Abstract P5-13-03: Alpelisib + endocrine therapy (ET) in patients with hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2-), <i>PIK3CA</i>-mutated advanced breast cancer (ABC) previously treated with cyclin-dependent kinase 4/6 inhibitor (CDK4/6i): Biomarker analyses from the Phase II BYLieve study

Abstract Introduction: Alpelisib (ALP, a PI3Kα inhibitor and degrader) + ET has demonstrated efficacy in patients (pts) with HR+, HER2-, PIK3CA-mutated ABC progressing on/after CDK4/6i + ET in the ongoing Phase II BYLieve study (NCT03056755). The primary endpoint, in pts with centrally confirmed PIK3CA mutation in tumor tissue (modified full analysis set [mFAS]), was met in Cohorts A (ALP + fulvestrant [FUL]) and B (ALP + letrozole [LET]) with 50.4% (95% CI, 41.2%-59.6%) of 121 pts and 46.1% (36.8%-55.6%) of 115 pts alive and without disease progression at 6 mo, respectively. Activating PIK3CA mutations, occurring in ~40% of tumors, confer worse prognosis in the advanced setting. Here, we assess the correlation between progression-free survival (PFS) and other baseline biomarkers among pts in BYLieve Cohorts A and B. Methods: Enrolled pre-/postmenopausal women with HR+, HER2-, PIK3CA-mutated ABC received CDK4/6i + aromatase inhibitor (Cohort A) or FUL (Cohort B), as immediate prior treatment (Tx). Pts received ALP 300 mg PO QD + FUL 500 mg IM Q28D and C1D15 (Cohort A) or ALP 300 mg PO QD + LET 2.5 mg PO QD (Cohort B) as study Tx. This exploratory analysis of baseline biomarkers identified gene alterations using an error-connected sequencing ctDNA assay (Novartis PanCancer gene-panel) in plasma samples from pts with a centrally confirmed PIK3CA mutation in tumor tissue from Cohorts A and B. PFS was estimated using the Kaplan-Meier method in subgroups of pts based on high (≥10%) or low (&lt;10%) ctDNA fraction, high (≥10) or low (&lt;10) tumor mutation burden (TMB), and amplification (amp) status of chromosomes (chr) 8 or 11 amplicons (frequently observed in breast cancer and usually associated with early relapse). Results: In Cohorts A and B, 102 of 127 enrolled pts and 97 of the 126 enrolled pts were included in this analysis, respectively. In this analysis, median PFS (mPFS) was 7.3 mo in Cohort A and 5.7 mo in Cohort B, consistent with observations in the mFAS (Table 1). Pts with a low ctDNA fraction or TMB at baseline had longer mPFS than pts with high ctDNA fraction or TMB (Table 1). In addition to PIK3CA alterations observed in 73% of tumors from pts in Cohorts A and B, the most frequently altered genes in the studied population were ESR1 (27% and 28%) and TP53 (25% and 26%) in Cohorts A and B, respectively. Across both cohorts, pts whose tumors had loss of function alterations in genes known to mediate resistance to CDK4/6i (PTEN, RB1, CHD4, FAT1, NF1, ATM, CDKN2A-C) derived clinical benefit (partial response or stable disease) from ALP + ET. Amp of known amplicons on chr 8 (eg, FGFR1, TACC1) or 11 (eg, CCND1, FGF3, FGF4) were observed in 5%-8% of pts in Cohort A and 8%-11% of pts in Cohort B. Pts whose tumors had no gene amp on chr 8 and/or 11 had longer mPFS than pts whose tumors had an amp (Table 1). Similar results were observed when analyzing chr 8 (6.33 mo vs 4.92 mo) and chr 11 (6.33 mo vs 4.63 mo) separately in pts from Cohorts A plus B. Conclusion: In pts with HR+, HER2-, PIK3CA-mutated ABC, ALP was effective in the post-CDK4/6i setting regardless of ET partner and tumor genomic profile (including in presence of alterations in genes associated with CDK4/6i resistance). The absence of cross-resistance may allow pts with PIK3CA-mutated disease to benefit from ALP after disease progression on/after CDK4/6i. Table 1.mPFS in pts from BYLieve Cohorts A and B based on baseline biomarker statusCohort A (ALP + FUL)Cohort B (ALP + LET)nmPFS, mo (95% CI) nmPFS, mo (95% CI)Overall (mFASa)1,21217.3 (5.6-8.3)1155.7 (4.5-7.2)Patients with available plasma sample for the biomarker analysisb1027.37 (5.60-8.67)975.70 (3.77-7.33)ctDNA fractionHigh (≥10%)655.60 (4.00-7.37)685.43 (3.70-7.20)Low (&lt;10%)37NE297.33 (5.83-11.00)TMBHigh (≥10)103.85 (0.47-8.33)154.60 (1.67-5.73)Low (&lt;10)716.13 (5.43-9.60)645.63 (3.73-7.50)Chr 8 or 11 amplificationWith174.23 (1.70-6.13)213.77 (1.93-7.37)Without858.40 (5.70-9.60)766.00 (5.20-7.53)Chr, chromosome; ctDNA, circulating tumor DNA; NE, not estimable; mPFS, median progression-free survival; TMB, tumor mutation burden. aPatients who received at least 1 dose of study treatment and had centrally confirmed PIK3CA mutation by a Novartis-designated laboratory were included in the mFAS. bOnly includes patients with centrally confirmed PIK3CA mutation in tumor tissue. 1. Rugo HS, et al. Lancet Oncol. 2021;22(4):489-498; 2. Rugo HS, et al. SABCS 2020. Poster PD2-07. Citation Format: Dejan Juric, Nicholas Turner, Aleix Prat, Stephen Chia, Eva M Ciruelos, Manuel Ruiz-Borrego, Pamela Drullinsky, Florence Lerebours, Thomas Bachelot, O. Alejandro Balbin, Mukta Joshi, Estelle Roux, Christina H Arce, Murat Akdere, Hope S Rugo. Alpelisib + endocrine therapy (ET) in patients with hormone receptor-positive (HR+), human epidermal growth factor receptor 2-negative (HER2-), PIK3CA-mutated advanced breast cancer (ABC) previously treated with cyclin-dependent kinase 4/6 inhibitor (CDK4/6i): Biomarker analyses from the Phase II BYLieve study [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P5-13-03.