ctDNA-based MRD detection in unresectable NSCLC undergoing curatively intended chemoradiotherapy and durvalumab.

8011 Background: Durvalumab consolidation after chemoradiotherapy (CRT) has improved clinical outcomes in patients with unresectable stage III non-small cell lung cancer (NSCLC). Despite durvalumab treatment, a substantial proportion of patients relapse, while up to 20% achieve long-term survival without durvalumab. Circulating tumor DNA (ctDNA)-based minimal residual disease (MRD) detection has shown promise as a tool for risk-adaptive and personalized treatment strategies in resectable NSCLC but is still underexplored as a biomarker in patients with stage III NSCLC treated with CRT and consolidative durvalumab. Methods: The DART study is a multicenter phase II clinical trial enrolling 86 patients with unresectable stage III NSCLC. All patients received two cycles of platinum-doublet chemotherapy concomitant with radiotherapy to a total dose of 60–66 Gy, followed by durvalumab. We prospectively collected serial plasma samples from all patients at baseline(before CRT), at the initiation of durvalumab (one month post-CRT), and at predefined timepoints during durvalumab treatment. Plasma samples were analyzed using a novel tumor-agnostic ctDNA MRD assay (MEDICOVER Genetics), tailored to each patient’s cancer biomarker profile. This hybrid capture-based assay leverages genomic information in cell-free DNA from selected coding regions in 293 genes to classify plasma samples as positive (“ctDNA detected”) or negative (“ctDNA not detected”). Here, we present results of the first MRD analysis conducted in the trial, involving a total of 138 plasma samples from 20 patients who completed all scheduled blood draws. Results: The baseline ctDNA detection rate was 91.8% in the entire study population and 73.7% in the 20 patients with completed longitudinal MRD analyses. Detectable ctDNA at baseline varied according to stage and histology and was not associated with PFS. Nine patients had detectable ctDNA in at least one plasma sample during the first four months after CRT, which was significantly associated with shorter progression-free survival (PFS) (HR: 4.7; 95% CI: 1.6–13.1; p = 0.004). When assessing specific timepoints, patients with detectable ctDNA four months post-CRT had shorter PFS compared to patients without detectable ctDNA at this timepoint (HR: 3.77; 95% CI: 1.32–10.74; p = 0.013). In contrast, detectable ctDNAone month post-CRT was not associated with shorter PFS (HR: 2.23; 95% CI: 0.78–6.36; p = 0.13). Preliminary overall survival data indicate that detectable ctDNA during the first four months post-CRT significantly increased the odds of death within 24 months (OR: 16.48; 95% CI: 1.29–1000.51; p = 0.017). Conclusions: Detection of ctDNA during consolidative durvalumab after CRT using a novel tumor-agnostic MRD assay was associated with inferior outcomes, demonstrating the potential of ctDNA as a biomarker to identify high-risk patients for tailored interventions. Clinical trial information: NCT04392505 .