BACKGROUND: Glioblastoma multiforme (GBM) is the most common and aggressive adult glioma (16-month median survival). Its immunosuppressive microenvironment limits the efficacy of immune checkpoint inhibitors (ICIs). OBJECTIVES: To assess the effects of the ICIs antibodies anti-programmed cell death 1 (anti-PD-1) and anti-programmed cell death ligand 1 (anti-PD-L1) in treating adults with diffuse glioma. SEARCH METHODS: We searched CENTRAL, MEDLINE, Embase, and clinical trials registers on 8 March 2024. SELECTION CRITERIA: We included randomised controlled trials (RCTs) evaluating adults with diffuse glioma treated with anti-PD-1/PD-L1 compared to placebo or other therapies used alone or with other ICIs. Primary outcomes were overall survival (OS), progression-free survival (PFS), and serious adverse events (SAE). Secondary outcomes were overall response rate (ORR), quality of life (QoL), and less serious AEs. DATA COLLECTION AND ANALYSIS: We followed standard Cochrane methods. MAIN RESULTS: We included seven RCTs evaluating anti-PD-1 treatment in recurrent (N = 4) and newly diagnosed (N = 3) grade 4 glioma participants. The analysis encompassed 1953 participants; sample sizes ranged from 35 to 716. Meta-analyses were not possible due to heterogeneity and the small number of studies. Most trials had high risk of bias. Nivolumab versus bevacizumab in people with recurrent GBM (1 trial, 369 participants) Nivolumab probably does not increase OS (hazard ratio (HR) 1.04, 95% confidence interval (CI) 0.83 to 1.30; 1.3% more, 95% CI 6.30 fewer to 7.80 more; 369 participants; moderate-certainty evidence) or PFS (HR 1.97, 95% CI 1.57 to 2.48; 16.40% more, 95% CI 12.40 more to 19.00 more; 369 participants; moderate-certainty evidence). The evidence for SAE is very uncertain (risk ratio (RR) 1.20, 95% CI 0.74 to 1.92; 347 participants). Nivolumab probably does not increase ORR (RR 0.34, 95% CI 0.18 to 0.63; 309 participants; moderate-certainty evidence), but may not increase less serious AEs (RR 1.03, 95% CI 0.96 to 1.10; 347 participants; low-certainty evidence). Nivolumab plus bevacizumab 10 mg/kg versus nivolumab plus bevacizumab 3 mg/kg in people with recurrent GBM (1 trial, 90 participants) Nivolumab plus bevacizumab 10 mg/kg may not increase OS (HR 1.39, 95% CI 0.86 to 2.25; 9.90% more, 95% CI 5.20 fewer to 18.80 more; 90 participants; low-certainty evidence). The evidence for PFS (HR 1.23, 95% CI 0.78 to 1.93; 5.80% more, 95% CI 8.20 fewer to 14.20 more; 90 participants) and SAE (RR 1.19, 95% CI 0.79 to 1.79; 90 participants) is very uncertain. Nivolumab may not increase less serious AEs (RR 1.02, 95% CI 0.96 to 1.09; low-certainty evidence; 90 participants). Pembrolizumab plus bevacizumab versus pembrolizumab in people with recurrent GBM (1 trial, 80 participants) The evidence for OS (HR 1.03, 95% CI 0.65 to 1.63; 0.30% more, 95% CI 7.60 fewer to 2.90 more; 80 participants), PFS (HR 0.97, 95% CI 0.61 to 1.54: 0.40% fewer, 95% CI 9.20 fewer to 2.80 more; 80 participants), SAE (RR 1.32, 95% CI 0.75 to 2.42; 80 participants), and ORR (RR 12.76, 95% CI 0.77 to 210.27; 80 participants) is very uncertain. Pembrolizumab plus bevacizumab may not increase less serious AEs (RR 1.04, 95% CI 0.96 to 1.13; 80 participants; low-certainty evidence). Neoadjuvant (before surgical resection) and adjuvant (after surgical resection) pembrolizumab versus adjuvant-only pembrolizumab in people with recurrent GBM (1 trial, 35 participants) The evidence for OS (HR 0.39, 95% CI 0.17 to 0.92; 25.20% fewer, 95% CI 37.10 fewer to 2.10 fewer; 35 participants), PFS (HR 0.43, 95% CI 0.20 to 0.91; 30.10% fewer, 95% CI 52.20 fewer to 3.60 fewer; 35 participants), and SAE (RR 1.00, 95% CI 0.31 to 3.28; 32 participants) is very uncertain. Nivolumab plus radiotherapy versus temozolomide plus radiotherapy in people with newly diagnosed unmethylated GBM (1 trial, 560 participants) Nivolumab plus radiotherapy probably does not increase OS (HR 1.31, 95% CI 1.09 to 1.58 months; 8.30% more, 95% CI 2.80 more to 12.90 more; 560 participants) and PFS (HR 1.38, 95% CI 1.15 to 1.65 months; 7.50% more, 95% CI 3.60 more to 10.30 more; 560 participants; moderate-certainty evidence). The evidence for SAE is very uncertain (RR 0.87, 95% CI 0.65 to 1.18; 553 participants). It may not increase ORR (RR 1.08, 95% CI 0.43 to 2.69; 560 participants; low-certainty evidence) and probably does not increase less serious AEs (RR 1.00, 95% CI 0.96 to 1.04; 560 participants; moderate-certainty evidence). The evidence for time to deterioration of QoL is very uncertain (HR 0.76, 95% CI 0.59 to 0.99; 560 participants). Nivolumab plus temozolomide plus radiotherapy versus placebo plus temozolomide plus radiotherapy in people with newly diagnosed methylated GBM (1 trial, 716 participants) Nivolumab plus temozolomide plus radiotherapy probably does not increase OS (HR 1.10, 95% CI 0.92 to 1.32; 3.50 more, 95% CI 3.80 fewer to 9.60 more; 716 participants) and PFS (HR 1.10, 95% CI 0.92 to 1.32; 3.00 more, 95% CI 3.50 fewer to 7.90 more; 716 participants), and probably increases SAE (RR 2.91, 95% CI 2.05 to 4.12; 709 participants; moderate-certainty evidence). It does not increase less serious AEs (RR 1.02, 95% CI 1.00 to 1.04; 709 participants; high-certainty evidence). Adjuvant nivolumab plus temozolomide versus temozolomide in older people with GBM (1 trial, 103 participants) Nivolumab plus temozolomide probably does not increase OS (HR 0.85, 95% CI 0.54 to 1.33; 3.10 fewer, 95% CI 15.80 fewer to 3.60 more; 103 participants; moderate-certainty evidence) and PFS (HR 0.77, 95% CI 0.49 to 1.19; 5.40 fewer, 95% CI 19.10 fewer to 2.40 more; 103 participants; moderate-certainty evidence). The evidence for SAE is very uncertain (RR 1.58, 95% CI 0.88 to 2.81; 103 participants). The evidence for QoL is very uncertain (results only reported graphically; 103 participants). AUTHORS' CONCLUSIONS: In recurrent GBM, nivolumab alone probably has no benefit. Anti-PD1 plus bevacizumab may also be ineffective based on low- to very low-certainty evidence. Neoadjuvant plus adjuvant pembrolizumab may improve OS and PFS, but this was based on only one small trial and very low-certainty evidence. In newly diagnosed GBM, nivolumab plus radiotherapy in unmethylated and plus radiotherapy plus temozolomide in methylated GBM probably has no benefit. In older participants, adjuvant nivolumab probably offers no benefit.