Abstract P3-11-20: Treatment sequencing of HR+/HER2- metastatic breast cancer (mBC) patients based on PIK3CA alteration status - A retrospective analysis of a US clinicogenomics database

Abstract

Abstract Background: The PIK3CA gene encodes phosphatidylinositol-3-kinase (PI3K-α), which plays a central role in the growth, division, and many cellular processes critical for tumorigenesis. PIK3CA alteration is present in up to 40% of patients with HR+/HER2- mBC. Recent advances in therapies targeting the PI3K signaling pathway in breast cancer demonstrate improved outcomes; however, additional understanding on the implication of PIK3CA mutations in treatment sequencing decisions for patients with HR+/HER2- mBC is required. Methods: A retrospective analysis of the Flatiron Health/Foundation Medicine Clinico-Genomics Database, a nationwide database comprising de-identified patient-level structured and unstructured clinical data curated via technology-enabled abstraction and linked to genomic information, was conducted to characterize treatment sequencing among HR+/HER2- mBC patients, stratified by PIK3CA alteration status (PIK3CA+, PIK3CA-), as well as timing of NGS testing. The cohort included patients with follow-up through December 31, 2018, with no patients having received a PI3K inhibitor for mBC during this time. Patients in the cohort had ≥2 clinic visits since Jan 1 2011, chart confirmed diagnosis of HR+/HER2- mBC, and ≥1 Foundation Medicine next-generation sequencing (NGS) test conducted no earlier than 30 days before the mBC diagnosis date (index date). Treatment sequencing analysis included a sub-analysis by de-novo or relapsed mBC and based on sites of metastases and menopausal status. Results are presented as PIK3CA+ vs PIK3CA-, respectively (where applicable). Results: A total of 3,804 BC patients were in the database at the time of the analysis. After applying inclusion criteria, 1,293 patients were included for analysis (521 (40.3%) PIK3CA+, 772 (59.7%) PIK3CA-), median age 59yrs (Q1-Q3: 49-66), 98.8% female. PIK3CA+ patients were slightly older (median 60 vs. 56yrs). The majority of patients received their NGS test during or after 2L of mBC therapy (58% vs 54.9%), while only 5.6% of PIK3CA+ and 9.3% of PIK3CA- patients received NGS before commencing 1L therapy. The most common 1L mBC regimens were chemotherapy (CT) (36%), endocrine therapy (ET) (35%) and CDK4/6-based regimen (CDK4/6) (18%). PIK3CA status by NGS did not have a significant influence on treatment sequencing, however PIK3CA- patients had numerically higher CT than ET as 1L regimen. Common treatments were similar in 2L mBC, with less use of ET and increased use of CDK4/6 across groups. When assessing by menopausal status, the top three 1L treatments were unchanged, however there was a slight increased use of ET (43% vs 35%) and CDK4/6 (21% vs 20%) in post-menopausal patients. The most common sequence of treatments from 1L to 2L (post mBC diagnosis) was two lines of ET for the PIK3CA+ patients (14%) and two lines of CT for the PIK3CA- patients (19%). Two lines of CT was also the most common sequence for 2L to 3L for both groups (12% vs 18%). Remaining treatment sequences demonstrated significant heterogeneity and similar patterns in subgroups assessed. Conclusions: This analysis demonstrated significant heterogeneity of treatment patterns and sequences for both the PIK3CA+ and PIK3CA- groups, which may be expected given the limited available targeted therapies for patients with PIK3CA mutations during the study period. Mutation status is known late in the course of metastatic disease and does not appear to be associated with choice of treatment at this time. Citation Format: Stuart J Turner, Iris Wang, Jinhee Park, Hemanth Kanakamedala, Ines Lorenzo. Treatment sequencing of HR+/HER2- metastatic breast cancer (mBC) patients based on PIK3CA alteration status - A retrospective analysis of a US clinicogenomics database [abstract]. In: Proceedings of the 2019 San Antonio Breast Cancer Symposium; 2019 Dec 10-14; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2020;80(4 Suppl):Abstract nr P3-11-20.