Most patients with glioblastoma are older than 60 years, but treatment guidelines are based on trials in patients aged only up to 70 years. We did a randomised trial to assess the optimum palliative treatment in patients aged 60 years and older with glioblastoma.

Patients with newly diagnosed glioblastoma were recruited from Austria, Denmark, France, Norway, Sweden, Switzerland, and Turkey. They were assigned by a computer-generated randomisation schedule, stratified by centre, to receive temozolomide (200 mg/m² on days 1–5 of every 28 days for up to six cycles), hypofractionated radiotherapy (34·0 Gy administered in 3·4 Gy fractions over 2 weeks), or standard radiotherapy (60·0 Gy administered in 2·0 Gy fractions over 6 weeks). Patients and study staff were aware of treatment assignment. The primary endpoint was overall survival. Analyses were done by intention to treat. This trial is registered, number ISRCTN81470623.

342 patients were enrolled, of whom 291 were randomised across three treatment groups (temozolomide n=93, hypofractionated radiotherapy n=98, standard radiotherapy n=100) and 51 of whom were randomised across only two groups (temozolomide n=26, hypofractionated radiotherapy n=25). In the three-group randomisation, in comparison with standard radiotherapy, median overall survival was significantly longer with temozolomide (8·3 months [95% CI 7·1–9·5; n=93] vs 6·0 months [95% CI 5·1–6·8; n=100], hazard ratio [HR] 0·70; 95% CI 0·52–0·93, p=0·01), but not with hypofractionated radiotherapy (7·5 months [6·5–8·6; n=98], HR 0·85 [0·64–1·12], p=0·24). For all patients who received temozolomide or hypofractionated radiotherapy (n=242) overall survival was similar (8·4 months [7·3–9·4; n=119] vs 7·4 months [6·4–8·4; n=123]; HR 0·82, 95% CI 0·63–1·06; p=0·12). For age older than 70 years, survival was better with temozolomide and with hypofractionated radiotherapy than with standard radiotherapy (HR for temozolomide vs standard radiotherapy 0·35 [0·21–0·56], p<0·0001; HR for hypofractionated vs standard radiotherapy 0·59 [95% CI 0·37–0·93], p=0·02). Patients treated with temozolomide who had tumour MGMT promoter methylation had significantly longer survival than those without MGMT promoter methylation (9·7 months [95% CI 8·0–11·4] vs 6·8 months [5·9–7·7]; HR 0·56 [95% CI 0·34–0·93], p=0·02), but no difference was noted between those with methylated and unmethylated MGMT promoter treated with radiotherapy (HR 0·97 [95% CI 0·69–1·38]; p=0·81). As expected, the most common grade 3–4 adverse events in the temozolomide group were neutropenia (n=12) and thrombocytopenia (n=18). Grade 3–5 infections in all randomisation groups were reported in 18 patients. Two patients had fatal infections (one in the temozolomide group and one in the standard radiotherapy group) and one in the temozolomide group with grade 2 thrombocytopenia died from complications after surgery for a gastrointestinal bleed.

Standard radiotherapy was associated with poor outcomes, especially in patients older than 70 years. Both temozolomide and hypofractionated radiotherapy should be considered as standard treatment options in elderly patients with glioblastoma. MGMT promoter methylation status might be a useful predictive marker for benefit from temozolomide.

Merck, Lion’s Cancer Research Foundation, University of Umeå, and the Swedish Cancer Society.