Integrating single-cell and bulk RNA data to explore the prognostic significance of LIM domain family genes in colorectal cancer and unveil their immune landscape.

BACKGROUND: Colorectal cancer (CRC) is one of the most prevalent and lethal malignancies worldwide. The LIM domain family genes have been found to play critical roles in various cancers, including CRC. However, their specific prognostic implications in CRC remain unclear. METHODS: Through scRNA-seq, cells were stratified into high- and low-LIM score groups by AUCell, and DEGs were identified. WGCNA in the TCGA-CRC cohort identified LIM domain family-correlated modules. Overlapping genes between module genes and DEGs were used to build a risk model via LASSO and multivariate Cox regression. High- and low-LIMRS groups were compared for prognosis, tumor microenvironment, function, and drug sensitivity. In vitro experiments validated RGS16's effect on Erlotinib sensitivity. RESULTS: AUCell revealed that high-LIM-score cells were mainly fibroblasts. A nine-gene signature (CRYAB, RGS16, CSRP2, RCAN2, TPM1, EGFL7, PLS3, PDGFRA, PLAU) stratified CRC patients into high- and low-LIMRS groups with distinct prognoses. After FDR correction, the high-LIMRS group showed a higher type II IFN response, whereas TIDE and IPS analyses suggested greater potential ICI responsiveness in the low-LIMRS group. The calcium signaling pathway emerged as a key regulator. Drug sensitivity analysis identified erlotinib, fulvestrant, EPZ5676, and AZD1208 for low-LIMRS, and vinblastine, docetaxel, paclitaxel, vinorelbine for high-LIMRS. In vitro experiments confirmed that Erlotinib efficacy depends on RGS16 expression. CONCLUSION: The prognostic model we developed, based on LIM domain family genes, contributes to predicting the prognosis of CRC patients and provides preliminary biological evidence for further exploration of its clinical translational potential.