Impact of molecular and clinical variables on survival outcome with immunotherapy for glioblastoma patients: A systematic review and meta‐analysis

BackgroundGiven that only a subset of patients with glioblastoma multiforme (GBM) responds to immuno‐oncology, this study aimed to assess the impact of multiple factors on GBM immunotherapy prognosis and investigate the potential predictors.MethodsA quantitative meta‐analysis was conducted using the random‐effects model. Several potential factors were also reviewed qualitatively.ResultsA total of 39 clinical trials were included after screening 1317 papers. Patients with O6‐methylguanine‐DNA methyltransferase (MGMT) promoter methylation [hazard ratio (HR) for overall survival (OS) = 2.30, p < 0.0001; HR for progression‐free survival (PFS) = 2.10, p < 0.0001], gross total resection (HR for OS = 0.70, p = 0.02; HR for PFS = 0.56, p = 0.004), and no baseline steroid use (HR for OS = 0.52, p = 0.0002; HR for PFS = 0.61, p = 0.02) had a relatively significant favorable OS and PFS following immunotherapy. Patients with a Karnofsky Performance Status score < 80 (HR = 1.73, p = 0.0007) and undergoing two prior relapses (HR = 2.08, p = 0.003) were associated with worse OS. Age, gender, tumor programmed death‐ligand 1 expression, and history of chemotherapy were not associated with survival outcomes. Notably, immunotherapy significantly improved the OS among patients undergoing two prior recurrences (HR = 0.40, p = 0.008) but not among patients in any other subgroups, as opposed to non‐immunotherapy.ConclusionSeveral factors were associated with prognostic outcomes of GBM patients receiving immunotherapy; multiple recurrences might be a candidate predictor. More marker‐driven prospective studies are warranted.