1078 Background: The phosphatidylinositol-4,5-bisphosphate 3-kinase catalytic subunit alpha ( PIK3CA) gene is mutated in ~40% of patients (pts) with hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2−) advanced breast cancer (ABC). Alpelisib (ALP), an α-selective PI3K inhibitor and degrader, is indicated in combination with fulvestrant (FUL) for pts with HR+, HER2− ABC following progression on/after endocrine therapy (ET)-based treatments (tx). Primary analyses from the 3 cohorts of the Phase 2, open-label, noncomparative BYLieve study demonstrated the efficacy of ALP + ET in pts with HR+, HER2–, PIK3CA-mutated ABC who progressed on/after other tx, including immediately after a cyclin-dependent kinase 4/6 inhibitor (CDK4/6i). Presented here are efficacy and safety data from the full 3 cohorts of the BYLieve study. Methods: Pts in all 3 cohorts completed ≥18 mo follow-up plus 1 mo safety follow-up. Efficacy assessments included DoR, BOR, PFS, PFS2, and OS. Safety/tolerability of ALP were also assessed. Median (m) PFS, PFS2, and OS were assessed by Kaplan-Meier methodology per local investigator assessment. Results: Cohort A (ALP + FUL) comprised 127 pts whose immediate prior tx was CDK4/6i + aromatase inhibitor (AI); Cohort B (ALP + letrozole) enrolled 126 pts whose immediate prior tx was CDK4/6i + FUL. In Cohort C (ALP + FUL), 126 pts had disease progression on/after AI and received CT or ET as immediate prior tx. Median follow-up for the 3 cohorts was 21.8, 25.3, and 18.5 mo, respectively. mPFS in Cohorts A, B, and C was 8.0 mo (95% CI, 5.6-8.6), 5.6 mo (3.7-7.1), and 5.6 mo (5.4-8.1), respectively; mPFS2 was 15.2 mo (95% CI, 11.4-21.7), 13.0 mo (10.2-13.9), and 13.5 mo (11.5-17.3), respectively. mOS in Cohorts A, B, and C was 27.3 mo (95% CI, 21.3-32.7), 29.0 mo (24.5-34.8), and 20.7 mo (16.9-28.1), respectively. Efficacy of ALP + ET in 2nd and 3rd lines will be presented. In Cohorts A, B, and C, respectively, 29 (22.8%), 19 (15.1%), and 23 (18.3%) pts discontinued tx due to adverse events (AEs). The most common any grade AEs in all 3 cohorts (A; B; C) were diarrhea (n = 82, 64.6%; n = 86, 68.3%; n = 68, 54.0%), hyperglycemia (n = 76, 59.8%; n = 82, 65.1%; n = 85, 67.5%), nausea (n = 59, 46.5%; n = 69, 54.8%; n = 51, 40.5%), rash (n = 40, 31.5%; n = 39, 31.0%; n = 51, 40.5%), and fatigue (n = 39, 30.7%; n = 39, 31.0%; n = 44, 34.9%). Hyperglycemia was the most common grade ≥3 AE, observed in 37 (29.1%), 32 (25.4%), and 30 (23.6%) pts in Cohorts A, B, and C, respectively. Conclusions: After ≥18 mo follow-up, with mature data, ALP + ET demonstrated clinical activity in BYLieve Cohorts A, B, and C. Within the trial, ALP + ET showed median OS between 20.7 mo and 29.0 mo following prior CDK4/6i, chemotherapy, or ET. AEs associated with ALP were well defined and manageable, with no new safety signals observed in any cohort. Clinical trial information: NCT03056755 .