Abstract PD5-3: Phase I trial: PI3Kα inhibitor BYL719 plus aromatase inhibitor (AI) for patients with hormone receptor-positive (HR+) metastatic breast cancer (MBC)

Background: Phosphatidylinositol 3-kinase (PI3K) hyperactivation plays a role in endocrine therapy resistance. Adding an α-selective PI3K inhibitor (BYL719) to hormonal therapy may therefore overcome resistance in HR+ MBC. We report results from a phase I study to evaluate the safety and preliminary efficacy of BYL719 plus an AI in patients (pts) with HR+ MBC. Methods: This 3+3 dose-escalation trial studied daily oral BYL719 added to standard dose letrozole (L, Arm A) or exemestane (E, Arm B), and later examined intermittent dosing (Arm C, L + BYL719 every other week; Arm D, E + BYL719 on 5 of 7 days weekly). Pts with HR+ MBC, any/no PIK3CA mutation, and on L/E were eligible. A cycle (C) was 28 days (d). Endpoints were dose-limiting toxicity (DLT), tolerability (CTCAE 4.0), and efficacy (RECIST v1.1). Paired tumor biopsies were performed for genomic and proteomic correlatives. Serial plasma was collected to quantify cell-free (cf) DNA and mutant allele fraction. Results: 14 pts (median (M) age: 55 (30-69) yrs), 7 each on Arms A and B, received a M of 76d (6-312+) of BYL719 + L or E. All were evaluable for toxicity, 10 for response. PIK3CA status was mutant(MT)/wild-type(WT)/unknown in 8/5/1 pts. M number of prior MBC therapies was 2 (1-12) in Arm A, 6 (2-14) in Arm B. Arms had similar toxicities. On Arm A, BYL719 was given at 300mg daily (DL0) to 3 pts who completed the 28d DLT period. 2 pts had 3 distinct DLTs: maculopapular rash (N=1), hyperglycemia (N=1), abdominal pain (N=1). Dose was de-escalated (DL-1=250mg) with no DLT in 3 enrolled pts. On Arm B, DL0, 1 pt experienced DLT (maculopapular rash) of 3 initially enrolled pts. Arm B expansion at DL0 had 1 additional pt with DLT (rash). Clinically significant, treatment-related toxicities included grade (G)≥4: none; G3: maculopapular rash (N=8, including 1 pt treated at DL-1), hyperglycemia (N=1) and G1/2: fatigue (N=7), nausea (N=7), and hyperglycemia (N=6). Toxicity required 6 dose reductions in 4pts and discontinuation in 2 pts. M duration on study for PIK3CA MT vs. WT was 169.5d vs. 69.5d, respectively. In pts with PIK3CA –MT MBC evaluable for response (n=7), 6 had clinical benefit: 1 PR (pt heavily pre-treated, including prior L, MBC to liver, Arm A, now C10+ after DLT); SD (n=5, included -29.9%, -19%, -12%), and POD (n=1). In pts with PIK3CA -WT MBC evaluable for response (n=3), 2 had SD (no changes ≥+/-5%) and 1 had POD. Serial cfDNA analysis in 4 pts with SD or PR demonstrated a decrease of >90% in the PIK3CA mutant allele fraction on treatment. Due to toxicity seen with continuous BYL719, the study was amended to explore intermittent dosing schedules (Arm C, L; Arm D, E; DL0=250mg), with 5 pts enrolled, 3 of whom have completed the DLT period with no DLTs, and 1 pt with G3 rash. Correlative studies including serial cf DNA collection from these pts is ongoing. Conclusions: Continuously dosed BYL719 with L or E shows promising antitumor activity. Skin toxicity warranted evaluation of alternative schedules. Mutant allele fraction may be an early predictor of response and may serve as a pharmacodynamic marker during intermittent treatment. Safety, efficacy, and correlative data from study Arm C and Arm D will also be presented. Citation Format: Payal D Shah, Mary Ellen Moynahan, Shanu Modi, Betty Ann Caravella, Farrah M Datko, Stephen Zamora, Elizabeth Comen, Theresa Gilewski, Steven M Sugarman, Gabriella D9Andrea, Diana E Lake, Shari B Goldfarb, Sujata Patil, Anne Covey, Michael F Berger, Mario E Lacouture, Larry Norton, Clifford A Hudis, Jose Baselga, Sarat Chandarlapaty, Maura N Dickler. Phase I trial: PI3Kα inhibitor BYL719 plus aromatase inhibitor (AI) for patients with hormone receptor-positive (HR+) metastatic breast cancer (MBC) [abstract]. In: Proceedings of the Thirty-Seventh Annual CTRC-AACR San Antonio Breast Cancer Symposium: 2014 Dec 9-13; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2015;75(9 Suppl):Abstract nr PD5-3.