Integrative genomic profiling identifies MLPH as a candidate gene in prostate cancer

Abstract

Background Prostate cancer (PCa) is a highly heterogeneous malignancy with complex genetic underpinnings. This study integrates multi-omics data to prioritize candidate susceptibility genes and evaluate their functional and clinical significance in PCa pathogenesis. Materials and methods We integrated PCa GWAS summary statistics with GTEx v8 expression quantitative trait locus reference panels to perform cross-tissue and single-tissue transcriptome-wide association studies. Candidate signals were refined using conditional analysis, MAGMA and fastBAT gene-level tests, Summary data-based Mendelian randomization, and Bayesian colocalization. Tumor-context cis-eQTL evidence from TCGA-PRAD was incorporated to prioritize regulatory signals retained in prostate cancer tissues. Prioritized candidates were further assessed using transcriptomic datasets, Human Protein Atlas immunohistochemistry, single-cell RNA-seq analysis, histological grading, preoperative PSA, and established genomic risk signatures. Gene network and pathway enrichment analyses were performed to explore potential biological context. Results The integrative genetic analyses identified 23 consensus candidate genes supported by multiple association frameworks. SMR and Bayesian colocalization further narrowed the candidate list, and tumor-context cis-eQTL analysis in TCGA-PRAD retained MLPH as the final prioritized candidate. The lead variant rs7582964 was significantly associated with MLPH expression in PRAD tumor tissues. MLPH was upregulated in PCa tissues compared with normal prostate tissues in TCGA-PRAD and showed concordant expression patterns in an independent GEO cohort and Human Protein Atlas immunohistochemistry data. Single-cell transcriptomic analysis localized MLPH expression mainly to epithelial cells, with the strongest signal observed in tumor epithelial cells. Clinically, MLPH expression was associated with histological differentiation, with reduced expression in the most poorly differentiated tumors. Lower MLPH expression also correlated with higher preoperative PSA and higher Decipher-like genomic risk scores. Functional analyses linked MLPH to vesicle-mediated transport, exocytosis, and hormone-related signaling pathways. Discussion and conclusion This integrative genomic analysis prioritizes MLPH as a candidate susceptibility gene for PCa and links its regulatory signal to tumor-context expression, protein-level evidence, cellular localization, and clinically relevant molecular features. These findings support a potential role for MLPH in PCa biology, particularly in relation to vesicle trafficking and tumor differentiation.