Diagnostic and Prognostic Accuracy of MiRNAs in Pancreatic Cancer: A Systematic Review and Meta-Analysis

Abstract

Pancreatic cancer (PC) remains a significant contributor to global cancer mortality, with limited effective diagnostic and prognostic tools. MicroRNAs (miRNAs) have emerged as promising biomarkers for PC diagnosis and prognosis. A comprehensive literature search was conducted in PubMed, Web of Science, and Scopus. Studies reporting sensitivity, specificity or area under the curve (AUC) for miRNAs in PC diagnosis, as well as hazard ratios (HRs) for survival evaluations, were included. Data extraction and quality assessment followed PRISMA guidelines. Meta-analyses were conducted using appropriate statistical methods. The protocol is registered in PROSPERO. Diagnostic analysis included 290 evaluations, revealing an overall AUC of 0.8226 for PC diagnosis. Subgroup analyses showed varying accuracies, with blood and tissue specimens yielding higher AUC values. Promising miRNAs with AUC values above 0.8 included miR-320, miR-1290, miR-93, miR-25, miR-451, miR-20, miR-21, miR-223 and miR-122. Prognostic analysis encompassed 46 studies, indicating significant associations between miRNA expression and overall survival (OS) and progression-free survival (PFS). The combined HR for studies reporting OS HRs higher than one was 1.7613 (95% CI: 1.5394–2.0152, p < 0.0001; I² = 81.7%). Notable miRNAs with prognostic significance included miR-10, miR-21 and miR-221. Studies reporting OS HRs less than one had a pooled HR of 0.6805 (95% CI: 0.5862–0.7901, p < 0.0001; I² = 65.4%). MiRNAs hold promise as diagnostic and prognostic biomarkers for PC. Blood and tissue specimens offer superior diagnostic accuracy, and several miRNAs show potential for predicting patient outcomes.