In brief
- NeoCOAST-2 is a randomised phase 2 platform trial of perioperative treatments added to chemotherapy plus durvalumab in stage IIA to IIIB NSCLC.
- The 24-month event-free survival (EFS) rates were 70.5% with oleclumab, 75.7% with monalizumab and 77.9% with datopotamab deruxtecan (Dato-DXd).
- Clearance of circulating tumour DNA (ctDNA) at neoadjuvant cycle 3 day 1 occurred in 50.9%, 54.5% and 72.5% of patients in the three arms.
- Across arms, ctDNA clearance was associated with longer EFS than non-clearance (pooled hazard ratio 0.22; P=0.0000443).
- Patients with both ctDNA clearance and a pathological complete response (pCR) had a hazard ratio for EFS of 0.10 against those with neither.
Study at a glance
- Trial: NeoCOAST-2, randomised phase 2 platform trial
- Population: resectable stage IIA to IIIB NSCLC
- Arms reported: chemotherapy plus durvalumab with oleclumab (Arm 1, n=76), monalizumab (Arm 2, n=72) or Dato-DXd (Arm 4, n=54)
- Earlier primary analysis: pCR 20.3%, 25.7% and 35.2%
- Endpoints in this report: EFS and ctDNA clearance (prespecified secondary), ctDNA dynamics against outcome (exploratory)
- 24-month EFS: 70.5%, 75.7% and 77.9%
- ctDNA clearance at cycle 3 day 1: 50.9%, 54.5% and 72.5%, tumour-agnostic assay
- Clearance vs non-clearance: pooled EFS HR 0.22
- Not reported: safety, overall survival
Dato-DXd led on pCR, 24-month EFS and ctDNA clearance
Relapse remains common in resectable NSCLC after perioperative treatment, and NeoCOAST-2 was designed to test several additions to chemotherapy and durvalumab side by side. The primary analysis had already reported pCR rates of 20.3% with oleclumab, 25.7% with monalizumab and 35.2% with Dato-DXd, each given with standard chemotherapy plus the anti-PD-L1 antibody durvalumab.
The same order held for the secondary endpoints. The 24-month EFS rate was 77.9% in the 54 patients of the Dato-DXd arm, against 75.7% with monalizumab and 70.5% with oleclumab. The arms were not compared statistically with one another in the abstract, and the authors describe the Dato-DXd figures as numerically highest.
A pooled hazard ratio of 0.22 for EFS after early ctDNA clearance
ctDNA was measured with a tumour-agnostic assay. At the start of the third neoadjuvant cycle, clearance rates were 50.9%, 54.5% and 72.5% in Arms 1, 2 and 4.
Pooled across the three arms, patients whose ctDNA cleared had longer EFS than those whose ctDNA did not (95% CI 0.11 to 0.45; P=0.0000443). The association with outcome was an exploratory analysis, and clearance is an observed response rather than a randomised factor, so the hazard ratio describes an association with prognosis.
Clearance with pCR identified the lowest-risk group
The two markers carried more information together than apart. Patients with both ctDNA clearance and pCR had an EFS hazard ratio of 0.10 (P=0.002) compared with patients who had neither. The 95% confidence interval for this estimate ran from 0.02 to 0.42, so its precision is limited.
“These findings suggest combined assessment of ctDNA clearance and pCR may improve risk stratification, and that the numerically highest pCR4 and EFS rates with the Dato-DXd combination warrant further evaluation.”McGrath et al., Nature (2026)
The arms enrolled between 54 and 76 patients each. The authors frame the combined use of ctDNA clearance and pCR as a possible improvement in risk stratification, and they do not report any treatment decision being made on ctDNA status.
Sources
- Nature. ctDNA clearance and outcomes for resectable lung cancer from NeoCOAST-2 (2026-10-25). doi.org
Featuring Lung Summit faculty
This study was co-authored by Lung Summit faculty Tina Cascone and Patrick Forde.
This article was produced independently by the Lung Summit editorial team, without industry funding or input.
