Background: Our previous analysis of data from the prospective, observational GALAXY study (UMIN000039205) reported post surgical detection of molecular residual disease (MRD) to be prognostic of patient (pt) outcomes and the most significant risk factor for recurrence regardless of BRAF V600E status. Here, we present an updated analysis and correlation of ctDNA dynamics with outcomes in pts with radically resected, stage II-IV CRC from the GALAXY study. Methods: A personalized, tumor-informed assay (Signatera, Natera, Inc.) was used for the detection and quantification of ctDNA in serial plasma samples collected at 1, 3, 6, 9, 12, 18, and 24 months post surgery until recurrence. CT scans of chest/abdomen/pelvis were conducted every 6 months. Post curative-intent surgery, pts underwent either treatment with adjuvant chemotherapy (ACT; N = 1,000) or observation (N = 1,518). The primary endpoint was disease-free survival (DFS), defined as the time between the date of surgery and date of detection of relapse/death due to any cause. Results: Of the 3,034 CRC pts enrolled between May 2020 and November 2022 in GALAXY study, 2,518 pts met the inclusion criteria and were analyzed in this substudy. The median follow-up was 16.3 months (range 0.1-37 mos). During the post-op MRD window, ctDNA results were available for 2,093 pts, 309 (14.8%) of whom were ctDNA+ and 1,784 (85.2%) were ctDNA-. Pts who were ctDNA+ during the MRD window (MRD+) had a significantly inferior DFS compared to MRD- pts (HR: 15.75, 95%CI: 12.59-19.68, p < 0.0001). Within the MRD+ group, a landmark analysis of ctDNA dynamics from MRD detection to 3-month time point revealed that pts who remained ctDNA+ were over 5-times more likely to recur compared to those who had ctDNA clearance (HR: 5.4, 95%CI: 3.58-7.67%, p < 0.0001). Among the 309 MRD+ pts, 181 received adjuvant therapy, 72.9% (132/181) of whom had ctDNA clearance. Notably, for those pts with subsequent ctDNA time points available, 68/126 (54%) had sustained clearance vs. 58/126 (46%) pts eventually returned ctDNA+. Pts with sustained clearance had remarkably better outcomes compared to those with transient ctDNA clearance (HR: 32.57, 95% CI: 9.94-106.76, p < 0.0001). Furthermore, we observed that among MRD+ pts treated with ACT, a 50% or greater decrease in ctDNA levels (mean tumor molecules/mL) at 6 months was associated with better DFS compared to pts with <50% decrease or increase in ctDNA levels (HR: 2.39, 95% CI: 1.32-4.34, p = 0.004). Conclusions: ctDNA-based detection of MRD as well as ctDNA dynamics in response to ACT were highly prognostic of pt outcomes. Ongoing randomized VEGA and ALTAIR studies in the CIRCULATE-Japan will establish clinical utility of ctDNA-guided adjuvant treatment. Clinical trial information: UMIN000039205.