PEGASUS ctDNA MRD assay guides adjuvant CAPOX→FOLFIRI escalation but misses relapse-free feasibility target in resected MSS colon cancer

PEGASUS (Phase 2, multicenter, single-arm; NCT04259944) tested a postoperative circulating tumor DNA (ctDNA) minimal residual disease (MRD) strategy in resected microsatellite-stable T4N0 (high-risk stage II) and stage III colon cancer (n=135): ctDNA+ patients received 3 months CAPOX with escalation to 6 months FOLFIRI if persistently ctDNA+, while confirmed ctDNA− patients de-escalated to 6 months capecitabine. ctDNA positivity at the postsurgery landmark was 26% (35/135) and predicted worse outcomes (3-year DFS 58% ctDNA+ vs 83% ctDNA−; HR 2.71, 95% CI 1.35–5.45; p=0.0036), but the primary endpoint failed—2-year relapse-free rate in ctDNA− patients at two subsequent liquid biopsies was 88% (90% CI 81–93), below the prespecified ≥92% threshold (only 100 ctDNA− cases vs 134 planned). Versus a matched TOSCA cohort (NCT00646607; typical oxaliplatin-based adjuvant therapy where longer oxaliplatin exposure drives substantially higher neuropathy), DFS looked similar while neurotoxicity was substantially lower—directionally consistent with IDEA data showing large neuropathy reductions with shorter oxaliplatin duration (grade ≥2 neuropathy ~14–17% with 3 months vs ~45–48% with 6 months).