EGFR amplified and overexpressing glioblastomas and association with better response to adjuvant metronomic temozolomide

Abstract

Background: Lack of robust predictive biomarkers, other than MGMT promoter methylation, makes temozolomide responsiveness in newly diagnosed glioblastoma (GBM) patients difficult to predict. However, we identified patients with long-term survival (35 months) within a group of newly diagnosed GBM patients treated with standard or metronomic adjuvant temozolomide schedules. We thus investigated possible molecular profiles associated with longer survival following temozolomide treatment. Methods: We investigated the association of molecular features with progression-free (PFS) and overall survival (OS). Human-derived GBM cancer stem cells (CSCs) were used to investigate in vitro molecular mechanisms associated with temozolomide responsiveness. Surgically removed recurrences allowed investigation of molecular changes occurring during therapy in vivo. Statistical analyses included one-and two-way analysis of variance, Student's t test, Cox proportional hazards, and the Kaplan-Meier method. All statistical tests were two-sided. Results: No association was found between survival and gene classifiers associated with different molecular GBM subtypes in the standard-treated group, while in metronomic-treated patients robust association was found between EGFR amplification/overexpression and PFS and OS (OS, EGFR-high vs low: hazard ratiodeath = 0.22, 95% confidence interval = 0.09 to 0.55, P =. 001). The result for OS remained statistically significant after Bonferroni correction (P interaction