Disease-free survival (DFS) and time to recurrence (TTR) with circulating tumor (ct) DNA–based decision for adjuvant treatment in colon cancer stage II (CIRCULATE): An AIO (KRK-0217)/ABCSG trial.

Abstract

LBA3500 Background: Adjuvant chemotherapy (ACT) provides limited benefit in unselected stage II colon cancer patients (pts). Postoperative ctDNA has higher prognostic value than classical clinical or molecular markers but has not been proven to predict benefit from ACT. Methods: Pts with UICC II, pMMR/MSS colon cancer were tested for ctDNA using an academic, tumor-informed, NGS based test (Stasik, Front Genet 2022). ctDNA positive (pos) pts were randomized (planned 2:1) to CHEMO (6 mo capecitabine [cape] or 3-6 mo cape+oxaliplatin) vs observation without ACT (OBS). ctDNA negative (neg) pts were randomized 1:4 to OBS vs OFF-STUDY. Patients in OBS and their investigators remained blinded to the ctDNA status. The primary endpoint was DFS in ctDNApos pts (CHEMO vs. OBS), the TTR was reported as 3-y recurrence rate. Differences were tested using one-sided log-rank tests. The planned sample size was 1,540 randomized pts assuming 10% ctDNApos. Because the funding period expired, the trial ended early and was finally analyzed. Results: From 06/2020 to 07/2025, 2,126 pts were screened at 138 sites in Germany/Austria. The ctDNApos rate was lower than expected among randomized pts (2.9%) and screened but not randomized pts (4.3%). Overall, 1,396 pts were randomized: 1,083 to OFF-STUDY, 287 to OBS (15 ctDNApos), and 26 pts to CHEMO (all ctDNApos). Median age was 64 y, 63% were male. Classical risk factors were present in 9.5% (4.8% pT4, 5.0% unplanned resection). In CHEMO, 21/26 pts (81%) started cape, 33% of them received oxaliplatin, median no of cycles was 6. There was one treatment-related death. DFS and OS were significantly higher in ctDNAneg vs ctDNApos pts (3-y-DFS 87% vs 52%, HR 0.23 [95%CI 0.13-0.43], p < 0.001; 3-y-OS 98% vs 88%, HR 0.18 [95%CI 0.05-0.61], p = 0.001). In the per-protocol (PP) analysis (excluding CHEMO pts not treated), CHEMO improved recurrence rates and DFS compared to ctDNApos/OBS (3-y recurrence rate19% vs 62%, HR 0.21 [95%CI 0.06-0.80], p = 0.009; 3-y-DFS 77% vs 38%, HR 0.28 [95%CI 0.09-0.94], p = 0.022). Cancer-specific survival (CSS) was numerically higher with CHEMO (3-y-CSS 100% vs 84%, HR 0.25 [95%CI 0.03-2.44]). In the ITT cohort (including the non-treated pts), the differences between arms were not significant (3-y recurrence rate 35% vs 62%, HR 0.48 [95%CI 0.17-1.33], p = 0.08; 3-y-DFS 61% vs 38%, HR 0.52 [95%CI 0.20-1.39], p = 0.12; 3-y-CSS 94% vs 84%, HR 0.40 [95%CI 0.07-2.38], p = 0.15). Conclusion: This is the first prospective randomized trial demonstrating that ctDNA guided ACT improves TTR and DFS in stage II colon cancer pts without clinical risk factors, supporting ctDNA testing for adjuvant decision making in the clinical practice. Limitations include the relatively low number of randomized ctDNApos pts, the sensitivity of the academic test developed 10 y ago and the reliance on PP analysis. Clinical trial information: NCT04089631 .