Prognostic value of the glasgow prognostic score for glioblastoma multiforme patients treated with radiotherapy and temozolomide

Abstract

To evaluate the prognostic value of the Glasgow Prognostic Score (GPS), the combination of C-reactive protein (CRP) and albumin, in glioblastoma multiforme (GBM) patients treated with radiotherapy (RT) and concurrent plus adjuvant temozolomide (GPS). Data of newly diagnosed GBM patients treated with partial brain RT and concurrent and adjuvant TMZ were retrospectively analyzed. The patients were grouped into three according to the GPS criteria: GPS-0: CRP < 10 mg/L and albumin > 35 g/L; GPS-1: CRP < 10 mg/L and albumin < 35 g/L or CRP > 10 mg/L and albumin > 35 g/L; and GPS-2: CRP > 10 mg/L and albumin < 35 g/L. Primary end-point was the association between the GPS groups and the overall survival (OS) outcomes. A total of 142 patients were analyzed (median age: 58 years, 66.2% male). There were 64 (45.1%), 40 (28.2%), and 38 (26.7%) patients in GPS-0, GPS-1, and GPS-2 groups, respectively. At median 15.7 months follow-up, the respective median and 5-year OS rates for the whole cohort were 16.2 months (95% CI 12.7-19.7) and 9.5%. In multivariate analyses GPS grouping emerged independently associated with the median OS (P < 0.001) in addition to the extent of surgery (P = 0.032), Karnofsky performance status (P = 0.009), and the Radiation Therapy Oncology Group recursive partitioning analysis (RTOG RPA) classification (P < 0.001). The GPS grouping and the RTOG RPA classification were found to be strongly correlated in prognostic stratification of GBM patients (correlation coefficient: 0.42; P < 0.001). The GPS appeared to be useful in prognostic stratification of GBM patients into three groups with significantly different survival durations resembling the RTOG RPA classification.