The impact of ERBB2 alterations in muscle-invasive bladder cancer after neoadjuvant chemotherapy

Abstract

Cisplatin-based neoadjuvant chemotherapy (NAC) has been a standard treatment for muscle-invasive bladder cancer (MIBC), although peri-operative management is evolving to include immune checkpoint inhibitor (ICI) and enfortumab vedotin (EV)-based regimens.^{1, 2} Despite these advances, access to EV and ICIs remains limited in much of the world, and cisplatin-based NAC continues to represent a standard option for a large proportion of patients globally. Identification of biomarkers associated with response to NAC therefore remains clinically important.

Both pathogenic ERBB2 mutations and gene amplification are well-recognized genomic alterations in MIBC.³ Prior studies have reported conflicting findings regarding the association of ERBB2 mutations and amplification with response to NAC in MIBC,^4-6 which may be due to heterogeneity in chemotherapy regimens, genomic sequencing methodologies and definitions of treatment response across these studies. These discordant results highlight the need for additional study to clarify the prognostic value of ERBB2 alterations in MIBC.

We previously reported that tumour mutational burden (TMB) ≥ 10 mutations per megabase (Mut/Mb) was associated with improvement in pathologic complete response rate (pCR), relapse-free survival (RFS) and overall survival (OS) in a retrospective cohort of 91 patients with MIBC treated with cisplatin-based NAC followed by radical cystectomy.⁷ In this cohort, all patients received NAC with either gemcitabine and cisplatin or standard or dose-dense methotrexate, vinblastine, doxorubicin and cisplatin (MVAC) across two campuses at our institution. Only four (4.4%) patients received adjuvant nivolumab, minimizing the potential for ICI-related confounding of survival outcomes in this cohort. Genomic profiling of pretreatment transurethral resection specimens was performed using the commercially available Tempus xT next-generation sequencing (NGS) platform. Associations between ERBB2 gene variants, TMB and clinical outcomes were assessed using logistic regression, Cox proportional hazards models and Kaplan–Meier techniques.

ERBB2 alterations were identified in 17/91 (18.7%) patients, including four pathogenic mutations (4.4%) and 13 amplifications (14.3%); all mutations were p.S310F. pCR was achieved in 29/91 (31.9%) patients overall. Among those with ERBB2 alterations, pCR occurred in 3/4 (75.0%) patients with a mutation and 6/13 (46.2%) patients with amplification. In aggregate, the presence of an ERBB2 alteration (mutation + amplification) was associated with improved likelihood of pCR compared to non-ERBB2 altered patients (OR 3.038, p = 0.048).

Given the favourable impact of ERBB2 alterations on pCR, we next assessed survival endpoints. Median follow-up for the entire cohort was 63.6 months; 38/91 (41.8%) relapsed. Five of 17 (29.4%) patients with an ERBB2 alteration relapsed. Median RFS was not reached in the ERBB2-altered group versus 54.7 months in the wild-type group (HR 0.50, p = 0.134; Figure 1A). Median OS was not reached versus 107.6 months, respectively (HR 0.381, p = 0.098; Figure 1B). Although we previously reported that TMB ≥ 10 Mut/Mb was associated with improved pCR, RFS and OS in this cohort, the presence of an ERBB2 alteration was not clearly associated with TMB ≥ 10 Mut/Mb (OR 2.217, p = 0.178).

To our knowledge, this is the first study to report the clinical impact of ERBB2 gene alterations detected by a widely available, contemporary NGS platform in a uniformly treated cohort of MIBC patients receiving cisplatin-based NAC. Although this study was not powered to detect definitive associations with survival endpoints, concordant trends towards improved RFS and OS were observed in patients with ERBB2 alterations. The lack of a clear association between ERBB2 alteration and high TMB further suggests that ERBB2-driven chemosensitivity may operate through mechanisms distinct from those associated with high TMB.

These findings should be interpreted in the context of several limitations, including modest sample size, retrospective design and a single-institution cohort. However, this cohort is among the largest reported to assess the prognostic value of ERBB2 alterations in NAC-treated MIBC, benefits from mature follow-up, uniform cisplatin-based NAC administration and a rigorous definition of pathologic response (ypT0ypN0). While applicability may be limited in regions where treatment paradigms have shifted to include peri-operative ICIs or EV-based regimens, these approaches remain inaccessible to a substantial proportion of patients globally, and cisplatin-based NAC is expected to remain standard of care in many settings. These findings also raise the question of whether ERBB2 alteration retains prognostic relevance in patients treated with contemporary peri-operative ICI and EV-based regimens, which warrants prospective investigation.

In summary, ERBB2 alterations detected by a routinely available NGS platform were associated with improved pCR and concordant trends towards improved RFS and OS in a uniformly treated cisplatin-based NAC cohort. These findings support the continued need to identify prognostic biomarkers in MIBC, particularly for the large proportion of patients globally who lack access to contemporary peri-operative regimens.

Conceptualization: EFB, JTS. Methodology: EFB, LCB, MM, ES, JTS. Data Analysis: EFB, LCB, MM, ES, JTS. Writing—original draft: EFB, JTS. Writing—review and editing: All authors. Funding acquisition: EFB.

M. Ciampricotti and E. Williams are employees of Tempus AI, Inc. The authors declare no other relevant conflict of interests.

The data that support the findings of this study are available from the corresponding author upon reasonable request.