Trajectory identification of high-risk subgroups after colorectal cancer surgery in patients with minimal residual disease: a single-center longitudinal study.

Background: Postoperative recurrence and distant metastasis are crucial for long-term survival in colorectal cancer (CRC) patients. Minimal residual disease (MRD) after surgery is linked to these outcomes, with a negative MRD usually suggesting low risk. However, some patients still face recurrence or metastasis. This study assessed the prognostic significance of circulating tumor DNA (ctDNA)-MRD in CRC and identified high-risk subgroups through analysis of its longitudinal trajectory. Methods: This study analyzed clinical data from (n=124) stage II-III CRC patients who underwent ctDNA-MRD testing at The First Affiliated Hospital of Kunming Medical University between January 2017 and December 2024. Patients were categorized into four groups based on postoperative ctDNA-MRD status and recurrence: true negative (TN), false negative (FN), true positive (TP), and false positive (FP). Utilizing the longitudinal trajectory results of ctDNA-MRD, in conjunction with prognostic data, we aimed to delineate the characteristics of high-risk subgroups among CRC patients. Results: (I) In this study, ctDNA-MRD negativity and alpha-fetoprotein (AFP) <7 ng/mL were protective factors for progression-free survival (PFS), while pathology node (pN1), and pN2 stages were risk factors. For overall survival (OS), protective factors included MRD negativity, AFP <7 ng/mL, carcinoembryonic antigen (CEA) <5 ng/mL, carbohydrate antigen 125 (CA125) <24 U/mL, and neutrophil-lymphocyte ratio (NLR) <3.8. Risk factors were albumin (ALB) <39 g/L, pN1, and pN2 stages. (II) Longitudinal MRD trajectory analysis revealed that ctDNA-MRD signals consistently increased in recurrent patients but decreased or disappeared in non-recurrent patients. The mean variant allele frequency (VAF) value was highest in the ctDNA-MRD-positive recurrent group, significantly more than that in the ctDNA-MRD-negative non-recurrent and recurrent groups (P=0.047), though detection breadth was highest in the ctDNA-MRD-positive recurrent group without statistical significance (P=0.20). Overall, total VAF was higher in recurrent than non-recurrent groups across all subgroups, but not significantly (P=0.71). Conclusions: ctDNA-MRD status serves as a crucial prognostic indicator for patients with CRC. Longitudinal monitoring of ctDNA-MRD is more effective than single-point assessments in identifying CRC patients at high risk.