Outcome evaluation in glioblastoma patients using different ranking scores: KPS, GOS, mRS and MRC.

Patient performance is an overall accepted independent prognostic factor in glioblastoma patients. Its estimation is essential for treatment planning, follow-up and clinical trials. Patient performance is mostly determined by usage of the Karnofsky Performance Score (KPS) for cancer patients. However, several other ranking scores have been developed specifically for patients with neurological diseases: Glasgow Outcome Score (GOS) for trauma patients, modified Ranking Score for stroke patients and Medical Research Council brain prognostic index (MRC) for brain tumour patients. The aims of this study were: (1) to compare these four performance scores in their ability to determine patient survival; and (2) to compare the prognostic value of performance with that of other prognostic factors. Univariate and multivariate survival analysis was used. Survival analysis revealed a high correlation to survival for all four scores. The maximum derivation of the curves was shown for the MRC and GOS. Performance had more clinical impact in determining patient survival than age and tumour resection. Differential treatment planning may need the formation of more than two patient groups. This was possible with the MRC, as well as the GOS and KPS. Forming more than three patient groups was not effective with any score.