RESUMO
BACKGROUND: Circulating tumour DNA (ctDNA)-based molecular residual disease (MRD) detection technology has been widely used for recurrence evaluation, but there is no agreement on the efficacy of assessing recurrence and overall survival (OS) prognosis, as well as the sensitivity and specificity of landmark detection and longitudinal detection. METHODS: We systematically searched Pubmed, Embase, Cochrane, and Scopus for prospective studies or randomized controlled trials that collected blood samples prospectively. The search period was from Jan 1, 2013, to Sept 10, 2023. We excluded retrospective studies. The primary endpoint was to assess the hazard ratio (HR) between circulating tumour DNA positive (ctDNA+) and negative (ctDNA-) for recurrence-free survival incidence (RFS), disease-free survival (DFS), progression-free survival (PFS), event-free survival (EFS), time to recurrence (TTR), distant metastasis-free survival (DMFS) or OS in patients with resectable cancers. We calculated the pooled HR of recurrence and OS and 95% confidence interval (CI) in patients with resected cancers using a random-effects model. Pooled sensitivity and specificity were estimated using the bivariate random effects model. FINDINGS: This systematic review and meta-analysis returned 7578 records, yielding 80 included studies after exclusion. We found that the HR of recurrence across all included cancers between patients with ctDNA+ and ctDNA- was 7.48 (95% CI 6.39-8.77), and the OS was 5.58 (95% CI 4.17-7.48). We also found that the sensitivity, area under the summary receiver operating characteristic curve (AUSROC) and diagnostic odds ratio (DOR) of longitudinal tests were higher than that of landmark tests between patients with ctDNA+ and ctDNA- (0.74, 95% CI 0.68-0.80 vs 0.50, 95% CI 0.46-0.55; 0.88 vs. 0.80; 25.70, 95% CI 13.20-45.40 vs. 9.90, 95% CI 7.77-12.40). INTERPRETATION: Postoperative ctDNA testing was a significant prognosis factor for recurrence and OS in patients with resectable cancers. However, the overall sensitivity of ctDNA-MRD detection could be better. Longitudinal monitoring can improve the sensitivity, AUSROC, and DOR. FUNDING: Special fund project for clinical research of Qingyuan People's Hospital (QYRYCRC2023006), plan on enhancing scientific research in GMU (GZMU-SH-301).