Integrated transcriptomic and network-based analysis identifies miR-3187-3p–SGK2 as a candidate co-dysregulated molecular pair in breast cancer

Breast cancer is a heterogeneous malignancy underpinned by complex genomic and transcriptomic alterations. Integrative analysis of large-scale transcriptomic data offers a powerful strategy for identifying robust molecular signatures associated with disease development and progression. This study aimed to identify and validate candidate molecular biomarkers in breast cancer through integrated transcriptomic and network-based approaches. Four breast cancer microarray datasets comprising 321 tissue samples (271 tumours and 50 normal breast tissues) were retrieved from the Gene Expression Omnibus (GEO) database and subjected to integrative analysis. Differential expression analysis was performed using the limma package, followed by weighted gene co-expression network analysis (WGCNA), protein–protein interaction (PPI) network construction, functional enrichment analysis, and competing endogenous RNA (ceRNA) network modeling. Candidate biomarkers were subsequently evaluated in independent public datasets and experimentally validated by RT-qPCR in 23 breast cancer and 14 normal breast tissue specimens. A total of 7,547 genes were analyzed across the integrated dataset. Applying predefined significance thresholds (|log 2 FC|>1.5; adjusted p < 0.05), 218 differentially expressed genes were identified, comprising 63 upregulated and 155 downregulated genes. Convergence analysis of WGCNA and differential expression results yielded 150 candidate genes for downstream investigation. PPI network topology analysis identified IGF1, IGFBP6, and CAV1 as hub genes, while pathway enrichment analysis further implicated AKT2, FOXO1, and PDK1, in the PI3K/AKT signaling pathway. ceRNA network analysis identified miR-3187-3p as a candidate microRNA potentially associated with SGK2. Validation in independent datasets confirmed significant dysregulation of SGK2 expression, while RT-qPCR demonstrated significant downregulation of both SGK2 and miR-3187-3p in breast cancer tissues compared with normal controls. Receiver operating characteristic analysis indicated promising diagnostic performance for miR-3187-3p, whereas SGK2 did not demonstrate statistically significant performance in the validation cohort. These findings should be interpreted cautiously and require validation in larger, independent cohorts. Through multi-layered transcriptomic integration, network-based modeling, independent cohort validation, and experimental confirmation, this study identified the miR-3187-3p/SGK2 pair as a candidate co-dysregulated molecular signature associated with breast cancer. These findings provide additional insight into the molecular network underlying breast cancer and identify candidate biomarkers for further mechanistic investigation and validation in larger, clinically well-characterized cohorts.