Introduction: Checkpoint inhibition has demonstrated clinical efficacy in a variety of solid tumors. Reports of PD-L1 expression in glioblastoma are highly variable (ranging from 6% to 88%) and its role as a prognostic marker has yielded conflicting results.
Methods: Using tissue microarrays, we compared five PD-L1 monoclonal antibodies (n=56) and validated expression (n=183) using quantitative immunohistochemistry and RNA in situ hybridization. Expression data from the TCGA and published studies were compared with clinical outcome. Multiplexed immunophenotyping was used to identify PD-L1+ cell populations in post-treatment glioblastoma.
Results: Using a 5% cut-off, PD-L1 expression was significantly associated with a poor prognosis in both histologically-defined (n=125, Log-rank p<0.001) and recurrent IDH-wildtype glioblastoma (n=60, Log-rank p=0.015). PD-L1 remained a significant negative prognosticator in Cox regression analysis (HR: 1.96, p=0.021). Analysis of TCGA data confirmed decreased OS in recurrent non G-CIMP glioblastoma (n=12, Log-rank p=0.023), but not in glioblastoma as a group (n=444, log rank p=0.135). PD-L1 RISH showed a significant correlation with IHC (p<0.0001). PD-L1 was observed in the proliferating perivascular stem cell and immune niche of post-treatment glioblastoma.
Conclusions: A 5% PD-L1 expression cut-off identified a subset of glioblastoma that is associated with a worse clinical outcome. This association remained significant within the newly-defined IDH-wildtype classification. These findings could have implications for patient stratification in future clinical trials of PD-1/PD-L1 blockade.
Patient Care: As checkpoint inhibition therapy continues to be investigated on the treatment of glioblastoma, a more complete understanding of PD-L1 expression in the disease is important in both prognosis. Additionally, accurate PD-L1 measurement may have implications for patient stratification in future trials of PD-1 and PD-L1 blockade.
Learning Objectives: 1. Establish which commercially available antibody most effectively detects PD-L1 via immunohistochemistry.
2. Determine the prevalence of PD-L1 expression in glioblastoma.
3. Determine if high degree of PD-L1 expression is associated with different patient outcome.