Diagnostic Role of Extracellular Vesicles in Cancer: A Comprehensive Systematic Review and Meta-Analysis

Background: Cancer-derived extracellular vesicles (EVs) are regarded to have significant function in most steps during cancer progression. This meta-analysis aims to investigate the accuracy of EVs as a biomarker in cancer diagnosis. Methods: The diagnostic efficacy of EVs for different cancers was assessed using pooled sensitivity and specificity, diagnostic odds ratio (DOR), and overall area under the curve (AUC) of the summary receiver operating characteristic (SROC). The positive likelihood ratio (PLR) and negative likelihood ratio (NLR) were verified to estimate the diagnostic efficacy of EV at a clinical level. Results: In all, 6,183 cancer patients and 2,437 healthy controls from 75 eligible studies reported in 42 publications were included in the study. The overall pooled sensitivity, specificity, PLR, NLR, and DOR were 0.62 (95% CI: 0.60-0.63), 0.76 (95% CI: 0.75-0.78), 3.07 (95% CI: 2.52-3.75), 0.34 (95% CI: 0.28-0.41), and 10.98 (95% CI: 7.53-16.00), respectively. Similarly, the AUC of the SROC was 0.88, indicating a high conservation of EVs as an early diagnostic marker. Furthermore, subgroup analysis suggested that the use of small EVs as a biomarker was more accurate in serum-based samples of nervous system cancer (p < 0.001). As a result, ultracentrifugation and quantification and size determination methods, such as Western blotting and ELISA were the most reliable identification methods for EV detection. We also indicated that increased secretion of EVs made them a capable biomarker for diagnosing cancer in elderly European individuals. Conclusions: Our study provides evidence that EVs are a promising non-invasive biomarker for cancer diagnosis. Well-designed cohort studies should be conducted to warrant the clinical diagnostic value of EVs.