RESUMEN
BACKGROUND: Glioblastoma multiforme is the most frequent malignant brain tumor in adults being marked with a very poor prognosis. Therapy concept implies concomitant radio-chemotherapy and facultative implantation of carmustine-eluted wafer. Current literature suggests microRNA 26a expression in glioblastoma to interact with alkylating chemotherapy. Subsequently, the aim of this study was to investigate the correlation of miRNA-26a expression and carmustine wafer implantation and its potential usefulness as a predictive marker for therapy response. METHODS: In total, 229 patients with glioblastoma multiforme were included into the final analysis. Of them, 80 cases were recruited from the Saarland University Medical Center for a retrospective matched-pair analysis stratified after therapy regime: One group (carmustine wafer group; n=40) received concomitant radio-chemotherapy with carmustine wafer implantation. The other group (control group; n=40) only received concomitant radio-chemotherapy. The results were confirmed by comparing them with an independent dataset of 149 patients from the TCGA database. All tumor specimens were evaluated for miRNA-26a expression, MGMT promoter methylation, and IDH1 R132H mutation status, and the results were correlated with the clinical data. RESULTS: Twenty-three patients in the carmustine wafer group showed low expression of miRNA-26a, while 17 patients showed a high expression. In the control group, 28 patients showed low expression, while 12 patients showed a high expression. The patients with high miRNA-26a expression in the carmustine wafer group were characterized by a significantly longer overall (hazard ratio [HR] 2.750 [95% CI 1.352-5.593]; p=0.004) and progression-free survival (HR 3.091 [95% CI 1.436-6.657]; p=0.003) than patients with low miRNA-26a expression. The 17 patients in the carmustine wafer group with high miRNA-26a expression showed a significantly longer progression-free survival (p=0.013) and overall survival (p=0.007) compared with the control group. There were no such correlations identified within the control group. TCGA datasets supported these findings. CONCLUSIONS: MiRNA-26a expression turned out to be a promising predictor of therapy response and clinical outcome in glioblastoma patients treated with carmustine wafer implantation. For evaluation of the role of miRNA-26a in a combined therapy setting, further studies are needed in order to translate general findings to the patient's individual situation.