Comprehensive analysis of fatty acid metabolism-related gene signatures for predicting prognosis in patients with prostate cancer

Abstract

Fatty acid metabolism (FAM) is an important factor in tumorigenesis and development. However, whether fatty acid metabolism (FAM)-related genes are associated with prostate cancer (PCa) prognosis is not known. Therefore, we established a novel prognostic model based on FAM-related genes to predict biochemical recurrence in PCa patients. First, PCa sequencing data were acquired from TCGA as the training cohort and GSE21032 as the validation cohort. Second, a prostate cancer prognostic model containing 10 FAM-related genes was constructed using univariate Cox and LASSO. Principal component analysis and t-distributed stochastic neighbour embedding analysis showed that the model was highly effective. Third, PCa patients were divided into high- and low-risk groups according to the model risk score. Survival analysis, ROC curve analysis, and independent prognostic analysis showed that the high-risk group had short recurrence-free survival (RFS), and the risk score was an independent diagnostic factor with diagnostic value in PCa patients. External validation using GSE21032 also showed that the prognostic model had high reliability. A nomogram based on a prognostic model was constructed for clinical use. Fourth, tumor immune correlation analyses, such as the ESTIMATE, CIBERSORT algorithm, and ssGSEA, showed that the high-risk group had higher immune cell infiltration, lower tumour purity, and worse RFS. Various immune checkpoints were expressed at higher levels in high-risk patients. In summary, this prognostic model is a promising prognostic biomarker for PCa that should improve the prognosis of PCa patients. These data provide new ideas for antitumour immunotherapy and have good potential value for the development of targeted drugs.