Glasgow prognostic score predicts prognosis for cancer patients in palliative settings: a subanalysis of the Japan–prognostic assessment tools validation (J-ProVal) study

The Glasgow prognostic score (GPS), which uses C-reactive protein and albumin levels, is a good predictor of prognosis in cancer patients undergoing anti-tumor therapy. The objective of this study was to investigate the correlation between GPS and survival among cancer patients in palliative settings, as findings in such populations have not been well described. This was a subanalysis of a multicenter, prospective, cohort study in patients who were adults, diagnosed with advanced cancer, and first referred to palliative care service in Japan. Patients who were not receiving anti-tumor therapy and who had undergone laboratory examinations were eligible. Clinical features were analyzed to investigate prognostic factors. A total of 1160 patients were enrolled (41.6 % female; median age, 72 years). The independent predictors were Eastern Cooperative Oncology Group performance status (ECOG PS) score of 4 (hazard ratio (HR), 1.54), liver metastasis (HR, 1.21), dyspnea (HR, 1.35), edema (HR, 1.25), prognostic performance index (HR, 1.56), neutrophil-lymphocyte ratio (HR, 1.43), and GPS of 2 (HR, 1.36). The sensitivity and specificity for 3-week prognosis of a GPS of 2 were 0.879 and 0.410. Median survival time with GPS of 0, 1, and 2 was 58 days (95 % confidence interval, 48–81), 43 days (37–50), and 21 days (19–24), respectively (log-rank test, p