Abstract IA20: Phase II study to evaluate the clinical efficacy and safety of MEDI4736 (durvalumab) in patients with glioblastoma (GBM): Results for cohort B (durvalumab monotherapy), bevacizumab-naïve patients with recurrent GBM

Abstract

Abstract Background: Outcome for patients with glioblastoma (GBM), the most common primary malignancy of the central nervous system, remains poor. Inhibition of PD-1 signaling has been shown to improve survival in an orthotopic, syngeneic murine GBM model. PD-L1 is expressed by GBM tumors, and its level of expression has been shown to correlate with patient outcome. Durvalumab (DUR) is a human IgG1 monoclonal antibody against PD-L1. Blockade of PD-1/PD-L1 has shown benefit among solid tumors; data implicate PD-1/PD-L1 signaling as a significant contributor to immunosuppression in glioblastoma (GBM). Bevacizumab (BEV) is an approved angiogenesis inhibitor for recurrent GBM; angiogenesis inhibition may promote antitumor benefit of immunotherapies. Methods: This ongoing phase II open-label study (NCT02336165) evaluates the safety and efficacy of DUR (10 mg/kg every 2 weeks) in 5 GBM cohorts with and without BEV. This report presents the safety and efficacy for Cohort B (BEV-naïve recurrent GBM; DUR monotherapy). The primary efficacy endpoint for Cohort B is progression-free survival at 6 months (PFS-6), based on modified RANO criteria by investigator assessment; secondary endpoints include safety/tolerability. A PFS-6 of 10% is the comparative historical benchmark (pre-BEV), based on published meta-analyses. The null hypothesis (PFS-6 ≤ 10%) was tested in the Intent-to-Treat (ITT) population against the alternative hypothesis (PFS-6 > 10%) at α = 0.05, using 90% confidence interval (CI). The null hypothesis was rejected if the lower bound of the CI was > 10%. ITT includes patients receiving any dose of DUR and having at least baseline and 1 post-baseline tumor assessment. Results: The first patient was dosed on 05Mar2015, and the data cutoff was 05Feb2016. Cohort B completed enrollment of 31 patients (male: 83.9%; mean age: 54.0 [24-77] years; baseline ECOG Performance Status (PS) 0: 51.6%, PS 1: 48.4%; baseline measurable lesions: 77.4%). Incidences of treatment-related adverse events (TRAEs) by maximum CTCAE grade (Gr) were Gr1: 35.5%; Gr2: 41.9%; Gr3: 9.7%; and Gr4/5: 0%. Most common TRAEs (≥3 (9.7%) patients) were fatigue, headache, hemiparesis, increased AST, and decreased platelets, WBCs and lymphocytes. Thirty patients were evaluable for efficacy (PFS), and 6 were progression free at 6 months. Kaplan-Meier estimate for PFS-6 was 20%. Conclusions: DUR appears to be well tolerated and shows activity in BEV-naïve recurrent GBM, although the primary PFS-6 endpoint was not met. Further studies are warranted. Citation Format: David A. Reardon, Thomas Kaley, Jorg Dietrich, Michael Lim, Gavin P. Dunn, Hui K. Gan, Timothy Cloughesy, Jennifer L. Clarke, Andrew Park, Mary Macri, Aileen Ryan, Toni Ricciardi, Vijay Reddy, Ralph Venhaus. Phase II study to evaluate the clinical efficacy and safety of MEDI4736 (durvalumab) in patients with glioblastoma (GBM): Results for cohort B (durvalumab monotherapy), bevacizumab-naïve patients with recurrent GBM [abstract]. In: Proceedings of the Second CRI-CIMT-EATI-AACR International Cancer Immunotherapy Conference: Translating Science into Survival; 2016 Sept 25-28; New York, NY. Philadelphia (PA): AACR; Cancer Immunol Res 2016;4(11 Suppl):Abstract nr IA20.