Abstract PD1-10: Immune Age Acceleration as a Biomarker for Immune-Related Adverse Events in Early Triple-Negative Breast Cancer Treated with Perioperative Pembrolizumab

Abstract Background: Immunotherapy is a cornerstone in the treatment of early stage, high-risk triple-negative breast cancer (TNBC), leading to improved pathological complete response rates and overall survival. However, a substantial proportion of patients experience immune-related adverse events (irAEs), which can be severe or even life-threatening. At present, there are no validated biomarkers capable of predicting the risk of high-grade irAEs among breast cancer patients undergoing immunotherapy. While therapeutic efficacy is largely dictated by tumor-intrinsic characteristics, evidence from studies in non-small cell lung cancer (NSCLC) suggests that irAE susceptibility is significantly influenced by the host’s baseline immune status. Advances in immunological profiling have enabled the quantification of immune system function by leveraging the physiological decline in immune competence that accompanies aging. These metrics—collectively referred to as immune age—offer a clinically meaningful assessment of immune system integrity that surpass chronological age in their clinical predictive value. Methods: Peripheral serum samples were collected from 145 treatment-naıve patients with stage I-III breast cancer (age range: 28-75 years). Baseline serum concentrations of 96 inflammatory proteins were quantified using the Olink inflammation panel. To derive immune age, we trained an elastic net linear regression model with cross-validation to predict chronological age from inflammatory protein profiles in post-surgical, treatment-naıve samples (n = 57), minimizing tumor-induced immune perturbation. The trained model was then applied to the full cohort to compute immune age. Immune age acceleration was defined as the difference between immune age and chronological age. Among 33 TNBC patients treated with perioperative pembrolizumab and chemotherapy, we compared immune age acceleration at baseline between those who did and did not develop high-grade (grades 3-4) irAEs. Results: The immune-age model accurately predicted chronological age in both training (post-surgery) and test (pre-surgery) cohorts, with Pearson correlation coefficients of 0.783 and 0.628, respectively. Immune age was positively associated with increased levels of inflammatory markers including CXCL9 and CCL11 (BH-adjusted P = 0.014 and P = 0.008, respectively), consistent with known hallmarks of immune aging. Across all measured inflammatory proteins, correlations with immune age were significantly stronger than with chronological age (paired t-test, P = 2.5 × 10−10), highlighting the superior capacity of immune age to capture individual baseline inflammatory status. Immune age acceleration followed a normal distribution centered around zero with a standard deviation of 4.7 years, reflecting the high inter-individual variability in the rates of immune aging. Among TNBC patients treated with perioperative pembrolizumab (n = 33), 13 (39.4%) developed grade 3-4 irAEs. Patients who developed high-grade irAEs exhibited significantly higher immune age acceleration at baseline (prior to treatment) compared to those who did not develop irAEs (t-test, P = 0.025), while chronological age showed no significant association with irAE development (median age: 44.0 vs. 46.9 years; P = 0.26). Conclusion: Immune age acceleration is associated with increased risk of severe irAEs in early-stage TNBC and may serve as a novel predictive biomarker. This immune-aging framework offers a host-centric approach to stratify risk prior to immunotherapy initiation, with potential implications for personalized surveillance and toxicity mitigation strategies. Citation Format: A. Alpert, M. Makarov, S. Mor, D. Aran, A. Shai. Immune Age Acceleration as a Biomarker for Immune-Related Adverse Events in Early Triple-Negative Breast Cancer Treated with Perioperative 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 PD1-10.