Real-world treatment patterns and outcomes among patients (pts) with second-line (2L) and third-line (3L) metastatic triple-negative breast cancer (mTNBC) in England using the Cancer Analysis System (CAS).

Abstract

1075 Background: TNBC is the most aggressive type of breast cancer due to rapid growth, metastasis, and recurrence post-treatment. This study aimed to assess real-world treatment patterns and survival of pts with mTNBC who received 2L and 3L therapy in England and report OS and PFS of pts receiving 2L therapies stratified by treatment-free interval from the curative setting. Methods: This retrospective study using the CAS database, included pts with mTNBC who received at least three systemic treatments (at least two in metastatic setting) for TNBC during the years 2012 to 2020. Cohort2L included pts with initial early-stage BC diagnosis and treatment (at least two systemic treatments prior to 2L). Cohort3L included pts initially diagnosed with either early-stage TNBC, or de novo advanced TNBC (at least two systemic treatments prior to 3L). The two cohorts are not mutually exclusive. The study outcomes were stratified by cohort (Cohort2L and Cohort3L) and treatment-free interval (<12 m versus ≥12 m from end of curative treatment to start of 1L treatment, Cohort2L only). Kaplan-Meier methods estimated progression-free survival (PFS) and overall survival (OS). PFS was proxied by ‘Time to treatment discontinuation or death’ (TTDD). Log-rank tests compared the distribution of OS and PFS for 2L stratified by treatment-free interval (<12 m versus ≥12 m). Results: Cohort2L included 606 pts and Cohort3L included 374 pts. Tumor morphology was similar across Cohorts. Pts at 3L had worse ECOG performance score compared to 2L, and more pts with de novo advanced TNBC had brain metastasis at any point after diagnosis than pts diagnosed with early-stage TNBC. Regimens at 2L for Cohort2L included capecitabine (32%), eribulin (16%), carboplatin and gemcitabine in combination (12%), and paclitaxel (10%). A similar distribution was seen for Cohort3L. Regimens at 3L (Cohort3L only), included eribulin (38%), capecitabine (16%), and paclitaxel (13%). Stratifying Cohort2L by treatment-free interval did not exhibit significant differences in PFS nor OS by log rank tests (Table). Conclusions: This nationwide study, in England, accentuates the significant unmet need in 2L and 3L therapy for mTNBC highlighted by the poor prognosis. The stratification by prior treatment-free interval from curative setting did not show a difference in OS or PFS for patients receiving 2L treatment. [Table: see text]