Perioperative circulating tumor DNA in esophageal cancer: prognostic signal in search of a decision framework

To the Editor, Esophageal cancer carries a high relapse risk despite neoadjuvant therapy (NAT) and esophagectomy. Standard predictors, including radiographic response, ypTNM stage, and tumor regression grade, incompletely reflect residual disease biology[1,2]. Circulating tumor DNA (ctDNA) has emerged as a marker of minimal residual disease (MRD) with potential to identify patients at risk for early relapse[3]. This commentary was prepared under direct author supervision using language editing support in compliance with the TITAN 2025 guidelines[4]. Shen et al present the most comprehensive synthesis of perioperative ctDNA in NAT-treated esophageal cancer to date[5]. Across >500 patients, detectable ctDNA after NAT or surgery was associated with significantly worse survival and reduced pathological complete response (pCR), identifying a high-risk subgroup despite multimodal therapy. Three findings are most relevant to surgical practice First, baseline ctDNA was not prognostic. Pretreatment positivity did not correlate with pCR or survival[5]. This contrasts with other malignancies and suggests that dynamic clearance rather than a single pre-NAT value better captures treatment sensitivity[3,6]. Second, post-NAT ctDNA defined poor responders. Persistent ctDNA after NAT was associated with approximately 3–4-fold higher risk of progression or death[5]. Prospective studies in esophageal squamous cell carcinoma (ESCC) confirm that ctDNA persistence strongly predicts early relapse and may outperform conventional imaging[7,8]. Third, postoperative ctDNA revealed occult residual disease. Detection following R0 resection was associated with a similar increase in relapse risk[5]. This mirrors colorectal cancer, where postoperative ctDNA identifies MRD and informs adjuvant therapy strategies[3,9]. Together, these findings suggest a clear model: ctDNA clearance indicates molecular remission; persistence reflects high relapse risk despite favorable pathology. Methodological strengths include prospective registration, PRISMA reporting, REMARK quality assessment, timepoint-based analysis (baseline/NAT/surgery), and inclusion of modern immunotherapy-based NAT regimens[5]. These features increase contemporary relevance. Limitations merit caution, as all available evidence remains observational with no randomized trials evaluating ctDNA-guided therapy in esophageal cancer. The number of ctDNA-positive cases was modest, introducing statistical uncertainty, and assay platforms varied (digital polymerase chain reaction vs tumor-informed next-generation sequencing) with heterogeneous thresholds for positivity. Moreover, the strongest signals were derived from ESCC cohorts treated with chemoradiotherapy, and it remains uncertain whether similar effects apply to adenocarcinoma in the current era of perioperative chemo-immunotherapy[1,8]. Clinical implications include (1) enhanced risk communication – persistent ctDNA may inform counseling even with favorable imaging; (2) optimized surveillance – positive patients may merit shorter follow-up intervals or early trial referral; and (3) integrated prognosis – combining ypTNM, regression grade, and ctDNA may outperform pathology alone[3,8]. Future priorities include standardization of ctDNA assays, modeling of ctDNA kinetics during NAT, randomized trials of ctDNA-guided adjuvant strategies, multimodal integration with radiomics and pathology, and cost-effectiveness evaluation. In conclusion, Shen et al provide compelling evidence that perioperative ctDNA is a strong prognostic biomarker after NAT and surgery in esophageal cancer[5]. While not yet ready for routine use, ctDNA represents a promising tool to shift surgical management toward molecularly informed decision-making.