Baseline eosinophil proportion is a useful predictor of immune-related adverse events following immune checkpoint inhibitor treatment for recurrent metastatic head and neck cancer.

Background: Immune checkpoint inhibitors (ICIs) occasionally cause immune-related adverse events (irAEs) in various organs. However, predictors of irAEs remain unidentified.

Objectives: We evaluated the predictors of irAEs and compared the outcomes of ICIs with and without irAEs in patients with recurrent/metastatic head and neck cancers (R/M HNCs).

Materials and methods: We retrospectively analyzed 157 patients with R/M HNCs who were administered an anti-PD-1 antibody between September 2014 and December 2022. We examined whether various pretreatment factors were associated with irAEs. The overall survival (OS) and progression-free survival (PFS) in patients with and without irAEs were analyzed.

Results: Overall, 44 patients (28.0%) developed irAEs. The survival curve estimated for patients with and without irAEs showed a significant difference in PFS (p = 0.018), but not in OS (p = 0.208). Multivariate analysis revealed significant differences in relative eosinophil counts (p < 0.001), TP (p = 0.014), and NLR (p = 0.002), which may be independent predictors of irAEs.

Conclusion: IrAEs may be associated with higher efficacy of ICIs and longer PFS. The relative eosinophil count may be predictors of irAEs and useful in routine medical practice. Using these biomarkers to predict irAEs will help predict ICI effects and manage irAEs.