Eosinophil Count as Predictive Biomarker of Immune-Related Adverse Events (irAEs) in Immune Checkpoint Inhibitors (ICIs) Therapies in Oncological Patients
Background: To date, no biomarkers are effective in predicting the risk of developing immune-related adverse events (irAEs) in patients treated with immune checkpoint inhibitors (ICIs). This study aims to evaluate the association between basal absolute eosinophil count (AEC) and irAEs during treatment with ICIs for solid tumors. Methods: We retrospectively evaluated 168 patients with metastatic melanoma (mM), renal cell carcinoma (mRCC), and non-small cell lung cancer (mNSCLC) receiving ICIs at our medical oncology unit. By combining baseline AEC with other clinical factors, we developed a mathematical model for predicting the risk of irAEs, which we validated in an external cohort of patients. Results: Median baseline AEC was 135/µL and patients were stratified into two groups accordingly; patients with high baseline AEC (>135/µL) were more likely to experience toxicity (p = 0.043) and have a better objective response rate (ORR) (p = 0.003). By constructing a covariance analysis model, it emerged that basal AEC correlated with the risk of irAEs (p < 0.01). Finally, we validated the proposed model in an independent cohort of 43 patients. Conclusions: Baseline AEC could be a predictive biomarker of ICI-related toxicity, as well as of response to treatment. The use of a mathematical model able to predict the risk of developing irAEs could be useful for clinicians for monitoring patients receiving ICIs.