DOCK3 Orchestrates Metastasis and Immune Microenvironment in Prostate Cancer

Ming ZhangJiaxue HanHaipeng ZhouQiao XiongXin ZhongPing Tan^*

• West China School of Medicine, West China Hospital, Sichuan University, Chengdu, China

introduction: Prostate cancer (PCa) is a leading cause of male cancer mortality, with metastasis and immune evasion posing major therapeutic challenges. DOCK3, a guanine nucleotide exchange factor implicated in cytoskeletal dynamics, is poorly characterized in PCa. This study investigates DOCK3's role in PCa metastasis and tumor immune microenvironment (TIME) remodeling. Methods : Multi-omics analyses integrated bulk RNA-seq from TCGA-PRAD (499 tumors/52 normals), scRNA-seq from GEO (45,325 cells), and genomic data. We performed:Differential expression analysis (DESeq2),Immune deconvolution (CIBERSORT,ssGSEA, xCell),WGCNA co-expression networks,Tumor mutational burden (TMB) assessment,Distant metastasis (M1 vs. M0) association studies,scRNA-seq clustering (Harmony/UMAP) and DE testing.Statistical significance thresholds: |log2FC|>1, padj<0.05. Results : DOCK3 expression was found to be significantly elevated in metastatic (M1) tumors compared to primary (M0) tumors (p<0.05) and demonstrated a strong positive correlation with a higher tumor mutational burden (TMB) in metastatic samples (p<0.001). Cellular specificity analysis revealed that DOCK3 was exclusively and highly enriched within malignant epithelial and stromal cells, specifically in Cluster 6, where it exhibited a log2 fold-change of 9.13 (padj<1e-200) and was expressed in 54% of cells, compared to a negligible presence in all other clusters. In the tumor microenvironment, elevated DOCK3 expression was associated with a significant increase in cytotoxic immune infiltration, notably of CD8⁺ T and Natural Killer cells, a finding consistently supported by multiple computational algorithms (all p<0.05). Clinically, a high level of DOCK3 was significantly associated with metastatic status (p<0.01), whereas high expression of CDKN3 was correlated with advanced disease features, including higher Gleason scores (3-5) and T-stage (T2-T4) (p<0.01). Furthermore, significant differences in immune infiltration patterns were observed between clusters. Pathway enrichment analysis of genes co-expressed with DOCK3, identified through the WGCNA Green Module, indicated significant involvement in biological processes such as cytoskeletal reorganization, muscle contraction, and metabolic pathways (FDR<0.01). Conclusion: DOCK3 drives PCa metastasis through cytoskeletal dynamics while paradoxically promoting an immunologically active microenvironment. Its tumor-specific expression and association with aggressive clinical features nominate DOCK3 as a novel biomarker for risk stratification and a promising therapeutic target for combinatorial immunotherapy in immunologically "cold" PCa.