A Prognostic Model in Metastatic or Recurrent Gastric Cancer Patients with Good Performance Status Who Received First-Line Chemotherapy.

PURPOSE: Good performance status is widely known as a superior prognostic predictor. However, some patients have large survival differences despite having good performance status that are influenced by certain prognostic factors. The purpose of this study was to explore baseline host- or tumor-related factors and to establish a prognostic model for metastatic or recurrent gastric cancer patients with good performance status who received first-line chemotherapy. METHODS: A total of 310 metastatic or recurrent gastric cancer patients with good performance status who received first-line chemotherapy were enrolled. Prognostic significance was determined using multivariate Cox regression analysis. Incorporating all pretreatment indicators, a prognostic model was established. Overall survival outcomes were compared with different risk groups using the Kaplan-Meier method and log-rank test. RESULTS: In multivariate analysis, no previous gastrectomy [hazard ratio (HR) = 1.42; 95% confidence interval (CI) = 1.08-1.85], number of distant metastatic sites (HR = 1.47; 95% CI = 1.11-1.96), bone metastasis (HR = 2.20; 95% CI = 1.16–4.18), liver metastasis (HR = 1.77; 95% CI = 1.31-2.39), and an elevated neutrophil lymphocyte ratio (HR = 1.37; 95% CI = 1.04-1.79) were independent prognostic factors of overall survival. Patients were categorized into three risk groups according to their risk scores. Median survival times for the low-risk (0 point), intermediate-risk (1-3 points), and high-risk (≥4 points) groups were 19.7, 10.7 and 5.1 months, respectively (P < .001). CONCLUSIONS: A prognostic model was developed that could facilitate risk stratification for metastatic or recurrent gastric cancer patients with good performance status who received first-line chemotherapy to help clinicians choose an applicable treatment based on the estimated prognosis.