Session Item

Dose-escalation in radiotherapy of glioblastoma has been a controversial topic. While some studies did not show oncologic benefits but higher rates of radionecrosis, several clinical protocols for dose escalation (e.g. with intraoperative radiotherapy or targeting tumor volumes identified by MR spectroscopy) have been initiated. In this analysis we used biomathematical modelling to hypothesize dose-response-relationship curves for clinically applied radiation schedules for the treatment of glioblastoma to hypothesize whether dose-escalation might possibly be promising in this clinical setting.

A literature research was performed to retrieve clinical data of glioblastoma patients treated with different radiation schedules and total treatment doses combined with temozolomide. Progression-free survival rates (PFS) at 6 months and 12 months were recorded. PFS at 6 months (weighted by the number of patients included in the respective study) was fitted with a linear-quadratic model and optimized manually. Dose-response-curves were calculated for PFS at 6 months.

27 studies were identified with different fractionation schedules and total treatment doses. In all studies, patients were treated with combined radiochemotherapy with temozolomide. The studies had included between 10 and 287 patients. PFS at 6 and 12 months ranged from 0.43 to 0.90 and 0.20 to 0.76, respectively. The linear-quadratic fit showed a good correlation with r=0.68 for PFS at 6 months. The applied fractionation schedules are in the steep part of the dose-response curve.

Clinical data of patients treated with radiotherapy with different fractionation schedules and total radiation dose were well fitted by a radiation dose response curve based on the linear quadratic model. PFS at 6 months showed that clinically applied radiation schedules are located in the steep part of the dose-response-curve. Thus, even limited dose escalation might be hypothesized to have a clinical benefit for the treatment outcome of glioblastoma patients.