Prognostic and diagnostic value of long non-coding RNAs in esophageal squamous cell carcinoma: a systematic review and meta-analysis

Abnormally-expressed long non-coding RNAs (lncRNAs) may be effective prognostic or diagnostic biomarkers for esophageal squamous cell carcinoma (ESCC). To overcome the shortcomings and inaccuracies of single studies, the clinical value of lncRNAs as predictive ESCC biomarkers was examined through systematic review and meta-analysis. Thirty-five studies, including 26 on clinicopathological features, 27 on prognosis, and 4 on diagnosis were selected from electronic databases. Among ESCC clinicopathological features, HOTAIR and MALAT-1 overexpression correlated with lymph node metastasis (OR = 3.29, 95% CI: 1.18-9.16, P = 0.02; OR = 1.77, 95% CI: 1.04-3.00, P = 0.04, respectively). HOTAIR and AFAP1-AS1 overexpression correlated with tumor-node-metastasis stage (OR = 6.93, 95% CI: 2.79-17.18, P < 0.0001; OR = 2.92, 95% CI: 1.66-5.13, P = 0.0002, respectively). HOTAIR upregulation and MEG3 downregulation correlated with shorter overall survival (HR = 2.01, 95% CI: 1.56-2.58, P < 0.00001; HR = 0.47, 95% CI: 0.25-0.88, P = 0.02, respectively). Evaluation of the clinical performance of all lncRNAs as ESCC predictors yielded a sensitivity of 81%, specificity of 74%, and diagnostic odds ratio of 17.44, with an AUC of 0.87 (95% CI: 0.84-0.90). These results highlight the potential prognostic and diagnostic value of ESCC-related lncRNAs.