Natural killer cell as a potential predictive biomarker for early immune checkpoint inhibitor-associated cardiovascular adverse events: a retrospective cohort study.

Background: Peripheral immune cells can predict responses to immune checkpoint inhibitor (ICI) therapy, but their relationship with early ICI-associated cardiovascular adverse events (CVAEs) is unclear. This study aimed to assess the predictive value of peripheral immune cells in early ICI-associated CVAEs. Methods: Single-cell RNA sequencing (scRNA-seq) dataset from the Gene Expression Omnibus database was used to explore immune cell changes associated with ICI-associated CVAEs. Patients who had received ICI therapy for three cycles at the First Affiliated Hospital of Chongqing Medical University between November 2020 and November 2022 were then included. Patients were stratified into CVAEs and no CVAEs groups and compared peripheral immune cell subsets. Univariate and multivariate regression analyses were conducted to identify CVAEs risk factors. Receiver operating characteristic (ROC) curve analysis determined optimal cutoff values for potential biomarkers. Propensity score matching (PSM) was used to validate the predictive value of baseline NK cell proportion for CAVEs. Results: ScRNA-seq data revealed decreased CD8+ T and B cell proportions in the CVAEs group, while NK cell proportions increased. Among 203 patients, dynamic changes in the proportion of total T cell, CD8+ T cell, and NK cell differed significantly between groups. Baseline NK cell proportion was identified as an independent risk factor for CVAEs (p=0.009). ROC analysis identified baseline NK cell proportion as a potential predictor of CVAEs (AUC 0.674). The optimal cutoff value was determined to be 16.4%, and this finding was confirmed following PSM. Conclusion: Baseline NK cell proportion was a potential predictor of early ICI-associated CVAEs.