Alpelisib (ALP), a breast cancer therapy, for PIK3CA-related overgrowth spectrum (PROS): A real-world data approach to a rare disease indication.

Abstract

e18694 Background: PIK3CA mutations seen in cancer also drive other disease, exciting interest in repurposing oncology therapies for non-oncology indications. ALP is a PI3Kα inhibitor approved for HR+, HER2– ABC with fulvestrant. Clinical benefit of ALP in PROS, a group of rare phenotypically varied disorders driven by mutations in PIK3CA, was found in 19 pts (Venot, 2018). EPIK-P1 (NCT04285723) collected real-world data (RWD) to confirm ALP’s efficacy and safety in PROS. In the primary analysis of the primary endpoint, 12/32 pts (37.5%; 95% CI, 21.1-56.3%) with complete cases responded (≥20% reduction in the sum of target lesion volume from index/treatment [tx] start). Adverse events (AEs) and tx-related AEs were experienced by 82.5% (n = 47) and 38.6% (n = 22) of pts, respectively. The most common tx-related AE was hyperglycemia (n = 7, 12.3%). No deaths were reported (Canaud, ESMO 2021 LBA23). Methods: EPIK-P1 retrospectively abstracted data from the medical charts of pts aged ≥2 y with PROS experiencing severe/life-threatening conditions receiving ALP under compassionate use. Secondary endpoints evaluated safety and clinical benefit. Results: 57 pts were enrolled and followed for a median of 18.1 mo. No pt in the primary analysis (complete cases, n = 32) had progressive disease (PD) at wk 24; 1 pt discontinued ALP prior to wk 24 and considered a nonresponder. Of the 12 responders, median duration of response (time from first documented response to first documented PD or death due to any cause) was not estimable. Median time to censoring (date of last adequate lesion assessment) was 63.3 wk (̃14.6 mo; range, 1 d-186.7 wk); 2 pts had no further imaging assessments, 10 pts did not experience an event and tx was ongoing at data cutoff. Time to censoring was > 6 mo in 7 pts and > 12 mo in 6 pts. 11/20 nonresponders (55.0%) had target lesion volume reduction ranging from 0.7 to 19.6%. No pt required rescue surgery between index and wk 24. At index all pts reported ≥1 PROS-related symptom (eg, fatigue, pain). At wk 24, 82.5% (n = 47) reported symptom improvement regardless of primary analysis response status (responders, 10/12, 83.3%; nonresponders, 16/20, 80.0%). Median daily dose was 50 mg for pediatrics and 250 mg for adults. Dose reductions or interruptions were experienced by 15.8% of pts (n = 9); AEs led to dose reductions and interruptions in 3.5% (n = 2) and 10.5% (n = 6) of pts, respectively. 52 pts (91.2%) remained on tx at data cutoff; among 5 pts (8.8%) who discontinued, pt decision was the leading cause. Conclusions: RWD from EPIK-P1 demonstrate alpelisib as well tolerated, providing clinical benefit to pts with PROS regardless of response status in the primary analysis of the primary endpoint. Continued RWD collection complements planned prospective, randomized, placebo-controlled studies, facilitating access to alpelisib as an emerging pharmacologic option for PROS. Clinical trial information: NCT04285723.