Background: In patients with MGMT-nonmethylated glioblastoma(GBM),primary chemotherapy with temozolomide(TMZ)is at best moderately effective.There is an urgent need for more effective therapies in this large subgroup of GBM.Since results of phase Ⅱ trials with the antiangiogenic agent bevacizumab(BEV)+/-irinotecan(IRI)are promising in recurrent GBM,the GLARIUS trial explored the efficacy of BEV/IRI as compared to standard TMZ in the first-line therapy of MGMT-non-methylated GBM.Methods: In the randomized,multicenter,open-label GLARIUS trial,adult patients with newly diagnosed,histologically confirmed and MGMT-non-methylated GBM received local radiotherapy(RT,30 x 2 Gy)and were randomized(2:1)for experimental therapy with BEV(10 mg/kg q2w)during RT followed by maintenance BEV(10 mg/kg q2w)+ IRI(125 mg/m2 q2w(without enzyme-inducing antiepileptic drugs(EIAEDs))or 340 mg/m2(with EIAEDs))or standard therapy with daily TMZ(75 mg/m2)during RT followed by 6 courses of TMZ(150-200 mg/m2/day for 5 days q4w).The primary endpoint was progression-free survival rate after 6 months(PFS-6)as determined by central neuroradiological review.Results: The intent-to-treat population included 170 patients(67.1%male,median age 56 years(range 25-78 years),48.8%complete resection rate,78.8%of patients with KPS 90%or higher); 116 patients received BEV/IRI,54 patients had TMZ.The frequencies of adverse events in both arms of the trial were within the expected range.The PFS-6 rate was significantly higher in the BEV/IRI arm(71.1%,95%CI 58.1-80.8%)than in the TMZ arm(26.2%,95%CI 13.1-41.4%,p<0.0001 logrank test).Conclusions: The significant and clinically meaningful increase in the primary endpoint PFS-6 upon BEV/IRI chemotherapy suggests that BEV/IRI is superior to standard TMZ therapy in newly diagnosed MGMT-nonmethylated GBM patients.