Prognostic value of circulating tumor DNA following local therapy in colorectal cancer with liver metastases.

Abstract

26 Background: Local therapies have the potential to cure a portion of patients with colorectal cancer with liver-only metastatic disease (CRLM). Peri-operative chemotherapy has not been shown to improve overall survival when added to local therapy in CRLM. Circulating-tumor DNA (ct-DNA) may have potential to help guide management decisions following local therapy. Methods: We conducted a retrospective analysis of health records of CRLM patients who had tumor-informed ct-DNA (Signatera) measurement following curative-intent local therapy prior to August 30th, 2023. Disease-free survival (DFS) was defined as the time from ct-DNA test until radiographic evidence of disease. Descriptive characteristics of the cohort, ct-DNA results and their impact on adjuvant therapy decisions were recorded. Survival analysis was carried out using R software v4.3.1 with survival and survminer packages. The Kaplan–Meier method was used to estimate the survival distribution. Log-rank test was used for comparison. A multivariable Cox proportional hazards model was used to assess prognostic factors associated with DFS ( coxph). Results: 27 patients with CRLM underwent local therapy followed by ct-DNA collection. The median age was 61 years, 67% were male, and 85% were White. Molecular characteristics: KRAS 31%, NRAS/BRAF 0%, microsatellite instability 4%. The median time from initial diagnosis to treatment of liver disease was 21.0 months. 63% of patients received less than 6 months of systemic therapy prior to local therapy. ct-DNA was checked prior to definitive therapy in 12 cases and was positive in 5 cases (detection rate of 42%). Liver-directed therapy involved: resection (n=15), ablation (6), radiation (2) or a combination (4). The median follow-up time from liver-directed therapy was 13.2 months. Post-local therapy chemotherapy was initiated prior to recurrence in 59% of cases. ct-DNA results contributed to immediate post-management decisions in 15% of cases. There were 14 radiographic recurrences (11 to liver only, 3 to other sites). Positive ct-DNA on initial testing following local therapy was associated with poorer DFS in univariate analysis and was the only prognostic factor identified on multivariable analysis. Conclusions: ct-DNA positivity following local therapy for CRLM is associated with earlier recurrence. Baseline testing before local therapy may be important for interpretation of testing results post-local therapy, given frequent missed disease detection. Additional clinical trial data is needed to explore ct-DNA-guided treatment decisions in CRLM. [Table: see text]