Characterization of alpelisib-associated hyperglycemia in metastatic breast cancer.

Abstract

1016 Background: For women with metastatic hormone receptor (HR)-positive, human epidermal growth factor 2 (HER2)-negative breast cancer, the combination of alpelisib and fulvestrant improves progression-free survival in those with PIK3CA mutations. Hyperglycemia is a major toxicity of PI3K inhibitors including alpelisib, which limits the clinical efficacy of these drugs due to interrupted/reduced dosing and discontinuation. In the SOLAR-1 trial that led to the Food and Drug Administration (FDA) approval of alpelisib, over 60% of patients developed hyperglycemia of any grade, and over 36% developed grade 3-4 hyperglycemia. Here we describe the incidence and treatment of alpelisib-associated hyperglycemia in a single center cohort. Methods: Patients with metastatic breast cancer who received alpelisib on a clinical trial or as part of standard care from 2013-2021 at Memorial Sloan Kettering Cancer Center were included in this retrospective study. Patient and tumor characteristics and pre-treatment body mass index (BMI), hemoglobin A1c, and serum glucose levels were abstracted from medical records. Alpelisib dose interruptions, reductions, or discontinuation was recorded as well as endocrinology consultation and use of anti-hyperglycemic agents. Date of progression and/or death were recorded where applicable. Results: 247 patients were included in this study, among whom 245 (99.1%) were female and 198 (80.1%) were white. 100 (40.5%) were treated on a clinical trial. Median baseline BMI was 25.4 kg/m2. Among 164 patients with baseline hemoglobin A1c levels available, 93 (56.7%) patients had normal hemoglobin A1c, 54 (32.9%) had prediabetes, and 17 (10.4%) had diabetes. 152 patients (61.5%) developed hyperglycemia of any grade; 56 (22.7%) developed grade 3 and 16 (6.5%) developed grade 4 hyperglycemia. The median time to onset of hyperglycemia was 16 days. BMI ≥25 kg/m2 or hemoglobin A1c ≥5.7% were strongly predictive of development of any-grade hyperglycemia ( p= 0.036 and p< 0.001, respectively) and grade 3-4 hyperglycemia ( p< 0.001 for both). Among those who developed hyperglycemia, 101 (40.9%) received treatment; 69 patients (27.9%) required only 1 anti-hyperglycemic agent whereas 9 (3.6%) required ≥3 anti-hyperglycemic agents. 49 (19.8%) were referred for endocrinology consult. In 66 patients (26.7%), alpelisib was held until resolution of hyperglycemia; 42 patients (17%) required dose reductions, and 11 (4.5%) discontinued alpelisib due to hyperglycemia. There was no significant difference in progression-free survival by hyperglycemia status or severity of hyperglycemia. Conclusions: Overweight BMI and hemoglobin A1c in the prediabetes or diabetes range were strongly predictive of developing alpelisib-associated hyperglycemia. Management of these co-morbidities prior to alpelisib treatment should be strongly considered.