BACKGROUND
Glioblastoma (GB) is the most common malignant primary brain tumor in adults and associated with a dismal prognosis. While the standard of care for newly diagnosed GB is defined in current treatment guidelines, recommendations for recurrent GB are less well established. The current systematic meta-analysis of recently published randomized controlled trials (RCTs) represents the strongest available evidence on targeted agents in patients with recurrent GB.

METHODS
Cochrane Library, Pubmed, MEDLINE (Ovid), ClinicalTrials.gov, WHO's International Clinical Trials Registry Platform and Google Scholar were searched ranging from the year 1954-2022 for RCTs of targeted therapies in patients with recurrent GB. Hazard Ratios (HRs) of overall survival (OS) and progression-free survival (PFS) were extracted to perform a random-effects meta-analysis.

RESULTS
16 RCTs (n=3,025 patients) were included in the meta-analysis. Experimental treatment was either compared to lomustine (CCNU) alone, in combination with CCNU/temozolomide (TMZ) to CCNU alone or to bevacizumab alone. In these three subgroups, targeted agents associated with improved OS compared to the control arm were regorafenib (RR= 0.50; 95% CI 0.33-0.75), Depatux- M+ TMZ (RR= 0.66; 95% CI 0.44-0.93) and rindopepimut + bevacizumab (RR= 0.53; 95% CI 0.32-0.88). Treatment with bevacizumab + CCNU (RR= 0.49; 95% CI 0.35-0.69) and regorafenib (RR= 0.65; 95% CI 0.44-0.95) were associated with improved PFS.

CONCLUSION
In this systematic meta-analysis, we provide the current highest level of evidence for the role of targeted therapies in recurrent GB. Even though some studies revealed a benefit either for OS and/or PFS, results have to be critically reviewed regarding initial distribution of prognostic factors, underlying molecular mechanisms, sample size and trial design. There is a need for more specific and personalized study designs using newly obtained tumor tissue and close monitoring of treatment responses to allow for potential modification of treatment if needed.