Radiotherapy is effective in patients with glioblastoma multiforme with a very limited prognosis and in patients above 70 years of age

Purpose/Objective: Treatment of patients with a glioblastoma multiforme (GBM) or astrocytoma grade 4 generally consists of surgery followed by radiotherapy. Different radiotherapy dose schedules are in use, depending on the prognosis of the patients, which is very limited in most cases. The recursive partitioning analysis as reported by the RTOG is based on prognostic factors like age, type of treatment, Karnofsky performance score, mental state and neurological functioning. It enables physicians to estimate prognosis for an individual patient. For GBM, 4 groups are identified, with a median survival of 17.9, 11.1, 8.9 and 4.6 months respectively. However, the outcome of this recursive partitioning is based on results of randomized trials, which exclude patients above 70 years of age. Purpose of this study was to evaluate whether recursive partitioning can predict prognosis adequately for all patients with a GBM treated in a single institution. In addition to this, the value of radiotherapy in patients with a very limited life expectancy was studied. Materials/Methods: We retrospectively evaluated all patients treated between 1990 and 2000 in our institution with a GBM. Astrocytomas with an oligodendroglial component were excluded from this analysis. For all patients, details of tumor localisation and treatment were recorded. Karnofsky performance score, mental state and neurological functioning were subtracted from the notes, if not mentioned specifically. Based on these details, patients (including patients > 70) were divided in the RTOG recursive partitioning groups. Survival was calculated from the date of operation. Results: In total, 208 patients with GBM were identified. Five patients were lost to follow up within 1 month after the operation, leaving 203 patients for analysis. The median age was 58 years, 73 were female and 130 were male. The number of patients in RTOG recursive partioning groups were 18, 86, 60 and 39 for group III, IV, V and VI, respectively. At time of analysis, only one patient was still alive. The median survival was 8.0 months for the total group. The median survival for the recursive partitioning groups is 13.4,10.6, 3.8, 2.1 months for group III, IV, V and VI respectively. The median survival for patients > 70 years was 3.7 as compared to 8.0 months for the group 50–70 years and 11.0 months for patients < 50 years. In the RTOG groups separately, patients > 70 years had a similar survival compared with patients of 50–70 years. Patients who underwent RT (66%) survived significantly longer with 10.6 months vs. 1.9 months for patients who did not receive radiotherapy (p<0.0001). Because the application of radiotherapy in group V is often subject of discussion, we analyzed this group separately. Of the 60 patients in group V, 50% received radiotherapy, with a median survival of 9.4 months for the irradiated patients in contrast to only 2.1 months for non-irradiated patients. In a multivariate analysis including RT and Karnofsky score, RT proved to be the only prognostic factor for survival (HR 8.9, 95% CI 3.6–21.8, p<0.001). Conclusions: The implication of recursive partitioning in a single institution is feasible and enables more reliable comparison of treatment results. Prognosis for patients above 70 years of age is not different from younger patients, when corrected for RTOG grouping. We would therefore suggest not to exclude these patients from radiotherapy. In view of the limited prognosis of patients in RTOG group V and VI, short radiotherapy regimens are recommended for these patients.