Circulating tumour DNA and molecular residual disease in oesophageal cancer: Clinical readiness, evidence gaps and trial-ready applications

Oesophageal cancer remains prone to relapse after curative-intent treatment, and conventional staging, imaging and pathology do not fully identify patients with occult residual disease. Circulating tumour DNA (ctDNA) and molecular residual disease (MRD) testing offer a serial tumour-derived signal that may refine risk assessment across the treatment pathway. In this critical review, supported by a structured literature search from 2015 to June 2026, we evaluate ctDNA-MRD through the lens of clinical readiness rather than technical feasibility alone. Current evidence shows that ctDNA positivity is most informative after treatment, when the key question shifts from baseline tumour burden to residual disease. Baseline ctDNA is useful mainly for establishing molecular trackability, but should not independently direct treatment. Post-neoadjuvant MRD can refine response assessment and support trial selection, especially when integrated with endoscopy, biopsy and imaging, but is insufficient to justify surgery omission or watch-and-wait outside protocolised studies. Postoperative MRD is currently the most mature application, consistently identifying patients at high recurrence risk beyond conventional pathology. Surveillance ctDNA can detect molecular relapse before radiographic recurrence in some patients, but ctDNA-positive/imaging-negative disease should be regarded as an investigational state rather than clinical relapse. Across settings, ctDNA-MRD currently functions primarily as a risk-assessment biomarker; prospective, ideally randomised, trials are needed to determine whether ctDNA-guided management improves survival, quality of life, toxicity burden or cost-effectiveness. Until then, it should not be used as a stand-alone basis for treatment escalation, de-escalation, organ preservation or pre-emptive therapy. © 2026 Elsevier B.V.