Abstract

Objective: To explore fatty acid metabolism-related genes and signature, which could predict survival outcomes of clear cell renal carcinoma (ccRCC) patients.

Materials and Methods: Transcriptional and survival data of fatty acid genes in ccRCC patients were retrieved from UCSC Xena and Geo DataSets. We first performed Lasso Cox regression analysis to identify survival-related genes. These genes were then used to construct metabolic-related gene signature and risk score. Enrichment analysis and immune component and chemotherapy response prediction were also performed.

Results: In total, five survival-related genes were identified: AGR2, HAO2, IGF2BP1, MCCD1 and OLFM4 (p < 0.05). A series of survival value analyses revealed survival-related signature and risk score, including KM analysis (training set: p < 0.001; test set: p = 0.008). Four clinical indexes (T stage, N stage, M stage, and pathology) were positively correlated with risk score. Time-dependent ROC analysis yielded AUC value of 0.813. Immune landscape analysis revealed that risk score was strongly correlated with TAM score and cytotoxic score. Patients with high risk score and TAM score or cytotoxic score had the shortest survival time. Finally, inhibition of fatty acid metabolism in human ccRCC cell line produced corresponding changes in five genes, consistent with our preliminary results.

Conclusion: We identified five survival-related genes (AGR2, HAO2, IGF2BP1, MCCD1 and OLFM4) in ccRCC patients. Our results also indicated that survival-related signature based on these genes is a potential robust prognostic biomarker for ccRCC in patients.