Center for Neuro-Oncology, Dana-Farber Cancer Institute
Purpose: Vascular endothelial growth factor (VEGF) is upregulated in glioblastoma and may contribute to immunosuppression. We performed a phase 2 study of pembrolizumab, a programmed death-1 (PD-1) blocking antibody alone or with the anti-VEGF antibody bevacizumab in recurrent glioblastoma with detailed analyses of biomarkers and patient neurologic function. Methods: Eighty bevacizumab-naive, recurrent glioblastoma patients were randomized to receive pembrolizumab with bevacizumab (cohort A, n=50) or pembrolizumab monotherapy (cohort B, n=30). The primary endpoint was six-month progression-free survival (PFS-6). Exploratory endpoints included evaluation of tumor PD-L1 expression, TIL density, immune activation gene expression signature and plasma cytokines with outcome. Changes in neurologic function were prospectively assessed using the Neurologic Assessment in Neuro-Oncology (NANO) scale. Results: Pembrolizumab alone or with bevacizumab was well tolerated but of limited benefit. For cohort A, PFS-6 was 26.0% (95% CI: 16.3, 41.5), median OS was 8.8 months (95% CI: 7.7, 14.2), ORR was 20% and median duration of response was 48 weeks. For cohort B, PFS-6 was 6.7% (95% CI: 1.7, 25.4), median OS was 10.3 months (95% CI: 8.5, 12.5) and ORR was 0%. Factors associated with worsened OS included baseline dexamethasone use and increased post-therapy plasma VEGF (cohort A) and wild-type IDH1, unmethylated MGMT and increased baseline PlGF and sVEGFR1 levels (cohort B), but tumor immune markers were not informative. The NANO scale effectively predicted neurologic function. Conclusions: Although well tolerated, pembrolizumab was ineffective both as monotherapy and with bevacizumab for recurrent glioblastoma. Nonetheless, radiographic responses to combinatorial therapy were durable. Baseline dexamethasone use and plasma cytokines but not tumor immunologic biomarkers were associated with outcome. Neurologic function evaluated by the NANO scale contributed to outcome assessment.