Abstract PS5-05-23: Real-world Treatment Patterns of Capivasertib in Metastatic Breast Cancer in the US

Abstract Background: In patients with hormone receptor-positive/human epidermal growth factor receptor 2-negative metastatic breast cancer (MBC) with ≥ 1 PIK3CA, AKT1, or PTEN tumor alterations, capivasertib plus fulvestrant is approved in the US for use after progression on at least one endocrine-based treatment. This study assessed real-world treatment patterns of capivasertib in the US. Methods: This retrospective cohort study used Flatiron Health electronic health record-derived database. Eligible patients were ≥18 years, diagnosed with MBC, initiated capivasertib between November 16, 2023 and July 31, 2024, and had at least 6 months (mos) of potential follow-up time from capivasertib initiation (index date). Real-world time to treatment discontinuation (rwTTD) and time to next line of treatment (rwTTNT) were assessed from the index date. Summary statistics were used to describe patient characteristics at baseline and treatment patterns. rwTTD and rwTTNT were estimated using Kaplan-Meier methods. The line of therapy (LOT) was defined using an algorithm that incorporated prior treatment use. Results: This study identified 412 patients with MBC who received capivasertib. The median age was 67 years (IQR 60, 75). The majority were female (98.8%), white (71.6%), and from community practice settings (83.0%). Prior to capivasertib initiation, sites of disease included: bone (91.0%), liver (44.2%), lung (28.2%), and brain (7.0%). PIK3CA, AKT1, or PTEN alterations were documented in 90.3% of patients: PIK3CA (75.0%), PTEN (14.3%), and AKT1 (10.0%) alterations. Co-mutations with ESR1 mutations were observed in 25.5% of patients. Prior therapies for MBC included: CDK4/6 inhibitors (87.1%), fulvestrant (66.0%), chemotherapy (31.8%), alpelisib (20.6%), and antibody-drug conjugates (14.1%) in any prior LOT. Most patients received capivasertib with fulvestrant (n=387, 93.9%); in 25 patients (6.1%), use with another endocrine therapy/combination or alone was recorded in the chart, but could not be verified due to data limitations. Among patients recorded as having received capivasertib with fulvestrant (n=387/412, 93.9%), more than half of patients received capivasertib in second-line (2L) (n=115, 29.7%) or 3L (n=102, 26.4%), though many were treated in 4L+ (n=146, 37.7%). 2L and 3L median rwTTNT was 7.1 mos (IQR 5.8, NE) and 6.9 mos (IQR 6.1, 7.8), respectively. 2L and 3L median rwTTD was 6.9 mos (IQR 5.3, NE) and 6.6 mos (IQR 5.2, 7.1), respectively (Table 1). Conclusions: Findings from this large database analysis of US patients demonstrate the effectiveness of capivasertib + fulvestrant in real-world practice. Clinical outcomes in 2L and 3L closely match those observed in the CAPItello-291 Phase 3 randomized controlled trial, which supported FDA approval of the capivasertib + fulvestrant regimen. Numerically improved outcomes were observed in patients who used capivasertib in earlier vs later line settings. Citation Format: K. H. Natsuhara, L. Park, S. Udayachalerm, R. Mireia, B. Murphy, B. Nordstrom, L. A. Huppert. Real-world Treatment Patterns of Capivasertib in Metastatic Breast Cancer in the US 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 PS5-05-23.