Integrating cell interaction with transcription factors to obtain a robust gene panel for prognostic prediction and therapies in cholangiocarcinoma.

The efficacy of immunotherapy for cholangiocarcinoma (CCA) is blocked by a high degree of tumor heterogeneity. Cell communication contributes to heterogeneity in the tumor microenvironment. This study aimed to explore critical cell signaling and biomarkers induced cell communication during immune exhaustion in CCA. We constructed empirical Bayes and Markov random field models eLBP to determine transcription factors, interacting genes, and associated signaling pathways involved in cell-cell communication using single-cell RNAseq data. We then analyzed the mechanism of immune exhaustion during CCA progression. We found that -positive macrophages with high levels of could interact with to promote the exhaustion of CD8 T cells in CCA. Transcription factors and can upregulate the expression of in -positive macrophages. Subsequently, we obtained a panel containing 54 genes through the model, which identified subtype S2 with high expression of immune checkpoint genes that are suitable for immunotherapy. Moreover, we found that patients with subtype S2 with a higher mutation ratio of had immune-exhausted genes, such as and Finally, we constructed a nine-gene eLBP-LASSO-COX risk model, which was designated the tumor microenvironment risk score (TMRS). Cell communication-related genes can be used as important markers for predicting patient prognosis and immunotherapy responses. The TMRS panel is a reliable tool for prognostic prediction and chemotherapeutic decision-making in CCA.