Prognostic value of the preoperative albumin-bilirubin score among patients with stages I-III gastric cancer undergoing radical resection: A retrospective study

Abstract

The albumin-bilirubin (ALBI) score was originally used to accurately assess liver dysfunction and predict the prognoses of patients with hepatocellular carcinoma. Following its more recent application to patients with gastrointestinal tumors, this study analyzed the prognostic value of the ALBI score in Chinese patients with advanced resectable (tumor-node-metastasis [TNM] stages I-III) gastric cancer (GC). This study investigated 1185 patients with advanced GC, including 429 with TNM stage I. The patients were divided into training and verifications groups (593 and 592 patients, respectively) in which these patients were classified as high risk (ALBI score ≥ -2.65) or low risk (ALBI score < -2.65). Univariate and multivariate Cox regression analyses were performed, and a visual survival prediction model (nomogram) was created. On Kaplan-Meier analysis, patients who were low-risk and high-risk according to their ALBI scores had significantly different survival rates in both the training and verification groups (p < 0.01). The difference was also significant when analyzing only patients with TNM stage I GC (p = 0.031). Univariate and multivariate analyses showed that the ALBI score (p = 0.014), age (p < 0.001), Nutritional Risk Screening 2002 score (p = 0.024), sarcopenia (p = 0.049), tumor differentiation (p = 0.027), and TNM stage (p < 0.001) were independent risk factors for survival. Our survival prediction model that incorporated the ALBI score accurately predicted the 5-year survival rate of Chinese patients with GC. Therefore, the ALBI score is a valid clinical indicator and good predictor of survival after surgery for progressive GC. Moreover, this score is simple to derive and does not burden patients with additional costs.

FIGURE 1

Flow chart for the inclusion and exclusion of gastric cancer (GC) patients with tumor‐node‐metastasis stages (TNM) I–III. Based on strict inclusion and exclusion criteria, 1185 GC patients with TNM I–III at the hospital between August 2014 and June 2019 were selected for the study. ALBI, albumin‐bilirubin; GC, gastric cancer; TNM, tumor‐node‐metastasis.

FIGURE 2

Kaplan–Meier curves of the overall survival of gastric cancer patients based on the albumin‐bilirubin (ALBI) score. (a) The overall survival curves of the training group (p < 0.01). (b) The overall survival curves of the verification group (p < 0.01). P values are calculated by log‐rank test.

FIGURE 3

Efficacy of albumin‐bilirubin (ALBI) scores for tumor‐node‐metastasis (TNM) stage I gastric cancer (GC) patients. (a) The overall survival (OS) curves for TNM stage I GC patients (p = 0.031). (b) Receiver operating characteristic (ROC) curve for 1‐, 2‐, and 3‐year overall survival (OS). (c) Time‐dependent area under curve (AUC) for assessing the accuracy of the ALBI score to predict OS. P values are calculated by log‐rank test. ALBI, albumin‐bilirubin; AJCC, American Joint Committee on Cancer; AUC, area under the curve; TNM, tumor‐node‐metastasis; OS, overall survival.

FIGURE 4

Novel nomogram model for gastric cancer patients. (a) Nomogram model for overall survival (OS). (b) Calibration curves of the nomogram for predicting 1‐, 2‐, and 3‐year OS. (c) Decision curve analysis of the nomogram model. (d) Time‐dependent area under the curve for assessing the predictive accuracy of the nomogram for OS. ALBI, albumin‐bilirubin; AUC, area under the curve; NRS 2002, Nutritional Risk Screening 2002 score; OS, overall survival; TNM, tumor‐node‐metastasis.