Real-world outcomes associated with tarlatamab for extensive-stage small cell lung cancer (ES-SCLC) in the Veterans Health Administration (VA).

Abstract

e20153 Background: ES-SCLC remains an aggressive malignancy with a very poor prognosis. Tarlatamab, a DLL3-targeted bispecific T-cell engager, improves survival in previously treated ES-SCLC. However, real world outcomes of tarlatamab use outside of large medical centers, and specifically in the VA, have not been reported. Examination of toxicity and efficacy outcomes in the VA allows examination of a national cohort characterized by an older age and higher comorbidities. Methods: We conducted a retrospective cohort analysis of patients treated with tarlatamab from 6/6/2024 to 1/13/2026 using the VA Corporate Data Warehouse. Annual SCLC incidence was obtained from the VA Cancer Registry System. Clinical data was obtained by manual chart review. Descriptive statistics were used to summarize patient demographics, disease characteristics, treatment patterns, clinical outcomes, and safety outcomes. Results: An estimated 908 cases of ES-SCLC were estimated to have been diagnosed during the study interval. 52 patients received tarlatamab at 23 distinct hospital sites in 17 states (2 patients were treated at 2 sites). Patient characteristics: 90% male; median age 68 (range 43-81); 73% White, 8% Black and 8% Latino; ECOG 0 = 6 (12%), 1 = 27 (52%), 2 = 14 (27%), 3 = 2 (4%), unknown 3 (6%); 69% ES-SCLC at diagnosis. At the start of tarlatamab, 24 (46%) and 21 (40%) had liver and brain metastasis. The number of starts increased from 2.5/month to 4.1/month between the first and second half of the study interval, 20 (38.5%) were started as second line treatment. Outcomes: 43 patients were evaluable for response. Disease control rate was 46% (95% CI, 32.5%-61.1%). Median progression free survival was 2.6 months (95% CI 2.0-6.0) and median overall survival (OS) 4.9 months (95% CI: 3.4-NR) with 12 month- OS rate 27.6% (95% CI, 14.0%–54.3%). Safety: Any grade CRS occurred in 21 patients (40%) and recurred in 5 (10%); 91% and 9% of CRS episodes occurred in cycle 1 or 2 respectively. Grade 3+ CRS occurred in 3 (6%). Glucocorticoid and tocilizumab were used in 38% and 33% respectively. No recurrent CRS was grade 3-5. ICANS occurred in 13 (25%) patients, with no recurrence. No association was found between ICANS incidence and baseline brain metastasis. Grade 5 toxic events: CRS 1, ICANS 2, hepatotoxicity 1. Conclusions: An increasing but small fraction of ES-SCLC patients have received tarlatamab in the VA. Among the first 52 patients treated in VA, both outcomes and toxicity were numerically less favorable than reported in the DeLLphi-304 study with lower PFS and OS and higher risk of ICANS and G3 CRS with fatal toxicity in 4 (8%) patients. Analysis of patients with ES-SCLC treated outside of VA may further inform the efficacy and safety of tarlatamab in Veterans and guide future use in VA.