Correlation of circulating tumor DNA and recurrence in patients with metastatic colorectal cancer with no evidence of disease: An open-label, prospective, phase II study.

3549 Background: Patients with metastatic colorectal cancer (mCRC) can achieve no evidence of disease (NED) after curative treatment. However, how to precisely guide the treatment after NED remained to be explored. Minimal residual disease (MRD) detection using circulating tumor DNA (ctDNA) analysis has shown to help identify patients’ risk of relapse in early-stage CRC. Thus we aimed to explore the feasibility of ctDNA by a tumor-informed personalized approach to predict recurrence and guide chemotherapy in mCRC with NED. Methods: We enrolled mCRC patients who achieved NED after local curative treatment from June 2022 to December 2023. Surgical or puncture tissues were collected. Blood samples were collected prior to curative treatment, at 1 month after curative treatment, and subsequently every 2–3 months until disease progression or completion of a 2-year follow-up period. Tumor-derived variants were identified by whole-exon sequencing of the tissue samples. Up to 50 highly ranked variants were selected for the personalized panel design, which was used to assess the MRD status of blood samples. Patients with negative MRD post curative treatment were followed, while those with positive MRD were subjected to investigator-determined adjuvant chemotherapy (ACT). Results: A total of 106 mCRC patients with NED after curative treatment were enrolled. The ctDNA was detected in 62 patients (92.5%, 62/67) pre curative treatment and in 55 patients (51.9%, 55/106) post curative treatment. At a median follow-up period of 252 days, 42 patients experienced recurrence and 33 patients (78.6%) had positive ctDNA post curative treatment. Patients with positive ctDNA post curative treatment demonstrated an inferior recurrence-free survival (RFS) than those with negative ctDNA (HR: 4.58, 95% CI: 2.18–9.64; P<0.001), which was not observed by CEA (HR: 1.87, 95% CI: 0.84–4.15; P=0.12). In patients with negative ctDNA after curative treatment, there was no difference in RFS between those treated with ACT or not (HR: 2.51, 95% CI: 0.48–13.04; P=0.25). During the dynamic longitudinal monitoring of ctDNA, the median RFS was 5.6 months, 15.5 months, 10.2 months and unreached in patients with persistently positive ctDNA, converted negative ctDNA, converted positive ctDNA, and persistently negative ctDNA, respectively (log-rank P for trend <0.001). All recurrent patients had at least one positive ctDNA detection before imaging confirmed recurrence, and the median lead time interval of ctDNA positivity to radiographic recurrence was 112 days (range, 3–441 days). Conclusions: Our results demonstrated that patients with ctDNA negative post curative treatment had superior RFS compared with those with positive ctDNA. Post curative treatment and dynamic ctDNA had the potential to guide ACT in mCRC patients with NED. The follow-up is still on-going. Clinical trial information: NCT05635630 .