Age and WHO Performance Score are major clinical prognostic factors in glioblastoma

In glioblastoma patients with similar prognostic factors—like age, performance status and extent of surgical resection—clinical evolution is still variable. In addition, the benefit of resection of recurrent tumours is uncertain. In this study, prognostic subgroups are established on a clinical basis and the benefit of surgery for recurrence is compared between them.
Recursive partitioning analysis on EORTC sample 26981 (n=239) identified clinical factors influencing Progression-Free Survival (PFS), in order to create prognostic subgroups, whose PFS and survival after recurrence were compared using Kaplan-Meier curves and log-rank tests. Results were validated on a sample from Liège University Hospital (n=92).
WHO Performance Status (WHO-PS) and age at first intervention (“age”) defined 4 subgroups. PFS was shorter in WHO-PS 2–3 than WHO PS 0–1 patients. In both groups, age was the second classifying factor (cut-off: 47.5y for WHO-PS 0–1, 60.5y for WHO-PS 2–3). The 4 subgroups had significantly different median PFS (77.5, 144, 234.5 and 346.5 days) and median overall survival (126, 330, 476 and 675 days). Survival after recurrence was correlated with PFS, and recurrence resection was associated with longer survival, regardless of the prognostic subgroup.
Clinical prognostic glioblastoma subgroups with distinct PFS exist and can be defined by WHO-PS and age. This classification can be used as a clinical basis to identify molecular patterns associated with prognosis. In addition, in all subgroups, this retrospective analysis suggests a benefit of surgical resection when feasible.