The dual role of baseline absolute eosinophil count in non-small cell lung cancer immunotherapy: a biomarker for enhanced efficacy and elevated risk of immune checkpoint inhibitor-related pneumonitis

Objectives: This retrospective study aimed to evaluate the association between baseline absolute eosinophil count (AEC) and both survival outcomes and the risk of immune checkpoint inhibitor-related pneumonitis (ICI-pneumonitis) in patients with advanced non-small cell lung cancer (NSCLC) receiving immune checkpoint inhibitor (ICI)-based therapy. Methods: We retrospectively enrolled 158 patients with advanced or recurrent NSCLC who received ICI-based therapy at our center. Patients were dichotomized into high- (≥ 120 cells/µL) and low- (< 120 cells/µL) baseline AEC groups. Baseline clinical characteristics, treatment regimens, and immune-related adverse events (irAEs) were collected. Associations between baseline AEC and progression-free survival (PFS) and overall survival (OS) were analyzed using Kaplan–Meier curves and multivariable Cox proportional hazards models. Chi-squared tests were used to compare response rates and irAE incidence between groups. A Fine-Gray competing-risk regression model was applied to identify predictors of ICI-pneumonitis, accounting for death or progression as competing events. Results: Patients with high baseline AEC showed significantly improved median PFS (mPFS, 24.3 vs. 12.4 months, p < 0.001) and median OS (mOS, not reached vs. 48.3 months, p = 0.008), together with a higher objective response rate (ORR, 59.1% vs. 40.0%, p = 0.018). Multivariable Cox models confirmed high baseline AEC as an independent favorable prognostic factor for both PFS (hazard ratio [HR]: 0.402, 95% confidence interval [CI]: 0.265–0.609; p < 0.001) and OS (HR: 0.483, 95% CI: 0.256–0.912; p = 0.025). The high-AEC group experienced a higher incidence of any irAEs (39.8% vs. 23.1%, p = 0.028) and ICI-pneumonitis (17.2% vs. 4.6%, p = 0.017). In the Fine-Gray competing-risk model, high baseline AEC remained independently associated with a higher cumulative incidence of pneumonitis (sHR: 4.33, 95% CI: 1.33–14.02; p = 0.015). Conclusion: Our findings suggest that baseline AEC may serve as a dual biomarker in NSCLC immunotherapy, demonstrating associations with both improved survival outcomes and an elevated risk of ICI-pneumonitis. These results highlight the need for vigilant monitoring in patients with high baseline AEC. Incorporating this readily available parameter into clinical assessment could help refine risk–benefit stratification for patients considering ICI-based therapy.