Abstract PS2-06-05: Comparing Cross-Toxicities and Efficacy of Everolimus and Alpelisib in Different Sequences in Metastatic Breast Cancer Patients

Abstract Background: Everolimus (EV) and alpelisib (AL), both approved for treating HR+/HER2- Metastatic Breast Cancer (MBC), inhibit different steps of the PI3K/AKT/mTOR (PAM) pathway. They share unique side effects (SE), many of which are considered class effects (e.g., hyperglycemia). It is unclear if the emergence of an SE during a PAM inhibitor (PAMi) treatment predicts the risk of the same SE during subsequent PAMi treatment. Such cross-incidence (CrIn) of SEs across PAMi is understudied and could help predict and prevent SEs. Similarly, the efficacy of a PAMi on a tumor resistant to a different PAMi hasn’t been well established, as these tumors are excluded from new PAMi clinical trials. Methods: The objectives of the study were to assess 1. CrIn of important AEs between AL and EV; 2. efficacies of AL after treatment with EV and vice versa, among patients with HR+/HER2- MBC. A retrospective study was conducted in 3 U.S. cancer institutes (Wilmot, Roswell Park, Duke). Patients with HR+/HER2-MBC with PIK3CA mutation who received both drugs were included. Groups (grps) were defined based on the PAMi sequence: EV followed by AL (EV-AL) or reverse, AL-EV. Progression-free survival (PFS) and Overall Survival (OS) were estimated using Kaplan-Meier method and compared between grps using log-rank test. Survival analyses were conducted for the 2nd PAMi in the sequence. Multivariable analyses were done using Cox Models. Results: Of 55 pts screened, 46 were eligible: 30 in EV-AL and 16 in AL-EV grp. Baseline (at start of 2nd PAMi) variables were similar between the grps. Median age (years) was 64 in EV-AL grp, 65 in AL-EV grp. 70% of EV-AL grp and 81% of AL-EV grp were white. 47% of EV-AL grp and 56% of AL-EV grp had visceral disease. 77% of EV-AL grp and 79% of AL-EV grp had ECOG performance status 0-1. 97% of EV-AL and 100% of AL-EV grp had prior CDK4/6 inhibitors. 23% of EV-AL grp and 12% of AL-EV grp had >2 lines of prior chemotherapy for MBC. Median AL use was 5.3 months (m) in EV-AL grp vs 4.5m in AL-EV grp (p= 0.15); Median EV use was 6.3m in EV-AL grp vs 3.8m in AL-EV grp (p=0.76). Fulvestrant was the most common endocrine therapy with AL (90% in EV-AL, 94% in AL-EV) vs exemestane with EV (67% in EV-AL, 75% in AL-EV). Overall, hyperglycemia (hypG) was the most common SE with AL (65%), followed by diarrhea (38%), rash (20%), and stomatitis (20%). On EV, patients had diarrhea (16%), stomatitis (16%), rash (14%) and hypG (11%). In EV-AL grp, among the 3 pts who had developed hypG on EV, all 3 developed hypG on AL (100% CrIn). CrIn of rash, diarrhea & stomatitis on AL after developing them on EV were 50%, 20% and 33% respectively. In AL-EV, CrIn of hypG, rash, diarrhea, & stomatitis on EV after having them on AL were 10%, 0%, 25%, & 33% respectively. Comparing the efficacy of 2nd PAMi among the grps, median PFS on AL was 6.9m (95% CI 4.2–10.8) in EV-AL grp vs 7.4m (95% CI 3.5–23.1) on EV in AL-EV grp (p=0.64). In multivariable analysis, risks of PFS event on 2nd PAMi were not significantly different between grps [Hazard ratio (HR) 0.95 (95% CI 0.37-2.4) p=0.91]. Median OS on AL was 15.9m (12.4-34.2) in EV-AL grp vs that on EV was 20.1m (13.2-46.7) in AL-EV grp (p=0.84). In multivariable analysis, the risks of death after starting 2nd PAMi were not significantly different between grps [HR 0.92 (95%CI 0.39-2.17), p=0.85]. Conclusion: This study has assessed CrIn of SEs across EV and AL use. Notably, all patients who developed hypG and half who developed rash on EV also developed them again on AL. These CrIn results could help clinicians use preemptive strategies (like metformin, topical steroids) while using 2nd PAMi. AL and EV therapy were each associated with mPFS >6m when used as 2nd PAMi, suggesting meaningful activity. No significant differences were observed in PFS or OS between these two drugs when used as the 2nd PAMi. Further validation of efficacy and CrIn of SEs is needed in a larger dataset. Citation Format: A. Dhakal, A. Shrestha, Z. Shah, S. Chinniah, A. Roy, T. Smith, S. Gandhi, H. Moore, A. Van Swearingen, M. Strawderman, D. Peterson, C. Anders, R. O'Regan. Comparing Cross-Toxicities and Efficacy of Everolimus and Alpelisib in Different Sequences in Metastatic Breast Cancer Patients [abstract]. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS2-06-05.