BACKGROUND: Colorectal cancer (CRC), a lethal tumor of the intestine, has lipotoxicity, which exerts a significant influence on the onset and development of intestinal disorders. This study provides an in-depth evaluation of lipotoxicity-related genes (LRGs) in the CRC treatment setting. METHODS: Differentially expressed genes (DEGs) within The Cancer Genome Atlas-CRC dataset and LRGs were intersected to obtain the LRG-DEGs. Subsequently, univariate Cox regression and a random survival forest model were conducted to select prognostic LRGs and construct a risk model for patients with CRC. Then, prognosis was further investigated via independent prognostic analysis, nomogram construction, enrichment analysis, immune microenvironment analysis, and drug sensitivity testing. Finally, reverse transcription quantitative PCR (RT-qPCR) was used to verify the expression levels of prognostic LRGs. RESULTS: The constructed risk model with five prognostic LRGs (PPARGC1A, CPT2, CXCL1, FABP4, and OFCC1) showed high prognostic effectiveness. The nomogram subsequently built based on risk score, age, T stage, and M stage showed strong prognostic power. Moreover, the enrichment of diverse pathways was observed across distinct risk groups, exemplified by the PPAR signaling pathway and complement and coagulation cascades. Analysis of the immune microenvironment revealed the strongest positive association between FABP4 and natural killer cells. Drug sensitivity testing identified efficacious drugs for patients with CRC, such as midostaurin and lenalidomide. Notably, RT-qPCR confirmed elevated expression levels for CXCL1 and OFCC1 and reduced levels for FABP4, PPARGC1A, and CPT2 in patients with CRC. CONCLUSION: Five prognostic LRGs were determined for CRC, and a new risk model was developed and validated, revealing the critical role of LRGs in CRC and improving our understanding of clinical interventions for this cancer type.
Xiong et al. (2026) studied this question.