Abstract
PURPOSE: The prognostic value of molecular residual disease (MRD) in non-small cell lung cancer (NSCLC) is well established, with treatment-guiding results anticipated. Here, we present updated analyses from our previously published cohort study of 261 patients with NSCLC undergoing complete resection. EXPERIMENTAL DESIGN: A total of 261 patients with stage I to III lung cancer who underwent radical surgery were enrolled. Enrolled patients underwent follow-up blood draws according to the predefined time points after surgery. As of December 31, 2023, with a median follow-up of 43.4 months, 948 postoperative blood samples were collected. RESULTS: Landmark and longitudinal MRD exhibited positive predictive values of 91.3% and 92.8%, respectively, with a median lead time of 5.2 months. Negative predictive values were 76.5% and 93.2%, respectively. Patients with landmark undetectable MRD could not benefit from adjuvant therapy through the updated follow-up (P = 0.529). Among the 13 patients with recurrent NSCLC and longitudinal undetectable MRD, seven (53.8%) had brain-only metastases, and four (30.8%) had no updated blood samples for over 6 months prior to recurrence. Besides, for those with longitudinal detectable MRD, higher maximum variant allele frequency (>0.55%) and ctDNA level (>13 hGE/mL) were associated with a high risk of short-term recurrence. Additionally, updated follow-up data further support that the peak time for detectable MRD was 18 months after landmark detection. CONCLUSIONS: These findings suggest the significant potential of MRD in guiding personalized treatment for NSCLC. Postoperative longitudinal undetectable MRD can indicate a cured population.