Performance of screening tests for esophageal squamous cell carcinoma: a systematic review and meta-analysis
Martin C. S. Wong, Yunyang Deng, Junjie Huang, Yijun Bai, Haoxiang Wang, Jinqiu Yuan, Lin Zhang, Hon Chi Yip, Philip Wai Yan Chiu
Abstract
BACKGROUND AND AIMS: This systematic review and meta-analysis aims to compare the pooled diagnostic accuracy of the currently available esophageal squamous cell carcinoma (ESCC) screening tests. METHODS: A comprehensive literature search of Embase and Medline (up to October 31, 2020) was performed to identify eligible studies. We pooled sensitivity, specificity, positive likelihood ratio, negative likelihood ratio, and diagnostic odds ratio for ESCC screening tools using a bivariate random-effects model. The summary receiver operating characteristic curves with area under the curve (AUC) were plotted for each screening test. RESULTS: We included 161 studies conducted in 81 research articles involving 32,209 subjects. The pooled sensitivity, specificity, and AUC of the major screening tools were respectively as follows: endoscopy (peroral endoscopy): .94 (95% confidence interval [CI], .87-.97), .92 (95% CI, .87-.95), and .97 (95% CI, .96-.99); endoscopy (transnasal endoscopy): .85 (95% CI, .70-.93), .96 (95% CI, .91-.98), and .97 (95% CI, .95-.98); microRNA: .77 (95% CI, .75-.80), .78 (95% CI, .75-.80), and .85 (95% CI, .81-.87); autoantibody: .45 (95% CI, .36-.53), .91 (95% CI, .89-.93), and .84 (95% CI, .81-.87); and cytology: .82 (95% CI, .60-.93), .97 (95% CI, .88-.99), and .97 (95% CI, .95-.98). There was high heterogeneity. CONCLUSIONS: The diagnostic accuracy seemed to be comparable between cytology and endoscopy, whereas autoantibody and microRNAs bear potential as future noninvasive screening tools for ESCC. To reduce ESCC-related death in high-risk populations, it is important to develop a more accurate and less-invasive screening test.