Update on the phase II SUMMIT trial: Neratinib + fulvestrant for HER2-mutant, HR-positive, metastatic breast cancer

Abstract

Background: Somatic mutations in HER2 (ERBB2) are oncogenic and may confer resistance to endocrine therapy in hormone receptor-positive (HR+) metastatic breast cancer (MBC). Neratinib, an irreversible pan-HER tyrosine kinase inhibitor, has single-agent clinical activity in HER2mutant MBC and is synergistic with fulvestrant in preclinical studies. In SUMMIT (NCT01953926), a phase 2 basket trial of neratinib in HER2-mutant solid tumors, the HR+ MBC cohort had encouraging outcomes with neratinib plus fulvestrant (N+F) [Oct 2018 datacut; Smyth et al. SABCS 2018; #PD3- 06]. Updated results from this cohort will be reported, detailing the impact of prior therapies on clinical efficacy of N+F in HER2-mutant HR+ MBC. Methods: HR+HER2 non-amplified MBC patients with locally detected HER2 mutations received oral neratinib 240 mg once daily and intramuscular fulvestrant (labeled dose). Loperamide prophylaxis was mandatory during cycle 1. Efficacy endpoints were: objective response rate (ORR); clinical benefit rate (CBR); duration of response (DOR); and progression-free survival (PFS). Response was assessed by RECIST 1.1 and/or PET Response Criteria. Results: A total of 47 patients were enrolled between Jul 2013 and Jan 2019. Most (79%) patients had visceral disease and were heavily pretreated; the median number of prior metastatic therapies was 3 (range 1-11). ORR was 30%, CBR was 47% and median DOR in the 14 patients with a confirmed response was 9.2 months. Median PFS was 5.4 months. Twenty-five patients (53%) had received prior fulvestrant and 20 (43%) had received prior CDK4/6 inhibitor (CDK4/6i) therapy, with ORRs of 16% and 30%, respectively. Diarrhea was the most common adverse event (grade 3, 11%; median duration, 1.5 days) but did not lead to treatment discontinuation. Conclusions: N+F is clinically active in heavily pretreated HER2-mutant HR+ MBC patients, including in those who had received prior fulvestrant and CDK4/6i therapy, where some durable responses were noted. No new safety signals were identified; the incidence of diarrhea was similar to that seen with single-agent neratinib and was not dose limiting.