Background: Patients (pts) with recurrent glioblastoma (rGB) have a poor prognosis, and no treatment option demonstrated to improve survival in a randomized trial. Axitinib (AXI), an oral VEGFR 1-3 inhibitor has demonstrated single agent activity in rGB and reduces the need for corticosteroids (CS). Avelumab (AVE) is a fully human anti-PD-L1 IgG1 antibody with clinical activity in various tumor types. Combination of AXI and AVE may improve the outcome of pts with rGB. Methods: This open-label, dual-strata, single-center phase 2 clinical trial investigated the activity of AXI plus AVE in adult pts with rGB following prior surgery, RT and temozolomide. Pts were stratified according to their baseline use of CS. Pts without baseline need for CS initiated treatment with AXI (5 mg oral BID) plus AVE (10 mg/kg IV Q2W) (cohort-1). Pts in need of CS initiated AXI as a monotherapy; AVE could be added to AXI after 6 wks if the CS dose could be tapered to a physiologic dose level or less (cohort-2). Six-month-PFS served as the primary endpoint (with a prespecified threshold of ≥ 50% for cohort-1) according to Fleming one-stage design. Results: Between Jun 2017 and Aug 2018, 54 pts (27 per cohort) were enrolled (med age 55 y [range 19-75]; 63% male; 91% WHO PS 0-1). All pts in cohort-1 and 16 pts (59%) in cohort-2 received at least 1 dose of AVE. The 6-month-PFS was 18% (95% CI 4-33) in both cohorts. At the time of analysis, 2 pts were progression-free and continuing study treatment. Median OS in cohort-1 and -2 was respectively 26 wks (95% CI 21-32) and 18 wks (95% CI 14-22). No clear relation was found between baseline cognitive functioning (Cogstate subtests) and PFS/OS. The best overall response rate (iRANO) was 41% and 26% respectively for pts in cohort-1 and -2. The most frequent all-grade treatment-related adverse events (TRAE) were dysphonia (67%), lymphopenia (50%), diarrhea (48%), hypertension (48%), and fatigue (46%). The incidence of grade 3-4 TRAE was 30%; there were no grade 5 AE. Conclusions: The combination of AVE plus AXI is sufficiently well tolerated but did not meet the threshold for activity justifying further investigation in an unselected population of patients with rGB. Clinical trial information: NCT03291314