Abstract Introduction: Patients with la/m TNBC have poor prognosis and limited effective treatment options. Clinical guidelines recommend testing for biomarkers, including PD-L1 expression, to inform clinical decision making and eligible treatments. This study describes treatment patterns and clinical outcomes in patients with la/m TNBC treated in US community and academic health systems. Methods: This observational, cohort study used deidentified person-level NorstellaLinQ EMR/lab and supplemental claims data. Patients were required to have diagnosis of la/m TNBC followed by initiation of 1L mTNBC anti-cancer treatment between 1/1/ 2020 and 12/31/2023. Patients were followed from initiation of 1L treatment (index date) until earliest of death, end of EMR, or end of study period (12/31/2024). Patients were stratified by PD-L1 status closest to the index date. Tumor PD-L1 status were assessed using the PD-L1 assays in EMR laboratory results or evidence of PD-L1 tumor status in unstructured clinical notes. Clinical outcomes included real-world time on treatment (rwTOT) and real-world time to next treatment (rwTTNT). Results: A total of 724 patients met criteria for analysis (PD-L1 Negative: 163, PD-L1 Positive: 398, Unknown/did not receive IO in la/m setting: 163). Mean(sd) age at index: 59(13) years. Demographics were comparable across PD-L1 Negative/Positive subgroups. Less than 5% of study patients were exposed to IO therapy in (neo)adjuvant setting. Among all patients, there were over 30 distinct 1L treatment agent-based regimens observed. Among patients with PD-L1 negative tumors, most common 1L regimens were chemotherapy monotherapy (“chemo mono”) (73.6%) and chemotherapy + immunotherapy (chemo + IO) (17.2%)). Among patients with PD-L1 positive tumors, most common 1L regimens were chemo + IO (60.6%) and chemo mono (31.7%). Approximately 34% of study patients cycled through at least 3 LOTs during the study period. rwTOT and rwTTNT were described in Table 1. Conclusions: In this real-world study of US patients with la/m TNBC and confirmed PD-L1 status, there is variability in 1L, 2L and 3L regimens. The majority of the PD-L1 positive cohort received 1L chemo (+/- IO) followed by chemo + IO or ADCs in 2L/3L. Negative PD-L1 cohort was largely prescribed 1L chemo followed by ADCs or chemo in 2L/3L, while there is a proportion of patients who used chemo+ IO combination therapy that is higher than previously published literature. The proportion of patients receiving subsequent LOTs was low, highlighting the need for new and efficacious frontline treatment options. Citation Format: X. Wang, Y. Zhang, A. Petrilla, S. McElwee, B. Momani, J. Dinoso, B. Stwalley, C. Lai, A. Brufsky. Variations in lines of treatment and clinical outcomes in US health system patients with locally advanced, inoperable or metastatic triple negative breast cancer (la/m TNBC) abstract. In: Proceedings of the San Antonio Breast Cancer Symposium 2025; 2025 Dec 9-12; San Antonio, TX. Philadelphia (PA): AACR; Clin Cancer Res 2026;32(4 Suppl):Abstract nr PS5-02-27.
Wang et al. (2026) studied this question.