Abstract
Background: Prostate cancer remains one of the leading causes of cancer-related mortality in men worldwide. The treatment of it is currently based on surgical removal, radiotherapy, and hormone therapy. It is crucial to improve therapeutic prospects for the diagnosis and treatment of prostate cancer via drug target screening. Methods: We integrated eQTL data from the eQTLGen Consortium and pQTL data from UK Biobank Proteome Plasma Proteins (UKB-PPP) and deCODE health datasets. MR analyses (SMR, heterogeneity in dependent instruments (HEIDI), IVW, Wald ratio, weighted median, and MR-Egger) were used to screen candidate genes associated with prostate adenocarcinoma (PRAD) risk. Candidate genes were further verified through TCGA-based gene expression profile, survival analysis, and immune microenvironment evaluations. TIDE analysis was utilized to investigate gene immunotherapy response. Single-cell RNA sequencing data from the GSE176031 dataset were used to investigate the gene expression patterns. The Drug Bank, Therapeutic Target Database and Drug Signatures Database were utilized to predict targeted drugs for candidate genes. Results: MTHFD1 and LGALS4 were identified as promising therapeutic targets for PRAD, with evidence provided at multi-omics levels. LGALS4 was predominantly expressed in malignant cells and was correlated with enhanced immune checkpoint pathways, increased TIDE scores, and immunotherapy resistance. In contrast, MTHFD1was expressed in both tumor and microenvironmental cells and was associated with poor survival. Drug target prediction suggested that there are no currently approved drugs specifically targeting MTHFD1 and LGALS4. Conclusions: Our study identified MTHFD1 and LGALS4 as potential preventive targets for PRAD. However, future experiments are warranted to assess the utility and effectiveness of these candidate proteins.