Genomic determinants of response to alpelisib plus fulvestrant in the SOLAR-1 trial.

Approximately 40% of patients with hormone receptor (HR)-positive, human epidermal growth factor receptor 2 (HER2)-negative advanced breast cancer (ABC) have PIK3CA alterations, which contributes to endocrine therapy resistance. Alpelisib, an α-selective phosphatidylinositol 3-kinase inhibitor and degrader, given in combination with fulvestrant, is approved for the treatment of PIK3CA-mutated, HR-positive, HER2-negative ABC, based on the results of the SOLAR-1 trial (NCT02437318). Aside from PIK3CA, other gene alterations are associated with poor prognosis and limited response to treatment in this patient population. In this retrospective analysis, we carried out tissue-based next-generation sequencing of 398 patients (237 PIK3CA-altered, 161 PIK3CA-wild type) from SOLAR-1. Progression-free survival (PFS) correlative analysis was carried out in the PIK3CA-altered cohort. PIK3CA-altered and PIK3CA-wild type tumors had distinct genomic profiles. In the PIK3CA-altered cohort, patients who received alpelisib plus fulvestrant had a median PFS of 11.01 months versus 5.55 months for those receiving placebo plus fulvestrant (P = 0.0004). Patients in the lowest tumor mutational burden quartile as well as those with FGFR1 or FGFR2 alterations derived greater PFS benefit from alpelisib plus fulvestrant versus placebo plus fulvestrant [18.5 versus 3.22 months: hazard ratio (HR) 0.38, 95% confidence interval (CI) 0.21-0.68; FGFR1 12.71 versus 3.75 months: HR 0.38, 95% CI 0.17-0.81, P = 0.32; FGFR2 9.63 versus 2.78 months: HR 0.31, 95% CI 0.1-0.94, P = 0.29]; patients with MYC or RAD21 alterations derived limited PFS benefit. Cox and multitask machine learning models identified lower Eastern Cooperative Oncology Group performance status, prior cyclin-dependent kinase 4/6 inhibitor (CDK4/6i) treatment, and PTEN or TP53 alterations among the most deleterious factors for PFS in the PIK3CA-altered cohort. Alpelisib plus fulvestrant provides clinical benefit for patients with PIK3CA-altered, HR-positive, HER2-negative ABC across a range of concomitant alterations, including those previously implicated in endocrine therapy or CDK4/6i resistance. Machine learning models can identify factors including gene mutations that influenced PFS.