Steroid receptor expression and overall survival in breast cancer patients with ER+ bone metastasis: A retrospective review.

Abstract

1077 Background: Endocrine therapy (ET) resistance is common in estrogen receptor-positive (ER+) metastatic breast cancer (BC), where bone metastases (BMET) are usually the first sign of spread. ER signaling and ET effects can depend on other steroid hormones receptors (SHRs), such as progesterone receptors (PR) and androgen receptors (AR). However, the roles of these receptors in ER+ BC BMET are underexplored. To address this gap, PR and AR protein expression in HER2-/ER+ BMET and associations with overall survival (OS) were examined. Methods: In a retrospective analysis on BC BMET samples analyzed at Caris Life Sciences, n = 2038 HER2- BMETS were identified by immunohistochemistry (IHC) (≤1+intensity or ≤10% staining, or 2+ & > 10% with CISH-null reflex test). HER2- ER+ (by IHC, ≥1+ & ≥1%) BMETs (“ER+ BMET” n = 1700; 84.5% of total) were then examined for prevalence of IHC+ SHR expression (PR: ≥1+ & ≥1%; AR: ≥1+ & ≥10%) and associated pathogenic/likely pathogenic ESR1 or PIK3CA gene mutations (mut). Associations of SHR status with clinical outcomes were tested by inferring OS from biopsy collection or start of therapy to last contact. Results: Most ER+ BMET expressed PR (59.3%) or AR (87.1%). Only 9.4% of PR+/ER+ BMET were AR-null, while 38.2% of AR+/ER+ BMET were PR-null. Overall, “triple positive” (AR+/PR+/ER+) BMET comprised the largest group (53.7%), followed by PR-null AR+/ER+ BMET (33.2%). AR-null ER+ BMET with (5.6%) or without PR (7.3%) were less common. AR+ status was associated with better outcomes for ER+ BMET patients (pts) with longest OS for “triple positive”, while “loss” of PR was associated with shorter OS (Cox proportional hazards ratio (HR) = 1.37, p < 0.0001). AR-null status was associated with worse OS compared to “triple positive” regardless of PR status (AR-/PR- HR = 1.66, p < 0.001; AR-/PR+ HR = 1.85, p < 0.0001). In ER+ BMET with ESR1mut (16.3%), OS for triple positive tumors, which were more prevalent (74.0%) due to increased PR expression, was reduced (HR = 1.92, p < 0.0001 vs ESR1wt ). For PIK3CA , “loss” of PR abrogates benefits associated with AR+ status only in PIK3CAmut pts (48.7%) with patterns between SHR groups otherwise maintained. Among ER+ BMET pts who received aromatase inhibitors or fulvestrant, AR+ status was associated with the best OS, regardless of PR status, where treatment was associated with significantly longer OS (vs no treatment) across SHR groups, except in PR+/AR-null. For ER+ BMET pts treated with CDK4/6 inhibitors, OS was highest in the “triple positive” cohort, with onlyAR+ SHR groups demonstrating improved OS with treatment. Conclusions: Based on this analysis, AR expression is more prognostic of OS than PR, regardless of treatment, in pts with ER+ BC BMETs, with “triple positive” BMETs generally associated with the best OS. Research on SHRs as mediators vs biomarkers of risk in ER+ BC BMETs is needed to provide direction for possible therapeutic targeting.