Association between circulating tumor DNA and the prognosis of solid tumor patients receiving neoadjuvant therapy: a meta-analysis

Abstract

Circulating tumor DNA (ctDNA) has emerged as a promising noninvasive biomarker in solid tumors, yet its prognostic value following neoadjuvant therapy (NAT) warrants systematic validation. This study systematically searched databases such as PubMed, Embase, and the Cochrane Library, with a search period ranging from database establishment to October 2024. Two researchers independently screened the studies according to the inclusion/exclusion criteria, extracted data, and evaluated the quality of the included evidence, Stata 17.0 software (RRID:SCR_012763) was used to assess the correlation between positive ctDNA status and disease-free survival (DFS)/relapse-free survival (RFS), overall survival (OS), progression-free survival (PFS), event-free survival (EFS). Subgroup analysis was performed stratified by study design, detection time, region, tumor type, immunotherapy, detection methods, and statistical approaches (univariate vs multivariate analyses). The hazard ratio (HR) was used to analyze survival outcomes. The quality of cohort studies was assessed using the NOS, RCT studies were evaluated using ROB 2.0, and non-RCT studies were rated using ROBINS-I. Publication bias was assessed using a combination of Begg’s test and funnel plots. Sensitivity analysis was performed to evaluate the robustness of the results. Ultimately, 41 studies involving 3058 solid tumor patients were included in the analysis. The meta-analysis revealed a significant correlation between positive ctDNA status and worse survival outcomes, including DFS/RFS [HR = 8.60; 95% CI: 6.06–12.21], OS [HR = 7.00; 95% CI: 4.28–11.45], PFS [HR = 3.85; 95% CI: 2.35–6.30], EFS [HR = 3.40; 95% CI: 1.67–6.95]. In the subgroup analysis stratified by tumor type, ctDNA consistently predicted worse survival outcomes for patients with rectal cancer [DFS/RFS: HR = 9.23; 95% CI = 4.29–19.86; OS: HR = 9.72; 95% CI = 2.89–32.66], breast cancer [DFS/RFS: HR = 6.97; 95% CI = 4.20–11.56; OS: HR = 3.61; 95% CI = 1.76–7.40], esophageal cancer [DFS/RFS: HR = 10.83; 95% CI = 2.64–44.44; OS: HR = 2.85; 95% CI = 1.45–5.61], urothelial cancer [DFS/RFS: HR = 27.92; 95% CI = 12.16–64.09; OS: HR = 21.29; 95% CI = 10.51–43.11], lung cancer [DFS/RFS: HR = 4.59; 95% CI = 1.04–20.29; OS: HR = 25.00; 95% CI = 6.75–92.66], and pancreatic cancer [DFS/RFS: HR = 5.02; 95% CI = 1.23–20.51; OS: HR = 11.70; 95% CI = 1.26–108.88], colorectal liver metastasis [DFS/RFS: HR = 5.36; 95% CI = 2.63–10.90; OS: HR = 4.20; 95% CI = 1.50–11.78] receiving neoadjuvant therapy. However, in melanoma, ctDNA detection was not associated with DFS/RFS [HR = 10.8; 95% CI = 0.81–144.09]. In gastric cancer, ctDNA testing was associated with poor EFS [HR = 3.00; 95% CI = 1.30–6.91] but not with OS [HR = 7.17; 95% CI = 0.98–52.57]. Similarly, ovarian cancer was not associated with OS [HR = 2.46; 95% CI = 0.53–11.36]. Furthermore, subgroup analysis demonstrated that regardless of study design, detection time, region, immunotherapy, detection method, or univariate/multivariate analysis results, DFS/RFS, OS, PFS, and EFS were significantly poorer when ctDNA was detected positively. These findings suggest that ctDNA has potential as a prognostic biomarker for DFS/RFS, OS, PFS, and EFS in patients with solid tumors receiving NAT, although its prognostic performance may vary according to tumor type and clinical context.