Background: For patients (pts) with colon cancer (CC), the detection of circulating tumor DNA (ctDNA) is associated with persistent disease after resection and outperforms traditional clinical and pathological features in prognosticating recurrence risk. We hypothesized that for pts with low-risk stage II CC, a positive ctDNA status after resection may identify those who benefit from adjuvant chemotherapy. Methods: In this prospective phase II/III clinical trial, pts with resected stage II CC without traditional high risk features and whom the evaluating oncologist deems suitable for active surveillance (i.e., not needing adjuvant chemotherapy) were randomized 1:1 to two arms: standard-of-care/observation (Arm A), or ctDNA assay directed therapy (Arm B). Postoperative blood was analyzed for ctDNA with the Guardant LUNAR assay, covering CC-relevant mutations and CC-specific methylation profiling. Arm B pts with ctDNA detected were treated with 6 mos of adjuvant (CAPOX or FOLFOX) chemotherapy. Primary endpoint for the phase II study was clearance of ctDNA at the 6-mo timepoint. A 1-sided Fisher exact test was used to compare ctDNA clearance between Arm A and Arm B among the first 16 pts with ctDNA detected at baseline. If p>.35, the study would be stopped for futility of ctDNA clearance but would otherwise continue to phase III if p≤.35. We present the results of the pre-planned phase II analysis. Results: 635 pts were randomized (Arm A: 318; Arm B: 317). One pt with ctDNA detected in Arm B declined protocol-directed chemotherapy but was included in the intention to treat analysis. Among the first 16 pts with ctDNA detected at baseline for the primary endpoint analysis, clearance of ctDNA after 6 mos was observed in 3/7 pts (43%, 95% CI 10-82%) in the control arm and in 1/9 pts (11%, 95% CI 0.3-48%) in the experimental arm after chemotherapy (p=.98). There were no unanticipated toxicities in those treated with chemotherapy. Conclusions: The phase II endpoint was not met and further enrollment has been halted based upon pre-specified study stopping rules utilizing the original assay. No improvement in ctDNA clearance was observed after 6 mos of chemotherapy for pts with ctDNA detected following resection of stage IIA CC. Future trials evaluating ctDNA as an integral biomarker for minimal residual disease determination must account for assay specificity in this pt population. Support: U10CA180868, -180822; UG1CA189867; Guardant Health. Clinical trial information: NCT04068103.