Background: Continuing trastuzumab beyond disease progression was one of the most important paradigm changes in Oncology and is recommended by all guidelines. The reason behind it is that metastatic breast cancer (MBC) progresses much faster when the HER2 pathway is not blocked. However, the amount and robustness of data supporting this recommendation is limited, leading to lack of approval and/or access to this therapeutic strategy in many countries. The present study aims to provide additional data supporting that continuing trastuzumab (Tras) + chemotherapy (CT) may substantially improve outcomes of patients with MBC. Methods: We conducted a retrospective cohort study, using an U.S Electronic Health Record-derived de-identified database (Flatiron Health). Patients with MBC who were previously treated with anti-HER2 therapies and initiated as third line treatment CT combined with trastuzumab-based therapy (Tras+CT) or CT alone (index date) from 01/01/12-31/12/20 were included. An intention-to-treat approach was used to estimate the direct average treatment effects (ATEs) of initiating Tras+CT versus CT only on the Hazard Ratios for overall survival (OS) and real world progression free survival (rwPFS). Potential sources of confounding were identified using directed acyclic graphs and included clinical characteristics at index date as well as prior duration of similar treatments. Propensity score of received Tras+Ct vs CT was estimated using covariate balancing propensity score methodology. ATE was estimated using inverse probability of treatment weighting (IPTW) for Cox-proportional hazard models. Hazard ratios 95% confidence intervals were estimated using empirical bootstrap. Results: Three hundred and thirty seven patients initiated either treatment strategy (median age 60 years old), of these 288 patients initiated Tras+CT (49% trastuzumab based regimens, 43% T-DM1 based regimens and 8% including newer agents) and 49 received CT only. The median OS and median rwPFS were 23 months and 6 months for Tras+CT treated patients and 11 months and 5 months for CT treated patients. In the weighted population the median OS and median rwPFS were 19 months and 6 months for Tras+CT treated patients and 10 months and 5 months for CT treated patients resulting in hazard ratios of 0.29 [0.15-0.54] and 0.69 [0.45-1.06] for OS and rwPFS respectively. Conclusion: Including trastuzumab-based therapy in addition to CT in the third line treatment (Trast+CT) was associated with markedly improved survival outcomes compared with CT only. This study represents how HER2 blockade is maintained in third line in the U.S. over the period of time covered by the study. Limitations of the study could be unmeasured confounding factors, and the potential channeling of patients with lower socio-economic status towards CT treatment only. The latter could be similar to what is happening to patients treated in countries where continuing trastuzumab beyond progression is not accessible. This study provides further and strong support for this treatment strategy that should be accessible to all MBC patients across the world. As next steps, these results will be confirmed using other real-world datasets from different countries, including countries without current access to trastuzumab beyond progression.

Citation Format: Thibaut Sanglier, Ryan Ross, Tianlai Shi, Joao Mouta, Fatima Cardoso. Comparative effectiveness of initiating chemotherapy with or without trastuzumab based regimens as third line treatment for metastatic breast cancer [abstract]. In: Proceedings of the 2021 San Antonio Breast Cancer Symposium; 2021 Dec 7-10; San Antonio, TX. Philadelphia (PA): AACR; Cancer Res 2022;82(4 Suppl):Abstract nr P2-13-22.