PD-1/PD-L1 checkpoint blockades have achieved significant progress in several kinds of tumours. Pembrolizumab, which targets PD-1, has been approved as a first-line treatment for advanced non-small cell lung cancer (NSCLC) patients with positive PD-L1 expression. However, PD-1/PD-L1 checkpoint blockades have not achieved breakthroughs in treating glioblastoma because glioblastoma has a low immunogenic response and an immunosuppressive microenvironment caused by the precise crosstalk between cytokines and immune cells. A phase III clinical trial, Checkmate 143, reported that nivolumab, which targets PD-1, did not demonstrate survival benefits compared with bavacizumab in recurrent glioblastoma patients. Thus, the combination of a PD-1/PD-L1 checkpoint blockade with RT, TMZ, antibodies targeting other inhibitory or stimulatory molecules, targeted therapy, and vaccines may be an appealing solution aimed at achieving optimal clinical benefit. There are many ongoing clinical trials exploring the efficacy of various approaches based on PD-1/PD-L1 checkpoint blockades in primary or recurrent glioblastoma patients. Many challenges need to be overcome, including the identification of discrepancies between different genomic subtypes in their response to PD-1/PD-L1 checkpoint blockades, the selection of PD-1/PD-L1 checkpoint blockades for primary versus recurrent glioblastoma, and the identification of the optimal combination and sequence of combination therapy. In this review, we describe the immunosuppressive molecular characteristics of the tumour microenvironment (TME), candidate biomarkers of PD-1/PD-L1 checkpoint blockades, ongoing clinical trials and challenges of PD-1/PD-L1 checkpoint blockades in glioblastoma.