIBC tumors in the initial genomic analysis. (B) Lollipop plot of ERCC2 missense mutations in primary MIBC in the combined genomic cohort. (C) Oncoprint comparing the frequency of mutations in ERCC2 and the 10 most commonly altered genes between primary and secondary MIBC patients in the prospective validation cohort. Genomic alteration frequencies include novel missense mutations of unknown significance. (D) Volcano plot of the association between mutations and secondary versus primary MIBC in the combined genomic cohort, shown as –log10 p value versus log odds ratio. Results are from exact logistic regression. Bubble size is proportional to the total number of alterations. The horizontal dotted line indicates an unadjusted p value of <0.05. MIBC = muscleinvasive bladder cancer; NAC = neoadjuvant chemotherapy.