There is increasing evidence that the presence of an ongoing systemic inflammatory response is associated with poor outcome in patients with advanced cancer. The aim of the present study was to examine whether an inflammation-based prognostic score (Glasgow Prognostic score, GPS) was associated with survival, in patients with inoperable gastro-oesophageal cancer. Patients diagnosed with inoperable gastro-oesophageal carcinoma and who had measurement of albumin and C-reactive protein concentrations, at the time of diagnosis, were studied (n=258). Clinical information was obtained from a gastro-oesophageal cancer database and analysis of the case notes. Patients with both an elevated C-reactive protein (>10 mg l(-1)) and hypoalbuminaemia (<35 g l(-1)) were allocated a GPS score of 2. Patients in whom only one of these biochemical abnormalities was present were allocated a GPS score of 1, and patients with a normal C-reactive protein and albumin were allocated a score of 0. On multivariate survival analysis, age (hazard ratio (HR) 1.22, 95% CI 1.02-1.46, P<0.05), stage (HR 1.55, 95% CI 1.30-1.83, P<0.001), the GPS (HR 1.51, 95% CI 1.22-1.86, P<0.001) and treatment (HR 2.53, 95% CI 1.80-3.56, P<0.001) were significant independent predictors of cancer survival. A 12-month cancer-specific survival in patients with stage I/II disease receiving active treatment was 67 and 60% for a GPS of 0 and 1, respectively. For stage III/IV disease, 12 months cancer-specific survival was 57, 25 and 12% for a GPS of 0, 1 and 2, respectively. In the present study, the GPS predicted cancer-specific survival, independent of stage and treatment received, in patients with inoperable gastro-oesophageal cancer. Moreover, the GPS may be used in combination with conventional staging techniques to improve the prediction of survival in patients with inoperable gastro-oesophageal cancer.