Manuel Ruiz Borrego, Pablo Tolosa, Salvador Blanch, Adela Fernández, Ander Urriticoechea, Isabel Blancas, Cristina Saura, Beatriz Rojas, Begoña Bermejo, Jose Ponce, Maria Gión, Elisenda Llabres, Elena Galve, Juan Fernando Cueva, Ana López, José L Alonso-Romero, Santiago González-Santiago, Eduardo Martínez De Dueñas, Fernando Gomez Peralta, Eva Ciruelos, José Manuel Pérez-García, Antonio Llombart-Cussac, and Javier Cortés
Background: HG is an on-target AE of PI3K inhibition, reported in 63.7% (36.6% G≥3) of HR[+]/HER2[-] ABC patients (pts) treated with ALP plus fulvestrant in SOLAR-1. HG was the most frequent adverse event (AE) leading to ALP discontinuation (6.3%). MET reduces systemic insulin resistance and suppress PI3K and Ras signaling. METALLICA is assessing the prophylactic use of MET for prevention of ALP-induced G3-4 HG in PIK3CA-mutated, HR[+]/HER2[-] ABC pts with normal fasting glycemia or prediabetic criteria. Methods: This is an open-label, single-arm, two-cohort, phase 2 trial. Pts aged ≥18 years, ECOG PS of 0–1, and PIK3CA-mutated, HR[+]/HER2[-] ABC, progressing to an aromatase inhibitor (AI)-containing regimen, ≤2 previous endocrine therapy (ET) and ≤1 prior chemotherapy regimens for ABC were eligible. Pts were enrolled into cohorts according to glycemia at baseline: (A) pts with normal fasting glycemia < 100 mg/dL and glycosylated hemoglobin (HbA1c) < 5.7%; (B) pts with prediabetic fasting glycaemia 100–140 mg/dL and/or HbA1c 5.7–6.4%. Pts received oral ALP 300 mg/day, starting from C1D8, in combination with ET; fulvestrant, letrozole, or exemestane as per standard of care; and oral MET 1000 mg/day on days 1-3 and 2000 mg/day thereafter. The primary endpoint was G3-4 HG incidence as per NCI-CTCAE v.4.03 at 2 first cycles of treatment. Assessment of glycemia was performed by rigorous self-monitoring blood glucose and local laboratory confirmation in fasting conditions. Secondary endpoints included objective response (ORR), clinical benefit rate (CBR), duration of response (DoR), progression-free survival (PFS), and safety. Sample size was based on a Simon’s two-stage design in cohorts A (H0: G3-4 HG ≥25%; H1: G3-4 HG ≤10%) and B (H0: G3-4 HG ≥40%; H1: G3-4 HG ≤15%). We planned to attain 80% power at the nominal one-sided α level of 0.05 for each cohort. Results: Between Aug 30, 2020, and Mar 10, 2022, 68 pts were enrolled at 18 sites (48 cohort A, 20 cohort B). Median age was 55 (range, 29–79) years and 58.8% pts had visceral disease and an ECOG PS 0. A total of 66 (97.1%) pts had been previously treated with a CDK4/6i and 13 (19.2%) pts had received chemotherapy for advanced disease. Sixty-three (92.6%) pts received fulvestrant as ET (45 cohort A, 18 cohort B). With a median follow-up of 8 (range, 1.6–14.9) months, 28 (41.2%) pts remain on study treatment. Disease progression was the main reason for discontinuation, reported in 32 (47.1%) pts. The primary endpoint of the study was reached, with 1 (2.1%) pts (95%CI, 0.8–9.5; p < 0.001) in cohort A and 3 (15%) pts (95%CI, 4.5–33; p = 0.012) in cohort B experiencing a G3-4 HG episode over the 2 first cycles of treatment. For patients on fulvestrant, G3-4 HG rates were 1 pts (2.2%) and 3 (16.7%) pts for cohorts A and B, respectively. No ALP discontinuation related to HG was reported during the first 2 treatment cycles. Median PFS in all patients was 7.4 months (95%CI, 6–NA). Among pts with measurable disease, ORR was 14 (36.8%) pts (95%CI, 21.8–54). At the time of this analysis, DoR and CBR were still immature. The most common AEs were diarrhea (67.6%; 13.2% G≥3), nausea (67.6%; 0% G≥3), and fatigue (45.6%; 2.9% G≥3). Serious AEs occurred in 15 (22.1%) pts. The main serious AEs were rash (2.9% G≥3) and vomiting (1.5% G≥3). No additional pts reported G≥3 HG after the first 2 cycles. The dose of ALP was reduced according to the protocol in 19 (27.9%) pts. Eight (11.8%) pts permanently discontinued ALP due to AEs, none of whom related to HG. No treatment-related deaths were reported. Conclusions: Prophylactic use of MET substantially reduced the incidence and severity of ALP-related HG with no additional toxicities and could be a new standard for PIK3CA-mutated, HR[+]/HER2[-] ABC pts receiving ALP plus fulvestrant or other ET. Citation Format: Manuel Ruiz Borrego, Pablo Tolosa, Salvador Blanch, Adela Fernández, Ander Urriticoechea, Isabel Blancas, Cristina Saura, Beatriz Rojas, Begoña Bermejo, Jose Ponce, Maria Gión, Elisenda Llabres, Elena Galve, Juan Fernando Cueva, Ana López, José L Alonso-Romero, Santiago González-Santiago, Eduardo Martínez De Dueñas, Fernando Gomez Peralta, Eva Ciruelos, José Manuel Pérez-García, Antonio Llombart-Cussac, Javier Cortés. Metformin (MET) for the prevention of Alpelisib (ALP)-related Hyperglycemia (HG) in PIK3CA-mutated, Hormone Receptor-Positive (HR[+]) HER2-Negative (HER2[-]) Advanced Breast Cancer (ABC): The METALLICA study. [abstract]. In: Proceedings of the 2022 San Antonio Breast Cancer Symposium; 2022 Dec 6-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2023;83(5 Suppl):Abstract nr PD8-02.