Background: Thetype of radiological progression following bevacizumab treatment failure is related to outcome in patients with progressive glioblastoma. In an exploratory analysis of the randomized, placebo-controlled phase III clinical trial in newly diagnosed glioblastoma of bevacizumab (bev) plus radiotherapy/temozolomide (AVAglio) we aimed to investigate whether these progression types (PT) are induced by therapy (anti-angiogenic versus sole cytotoxic chemotherapy), are related to outcome and may therefore reflect a particular resistance mechanism. Methods: MRI scans (baseline until IRF progress) of 621 patients (bev arm, n = 299; placebo arm, n = 322) were available for analysis in sufficient detail and quality. PT were categorized according to contrast enhancement behavior in T1 and T2 hyper-intense signal change as published before (Nowosielski et al, Neurology 2014). Frequencies of the 5 PTs (cT1 flare-up, classic T1 progress, T2-diffuse, T2-circumscribed and primary non-responder), time to development (PFS) and overall survival (OS) were assessed within each treatment arm. Results: Classic T1 progress and primary non responder were more common in the placebo than in the bev arm (56.8 % versus 25.1% and 21.7% versus 3.3%, respectively.) In the bev arm, cT1 flare up and the T2 circumscribed PT were more common (37.5% versus 10.9% and 21.7%. versus 3.4%, respectively). T2 diffuse was equally distributed between the treatment arms and showed longer PFS compared to the other PTs in both arms. In the bev arm, cT1 flare up as well as T2 diffuse showed longer OS than all the other PTs (17.8 months vs median in T2 diffuse not yet reached). Conclusions: Exploratory analysis of AVAglio imaging data showed that radiologic progression types seem to be specific for anti-angiogenic or temozolomide treatment failure. These progression types show differences in PFS and OS and may therefore reflect a particular resistance mechanism that may guide further post progression therapy. Clinical trial information: NCT00943826