Cross-cohort transcriptional and network signatures of human papillomavirus-associated cancers

Introduction: Persistent human papillomavirus (HPV) infection contributes substantially to the global burden of virus-associated cancers, including those of the cervix, vulva, anus, penis, and oropharynx. HPV-positive tumors frequently exhibit distinct clinical characteristics. The molecular mechanisms and biomarkers that unify HPV-associated cancers across various tissue types remain poorly characterized. Materials and methods: Sixty bulk RNA-seq datasets representing three tumor types were retrieved and processed uniformly. A meta-analysis was performed to assess transcriptomic similarity among the tumor cohorts. Differential gene expression analysis was carried out to characterize HPV-positive and HPV-negative expression patterns. Weighted gene co-expression network analysis (WGCNA) was performed to identify HPV-associated gene modules. Hub genes were selected using Module Membership and Gene Significance criteria and further prioritized through cohort-specific protein–protein interaction analysis. Results: Differential expression analysis identified distinct transcriptional differences associated with HPV status across all three cohorts, although the magnitude of differential expression varied between tumor types. Cross-cohort meta-analysis identified a 2959-gene transcriptional program reaching genome-wide significance under a fixed-effect model, with 557 genes remaining significant under a heterogeneity-adjusted random-effects model. Gene set enrichment analysis identified 1578 pathways reaching false discovery rate (FDR)-corrected significance, predominantly related to immune response, DNA repair, and DNA replication. WGCNA identified HPV-associated co-expression modules whose hub genes were enriched for immune regulation, adaptive immune activation, and antiviral host-defense processes. Conclusions: Integrated differential expression, heterogeneity-aware meta-analysis, and co-expression network analyses identified transcriptional patterns associated with HPV status across the included cohorts. These findings provide candidate genes and biological pathways for future validation and functional investigation in independent cohorts of HPV-associated cancers.