Radiotherapy for Glioblastoma in the Elderly

F. Keime-Guibert  NEJM 2007;356:1527

Patients 70 years of age or older with a newly diagnosed anaplastic astrocytoma or glioblastoma and a Karnofsky performance score of 70 or higher were randomly assigned to receive supportive care only or supportive care plus radiotherapy (focal radiation in daily fractions of 1.8 Gy given 5 days per week, for a total dose of 50 Gy).
We randomly assigned 85 patients from 10 centers to receive either radiotherapy and supportive care or supportive care alone. The trial was discontinued at the first interim analysis, which showed that with a preset boundary of efficacy, radiotherapy and supportive care were superior to supportive care alone. A final analysis was carried out for the 81 patients with glioblastoma (median age, 73 years; range, 70 to 85). At a median follow-up of 21 weeks, the median survival for the 39 patients who received radiotherapy plus supportive care was 29.1 weeks, as compared with 16.9 weeks for the 42 patients who received supportive care alone. The hazard ratio for death in the radiotherapy group was 0.47 (95% confidence interval, 0.29 to 0.76; P=0.002). There were no severe adverse events related to radiotherapy. The results of quality-of-life and cognitive evaluations over time did not differ significantly between the treatment groups.

Conclusions Radiotherapy results in a modest improvement in survival, without reducing the quality of life or cognition, in elderly patients with glioblastoma.

This study shows that the addition of radiotherapy to supportive care prolongs survival and does not reduce the health-related quality of life or cognitive function of patients with newly diagnosed glioblastoma who are 70 years of age or older. The 16.9-week median survival of patients in our study who received only supportive care is consistent with the median survival reported more than two decades ago for younger patients treated with supportive care alone. Conversely, the 12.2-week survival benefit with radiotherapy in the older patients in our study is about half the survival gain reported in the two earlier studies (22 and 24 weeks), which compared conventional radiotherapy (a total dose of 45 to 60 Gy, given in fractions of 1.7 to 2.0 Gy) with supportive care in a younger population

We selected a conventional 50-Gy schedule to minimize the age-related risk of radiation-induced neurotoxicity. With this schedule, there were no cases of delayed neurotoxicity,but the short survival of the patients in our study may have precluded the development of late toxicity. The optimal dose of radiotherapy in elderly patients remains undetermined. It is unclear whether a total dose of 60 Gy would increase the survival benefit of radiotherapy in older patients, as it does in younger patients. In our trial, the 29.1-week median survival and 15% rate of discontinuation of radiotherapy compare favorably with the 22.1-week survival and 26% rate of discontinuation reported in a prospective study in which a dose of 60 Gy was delivered in 30 fractions over a period of 6 weeks in older patients. In that study, an abbreviated course of radiotherapy (40 Gy in 15 fractions over a period of 3 weeks), as compared with the 60-Gy schedule, resulted in a similar median survival (24.3 weeks vs. 22.1 weeks), but a lower rate of premature discontinuation of radiotherapy (10% vs. 26%).

Since the goal of the treatment of glioblastoma in older patients is palliation, the quality of life is relevant. The evaluation of health-related quality of life in patients with malignant glioma is notoriously difficult.The main limitation is that severely ill patients with a rapidly progressive, fatal disease are not always compliant with such evaluations. To our knowledge, only one prospective study in the elderly attempted to evaluate health-related quality of life sequentially in patients with glioblastoma, but the data could not be analyzed because the compliance rate was too low (15 to 21% immediately after radiotherapy).17 In our study, the rate of compliance with the questionnaires was similar to that in the few reported studies of patients with glioblastoma. We were able to conduct a meaningful analysis of the data obtained from the first four follow-up evaluations (135 days of follow-up). Thereafter, the number of patients was too small for a reliable analysis.

At baseline, scores for the health-related quality of life did not differ significantly between the two groups and were similar to scores reported previously.During and after treatment, scores on several evaluation scales (physical, cognitive, social, fatigue, and motor dysfunction) progressively declined in both groups, although the global score for health-related quality of life did not change significantly over time in either group. We did not observe the mild-to-moderate improvement on several scales that was reported in younger patients after radiotherapy, with or without chemotherapy.

The evaluation of neuropsychiatric symptoms and cognitive function is associated with the same compliance limitation as the health-related quality-of-life analysis. The Neuropsychiatric Inventory did not show a time or treatment effect, particularly in the case of depression. In contrast, the cognitive evaluation (the MMSE and the initiation subscale of the MDRS) showed significant deterioration over time in both groups. However, as compared with supportive care alone, radiotherapy plus supportive care did not have a detrimental effect on cognitive function.

The population of elderly patients with cancer is underrepresented in clinical trials. Likely causes of this underrepresentation are study-imposed restrictions, coexisting conditions, concern about the toxic effects of treatment, patient and family preferences, and the reluctance of physicians to enroll elderly patients in clinical trials. Nevertheless, our study shows that these barriers may be overcome, even in trials that involve a rapidly progressive, fatal disease, such as glioblastoma, and a palliative-care comparison group.

In conclusion, radiotherapy increases the median survival of elderly patients with glioblastoma who have a good performance status at the start of treatment. As compared with supportive care, radiotherapy in such patients does not cause further deterioration in the Karnofsky performance status, health-related quality of life, or cognitive functions, but the survival benefit is modest.