Purpose: Existing salvage therapies for recurrent glioblastoma (rGBM) have limited efficacy, with a median survival of approximately 9 months. Given the complex molecular heterogeneity of GBM, single-target approaches have consistently failed as a treatment strategy. We conducted a phase I clinical trial to assess the feasibility, safety, and efficacy of a genomically tailored multiagent regimen in 30 adults with surgically treated rGBM.
Patients and methods: Adults with IDH wild-type GBM (n = 29) or grade 4 IDH-mutant astrocytoma (n = 1) were consented and underwent clinically indicated surgery for recurrent disease. Comprehensive genomic profiling was performed on the recurrent tumors, and results for each patient were discussed at an individualized molecular tumor board to determine a personalized treatment regimen combining up to four FDA-approved drugs, including one cytotoxic agent as the backbone.
Results: A total of 12 drugs were used in 18 combinations-the most common regimen was lomustine, afatinib, and abemaciclib (n = 8). The most common toxicities included cytopenias, rash, and gastrointestinal symptoms, requiring frequent dose reductions. Measured from surgery at trial enrollment, progression-free survival at 6 months was 40%, overall survival (OS) at 9 months was 73%, and median OS was 12.7 months. After trial therapy, genomic profiling performed on subsequent recurrent tumor specimens identified genetic evolution corresponding to putative treatment resistance mechanisms.
Conclusions: Implementation of individualized treatment regimens in a timely fashion was feasible for patients with surgically resectable rGBM. Overall efficacy was not significantly improved compared with a contemporary patient cohort treated without experimental regimens, with full dosing of most combination therapies limited by toxicities.
©2026 The Authors; Published by the American Association for Cancer Research.