CE3 COMPARATIVE EFFECTIVENESS AMONG SINGLE AGENT CHEMOTHERAPEUTICS IN THE TREATMENT OF TNBC USING TIME TO NEXT TREATMENT AS A SURROGATE FOR EFFICACY

There is an extensive list of FDA approved and NCCN recommended single agent chemotherapies (SAC) for the treatment of metastatic triple negative breast cancer (mTNBC) with similar levels of evidence and disease activity based upon clinical trials with highly variable design. The objective of this study was to describe the real-world frequency of SAC by line of therapy (LOT) comparing their effectiveness by measuring time to next treatment (TNT). Female TNBC patients, identified by excluding all patients who received any HER2 targeted therapy or endocrine therapy, with evidence of continuous medical and/or pharmacy claims were identified in the Symphony Health Integrated Dataverse (SHIDV) database. Inclusion criteria required a minimum of one non-diagnostic claim for breast cancer and one distant metastasis (ICD-9/10 code) and having initiated at least one SAC from 01/01/2013 to 12/31/2017. The proportions of patients receiving each agent in first-line (1L), second-line (2L), and ≥ third-line (3L+) were calculated. TNT was measured using the Kaplan Meier method. 1795 mTNBC patients met the study criteria; mean age 59.8 years, 58% commercially insured. The top 3 most frequently prescribed SAC by LOT: (1L) capecitabine (15%), nab paclitaxel (7%), carboplatin (6%); (2L) capecitabine (17%), paclitaxel (15%), eribulin (11%); (3L+) eribulin (17%), capecitabine (10%), gemcitabine (9%). TNT (months): (1L) capecitabine (9.7), carboplatin (6.4) and liposomal doxorubicin (6.4); (2L) capecitabine (8.5), vinorelbine (6.0), carboplatin (5.6); (3L+) carboplatin (6.1), capecitabine (6.0), liposomal doxorubicin (4.7). Our research demonstrates significant variance between the frequencies of individual SAC use and their respective TNT. TNT is a surrogate for progression-free survival and is likely influenced by toxicity and patient choice. Preserved TNT in later LOT and the practice of SAC sequencing may result in no survival difference by respective sequence but research is warranted to test this hypothesis.