Preventing alpelisib-related hyperglycaemia in HR+/HER2−/PIK3CA-mutated advanced breast cancer using metformin (METALLICA): a multicentre, open-label, single-arm, phase 2 trial

Abstract

Hyperglycaemia is an early and frequent adverse event during alpelisib treatment. METALLICA aimed to evaluate prophylactic metformin to prevent or reduce hyperglycaemia occurrence in patients with HR+/HER2-/PIK3CA-mutated advanced breast cancer (ABC). Between August 13th, 2020 and March 23rd, 2022, this 2-cohort, phase 2, multicentre, single-arm trial (NCT04300790) enrolled patients with HR+/HER2-/PIK3CA-mutated ABC: cohort A, normal glycaemia (fasting plasma glucose <100mg/dL [<5.6mmol/L] and HbA1c <5.7%), and cohort B, prediabetes (fasting plasma glucose 100-140mg/dL [5.6-7.8mmol/L] and/or haemoglobin A1C [HbA1c] 5.7-6.4%). Participants were at least 18 years old, with Eastern Cooperative Oncology Group performance status of 0-1, and up to two prior lines of endocrine therapy (ET) for ABC. Alpelisib plus ET were administered in 28-day cycles after initiation of prophylactic metformin plus ET. Primary endpoint was the incidence of grade 3-4 hyperglycaemia over the first 8 weeks. Secondary endpoints included safety, progression-free survival (PFS), objective response rate (ORR), and clinical benefit rate (CBR). The primary objective for cohort A and B is met with ≤7 (14.6%) and ≤4 (20%) patients with grade 3-4 hyperglycaemia over the first 8 weeks, respectively. 233 patients were screened, and 68 (20.2%) patients were enrolled in cohorts A (n=48) and B (n=20). Median follow-up was 7.8 months (IQR 1.4-19.6). Over the first 8 weeks, one (2.1%) of 48 patients in cohort A (95% CI: 0.5-11.1; P<0.0001), and three (15.0%) of 20 patients in cohort B (95% CI: 5.6-37.8; P=0.016) had grade 3-4 hyperglycaemia. Serious treatment-related adverse events occurred in seven patients (10.3%). The most common were rash (two [2.9%]), vomiting (two [2.9%]), and diarrhoea (two [2.9%]). Discontinuation of alpelisib caused by AEs was reported in nine patients (13.2%), none caused by hyperglycaemia. At data cutoff (15 June, 2022), no treatment-related deaths were observed. In the full analysis set, median PFS was 7.3 months (95% CI: 5.9-not reached), ORR was 20.6% (95% CI: 11.7-32.1%), and CBR was 52.9% (95% CI: 40.4-65.2). In HR+/HER2-/PIK3CA-mutated ABC, prophylactic metformin before alpelisib plus endocrine treatment has low incidence and severity of alpelicib-induced hyperglycaemia. Novartis Pharmaceuticals.