Refining outcomes in technically resectable colorectal liver metastases: A simplified risk score and the role of preoperative chemotherapy

Background To improve surgical outcomes in patients with colorectal liver metastasis (CRLM) after curative liver resection, it is essential to provide individualized treatment using a simple, practical risk score, as conventional ones are cumbersome. We aimed to establish such risk score to predict CRLM prognosis post-resection. Methods This study involved 115 patients who underwent initial curative liver resection for CRLM across multiple centers. Risk factors for recurrence and survival were analyzed using Cox proportional hazards models and log-rank tests. The predictive performance of the established prognostic criteria for recurrence-free survival (RFS) and overall survival (OS) was assessed. Model performance was compared with established prognostic tools (the Beppu nomogram and Fong’s clinical risk score) using the Akaike Information Criterion (AIC). We also examined the outcomes in subgroups based on their response to preoperative chemotherapy. Results The combination of CRLM count (≥ 3) and largest tumor diameter (≥ 5 cm) was identified as the only independent risk factor for both recurrence (hazard ratio [HR] = 2.05, p = 0.007) and survival (HR = 2.24, p = 0.017). The model demonstrated comparable AIC values for recurrence (609.61) and survival (327.68) relative to the Beppu nomogram (AIC = 611.34) and Fong’s score (AIC = 327.25), respectively. Among patients who received preoperative chemotherapy (n = 72), those classified as high-risk had significantly poorer RFS (HR = 3.11, p < 0.001) and OS (HR = 2.80, p = 0.010) than their lower-risk counterparts did. Moreover, patients with progressive disease during chemotherapy had significantly worse outcomes than those who achieved tumor control. Conclusions This simple two-parameter model offers a practical prognostic tool after CRLM resection, with the inclusion of chemotherapy response assessment further enhancing risk stratification and outcome prediction.