Abstract PS1-01-29: Real-world immune toxicity and survival outcomes by age in triple-negative breast cancer patients receiving pembrolizumab

Abstract In triple-negative breast cancer (TNBC), pembrolizumab combined with chemotherapy has become a standard of care in both neoadjuvant and adjuvant settings, as established by the KEYNOTE-522 trial. Despite these clinical advances, immune checkpoint inhibitors (ICIs) are associated with immune-related adverse events (irAEs) that can affect multiple organ systems and result in substantial morbidity. In other cancer types, older patients have been shown to experience higher rates of irAEs, potentially due to age-related immune dysregulation and a higher burden of comorbidities, while younger patients often face distinct survivorship challenges. However, real-world data on age-specific irAE patterns and outcomes in TNBC patients receiving pembrolizumab are lacking. To date, no studies have specifically evaluated the differential impact of pembrolizumab-based therapy on irAE incidence, severity, and outcomes across age groups in TNBC populations. This is a retrospective cohort study using TriNetX, TNBC patients treated with pembrolizumab between 2016 and 2024. Two age-stratified cohorts were created: younger (≤54 years) and older (≥55 years) of 1,661 patients per group after propensity score match. irAEs covering endocrine, hepatic, pulmonary, hematologic, and gastrointestinal toxicities were identified. Corticosteroid therapy and hospitalization were assessed as markers of toxicity severity. Kaplan-Meier survival curves were used. Older patients had a higher incidence of irAEs at 1 month (OR 1.43, 95% CI 1.01-2.04, p=0.048) and 3 months (OR1.43, 95% CI 1.11-1.79, p=0.0049), mainly due to thyroiditis (OR 1.67, 95% CI 1.08-2.13 p=0.017). Differences in irAE rates were no longer significant at 6 or 12 months. Despite lower irAE rates, younger patients had higher dexamethasone use at 3 months (OR 1.64, 95% CI 1.1-2.3, p=0.0047) and significantly worse survival (log-rank p=0.0071; HR 1.4, 95% CI 1.1-1.9). A composite marker of corticosteroid use and hospitalization showed younger patients consistently experienced more severe toxicities and poorer survival at all timepoints. At 1 month: OR 1.45, 95% CI 1.03-2.17, p=0.03; HR 1.3, 95% CI 1.01-1.9, log-rank p=0.04. At 3 months: OR 1.57, 95% CI 1.13-2.1, p=0.0087; HR 1.4, 95% CI 1.08-1.88, log-rank p=0.0105. At 6 months: OR 1.38, 95% CI 1.0-1.9, p=0.048; HR 1.3, 95% CI 1.03-1.69, log-rank p=0.027. At 12 months: OR 1.57, 95% CI 1.13-2.1, p=0.0063; median survival 171 days for younger vs. not reached in older group (log-rank p=0.0023; HR 1.49, 95% CI 1.1-1.8). At 3 months, total corticosteroid use and hospitalization remained higher in younger patients (OR 1.4, 95% CI 1.02-2.1, p=0.0359; HR 1.36, 95% CI 1.01-1.83; log-rank p=0.046). At 12 months, younger patients still had lower median survival (212 days vs. not reached; log-rank p=0.11; HR 1.2, 95% CI 0.95-1.59). While all-cause mortality was higher in older patients (p0.0001), hospice referrals were similar between groups. In this real-world cohort of TNBC patients treated with pembrolizumab, older patients experienced a higher incidence of early irAEs—particularly thyroiditis—but demonstrated superior survival. In contrast, younger patients had fewer documented irAEs but showed greater corticosteroid use, more frequent hospitalizations, and significantly worse survival across all timepoints when toxicity severity was considered. These findings suggest that irAE incidence alone may not fully capture clinical burden or risk and that age-related differences in immune response, toxicity manifestation, and management practices may contribute to divergent outcomes. Further investigation is warranted to elucidate the immunologic mechanisms of aging in the context of ICI use and safety to guide toxicity monitoring and supportive care strategies. Citation Format: G. Gorecki, O. Abioye, K. Babu, R. Srinishant, N. Gupta, C. Hilton. Real-world immune toxicity and survival outcomes by age in triple-negative breast cancer patients receiving pembrolizumab 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 PS1-01-29.