Prognostic value of ESR1 mutation status in patients with metastatic breast cancer receiving second-line therapy: A real-world evidence study.

e13140 Background: ESR1 mutations emerge during endocrine therapy and are associated with endocrine resistance in HR+/HER2-metastatic breast cancer (mBC). The independent prognostic value of ESR1 mutations in patients starting second-line (2L) therapy remains unclear in real-world settings. Methods: We conducted a retrospective cohort study using Flatiron Health EHR-derived de-identified database (range: 02/01/2015-12/31/2024) of patients with HR+/HER2- mBC receiving first-line (1L) CDK4/6 inhibitor plus aromatase inhibitor and had ESR1 testing within 90 days of 2L therapy initiation. ESR1 status was classified as Mutated (any positive test within window), or Not Detected (ND, negative test with no prior positive). Primary endpoints were real-world progression-free survival (rwPFS; composite of progression or death) and overall survival (rwOS). Multivariable Cox proportional hazards models adjusted for age, ECOG, calendar year, demographics, metastatic burden, prior adjuvant AI, treatment-free interval,1L duration, practice setting, and sample type (Model A), with additional adjustment for 2L treatment (Model B). Stratified analyses by sample type (blood vs tissue) were also performed. Results: Of 40,437 patients identified, 480 met eligibility criteria with valid ESR1 results (170 ESR1 Mutated 35.4%, 310 ESR1 ND 64.6%). Blood samples comprised 56% of tests, tissue 42%, other 2%. Patients with ESR1 mutation detected had significantly longer treatment-free interval (median 34 vs 10 months, p = 0.023) and 1L duration (20 vs 12 months, p < 0.001) than patients without ESR1 mutation detected. Median rwPFS was 5.9 months (95% CI: 5.1-9.0) for ESR1 Mutated vs 6.2 months (5.3-8.5) for ESR1 ND; median rwOS was 19.4 months (17.0-26.4) vs 24.0 months (19.3-30.5). In multivariable models, ESR1 Mutated status was associated with significantly worse rwPFS (Model A: HR = 1.60 1.11-2.31, p = 0.011; Model B: HR = 1.79, 1.21-2.64, p = 0.004) and rwOS (Model A: HR = 1.76, 1.14-2.72, p = 0.011; Model B: HR = 1.93, 1.21-3.09, p = 0.006). Sample type stratification demonstrated the prognostic effect was significant in blood samples (rwPFS Model B: HR = 2.27, 1.36-3.78, p = 0.002; rwOS Model B: HR = 2.09, 1.26-3.47, p = 0.004) but not in tissue samples (rwPFS Model B: HR = 1.20, 0.60-2.39,p = 0.60; rwOS Model B: HR = 1.77, 0.77-4.09, p = 0.18). Conclusions: In this real-world cohort of patients with HR+/HER2- mBC, ESR1 Mutated status demonstrated significant prognostic value for worse rwPFS and rwOS after 1L CDK4/6i plus AI therapy, with effects independent of 2L treatment choice. Sample type stratification revealed that the prognostic signal was driven by blood-based testing, with no effect observed in tissue samples. These results have important implications for interpretation of ESR1 status.