Association and interaction network analysis of immune cell subgroups in cervical cancer microenvironment based on Mendelian randomization.

BACKGROUND

The interactions among cell subgroups in the cervical cancer immune microenvironment play crucial roles in tumor development, but their causal relationships remain unclear.

METHODS

This study employed Mendelian randomization to analyze causal associations between immune cell subgroups and cervical cancer. Multiple statistical methods, including inverse variance weighted, weighted median, and simple mode approaches, were used to evaluate effect sizes. Hierarchical clustering, UMAP, and t-SNE were applied for cell subgroup classification, combined with MIF signaling pathway analysis for cell-cell interaction networks.

RESULTS

Most immune cell subgroups showed effect estimates close to 1.000 (95%CI: 0.997-1.002) with statistical significance (p < 0.05). Hierarchical clustering analysis revealed eight major cell populations: regulatory T cells, T cells, epithelial cells, natural killer cells, monocytes, ciliated epithelial cells, B cells, and fibroblasts. Cell-cell interaction network analysis demonstrated extensive connectivity among immune cells and between immune and epithelial cells, with particularly strong interactions between monocytes and other immune cells. MIF signaling pathway analysis further confirmed the close relationship between regulatory T cells and T cells.

CONCLUSION

This study systematically revealed the causal associations among cell subgroups in the cervical cancer immune microenvironment using Mendelian randomization, providing new insights into understanding tumor immune microenvironment regulation mechanisms and potentially offering theoretical basis for optimizing cervical cancer immunotherapy strategies.