Background Cancer-associated fibroblasts (CAFs) are essential components of the tumor microenvironment (TME), contributing to tumorigenesis, progression, and resistance to therapy. However, the functional diversity of CAF subpopulations and their role in tumor progression and patient prognosis remain poorly understood. This study aims to explore CAF heterogeneity and their functional roles in the TME using single-cell RNA sequencing (scRNA-seq) and multi-omics data analysis. Methods scRNA-seq data were analyzed to cluster CAF subpopulations in the TME, with key genes identified through functional annotation. Differentially expressed genes were analyzed, and prognostic genes were selected via Cox and LASSO regression. A risk score model (RiskScore) was developed for survival prediction and immune therapy sensitivity evaluation. Core CAF genes were examined using siRNA interference, qPCR, and Western blotting. Drug sensitivity was assessed to explore the clinical relevance of these genes. Results Four CAF subpopulations (CAF-0, CAF-1, CAF-2, CAF-3) were identified, revealing differences in key tumor-associated signaling pathways (e.g., MYC, WNT, TGF-beta). Thirteen core genes related to prognosis were identified, and a RiskScore model was developed, showing significantly worse survival rates for high-risk patients (p < 0.001) and features of immune suppression, including increased M0 macrophage infiltration. Drug sensitivity analysis indicated that core genes (e.g., KLRB1, MAP1B) were linked to drug sensitivity, suggesting potential biomarkers for targeted therapy. Experimental validation showed that knockdown of the HIP1R gene significantly reduced tumor cell expression, confirming its critical role in tumor development. Conclusion This study offers a comprehensive analysis of CAF heterogeneity and its impact on TME, patient prognosis, and drug sensitivity. The developed RiskScore model provides theoretical support for personalized treatment based on CAF-related genes, offering new insights into CAF-driven tumor progression and potential targets for precision oncology and immunotherapy.