The preoperative albumin-to-carcinoembryonic antigen ratio (ACR) predicts prognosis and facilitates risk stratification in gastric cancer: a retrospective cohort study.

[AIM] This study aimed to evaluate the association between the preoperative albumin-to-carcinoembryonic antigen ratio (ACR) and clinicopathological characteristics as well as the association between ACR and prognosis in gastric cancer, so as to provide evidence for improved risk stratification and personalized management.

[METHODS] Clinicopathological data of gastric cancer patients who underwent radical gastrectomy at Shanxi Province Cancer Hospital between December 2015 and January 2017 were retrospectively reviewed. Following patient follow-up, Cox proportional-hazards regression was used to identify prognostic factors for overall survival (OS) and disease-free survival (DFS). An ACR-incorporated nomogram was developed, and its predictive accuracy was evaluated using the concordance index (C-index) and calibration curves. Its discriminative ability was further compared against that of the conventional TNM staging system.

[RESULTS] Among the 1,161 enrolled patients, 192 (16.5%) and 969 (83.5%) were classified into low- and high-ACR groups, respectively. Low ACR was correlated with adverse pathological features and inferior survival outcomes. Both OS and DFS were significantly shorter in the low-ACR group (both < 0.001). Multivariate analysis identified high ACR as an independent protective factor for OS (HR = 0.741, 95% CI: 0.606-0.906; = 0.003) and DFS (HR = 0.809, 95% CI: 0.655-0.998; = 0.048). Time-dependent receiver operating characteristic (ROC) analysis confirmed the good predictive performance of the ACR-based model in both training and validation sets. The nomogram demonstrated superior predictive accuracy (C-index: 0.748 for OS, 0.730 for DFS) compared with the TNM staging system.

[CONCLUSION] Preoperative low ACR is significantly associated with aggressive tumor biology and poor survival in gastric cancer. The ACR-based nomogram serves as a clinically useful tool for prognostic prediction, risk stratification, and the guidance of personalized therapy.