Abstract P2-15-09: Systemic corticosteroid use in patients with metastatic triple-negative breast cancer treated with first-line therapy in the United States

Background: Corticosteroids (CS) have immunosuppressive properties that may diminish the efficacy of cancer immunotherapy (CIT). In retrospective analyses of clinical trial data, the use of CS for immune-mediated adverse events after starting CIT does not clearly appear to adversely affect outcomes; however, patients (pts) receiving CS at the start of CIT may have shorter overall and progression-free survival (Arbour, J Clin Oncol, 2018). With the increasing use of CIT for pts with metastatic triple-negative breast cancer (mTNBC) following the approval of atezolizumab in combination with nab-paclitaxel, it is important to understand the landscape of CS use as part of first-line (1L) treatment. Here we describe patterns of CS use in pts receiving 1L therapy for mTNBC, with a focus on taxanes. Methods: A total of 974 pts with confirmed mTNBC diagnosed from Jan 1, 2011, to Jul 5, 2018, and treated in the 1L setting from the de-identified Flatiron Health electronic health record-derived database were included in the study. Pts were categorized as having no CS use during 1L treatment, CS use at the start of 1L treatment or CS use after the start of 1L treatment. Results: The median age at mTNBC diagnosis was 60 years. The majority of pts were female (99%), treated at community centers (94%), and had recurrent disease (59%). Among the most common regimens in 1L were those containing capecitabine (n=208), paclitaxel (n=204), or nab-paclitaxel (n=131) (Table 1). Capecitabine was administered as monotherapy more frequently than paclitaxel and nab-paclitaxel (86% vs 67% and 52%, respectively). Most pts receiving capecitabine did not receive CS during 1L therapy (79%), whereas the majority of pts receiving paclitaxel and nab-paclitaxel did (>95% and 78%, respectively). However, among pts treated with nab-paclitaxel, CS use during 1L was more common in pts receiving combination regimens (88%) than those receiving monotherapy (67%). In general, pts who were treated with paclitaxel or nab-paclitaxel and initiated CS use at the start of 1L treatment continued it throughout their 1L treatment. CS was given (orally or intravenously) with 92% of paclitaxel administrations, 67% of nab-paclitaxel monotherapy administrations, 77% of nab-paclitaxel combination therapy administrations, and 81% of administrations when agents were given in combination with capecitabine. Among pts who received nab-paclitaxel monotherapy, pts who used CS during 1L tended to be younger and less likely to have visceral metastases or recurrent disease than pts who did not use CS. Conclusions: Overall, CS use was common at 1L initiation of most regimens used for the treatment of mTNBC, with the exception of capecitabine. Among pts who were treated with taxanes, those who initiated CS at 1L treatment tended to continue CS treatment for the duration of 1L. Despite the lack of requirement for the use of CS with nab-paclitaxel, the majority of pts treated with nab-paclitaxel monotherapy still received CS at the start of 1L treatment, suggesting use as premedication. None of the assessed pt characteristics were associated with CS use in pts who were treated with nab-paclitaxel monotherapy; however, due to widespread use of CS in the population and small sample size, definitive conclusions are difficult. Larger data sets with reasons for CS use are needed to further understand the observed patterns. Citation Format: MK Downer, Patricia Luhn, David Miles, Matthew Kent, Kenneth Russell, Joyce O’Shaughnessy. Systemic corticosteroid use in patients with metastatic triple-negative breast cancer treated with first-line therapy in the United States [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 P2-15-09.