CheckMate 77T analysis links ctDNA clearance to nivolumab benefit in resectable NSCLC

August 12, 2026 at 05:00 AM2 min

This digest was compiled by AI from multiple sources — links to the originals are below.

A biomarker analysis of the CheckMate 77T trial found pre-surgical clearance of circulating tumour DNA predicted pathologic complete response and longer event-free survival in patients with resectable non-small cell lung cancer treated with perioperative nivolumab. Among 98 nivolumab-treated patients with evaluable biomarkers, 66% achieved ctDNA clearance before surgery and half of those had pathologic complete response, versus 38% clearance and 12% pCR in the placebo arm. Patients who became molecular residual disease-positive after surgery all had disease recurrence.

ctDNA Clearance and Pathologic Response

Among the 98 nivolumab-treated patients with evaluable biomarkers, 83 (85%) had detectable ctDNA before neoadjuvant treatment and 90 (92%) at treatment completion. In the placebo arm, 75 of 92 patients (82%) had detectable ctDNA at start and 78 (85%) at completion. Pre-surgical ctDNA clearance occurred in 50 of 76 (66%) nivolumab-treated patients with paired evaluable samples, and 25 of those 50 (50%) achieved pathologic complete response. In the placebo group, 24 of 64 (38%) had clearance and 3 of 24 (12%) achieved pCR.

Molecular Residual Disease After Surgery

Four of 48 (8%) patients in the nivolumab group and 9 of 44 (20%) in the placebo group who were negative for molecular residual disease after surgery and before adjuvant treatment initiation became positive during adjuvant treatment. All patients who converted to MRD-positive had disease recurrence. The analysis included 190 biomarker-evaluable patients among 461 randomized in CheckMate 77T, representing 41% of the trial population.

Driver Gene and Machine-Learning Predictors

In the biomarker-evaluable subset, event-free survival appeared prolonged with nivolumab versus placebo in patients with alterations in KEAP1, STK11, CDKN2A and/or SMARCA4, with hazard ratio 0.48 (95% CI 0.28–0.83) among 60 nivolumab and 45 placebo patients. A machine-learning model trained on biomarker-evaluable patients identified pre-surgical ctDNA clearance, non-N2 NSCLC, pathologic complete response, squamous tumour histology, and nivolumab treatment as top predictors of prolonged EFS. The CheckMate 77T primary analysis previously reported EFS hazard ratio 0.58 (97.36% CI 0.42–0.81; P<0.001) for perioperative nivolumab versus placebo.