Summary
In the phase 2 NeoCOAST-2 trial, clearance of circulating tumour DNA during treatment was associated with longer event-free survival in resectable non-small-cell lung cancer. The Dato-DXd combination had the highest reported ctDNA-clearance and 24-month event-free-survival rates among the three arms described.
A phase 2 report from the randomised NeoCOAST-2 trial found that circulating tumour DNA (ctDNA) clearance during treatment was associated with longer event-free survival in people with resectable non-small-cell lung cancer (NSCLC). The study included patients with stage IIA–IIIB disease and compared three perioperative treatment combinations. The findings, published in Nature on 25 October 2026, also showed the highest reported ctDNA-clearance and 24-month event-free-survival rates in the arm that included datopotamab deruxtecan (Dato-DXd).
Treatment arms and clinical outcomes
All three groups received standard chemotherapy and durvalumab, an anti-PD-L1 antibody, with an additional drug: oleclumab in Arm 1 (76 patients), monalizumab in Arm 2 (72 patients), or Dato-DXd in Arm 4 (54 patients). At the trial’s primary analysis, pathological complete response (pCR) rates were 20.3%, 25.7% and 35.2%, respectively. A pCR is a tissue-based assessment of whether viable cancer remains in the surgical specimen after treatment.
The new report presents prespecified secondary outcomes. The 24-month event-free-survival (EFS) rates were 70.5% in Arm 1, 75.7% in Arm 2 and 77.9% in Arm 4. EFS measures the time patients remain free of events defined by the trial.
The researchers also assessed ctDNA clearance using a tumour-agnostic assay. Unlike a test designed around mutations identified in an individual patient’s tumour, a tumour-agnostic assay is not tailored to that tumour. At neoadjuvant Cycle 3 Day 1, clearance rates were 50.9% in Arm 1, 54.5% in Arm 2 and 72.5% in Arm 4.
What the blood test adds
Patients with ctDNA clearance had longer EFS than those without clearance in a pooled analysis of the reported arms (hazard ratio 0.22; 95% confidence interval 0.11–0.45; P = 0.0000443). A hazard ratio below 1 indicates a lower rate of trial-defined events in the clearance group during the analysis period. Patients with both ctDNA clearance and pCR also had more favourable EFS than patients with neither (hazard ratio 0.10; 95% confidence interval 0.02–0.42; P = 0.002).
These results suggest that a blood-based measure of tumour response could complement the later tissue assessment and help distinguish patients with different outcomes. The combination of ctDNA clearance and pCR is a potential way to improve risk stratification, but the reported analyses of ctDNA dynamics and outcomes were exploratory.
Arm 4 had the numerically highest pCR, ctDNA-clearance and 24-month EFS rates among the three combinations reported. The researchers say those results warrant further evaluation; the phase 2 findings are not, by themselves, confirmation that this regimen is superior to the others.