C-reactive protein-to-albumin ratio predicts outcome of neoadjuvant chemotherapy for colorectal liver metastases: A multicenter prediction model.

Abstract

Neoadjuvant chemotherapy for patients with colorectal liver metastasis often results in varying outcomes. This study aimed to develop a clinical prediction model incorporating the C-reactive protein-to-albumin ratio to determine neoadjuvant chemotherapy efficacy in patients with colorectal liver metastasis METHODS: This retrospective study included 1,335 patients with colorectal liver metastasis who received neoadjuvant chemotherapy. Patients were classified into high and low C-reactive protein-to-albumin ratio groups on the basis of an optimal cutoff value. Multivariate regression and Cox proportional hazards models were used to examine associations between clinical factors and progression-free survival. A prediction model was constructed and validated both internally and externally. The optimal C-reactive protein-to-albumin ratio cutoff value was 0.234. Patients with a high C-reactive protein-to-albumin ratio had a greater risk of 3-year disease progression and shorter 5-year survival (both P < .05). The final model incorporated C-reactive protein-to-albumin ratio value, clinical risk score, and tumor differentiation. It demonstrated excellent performance in predicting progression-free survival (C-index = 0.69) and OS (C-index = 0.72), outperforming the clinical risk score alone in both internal and external validations. The C-reactive protein-to-albumin ratio is an effective independent predictor for patients with colorectal liver metastasis receiving neoadjuvant chemotherapy. The clinical prediction model exhibited superior predictive ability, potentially improving individualized treatment decisions patients with colorectal liver metastasis.