Mapping autophagy-lncRNA co-expression networks in early ovarian carcinoma by stage-resolved transcriptomic modelling

Early-stage ovarian carcinoma exhibits heterogeneous molecular programming that influences tumor progression and therapeutic vulnerability. Using RNA-seq data from 96 tumors (GSE101108), we developed an integrative bioinformatics workflow that combines DESeq2, ssGSEA, and weighted gene co-expression network analysis (WGCNA) to identify autophagy-associated biomarkers and long non-coding RNA (lncRNA)- driven co-expression networks across FIGO stage I-II ovarian tumors. Differential expression analysis (DESeq2) identified 140 stage-dependent genes, with the lncRNA LOC105374020 emerging as the most upregulated transcript in FIGO II tumors. ssGSEA revealed elevated autophagy activity in FIGO II tumors, coinciding with the enrichment of MHC class I antigen-processing pathways, suggesting adaptive remodeling of immune surveillance. WGCNA uncovered a lncRNA-enriched module strongly correlating with autophagy intensity (r = 0.78, FDR = 1.8 × 10 −19 ). Within this module, LOC105374020 displayed high intramodular connectivity, making it a potential regulator of early tumor adaptation. These results suggest lncRNAs embedded within autophagy-linked modules as candidate biomarkers of early progression and immune modulation. Collectively, this study outlines a systems-level framework integrating autophagy, immunity, and lncRNA regulation in early ovarian carcinoma. While the findings highlight promising biomarker candidates, additional external validation and functional assays will be essential to translate these signatures for prognostic or therapeutic use. • Integrative RNA-seq workflow captures stage-specific autophagy dynamics in ovarian cancer. • SsGSEA reveals elevated autophagy activity and immune modulation in FIGO II tumors. • WGCNA identifies lncRNA-enriched modules linked to autophagy intensity. • LncRNA LOC105374020 emerges as a top candidate regulator of early progression. • Systems-level modelling provides a framework for network-based biomarker discovery.