Impact of HER2 low status on pathologic response after neoadjuvant chemotherapy in TNBC: A large scale retrospective cohort study.

Abstract

571 Background: HER2 low status, defined as HER2 1+ or HER2 2+ and nonamplified by in-situ hybridization, has demonstrated the ability to identify a population of patients with triple-negative disease who benefit from trastuzumab deruxtecan in the metastatic setting. The implications of HER2 low status in the early-stage setting is unclear. This study evaluated whether HER2 low status impacts rates of pathologic complete response in response to neoadjuvant chemotherapy (NAC). Methods: Using the national cancer database (NCDB), patients with clinically invasive non-metastatic breast cancer between 2010-2021 were retrospectively identified (n=1,926,979). Patients with unknown receptor status or who did not have triple negative breast cancer (n=1,755,897) and those who did not have surgery (n=9,849) were excluded. Clinicopathologic, treatment, and outcome variables were compared using chi-square and anova tests. Subgroup analysis was performed among patients who received NAC and had available pathologic response results. Multivariable binary logistic regression was performed to assess clinical variables associated with pCR after NAC. Cox proportional hazards model was performed to assess clinical variables associated with overall survival (OS) in TNBC receiving NAC. Results: Of the 161,233 individuals eligible for analysis, 79,268 (49.4%) had HER2 low disease. The proportion of HER2 0, HER2 1+ and HER2 2+/ish negative (HER2 2+) were 50.6, 34.8, and 14.6% respectively. Overall, 49,994 (31%) individuals received NAC with an overall pCR rate of 42.9%. The pCR rate was significantly lower in those with HER2 2+/ish negative compared to those with HER2 0 TNBC at 39.9 vs. 44.2%, p <0.001. On multivariable analysis, HER2 1+ status trended towards a lower likelihood of pCR R 0.95, 95% CI 0.91-1.00, p=0.06) while HER2 2+/ish negative status was significantly associated with a lower likelihood of pCR (OR 0.87, 95% CI 0.82-0.94, p<0.001) compared to HER2 0 disease. Cox proportional hazards model analysis demonstrated that the strongest clinical factor for worse OS was non-pCR with HR 3.76, 95% CI 3.48 – 4.05, p<0.001 while neither HER2 1+ or HER2 2+/ish negative was associated with worse OS. Conclusions: In this analysis, tumors that were HER2 2+/ish negative were associated with a significantly lower pCR rate after NAC compared with HER2 0 tumors. This is the first large-scale study to demonstrate that HER2 low status may be prognostically unfavorable in early-stage TNBC. Examination of novel neoadjuvant therapeutic approaches tailored based on HER2 status including trastuzumab deruxtecan may be warranted to improve pCR rates and outcomes in patients with HER2-low early-stage breast cancer. pCR rate stratified by HER2 low status. HER2 0 HER2 1+ HER2 2+ n 81605 56109 23519 Neoadjuvant 26458 32.4% 16841 30.0% 6695 28.5% pCR 11690 44.2% 7064 41.9% 2672 39.9%