Quantitative radiographic analysis of phase II and III trials in recurrent glioblastoma treated with VB-111 with or without bevacizumab or bevacizumab monotherapy.

Abstract

2018 Background: VB-111 is a non-replicating adenovirus carrying a pro-apoptotic transgene for TNFR1/Fas under the control of a modified murine promoter to pre-proendothelin 1. The transgene is expressed only in angiogenic endothelial cells, and therefore VB-111 results in targeted apoptosis of neovascular vessels. The current study characterizes the quantitative radiographic results and impact on OS in phase 2 and 3 trials of recurrent glioblastoma (GBM) patients treated with VB-111 with or without bevacizumab (BV) or BV monotherapy. Methods: MRI data from a phase 2 (NCT01260506) and randomized, double arm, controlled phase 3 (GLOBE; NCT02511405) trial of VB-111 in recurrent GBM were used in current study: Arm A) VB-111 monotherapy until progression followed by combination VB-111 and bevacizumab (BV) (“Primed Combination”; Phase 2; N = 24); Arm B) VB-111 in combination with BV (“Unprimed Combination”; Phase 3; N = 124) and Arm C) BV monotherapy (“Control”; Phase 3; N = 120). Contrast enhanced T1-weighted digital subtraction was used to quantify tumor volume at all time points. Results: Baseline tumor volume was not significantly different between patient cohorts (Kruskal-Wallis; P = 0.1482; median~20mL). Continuous measures of baseline tumor volume were prognostic for OS in all treatment groups when controlling for therapy and age (Cox, P < 0.001, HR = 1.02). In patients with small tumors ( < 25mL), the “primed combination” cohort (Arm A) from the phase 2 trial had a significant OS advantage compared to both upfront combination of VB-111 and BV (Arm B; P = 0.0094, HR = 0.5328; median OS = 7mo vs. 15mo) as well as BV alone (Arm C; P = 0.0248, HR = 0.5776; median OS = 8.5mo vs. 15mo). Patients with a radiographic response ( > 65% reduction) had a significant survival difference from non-responders when controlling for age, baseline tumor volume, and treatment arm ( P = 0.0014, HR = 0.5822). Notably, in responders to VB-111 monotherapy or combination therapy after priming with VB-111 exhibited characteristic, expansive areas of necrosis in areas of initial disease. Conclusions: Small recurrent tumors have a significant OS advantage when “priming” with VB-111 monotherapy prior to combination VB-111 and BV at recurrence. Patients responding to VB-111 exhibit specific imaging characteristics related to the drug mechanism of action. Clinical trial information: NCT02511405; NCT01260506.