Circulating microRNA-21 as a Non-Invasive Diagnostic Marker for Colorectal Cancer: A PRISMA-Compliant Systematic Review and Meta-Analysis with Independently Reconstructed Contingency Data

Abstract

Background. Circulating microRNAs (miRNAs) are short, nuclease-resistant regulatory RNAs that persist in serum and plasma, a property that has made them attractive candidates for blood-based cancer testing. miR-21 is the oncomiR most often found up-regulated in colorectal cancer (CRC), and many groups have proposed measuring it in blood as a screening or monitoring tool. Reported accuracy, however, varies widely from one study to the next, which has left its true diagnostic value unsettled. Methods. We searched five databases for diagnostic studies of circulating miR-21 in CRC that supplied, or allowed reconstruction of, a full two-by-two table. Eighteen studies (1,217 cases; 962 controls) met the criteria. For every study we rebuilt the contingency table from the reported accuracy and group sizes, recomputed sensitivity and specificity with exact Clopper–Pearson intervals, and pooled the estimates with DerSimonian–Laird random-effects models on the logit scale. A Moses–Littenberg summary ROC (SROC) curve gave the area under the curve (AUC), Deeks' test screened for small-study effects, and a specimen-stratified analysis probed heterogeneity. Results. Pooled sensitivity was 0.75 (95% CI 0.69–0.81) and pooled specificity 0.81 (95% CI 0.76–0.86). The positive and negative likelihood ratios were 4.04 and 0.30, the diagnostic odds ratio 14.9 (95% CI 9.4–23.7), and the SROC AUC 0.87. Heterogeneity was high (I² = 78% for sensitivity, 70% for specificity). Deeks' funnel plot showed no significant asymmetry (p = 0.91). Plasma-based assays were more specific than serum-based ones (0.87 vs 0.79). Conclusions. Blood miR-21 rules CRC in more reliably than it rules it out, so a single measurement is not sufficient as a stand-alone screen. Its accuracy is good enough, though, to justify a place inside standardised multi-marker liquid-biopsy panels. The reproducible analysis pipeline reported here transfers directly to other circulating-miRNA markers.