Modified Glasgow prognostic score predicts the prognosis of patients with advanced esophageal squamous cell carcinoma: A propensity score‐matched analysis

Abstract

Objective

The aim of this study is to evaluate the prognostic value of the modified Glasgow prognostic score (mGPS) in advanced esophageal squamous cell carcinoma (SCC) patients.

Methods

The study enrolled 311 patients with advanced esophageal SCC from January 2012 to December 2018. Univariate and multivariate analyses were calculated by the Cox proportional hazards regression model in advanced esophageal SCC patients. The Kaplan–Meier method was used to evaluate the ability of the mGPS for survival rates. Propensity score‐matched (PSM) analysis was carried out to balance imbalanced variables.

Results

The Cox proportional hazards analysis showed that factors including M stage, ECOG, mGPS group, and sex were identified as independent predictors. The mGPS presented a good level of overall survival (OS) prediction with a risk‐adopted classification for advanced esophageal SCC patients. The survival rates in advanced esophageal SCC patients with mGPS 0, 1, and 2 were 18.8%, 8.4%, and 4.2%, respectively ( p < 0.001). Moreover, before and after PSM, the mGPS was associated with 3‐year survival rates of advanced esophageal SCC patients in the Kaplan–Meier survival analysis. In addition, the mGPS for OS prediction demonstrated better performance than sex and ECOG score. The area under curve (AUC) of the mGPS combined with M stage for the prognosis of advanced esophageal SCC was 0.677 (0.592–0.763).

Conclusion

The mGPS is a cost‐effective, accessible tool capable of prognosticating in this cohort. It could be a useful surveillance system of prognosis in advanced esophageal SCC patients.

Abstract

The aim of this study was to evaluate the prognostic value of the modified Glasgow prognostic score (mGPS) in advanced esophageal squamous cell carcinoma (SCC) patients. The mGPS was determined as follows: mGPS 0 = CRP (C‐reactive protein) ≤ 10 mg/L, mGPS 1 = CRP > 10 mg/L, and albumin >3.5 g/dl, and mGPS 2 = CRP > 10 mg/L and albumin <3.5 g/dl. In this study, there were only 24 cases in the mGPS 2, therefore the mGPS 1 and mGPS 2 were combined. The mGPS group was determined as follows: mGPS group 0 = mGPS 0; the mGPS group 1 was defined by the composition of mGPS 1 and mGPS 2. Considering the risk factors were imbalanced between mGPS group 1 and mGPS group 0, we applied a 1:1 propensity score‐matched ratio to minimize these differences. Before matching, the 3‐year survival was 18.8% in the mGPS group 0 versus 7.6% in the mGPS group 1 ( p < 0.001). After matching, the 3‐year survival was 19.6% in the mGPS group 0 versus 7.1% in the mGPS group 1 ( p < 0.001). The mGPS is a cost‐effective, accessible tool capable of prognosticating in this cohort. It could be a useful surveillance system of prognosis in advanced esophageal SCC patients.