Background Circulating tumor DNA (ctDNA) assays are emerging as promising non-invasive tools for colorectal cancer (CRC) screening. This updated systematic review and meta-analysis evaluated the diagnostic accuracy of blood-based ctDNA assays for early CRC detection in asymptomatic adults. Methods We included prospective and cross-sectional diagnostic accuracy studies comparing ctDNA assays (mutation, methylation, or fragment-based) against colonoscopy or histopathology. Study-level 2×2 data were extracted to calculate sensitivity and specificity. Pooled estimates were derived using a Reitsma bivariate random-effects model. Secondary measures like diagnostic odds ratio (DOR), positive and negative likelihood ratios (PLR, NLR), positive predictive value (PPV), and negative predictive value (NPV) were synthesized on transformed scales and back-transformed for reporting. Subgroup analyses by assay type and meta-regression explored heterogeneity. Risk of bias was assessed with QUADAS-2 and certainty of evidence with GRADE. Results Ten studies met inclusion criteria, nine contributing to pooled estimates. The pooled sensitivity was 0.887 (95% CI, 0.848–0.920) and specificity 0.913 (95% CI, 0.898–0.928). The SROC curve yielded an AUC of 0.97, indicating excellent discrimination. Secondary pooled metrics included a DOR of 83.9, PLR of 9.2, and NLR of 0.12. PPV was low (99%) at screening prevalences of 0.5–2%. Subgroup analyses by assay type revealed no significant differences in diagnostic accuracy. Conclusions Blood-based ctDNA assays show high pooled sensitivity, specificity, and discrimination for CRC detection, but limited predictive value at screening prevalence and study heterogeneity constrain clinical adoption. Large, standardized, population-based studies are warranted.
Guntupalli et al. (Mon,) studied this question.