Introduction:

The ERBB2 gene encodes the Human Epidermal Growth Factor 2 (HER2) receptor tyrosine kinase. ERBB2 alterations have been shown to play a role tumorigenesis and impact prognosis in numerous malignancies. Antibody-drug conjugates targeting HER2 are currently under investigation for advanced bladder cancer and may have a roll in earlier stage disease. The evidence regarding the impact of ERB22 alterations on non-muscle invasive bladder cancer (NMIBC) is conflicting. We sought to better characterize the genomic landscape associated with ERBB2 alterations in NMIBC cohort and determine the potential impacts on prognosis.  

Methods:

An institutional database of patients with NMIBC who underwent targeted exome sequencing with a 505 gene panel (n = 503) was queried for ERBB2 alterations. Samples were excluded if they were from patients with multiple specimens and displayed discordance in ERBB2 alteration status (n = 2), or were listed as being obtained from metastatic sites (n = 20).  Demographic and clinicopathologic features were evaluated, Chi-Squared analysis was used to compare differences in co-alteration frequency and clinicopathologic features. Survival analysis was performed using the Kaplan-Meyer method and compared using the Log-Rank test. The Benjamini-Hochberg procedure was used to estimate the false discovery rate (q), significant p-values were reported when q<0.05 

Results:

A total of 493 patients were included. ERBB2 alterations were seen in n = 80 patients (16.3%), the most common alterations were missense mutations (11.2%) followed by amplifications (3.65%). 67.4% of missense mutations were in putative drivers. ERBB2 altered tumors were enriched in 47 distinct co-alterations including ERCC2 (38.8% vs 10.8%, p<10-8), CDK12 ( 27.1% vs 4.6%, p<10-8), E2F3 (20.0% vs 5.3% p< 0.0001), FAT1 (28.2% vs 10.3%, p< 0.0001) MSH6 (10.6% vs 1.32%, p<0.0001), and PMS1 (8.2% vs 0.6%, p<0.001). ERBB2 and FGFR3 alterations were mutually exclusive (p<10-10).  While ERBB2 altered tumors appeared enriched for high-risk pathologic features (High grade 96.3% vs 87.1%, pT1 55.6% vs 39.7%, CIS 40.7% vs 34.15%) these differences were not significant after multiple comparison correction. ERBB2 altered tumors showed favorable recurrence free survival (RFS), with median RFS not reached vs RFS of 26 months (95% CI, 22 – 37) for unaltered tumors (p<0.05) 

Conclusion:

ERBB2 alterations are common in NMIBC and are associated with a distinct pattern of co-alterations. Notably, FGFR3 and ERBB2 alterations appear to occur in a mutually exclusive fashion, suggesting that pursuit of distinct targeted therapies for these groups may be necessary. Although the literature regarding the impact of ERBB2 alterations on prognosis in NMIBC is mixed, in the cohort presented here, alterations were associated with an improved RFS relative to the unaltered genotype.

Funding: N/A