MicroRNAs as novel biomarkers for pancreatic cancer diagnosis: a meta-analysis based on 18 articles

Zhongyang Ding,Haorong Wu,Jiaming Zhang,Guorong Huang,Dongdong Ji
DOI: https://doi.org/10.1007/s13277-014-2133-4
2014-01-01
Abstract:Dysregulated microRNAs (miRNAs) have been reported to be associated with pancreatic cancer (PaC), suggesting that they may serve as useful novel diagnostic biomarkers for PaC. Various studies have been performed to investigate the diagnostic value of miRNAs for PaC but have obtained conflicting results. Therefore, this meta-analysis aims to comprehensively and quantitatively evaluate the potential diagnostic value of miRNAs for PaC. We systematically searched PubMed, Embase, Google Scholar, Cochrane Library, and Chinese National Knowledge Infrastructure for publications concerning the diagnostic value of miRNAs for PaC without language restriction. The quality of each study was scored using the revised Quality Assessment of Diagnostic Accuracy Studies (QUADAS-2). The summary receiver operator characteristic curve and other parameters were applied to check the overall test performance. Heterogeneity was tested with the I 2 test and publication bias was tested with the Deek’s funnel plot asymmetry test. This meta-analysis included 18 articles with a total of 2,036 patients and 1,444 controls. The pooled sensitivity was 82 % (95 % CI, 78–86 %); the specificity was 77 % (95 % CI, 73–81 %); the PLR was 3.6 (95 % CI, 3.0–4.4); the NLR was 0.23 (95 % CI, 0.18–0.29); the DOR was 16 (95 % CI, 10–24); and the AUC was 0.86 (95 % CI, 0.83–0.89). Subgroups analyses were also performed and revealed that there were significant differences between some subgroups: the multiple-miRNAs profiling-based assays, non-blood-based assays, and healthy control-based studies all showed higher accuracies in diagnosing PaC than that of their counterparts. This meta-analysis suggests that the use of miRNAs has potential diagnostic value with a relatively high sensitivity and specificity for PaC, particularly the use of multiple miRNAs for discriminating PaC patients from healthy individuals. More prospective studies on the diagnostic value of miRNAs for PaC are needed in the future.
What problem does this paper attempt to address?