Biom-35. identification and validation of potential biomarkers for decitabine-mediated sensitization to temozolomide in glioblastoma

Abstract Glioblastoma (GBM) is an aggressive form of brain cancer with limited treatment options and a poor prognosis. Our previous research demonstrated that the DNA hypomethylating agent decitabine (DAC) enhances the sensitivity to temozolomide (TMZ) of a subset of primary GBM cell lines derived from fresh surgical tissue specimens, and that this effect may be mediated by upregulation of DNA mismatch repair genes including MLH1. In this context, we aim to identify additional biomarkers for DAC-mediated sensitization to TMZ with potential clinical utility through in vitro and in vivo experiments. We performed single cell RNA sequencing on three DAC-sensitized and three DAC-desensitized cell lines and found 4 genes that exhibited significant differential expression: DCN, LUM, CRABP2 and NUPR1. We validated these findings using quantitative immunoblots, RT-qPCR and immunohistochemistry, finding that NUPR1 protein expression levels varied most significantly between DAC-sensitized and desensitized GBM tissues. Long-range direct methylation sequencing of non-amplified genomic DNA enriched using CRISPR/Cas9 demonstrated several CpG sites within the 4 genes that may be predictive of TMZ sensitization response after treatment with DAC. Finally, we performed mouse survival experiments using orthotopic patient-derived xenografts (PDXs) of the six GBM cell lines, with preliminary results showing that in selected PDX survival was significantly increased with combined DAC-TMZ versus TMZ monotherapy as predicted by in vitro assays. In conclusion, our study demonstrates that combining DAC with TMZ may be a potential treatment strategy for GBM in a subset of patients. There likely exist several biomarkers that can be used at the time of diagnosis, alone or in combination, to predict the response to epigenetic preconditioning using DAC for TMZ sensitization, which will be necessary for optimal patient selection in an early phase clinical trial. Further data and validation studies are necessary to strengthen these findings and to confirm their clinical relevance.