Prognostic and predictive significance of altered microRNA expression in pancreatic cancer

Introduction: Pancreatic cancer (PC) remains one of the most lethal malignancies with a dismal 5-year survival rate of 13%. MicroRNAs (miRNAs) have emerged as critical regulators in tumor progression, yet their clinical relevance in PC requires validation through large-scale datasets. Objective: This study aims to detect irregularities in miRNA expression within PC patients by analyzing data sourced from the Cancer Genome Atlas (TCGA) database. Methods: Expression profiles of 14 miRNAs in 183 PC patients were analyzed using data from the TCGA database, correlating them with survival, tumor-node-metastasis stage, tumor grade, recurrence, and lymph node metastasis. Molecular mechanisms were explored via target prediction and correlation analysis. Results: Reduced miR-107 expression and elevated miR-21 expression were each independently linked to worse overall survival (p<0.05). Lower miR-107 levels were correlated with advanced stage (Stage I vs. Stage II: p<0.05) and lymph node metastasis (p<0.05), whereas higher miR-21 levels were associated with higher T stage (T3+4 vs. T1+2: p<0.0001), advanced overall stage (Stage I vs. Stage II: p<0.0001), higher tumor grade (G3+4 vs. G1: p<0.001) and increased recurrence rates (*p<0.05). Mechanistically, the degradation of oncogenic miR-21—mediated by poly(A) polymerase-associated protein D5—was disrupted (p<0.001), leading to the overexpression of miR-21, thereby reducing the expression of tumor suppressor gene programmed cell death 4 (p<0.0001). MiR-107 was a candidate for targeting Notch2 and cyclin-dependent kinase 6 (p<0.05). Conclusion: MiR-21 and miR-107 are pivotal prognostic biomarkers in PC, warranting further exploration as therapeutic targets.