Post-Operative ctDNA (MRD) in Colorectal Cancer: What the DYNAMIC Trial Actually Showed

The clinical question

After a stage II colon cancer is surgically removed, the oncologist faces a recurring dilemma: who still harbors microscopic residual disease that will later surface as recurrence, and who is already cured by the operation alone? Historically, that judgment has rested on clinicopathologic risk features — T stage, lymphovascular invasion, tumor grade, the number of nodes examined, perforation or obstruction at presentation. These features are imperfect proxies. Many patients flagged as "high risk" never recur and are exposed to chemotherapy they did not need; others judged low risk relapse despite surgery alone. The DYNAMIC trial asked a sharper question: can a blood test for circulating tumor DNA (ctDNA) measured after surgery guide the decision to give or withhold adjuvant chemotherapy, and do so without worsening recurrence-free survival (RFS) [1]?

The analyte and the assay

The relevant analyte is circulating tumor DNA — fragments of tumor-derived DNA shed into the bloodstream, a subset of the broader pool of cell-free DNA (cfDNA). In the post-operative setting, detectable ctDNA is interpreted as a molecular signature of minimal (or measurable) residual disease (MRD): tumor material persisting despite apparently complete resection.

DYNAMIC used a tumor-informed assay strategy. Rather than screening blood for a fixed, universal panel, investigators first sequenced each patient's resected tumor to identify somatic mutations, then built a personalized assay (the Safe-SeqS approach) to track those specific mutations in serial plasma samples [1]. This design matters. A tumor-informed assay anchors the blood test to variants already known to be present in that individual's cancer, which improves specificity — a mutation detected in plasma is far less likely to be background noise when you have already confirmed it in the tumor.

Blood was drawn at weeks 4 and 7 after surgery. A patient was defined as ctDNA-positive if either sample was positive [1].

What a result can and cannot establish deserves plain statement. A positive post-operative ctDNA result indicates that tumor-derived DNA is detectable in the circulation, which carries a substantially elevated risk of recurrence — but it does not localize disease, quantify how much remains, or guarantee that any particular treatment will eradicate it. A negative result reflects the absence of detectable ctDNA at the assay's limit of detection. It is not proof of cure; micrometastatic disease below the detection threshold can exist. The test shifts probabilities; it does not deliver certainties.

The evidence

DYNAMIC was a multicenter, randomized trial conducted in Australia enrolling patients with resected stage II colon cancer. A total of 455 patients were randomized 2:1 to either a ctDNA-guided management arm (n=302) or standard clinicopathologic management (n=153) [1].

In the ctDNA-guided arm, treatment followed the blood result: patients who were ctDNA-positive at week 4 or 7 were offered adjuvant chemotherapy (an oxaliplatin-based or fluoropyrimidine regimen), while ctDNA-negative patients were not treated. In the standard-management arm, the decision to treat rested on conventional clinicopathologic criteria at the treating clinician's discretion [1].

Two results anchor the trial.

Chemotherapy use fell substantially. In the ctDNA-guided group, 15% of patients received adjuvant chemotherapy, compared with 28% in the standard-management group [1]. The strategy roughly halved chemotherapy exposure across the managed population.

Recurrence-free survival was not compromised. At 2 years, RFS was 93.5% in the ctDNA-guided group versus 92.4% in the standard-management group. The difference met the trial's prespecified non-inferiority margin (the hazard ratio and confidence interval were consistent with non-inferiority of the ctDNA-guided approach) [1].

The prognostic signal within the ctDNA-guided arm reinforced biological plausibility. Among ctDNA-positive patients who received chemotherapy, 3-year RFS was 86.4%; among ctDNA-negative patients who did not receive chemotherapy, 3-year RFS was 92.5% [1]. The lower recurrence-free rate in the treated, ctDNA-positive group is expected — these are the patients selected by a biomarker of residual disease — and underscores that a positive result identifies genuinely higher-risk biology.

Where this stands

It is worth pausing on what DYNAMIC represents. For years, the argument that post-operative ctDNA ought to guide adjuvant decisions rested on observational cohorts showing that ctDNA-positive patients recur more often. Those data established prognosis but could not answer the interventional question — whether acting on the result improves outcomes. DYNAMIC is the first randomized evidence that a ctDNA-informed de-escalation strategy can cut chemotherapy use without a detectable RFS penalty at two years [1]. Having a randomized answer, rather than a plausible inference, genuinely changes the texture of the conversation at the clinic desk.

That said, the honest reading of how far these findings travel requires care:

• The 2-year endpoint cuts both ways. A shorter follow-up horizon is precisely what made this trial deliverable on a reasonable timeline — but stage II colon cancer can recur late, and non-inferiority demonstrated at two years is not the same as non-inferiority demonstrated at five or ten. Longer follow-up is needed before anyone treats the question as closed, and clinicians weighing the data should hold that caveat front of mind.

• Overall survival was not the primary endpoint. DYNAMIC was designed and powered around recurrence-free survival; it does not establish an OS benefit or equivalence [1].

• The assay was tumor-informed and specific to the methodology used. Results obtained with one personalized, sequencing-based platform do not automatically transfer to every commercial MRD assay. Analytic performance, limit of detection, and sampling timing all vary between platforms.

• The population was narrow — resected stage II colon cancer only. Extrapolation to stage III disease, to rectal cancer, or to other settings is not supported by this trial.

• A negative result is not a clearance certificate. Patients managed without chemotherapy on the basis of a negative ctDNA result still recurred at a measurable rate [1].

None of this is a mark against DYNAMIC; it is simply the perimeter of what one well-conducted trial can prove.

What the field is now watching for is whether confirmatory randomized trials in other colorectal settings reproduce the de-escalation signal, whether longer follow-up preserves the recurrence-free equivalence seen at two years, and whether overall survival data eventually accompany the RFS findings. Running alongside those clinical questions is a quieter but equally consequential one: assay standardization. Until MRD platforms share comparable sensitivity thresholds and reporting conventions, a "ctDNA-negative" result will not mean the same thing from one laboratory to the next — and for a test meant to license withholding treatment, that is the kind of detail that decides whether a promising strategy becomes a dependable one.

---

References

1. Tie J, et al. Circulating Tumor DNA Analysis Guiding Adjuvant Therapy in Stage II Colon Cancer (DYNAMIC). N Engl J Med. 2022. PMID 35657320; PMCID PMC9701133.