Diagnostic Value of CircRNAs as Potential Biomarkers in Oral Squamous Cell Carcinoma: a Meta-Analysis

INTRODUCTIONCircular RNAs (CircRNAs), an emerging non-coding RNA, have been demonstrated to be involved in tumorigenesis, metastasis, and cancer progression, and could represent novel potential biomarkers for diagnosing oral squamous cell carcinoma (OSCC). However, no meta-analysis has investigated the diagnostic role of circRNAs in OSCC. Hence, to investigate whether circRNAs could serve as specific biomarkers for OSCC, the present systematic review and meta-analysis evaluated the diagnostic efficiency of circRNAs in patients with OSCC. MATERIALS AND METHODSA thorough search of online databases (Pubmed, Web of Science, Embase, and the Cochrane Library) was conducted to collect relevant studies up to March 30th, 2021. All eligible studies were case-control studies. The quality of each study was evaluated by the Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool. STATA (version 15.1) and Review Manager (version 5.4) were employed to conduct the meta-analysis, and the PRISMA statement was adopted in this study. RESULTSA total of 16 studies were included in the meta-analysis, with five studies on upregulated circRNAs, and 11 on downregulated circRNAs. The enrolled studies that met our eligibility criteria all derived from China. The pooled sensitivity (SEN), specificity (SPE), diagnostic odds ratio (DOR), positive likelihood ratio (PLR), negative likelihood ratio (NLR), and the area under receiver operating characteristics curve (AUC) with the 95% confidence intervals (95% CIs) were 0.74 (0.69-0.79), 0.79 (0.73-0.84), 10.74 (7.81-14.77), 3.50 (2.78-4.45), 0.33 (0.27-0.39) and 0.83 (0.79-0.86), respectively. The subgroup analysis demonstrated that serum, plasma, and saliva specimens had a better diagnostic performance than tissue samples, with a high value of sensitivity, specificity, DOR, and AUC values. The results also showed that the subgroups of upregulated circRNAs and a sample size of ≥100 manifested higher specificity, DOR, and AUC for cancer detection than downregulated circRNAs and a sample size of < 100. CONCLUSIONSA strong association was demonstrated between the dysregulated expression of circRNAs and the diagnosis of OSCC. Hence, circRNAs have the potential to function as promising biomarkers and therapeutic targets for OSCC. SYSTEMATIC REVIEW REGISTRATIONPROSPERO, number CRD42021256857.