Background: The aim of this study was to investigate the possible prognostic factors and predictive accuracy of the Glasgow Prognostic Score (GPS) for patients with unresectable locally advanced esophageal squamous cell carcinoma (LAESCC) treated with chemoradiotherapy.
Methods: One hundred forty-two patients were enrolled in JCOG0303 and assigned to the standard cisplatin and 5-fluorouracil (PF)-radiotherapy (RT) group or the low-dose PF-RT group. One hundred thirty-one patients with sufficient data were included in this analysis. A Cox regression model was used to analyze the prognostic factors of patients with unresectable LAESCC treated with PF-RT. The GPS was classified based on the baseline C-reactive protein (CRP) and serum albumin levels. Patients with CRP ≤1.0 mg/dL and albumin ≥3.5 g/dL were classified as GPS0. If only CRP was increased or only albumin was decreased, the patients were classified as GPS1, and the patients with CRP >1.0 mg/dL and albumin <3.5 g/dL were classified as GPS2.
Results: The patients' backgrounds were as follows: median age (range), 62 (37-75); male/female, 119/12; ECOG PS 0/1/2, 64/65/2; and clinical stage (UICC 5th) IIB/III/IVA/IVB, 3/75/22/31. Multivariable analyses indicated only esophageal stenosis as a common factor for poor prognosis. In addition, overall survival tended to decrease according to the GPS subgroups (median survival time (months): GPS0/GPS1/GPS2 16.1/14.9/8.7).
Conclusions: Esophageal stenosis was identified as a candidate stratification factor for randomized trials of unresectable LAESCC patients. Furthermore, GPS represents a prognostic factor for LAESCC patients treated with chemoradiotherapy.
Clinical trial information: UMIN000000861.
Keywords: Chemoradiotherapy; Esophageal stenosis; Glasgow Prognostic Score; Prognostic factor; T4 esophageal cancer.