ESR1 mutational landscape and impact of co-existing resistance variants on clinical outcomes in patients with metastatic breast cancer.

Abstract

1074 Background: Mutations in estrogen receptor 1 ( ESR1) confer resistance to aromatase inhibitors but may retain sensitivity to selective estrogen receptor degraders (SERD). Recently, elacestrant, an oral SERD, was approved for patients with HR+/HER2- ESR1mut metastatic breast cancer (MBC). In this study, we evaluated the genomic landscape of ESR1 alterations from a genomic database and also evaluated the impact of resistance-associated alterations on clinical outcomes. Methods: A large de-identified database of mutations detected in plasma cell-free DNA (cfDNA) from patients with HR+/HER2- MBC was assessed for ESR1 mutations and resistance-associated co-alterations. We used the Guardant 360 assay, a 70- to 74-gene targeted next generation sequencing panel and applied a breast cancer subtype classifier based on detected mutations (Bardia A, SABCS 2020). In addition, clinical outcomes were evaluated from a clinically annotated institutional dataset, linking genomic alterations with progression-free survival (PFS) and overall survival (OS) to single-agent SERD therapy for patients with MBC. Results: Among 11,456 patients with MBC and detectable cfDNA, 4,694 were classified as likely HR+/HER2-, of which 2,708 (58%) had ESR1 missense mutations, the majority of which occurred in the ligand binding domain (n = 2,690, 99%). Among the latter, alterations included 204 unique mutations, of which 55 were variants of uncertain significance (VUS). Multiple ESR1 mutations were detected in 979 (36%) patients. Among 567 patients classified as HR+/HER2- that underwent serial sampling, 136 (24%) had detectable ESR1 missense mutations that were previously undetectable. Within the ESR1mut population, 1,990 (74%) had co-existing non-synonymous, non-VUS alterations at genes associated with SERD resistance: PIK3CA (n = 1,330; 49%), ARID1A (n = 314; 12%), PTEN (n = 201; 7%), ERBB2 (n = 180; 7%), AKT1 (n = 139; 5%), and NF1 (n = 103; 4%); top amplifications were FGFR1 (n = 522; 19%), EGFR (n = 277; 10%), and MYC (n = 177; 7%). Among a clinically annotated HR+/HER2- subgroup (n = 350), 51 patients received SERD monotherapy, either fulvestrant (n = 9) or oral SERD on clinical trial (n = 42); among 28 patients with cfDNA variants associated with SERD resistance, median PFS and OS were lower compared to the 23 patients without resistance mutations (PFS: 2.4 vs 11.3 mo; HR = 0.44; 95% CI, 0.23-0.81; p = 0.008; and OS: 21.1 vs 31.4 mo; HR 0.47; 95% CI, 0.22-0.98; p = 0.045). Conclusions: ESR1 mutations that qualify patients to receive approved SERD therapy are detectable by plasma-based genotyping and can associate with various genomic co-alterations, including variants that may impact clinical outcomes. Further research is needed to confirm the impact of co-alterations in additional datasets and evaluate the role of SERD-based combination therapy to further improve clinical outcomes of patients with MBC.