Clinical implications of primary and metastatic tumor-informed clonal tracking for MRD assessment in oligometastatic colorectal cancer.

3601 Background: While minimal residual disease (MRD) assays informed by primary tumor sequencing have demonstrated prognostic value in locally advanced colorectal cancer (CRC), the additive clinical benefit of combined tracking of metastatic-specific private variants in oligometastatic CRC (mCRC) remains unclear. We aimed to evaluate whether a combined primary and metastatic tumor-informed approach improves MRD detection sensitivity and prognostic accuracy following curative resection. Methods: We enrolled 24 patients who underwent curative resection for liver and/or lung oligometastases between 2023 and 2025. A comprehensive tumor-informed panel was designed based on variants identified through whole-exome sequencing (WES) of both primary and metastatic tissues. Serial plasma ctDNA was analyzed at three time points: post-metastasectomy (P1), post-adjuvant chemotherapy (P2), and at the time of recurrence (P3). The clinical performance of a primary tumor-only (PT-only) assay was compared with a combined primary and metastatic tumor-informed (PT+MT) assay. Results: The median age was 65 years (range 33–83), and 18 (75%) were male. Primary tumor locations included the colon (83%) and rectum (17%). Fourteen (58%) patients presented with synchronous metastases. Metastatic sites involved the liver (n = 13), lung (n = 10), or both (n = 1). At P1, P2, and P3, MRD positivity rates using the PT-only assay were 54%, 47%, and 100%, respectively. The PT+MT assay identified only one additional patient as MRD-positive at P1 compared to the PT-only assay. At a median follow-up of 29.2 months (IQR 22.6–32.7), 16 patients (67%) experienced recurrence. The PT-only assay showed significant prognostic impact (HR for ctDNA positivity: 4.53; 95% CI, 1.41–14.57; P = 0.006), which was comparable to the PT+MT assay (HR: 4.46; 95% CI, 1.38–14.38; P = 0.007). While PT+MT tracking did not significantly improve detection sensitivity at P1, it effectively captured clonal dynamics, showing a significant enrichment of metastatic-specific private variants in plasma at recurrence (26%, P3) compared to P1 (17%) and P2 (13%). Conclusions: Integrating metastatic-specific private variants into MRD tracking revealed distinct clonal dynamics but did not significantly enhance clinical sensitivity or prognostic performance compared to a primary tumor-informed assay in patients with oligometastatic CRC. These findings suggest that current MRD assay designs based on primary tumor sequencing remain a robust standard, even in the metastatic setting after curative resection.