Cancer-Associated Fibroblast-Derived Gene Signature Predicts Prognosis and Exhibits Correlations with Aggressive...

Cancer-associated fibroblasts (CAFs) play vital roles in HCC initiation and progression via multicellular, stromal-dependent processes. Despite their acknowledged significance, the comprehensive role of CAF-related gene signatures in HCC remains underexplored. Given the established role of CAFs in HCC progression and therapy resistance, we aimed to develop a CAF-derived gene signature for prognosis prediction.

This comprehensive analysis used RNA sequencing data from The Cancer Genome Atlas database and a validation cohort from the International Cancer Genome Consortium database. Differential gene expression analysis identified CAF-related genes associated with HCC prognosis. A gene signature was developed using the least absolute shrinkage and selection operator algorithm and Cox regression analysis, and its prognostic value was evaluated through Kaplan-Meier survival analysis and receiver operating characteristic curve (ROC) analysis. The signature's associations with immune cell infiltration, chemotherapeutic response, and functional enrichment were investigated to further evaluate the prognostic value of CAF-related genes.

In total, 256 differentially expressed CAF-related genes were identified between samples with high and low CAF infiltration. After systematic analysis, a prognostic signature, including 23 genes, was constructed. Kaplan-Meier and ROC analyses demonstrated the robust predictive potential of our risk model for patients with HCC. Univariate/ multivariate Cox regression analyses further confirmed that the risk model was an independent prognostic factor for HCC in both the testing and validation cohorts. The signature stratified patients into high- and low-risk groups with distinct survival (logrank p < 0.001), achieving areas under the curve of 0.848 (1 year), 0.795 (3 years), and 0.781 (5 years). Patients in the high-risk group were more responsive to vinblastine, docetaxel, and navitoclax. Patients in the low-risk group had greater chemosensitivity to olaparib. Moreover, the stromal score was significantly higher in the high-risk group. Finally, the signaling pathways enriched in the high-risk group were mainly associated with HCC progression.

We established a CAF-related gene signature model as an independent prognostic predictor for HCC. This model may also guide adjuvant chemotherapy and targeted therapy, though further validation across diverse cohorts is needed to confirm its clinical utility.

The CAF-related gene signature exhibited good predictive potential concerning the prognosis of patients with HCC.