ctDNA Predicts Relapse in Early Cancer, but Clinical Action Remains Unclear

Minimal residual disease (MRD) after curative-intent surgery for solid tumours is a primary driver of disease recurrence. Traditional methods for detecting MRD, such as imaging and serum tumour markers, often lack the sensitivity to identify microscopic disease.¹ Circulating tumour DNA (ctDNA) analysis, which detects tumour-derived DNA fragments in the bloodstream, has emerged as a highly sensitive tool for MRD detection.² The presence of ctDNA post-surgery is consistently associated with an elevated risk of relapse across various cancer types, including colorectal, breast, lung, and gastric cancers.³ This prognostic capability has led to interest in using ctDNA to guide adjuvant therapy decisions.

What the studies show

Multiple prospective and retrospective studies have demonstrated the prognostic value of post-operative ctDNA. For instance, in stage II colorectal cancer, patients with detectable ctDNA after surgery exhibit a significantly higher risk of recurrence compared to ctDNA-negative patients. One meta-analysis reported a pooled hazard ratio for recurrence-free survival of HR 7.4 (95% CI 5.6 - 9.8) for ctDNA-positive versus ctDNA-negative patients.⁴ Similar observations have been made in early-stage breast cancer, where ctDNA detection post-surgery predicted recurrence with a hazard ratio often exceeding HR 5.0.⁵ In non-small cell lung cancer (NSCLC), post-operative ctDNA positivity has been linked to a 3- to 10-fold increased risk of relapse.⁶

The methodology for ctDNA detection typically involves next-generation sequencing (NGS) of plasma samples. This can include tumour-informed approaches, where a patient's primary tumour tissue is sequenced to identify somatic mutations, and then a personalised assay tracks these specific mutations in plasma. Alternatively, tumour-agnostic approaches screen for a broader panel of cancer-associated mutations without prior tumour sequencing. The choice of assay impacts sensitivity and specificity, with tumour-informed methods generally offering higher sensitivity for MRD detection. Blood samples are usually collected at specific time points post-surgery, often several weeks after resection to allow for clearance of ctDNA from the surgical procedure itself. The half-life of ctDNA in circulation is short, typically hours, making it a dynamic marker of disease burden. This rapid clearance allows ctDNA levels to reflect the presence of active, proliferating tumour cells.

Despite this strong prognostic association, evidence supporting the use of ctDNA to escalate or de-escalate adjuvant therapy outside of a clinical trial setting is limited. Several ongoing randomised controlled trials are investigating ctDNA-guided adjuvant therapy. For example, trials in colorectal cancer are randomising ctDNA-positive patients to either standard adjuvant chemotherapy or intensified regimens, while ctDNA-negative patients may be considered for de-escalation or observation.⁷ Preliminary data from some of these trials suggest that ctDNA-guided de-escalation in ctDNA-negative patients may be safe, but definitive results on escalation strategies are still pending.⁸

The primary limitation is the absence of level 1 evidence from large, randomised controlled trials demonstrating that acting on ctDNA positivity improves patient outcomes (e.g., overall survival or disease-free survival) beyond standard-of-care adjuvant therapy. While ctDNA can identify patients at high risk of relapse, it is not yet clear whether intensifying treatment in these patients provides a clinical benefit that outweighs the potential toxicities.⁹ Furthermore, standardisation of ctDNA assays, including sensitivity, specificity, and optimal timing of sampling, remains an area of active research. The analytical sensitivity required to detect MRD means that false positives, though rare, could lead to unnecessary treatment, while false negatives could provide false reassurance.¹⁰ The clinical utility of ctDNA also depends on the specific cancer type and stage. For instance, in very early-stage cancers with low recurrence risk, the absolute benefit of intensified therapy based on ctDNA positivity might be modest, even if the relative risk is high. Conversely, in higher-risk early-stage cancers, the potential for benefit from targeted intervention might be greater. The cost-effectiveness of routine ctDNA testing and subsequent therapy modifications also requires careful consideration, particularly in healthcare systems with resource constraints. The integration of ctDNA into routine clinical practice will necessitate clear guidelines on patient selection, assay interpretation, and subsequent therapeutic algorithms, all supported by robust clinical evidence.