Abstract
BACKGROUND: Circulating tumor DNA (ctDNA), as a liquid biopsy biomarker, is undergoing extensive evaluation for its clinical utility across multiple tumor management scenarios. This study aimed to assess the prognostic value of ctDNA in esophageal cancer (EC) patients at different treatment time points through a systematic review and meta-analysis. MATERIALS AND METHODS: A comprehensive search of the PubMed, Embase, and Cochrane Library databases from construction to October 2024 was conducted, and studies investigating the association between ctDNA and progression-free survival (PFS) and overall survival (OS) in EC patients were screened for inclusion. Primary outcomes included PFS/OS by ctDNA status at different time points (baseline, after neoadjuvant therapy, and during follow-up). Risk ratios (HRs) for PFS/OS with positive ctDNA tests at various time points were combined, and subgroup analyses were conducted for tumor-informed and non-tumor-informed testing of ctDNA. RESULTS: A total of 22 studies involving 1519 patients were finally included in this meta-analysis. In univariate analyses, detection of ctDNA was associated with poorer PFS at baseline (HR = 1.64, 95% CI:1.30-2.07), after neoadjuvant therapy (HR = 3.97, 95% CI: 2.68-5.88) and during follow-up (HR = 5.42, 95% CI:3.97-7.38). Similarly, detection of ctDNA at all time points was associated with poorer OS (at baseline: HR = 2.02, 95% CI:1.36-2.99; after neoadjuvant therapy: HR = 3.41, 95% CI: 2.08-5.59; and during follow-up: HR = 4.93, 95% CI:3.31-7.34). Similar PFS and OS outcomes were observed in multivariate analyses. The uni- and multivariate combined HRs for PFS/OS with ctDNA detected at different time points were numerically related as: baseline < after neoadjuvant therapy < during follow-up(PFS:1.90→4.07→5.22; OS:2.39→3.15→5.37). When ctDNA was detected in combined tumor-informed assays at baseline and after neoadjuvant therapy, most HRs for recurrence and mortality risk showed a trend toward higher values compared with non-tumor-informed assays. ctDNA test positivity predicted clinical recurrence an average of 4.53 months earlier (range: 0.98-11.6 months) than conventional radiological imaging techniques. CONCLUSIONS: Positive ctDNA testing was associated with poorer prognosis throughout the treatment period of EC patients, and the prognostic value of monitoring ctDNA status increased with time from baseline to follow-up. SYSTEMATIC REVIEW REGISTRATION: https://www.crd.york.ac.uk/PROSPERO/view/CRD42024612909, identifier CRD42024612909.