WGCNA-ML-MR Integration: Uncovering Immune-Related Genes in Prostate Cancer

Jing Lv ¹ Yuhua Zhou ¹ Shengkai Jin ¹ Chaowei Fu ¹ Bo Liu ² Yang Shen ³ Menglu Li ⁴ Yuwei Zhang ² Ninghan Feng ^1,4*

• ¹ Wuxi Medical College, Jiangnan University, Wuxi, Jiangsu Province, China
• ² School of Medicine, Nantong University, Nantong, Jiangsu Province, China
• ³ Nanjing Medical University, Nanjing, Jiangsu Province, China
• ⁴ Jiangnan University Medical Center (JUMC）, Wuxi, Liaoning Province, China

Background: Prostate cancer is one of the most common tumors in men, with its incidence and mortality rates continuing to rise year by year. Prostate-specific antigen (PSA) is the most commonly used screening indicator, but its lack of specificity leads to overdiagnosis and overtreatment. Therefore, identifying new biomarkers related to prostate cancer is crucial for the early diagnosis and treatment of prostate cancer.Methods: This study utilized datasets from the Gene Expression Omnibus (GEO) to screen for differentially expressed genes (DEGs) and employed Weighted Gene Coexpression Network Analysis (WGCNA) to identify driver genes highly associated with prostate cancer within the modules. The intersection of differentially expressed genes and driver genes was taken, and Kyoto Encyclopedia of Genes and Genomes (KEGG) and Gene Ontology (GO) enrichment analyses were performed. Furthermore, a machine learning algorithm was used to screen for core genes and construct a diagnostic model, which was then validated in an external validation dataset. The correlation between core genes and immune cell infiltration was analyzed, and Mendelian randomization (MR) analysis was conducted to identify biomarkers closely related to prostate cancer.Results: This study identified six core biomarkers: SLC14A1, ARHGEF38, NEFH, MSMB, KRT23, and KRT15. MR analysis demonstrated that MSMB may be an important protective factor for prostate cancer. In q-PCR experiments conducted on tumor tissues and adjacent non-cancerous tissues from prostate cancer patients, it was found that: compared to the adjacent non-cancerous tissues, the expression level of ARHGEF38 in prostate cancer tumor tissues significantly increased, while the expression levels of SLC14A1, NEFH, MSMB, KRT23, and KRT15 significantly decreased. To further validate these findings at the protein level, we conducted Western blot analysis, which corroborated the q-PCR results, demonstrating consistent expression patterns for all six biomarkers. IHC results confirmed that ARHGEF38 protein was highly expressed in tumor tissues, while MSMB expression was markedly reduced.Our study reveals that SLC14A1, ARHGEF38, NEFH, MSMB, KRT23, and KRT15 are potential diagnostic biomarkers for prostate cancer, among which MSMB may play a protective role in prostate cancer.