Neutrophil-to-Lymphocyte Ratio-Based Model for Predicting Immune-Related Adverse Events in Patients With Cancer Treated With Immune Checkpoint Inhibitors in Thailand.

Abstract

To evaluate whether baseline peripheral blood biomarkers-particularly the neutrophil-to-lymphocyte ratio (NLR)-predict immune-related adverse events (irAEs) in patients with solid tumors receiving immune checkpoint inhibitors (ICIs) and to identify optimal cutoff values for clinical risk stratification. We conducted a retrospective cohort study of adults with solid tumors treated with ICIs (anti-programmed death 1, anti-PD-L1, or anticytotoxic T-lymphocyte-associated antigen 4-based combinations) at Srinagarind Hospital, Khon Kaen University (2018-2024). irAEs were identified according to ASCO and National Comprehensive Cancer Network criteria. Clinically significant irAEs were defined as grade ≥3 events or those requiring hospitalization or permanent ICI discontinuation. Logistic regression assessed associations between baseline NLR and irAE risk, expressed as odds ratios (ORs) with 95% CIs. Predictive performance was evaluated using the area under the receiver operating characteristic curve (AUROC), and Cox regression assessed time to irAE onset. Among 240 patients (mean age, 62.5 ± 10.8 years; 67.5% male), 70 (29.2%) developed irAEs and 21 (8.8%) experienced clinically significant events during a median follow-up of 13 weeks. Baseline NLR ≥2.75 independently predicted any irAEs (adjusted OR, 2.44 [95% CI, 1.34 to 4.42]; <i>P</i> = .003; AUROC 0.695 [95% CI, 0.625 to 0.766]) and clinically significant irAEs (adjusted OR, 3.05 [95% CI, 1.09 to 8.49]; <i>P</i> = .033; 0.770 [95% CI, 0.637 to 0.871]) in models incorporating age and ICI regimen. Baseline NLR ≥2.75 was also associated with earlier irAE onset. Baseline NLR ≥2.75, integrated with age and ICI regimen, may facilitate early identification of patients at the highest risk of clinically significant irAEs. Prospective multicenter validation is warranted to confirm its clinical utility in real-world oncology practice.