Concordance of HER2+ status by IHC/ISH and ERBB2 status by NGS in a real-world clinicogenomic database and analysis of outcomes in patients (pts) with metastatic breast cancer (mBC)

• Published in: Journal of clinical oncology Vol. 39; no. 15_suppl; p. 1036
• Main Authors: Cho-Phan, Cheryl D., Snider, Jeremy, Zhang, Liangliang, McGregor, Kimberly, Schrock, Alexa Betzig, Castellanos, Emily
• Format: Journal Article
• Language: English
• Published: 20-05-2021
• ISSN: 0732-183X; 1527-7755
• DOI: 10.1200/JCO.2021.39.15_suppl.1036

Abstract only 1036 Background: HER2 overexpression/amplification measured by IHC or ISH is a predictive biomarker for HER2-targeted therapies. Next-generation sequencing (NGS) can identify ERBB2 amplification (amp) and mutations. We examined clinical characteristics, NGS testing patterns, and outcomes of pts treated with 1L HER2 therapy with HER2+ mBC based on ERBB2 amp status using a real-world (RW) clinico-genomic database (CGDB). Methods: Pts with mBC (HER2+ by IHC and/or ISH) treated with 1L HER2 therapy who had undergone NGS and were treated within the Flatiron Health (FH) network were eligible. Clinical characteristics and HER2 testing results were obtained via technology-enabled abstraction of clinician notes and radiology/pathology reports and linked to genomic data from Foundation Medicine (FMI) in the nationwide (US-based), de-identified FH-FMI CGDB. Demographic, clinical and genomic characteristics were summarized and stratified by concordance between HER2+ (IHC 3+ or ISH amp+) and ERBB2amp+ status [copy number (CN) ≥ 5]. NGS testing patterns and 1L HER2 therapy were characterized and stratified by concordance status. Concordance was assessed based on contemporaneous timing of paired test specimen collection dates (FMI NGS ≤ 30 days of HER2+ status). RW overall survival (rwOS) stratified by HER2+/ ERBB2 amp concordance was estimated with Kaplan-Meier analysis and adjusted Cox proportional hazards models. Results: Among 268 eligible pts, HER2+/ ERBB2amp+ concordance was 66% (176/268); concordance among contemporaneous paired specimens was 73% (106/145). Demographic and clinical features were overall well-balanced with most pts treated at community sites [94%, (252/268)]; the discordant (HER2+/ ERBB2amp-) group (95/268) had more pts with hormone receptor positive disease (73% vs 62%). Concordance by assay type varied; IHC+ only, IHC+/ISH+, and ISH+ only agreement was 72% (95/132), 76% (26/34) and 52% (50/96), respectively. A higher proportion of discordance (35% vs 19%) was seen in pts treated at community vs. academic sites. Median rwOS was 32.9 months (IQR 25.9-38.9) among concordant (HER2+/ ERBB2amp+) and 15.5 months (IQR 8.9-30.1) among discordant (HER2+/ ERBB2amp-) pts, aHR = 0.71 [95% CI: 0.48-1.03; p = 0.073]. Conclusions: Among RW pts with HER2+ mBC receiving 1L HER2 therapy, discordance between ERBB2amp and IHC/ISH HER2 testing methods was observed. Pts with tumors HER2+ by IHC and/or ISH but negative for ERBB2amp had a trend towards worse rwOS following receipt of HER2 therapy compared to concordant cases. Contemporaneous timing of specimen collection was associated with greater concordance. Future analyses on the additive value of ERBB2 CN as a predictive marker, and assessing factors that may affect discordance such as intratumor HER2 heterogeneity, tumor content, and biopsy site are warranted.