Unraveling the mechanisms of irAEs in endometrial cancer immunotherapy: insights from FAERS and scRNA-seq data.

Endometrial cancer (EC) is one of the most common malignancies in women. In recent years, immunotherapy has gradually become a significant treatment option. However, the mechanisms underlying immune checkpoint inhibitor (ICI)-related Adverse Events (AEs) remain poorly understood, posing significant challenges for optimizing clinical treatment strategies. This study aims to integrate the FAERS database and single-cell transcriptomic data to investigate potential mechanisms underlying PD-1 inhibitor-related AEs in EC immunotherapy, with a focus on exploring the PD-1-associated cell communication network and its potential compensatory activation pathways. Data related to AEs were extracted from the FAERS database. Disproportionality analyses, including Reporting Odds Ratio (ROR), Proportional Reporting Ratio (PRR), Bayesian Confidence Propagation Neural Network (BCPNN), and Multi-item Gamma Poisson Shrinker (MGPS), were used to quantify signals of immune-related AEs (irAEs) associated with ICIs. We compared the occurrence timing and characteristics of AEs across different drugs. Subsequently, scRNA-seq was performed to analyze the tumor microenvironment of EC, focusing on PD-1-high expressing cell populations. Cell Communication was analyzed and key receptor-ligand pairs were identified. From Q1 2004 to Q3 2024, 21,838,627 drug-related reports were retrieved from FAERS, including 2,202 related to ICIs. ICI-associated irAEs involved 26 organ systems, with general disorders, gastrointestinal disorders, and injury/poisoning as the top System Organ Class (SOC). Fatigue, product use issues, and diarrhea were the most reported Preferred Terms (PTs). PD-1 inhibitors were associated with faster onset of AEs compared to PD-L1 inhibitors and Weibull modeling indicated an early failure-type AE pattern for both treatments. Single-cell analysis further demonstrated that PD-1 was highly expressed in CD8 + cytotoxic T cells and Tfh cells, which communicated with other cells within the tumor microenvironment through key receptor-ligand pairs such as CXCL12-CXCR4 and CXCL16-CXCR6. These findings suggested that PD-1 inhibitors may induce AEs through compensatory activation of the CXCR4 and CXCR6 pathways. This study suggested that PD-1 inhibitors may contribute to irAEs in EC, potentially through compensatory activation of the CXCR4 and CXCR6 pathways. By integrating FAERS and scRNA-seq data, key receptor-ligand interactions were identified, providing preliminary insights that could inform future efforts to optimize immunotherapy efficacy and mitigate AEs. However, further validation through clinical studies and mechanistic research is needed to confirm these findings.