Longitudinal ctDNA monitoring and prediction of anti-EGFR rechallenge outcomes in RAS/BRAF wild-type metastatic colorectal cancer (mCRC): The REMARRY & PURSUIT trials.

3514 Background: Anti‐EGFR monoclonal antibody (mAb) rechallenge involves re‐administering EGFR blockade after a treatment‐free interval to exploit clonal evolution. Recent evidence suggests that circulating tumor DNA (ctDNA)–based selection may improve outcomes; however, the predictive value of longitudinal ctDNA monitoring remains uncertain. Methods: The REMARRY study evaluated plasma RAS (p RAS ) dynamics in patients with RAS/BRAF V600E wild‐type metastatic colorectal cancer (mCRC), ECOG PS 0–1, who had previously responded to anti‐EGFR therapy and experienced progression within two months of the last dose. p RAS status was assessed at progression on prior anti‐EGFR, before rechallenge, at cycle 3, and at discontinuation using BEAMing digital PCR. Patients meeting additional criteria—namely, p RAS ‐negative, refractory or intolerance to standard chemotherapies, and an anti‐EGFR–free interval of at least 4 months—were enrolled in the phase II PURSUIT trial, which administered panitumumab (6 mg/kg) plus irinotecan (150 mg/m²) biweekly. The primary endpoint was confirmed objective response rate (ORR) per RECIST v1.1. In parallel, participants underwent next‐generation sequencing (NGS) of ctDNA in the GOZILA trial at corresponding time points to identify resistance alterations in genes including RAS, BRAF, EGFR ‐ECD, MAP2K , ERBB2, and MET. Results: Between May 2019 and May 2021, 183 patients were enrolled in REMARRY, and 50 p RAS‐ negative patients before rechallenge were included in PURSUIT. The confirmed ORR was 14.0% (90% CI, 7.0–23.0%), with a median progression‐free survival (PFS) of 3.6 months and a median overall survival (OS) of 12.0 months. Although all patients were p RAS ‐negative at baseline, 10.0% converted to p RAS ‐positive by cycle 3 and 36.0% by discontinuation. Patients who were p RAS ‐positive immediately after prior anti‐EGFR had higher conversion rates at cycle 3 (42.9% vs. 6.3%, p = 0.010) and at discontinuation (85.7% vs. 32.3%, p < 0.001) compared with those initially negative. The ORR was 23.8% in patients remaining p RAS ‐negative versus 0% in those converting to p RAS ‐positive (p = 0.254). Moreover, NGS‐detected resistance alterations after prior anti‐EGFR were associated with no responses (0/13) compared to a 27.8% response rate (5/23) in patients without such alterations (p = 0.038), and correlated with shorter PFS (HR 3.297, p = 0.002) and OS (HR 4.569, p < 0.001). In 86% (7/8) of patients who were p RAS ‐positive post–anti‐EGFR, the identical p RAS codons re‐emerged during rechallenge, consistent with NGS findings. Conclusions: ctDNA status immediately after progression on prior anti‐EGFR therapy predicts subsequent response, thereby supporting clinical decision‐making and personalized therapy. Trial Registration: PURSUIT (jRCTs031190096), REMARRY (UMIN000036424), GOZILA (UMIN000029315). Clinical trial information: jRCTs031190096 .