Feasibility of circulating tumor DNA-based minimal residual disease (ctDNA-MRD)-guided adjuvant chemotherapy in patients with stage II-III gastric cancer (GC): An adaptive trial (MRD-GATE).

4061 Background: Although adjuvant chemotherapy (ACT) is the standard treatment for stage II-III GC, this strategy lacks precision treatment options, and many patients cannot tolerate the adverse events (AEs) of ACT. ctDNA-MRD detection has been shown to predict recurrence risk. The aim of this study (NCT06157216) was to evaluate the feasibility of MRD-guided treatment in these patients. Methods: Patients with stage II-III GC who underwent R0 resection and D2 gastrectomy were enrolled. Tumor-informed ctDNA-MRD testing was performed at baseline (28 days after surgery) and subsequently at 3, 6, 9, 12, 18, 24, 30, and 36 months after surgery. ACT was tailored according to MRD status: baseline MRD-negative (MRD (-)) patients received de-escalated therapy (observation for stage II and S-1 monotherapy for stage III) and switched to combined ACT (SOX, S-1 plus oxaliplatin, or XELOX, capecitabine plus oxaliplatin) if MRD became positive, while baseline MRD-positive (MRD (+)) patients underwent combined ACT. The primary endpoint was 3-year disease-free survival rate (yDFSr). Secondary endpoints included the treatment de-escalation rate, DFS by MRD status, cumulative recurrence risk (CRR), 3-year overall survival rate (yOSr) and safety. Results: 65 patients were enrolled, with a median age of 60 years (range: 34-83), and 83.1% (54/65) were male, 31 patients had stage II and 34 patients had stage III GC. At baseline, 21.5% patients (14/65) were MRD(+) and received combination ACT. Among the 51 baseline MRD(-) patients, 45 received de-escalated therapy at onset (9 received combination ACT after MRD conversion) and 6 received combination ACT. The median follow-up time was 13.3 (range: 9.3-15.7) months. In the intention-to-treat population, the overall 1-yDFSr was 86.2% (90% CI: 79.4-93.5%), the 1-yOSr was 96.9% (90% CI: 93.5-100%) and the CRR was 13.8% (95% CI: 6.5-24.7%). The treatment de-escalation rate was 69.2% (45/65, 95% CI: 56.6-80.1%). Grade 3-4 AEs occurred in 24.6% (95% CI: 14.8%-36.9%) of patients. Baseline MRD (+) patients had a shorter DFS compared to MRD (-) ones (1-yDFSr: 57.1% vs. 94.1%, HR = 9.66, 95% CI: 2.40-38.81, log-rank P < 0.0001). Patients with sustained MRD (-) had the best DFS (1-yDFSr: 100%), while those with sustained MRD (+) had the shortest DFS. Patients with MRD conversion from positive to negative or from negative to positive had intermediate DFS (1-yDFSrs: 72.7% and 70.0%, respectively). In 9 patients with recurrence, ctDNA-MRD positivity identified recurrence a median of 3.4 months earlier than radiology. Conclusions: MRD-guided ACT for stage II-III GC significantly reduced the ACT rate, and increased the de-escalated CT rate, resulting in good disease-free survival and fewer side effects. The results of this MRD-guided precision of ACT deserve to be confirmed by large randomized clinical trials. Clinical trial information: NCT05585580 .