Search

MGMT promoter methylation is the only accepted biomarker with prognostic role in GBM but its routine implementation is limited partly response to TMZ is heterogeneous, but also due to lack of effective alternative treatment options. Therefore, additional biomarkers are needed to enable better prediction of survival and to improve individualized treatment of GBM patients. A potential new biomarker is the epidermal growth factor receptor variant III (EGFRvIII). This constitutively activated deletion variant is present in approximately one third of all IDH wildtype GBM, but its relevance to treatment response is poorly understood. The aim of the present study was to analyze the impact of endogenous EGFRvIII expression on chemosensitivity and the mechanisms underlying any differential treatment response. EGFRvIII expression was associated with prolonged median overall survival but only for GBM patients with MGMT promoter methylated tumors. In line with this, we observed increased TMZ sensitivity of EGFRvIII+ and MGMT promoter methylated cells, which translated into improved survival in xenograft experiments. The increased TMZ sensitivity was associated with an elevated DNA damage induction accompanied by an increased expression of DNA mismatch repair (MMR) proteins in EGFRvIII+ cell lines and EGFRvIII+ GBM patient samples. Subsequently, only a moderate reduction in MMR protein expression resulted in a dramatic TMZ resistance, suggesting that EGFRvIII expression specifically sensitized MGMT deficient cells to TMZ treatment by upregulating MMR. Furthermore, EGFRvIII expression in GBM cell lines was accompanied by increased DNA damage, replication fork slowing, stalling and enhanced origin firing, implying replication stress. Targeting of EGFRvIII-dependent replication stress by irinotecan led to hypersensitivity of EGFRvIII+ cells. Taken together this study illustrates that EGFRvIII-induced upregulation of MMR and replication stress increases chemosensitivity thereby highlighting the vulnerability of EGFRvIII+ GBM to available treatments. These important data may also guide the development of new and more effective personalized strategies.

Despite many clinical efforts, the prognosis of GBM patients has not improved in the last 10 years. Epigenetic silencing of the O6-methylguanine-DNA-methyltransferase (MGMT) gene is the only predictive biomarker that is associated with temozolomide (TMZ) response. The establishment of new biomarkers, for personalized treatment is of significant interest. The epidermal growth factor receptor (EGFR) variant III (EGFRvIII) is expressed in approximately 20–30% of GBM. The aim of this study was to investigate if EGFRvIII is a predictive biomarker for TMZ response by analyzing the TMZ sensitivity of isogenic cell cultures and tumour xenografts with/without endogenous EGFRvIII expression. Using two, unique isogenic pairs of MGMT deficient GBM cells we observed that endogenous EGFRvIII expression was associated with a significant increase in TMZ sensitivity in vitro and in vivo. Knockdown of EGFRvIII restored TMZ resistance, confirming EGFRvIII dependence. We observed prolonged accumulation in S/G2-phase in EGFRvIII+ cells after TMZ treatment, indicating increased replication stress induced by TMZ and a significantly higher number of DNA double strand breaks. Because TMZ sensitivity in MGMT negative cells is predominantly determined by functional mismatch repair (MMR), we next analyzed MMR protein expression. We observed that EGFRvIII expression was associated with upregulated MMR proteins and that conversely, even a moderate decrease in MSH6 and MLH1 expression yielded a pronounced increase in TMZ resistance. In accordance with our findings, analysis of clinical data revealed that EGFRvIII positive GBM patients displaying MGMT promotor methylation have significantly better outcomes than EGFRvIII negative patients. In summary, we have shown that MGMT negative EGFRvIII expressing GBM cells and tumours are significantly more sensitive to TMZ compared to isogenic EGFRvIII negative counterparts due to upregulated MMR. Our data highlight a new mechanism of TMZ sensitivity and we conclude that EGFRvIII is a predictive biomarker for response to TMZ in MGMT methylated GBM patients.

BACKGROUND Targeting kinases as regulators of cellular processes that drive cancer progression is a promising approach to improve patient outcome in GBM management. The glycogen synthase kinase 3 (GSK-3) plays a role in cancer progression and is known for its pro-proliferative activity in gliomas. The anti-proliferative and cytotoxic effects of the GSK-3 inhibitor AZD2858 were assessed in relevant in vitro and in vivo glioma models to confirm GSK-3 as a suitable target for improved single agent or combination treatments. MATERIAL AND METHODS The immortalised cell line U251 and the patient derived cell lines GBM1 and GBM4 were used in in vitro studies including MTT, clonogenic survival, live cell imaging, immunofluorescence microscopy and flow cytometry to assess the cytotoxic and anti-proliferative effects of AZD2858. Observed anti-proliferative effects were investigated by microarray technology for the identification of target genes with known roles in cell proliferation. Clinical relevance of targeting GSK-3 with the inhibitor either for single agent or combination treatment strategies was determined by subcutaneous and orthotopic in vivo modelling. Whole mount mass spectroscopy was used to confirm drug penetration in orthotopic tumour models. RESULTS AZD2858 was cytotoxic at low micromolar concentrations and at sub-micromolar concentrations (0.01 - 1.0 μM) induced mitotic defects in all cell lines examined. Prolonged mitosis, centrosome disruption/duplication and cytokinetic failure leading to cell death featured prominently among the cell lines concomitant with an observed S-phase arrest. No cytotoxic or anti-proliferative effect was observed in normal human astrocytes. Analysis of the RNA microarray screen of AZD2858 treated glioma cells revealed the dysregulation of mitosis-associated genes including ASPM and PRC1, encoding proteins with known roles in cytokinesis. The anti-proliferative and cytotoxic effect of AZD2858 was also confirmed in both subcutaneous and orthotopic in vivo models. In addition, combination treatment with AZD2858 enhanced clinically relevant radiation doses leading to reduced tumour volume and improved survival in orthotopic in vivo models. CONCLUSION GSK-3 inhibition with the small molecule inhibitor AZD2858 led to cell death in glioma stem cells preventing normal centrosome function and promoting mitotic failure. Normal human astrocytes were not affected by treatment with the inhibitor at submicromolar concentrations. Drug penetration was observed alongside an enhanced effect of clinical radiotherapy doses in vivo. The reported aberrant centrosomal duplication may be a direct consequence of failed cytokinesis suggesting a role of GSK-3 in regulation of mitosis in glioma. GSK-3 is a promising target for combination treatment with radiation in GBM management and plays a role in mitosis-associated events in glioma biology.

Glioblastoma multiforme (GBM) is the most common malignant brain tumor in adults with an estimated 5-year survival of less 10%. The current standard of care involves maximal tumor resection followed by radiation and chemotherapy with the alkylating agent temozolomide (TMZ). Epigenetic silencing of the O6-methylguanine-DNA methyltransferase (MGMT) gene predicts response to TMZ therapy, but does not explain all of the heterogeneity in responses observed in the clinic. The establishment of additional molecular biomarkers, is therefore of significant interest.

INTRODUCTION: HOX genes are essential for embryonic development but are dysregulated in numerous cancers, including glioblastoma. High levels of HOX expression are characteristic of a subset of glioma stem cells (GSC) and have been associated with self-renewal capability and treatment resistance. We aimed to determine whether HOX proteins are a potential therapeutic target in glioblastoma. METHOD: HOX expression profiles were determined using quantitative RT-PCR. Patient-derived GSC, glioma cell lines and primary astrocytes were treated with HXR9, a peptide that disrupts the interaction between HOX proteins and their binding partner PBX. Cell viability was determined using the MTS assay. Combination peptide and radiation treatments were assessed by clonogenic assay. Sub-cutaneous xenografts of U87-MG cells were established in Balb/C Nude mice and tumours were treated with intratumoural injection of peptide and/or CT-guided fractionated radiotherapy. RESULTS: Patient-derived GSC showed variable HOX profiles, similar to previously reported HOX-high and -low signatures. HXR9 was potently cytotoxic in a number of GSC cultures, with IC50 values of 3-9 µM. In contrast, normal astrocytes were much more resistant (IC50 = 70 µM), as were the cell lines T98G and U87MG (70 and 35 µM, respectively). An additive cytotoxic effect was observed for combined HXR9 and radiation treatment in vitro, and HXR9 treatment resulted in delayed tumour growth in vivo, both alone and in combination with radiotherapy. CONCLUSION: The HOX inhibitor HXR9 showed potent cytotoxicity in cultured GSC and reduced the rate of tumour growth in vivo. Targeting HOX proteins is a potential adjuvant treatment for glioblastoma alongside current chemoradiotherapy.

BACKGROUND: Resistance of glioblastoma to standard chemoradiotherapy may be mediated, at least partially, by a population of inherently resistant glioma stem cells (GSCs) which are able to survive treatment and reconstitute the tumour. HOX genes are dysregulated in various cancers and a subset of GSCs show high levels of HOX expression. In addition, a HOX gene signature has been associated with treatment resistance in glioblastoma. Our goal was to determine the potential of targeting HOX function as a novel therapy for glioblastoma. METHODS: HOX expression profiles for patient-derived GSCs and established glioma cell lines were determined by quantitative RT-PCR. In vitro, cells were treated with the HOX inhibitor peptide HXR9 (or control peptide CXR9), with or without radiation two hours later. Subsequently, cells were incubated through two doubling times and the MTS assay was used to determine cell viability. In vivo, sub-cutaneous xenografts of U87-MG cells were established in Balb/c Nude mice. Treatments (peptide and/or radiation) or controls were administered in three doses over consecutive days. RESULTS: Analysis of HOX expression in GSC lines demonstrated up-regulation of genes across all four clusters, although expression patterns varied between cell lines. In contrast, the established glioma cell lines T98G and U87-MG showed lower overall HOX expression. The HOX inhibitor peptide HXR9 had potent cytotoxic activity in a number of GSC lines, with IC50 values between 3 and 5 µM. In contrast, T98G and U87-MG were more resistant (IC50s of 70 and 35 µM, respectively), as was a normal neural progenitor cell line NP1 (IC50 65 µM). In vitro, combinations of sub-lethal doses of HXR9 with radiation demonstrated an additive cytoxic effect on GSCs. In vivo, treatment with HXR9 peptide resulted in tumour growth delay when combined with radiotherapy in a mouse xenograft model. CONCLUSIONS: Knockdown of HOX protein function reduced viability of glioma cells in vitro, with greater potency in GSCs than established cell lines. Combination of HOX-inhibition with radiation or demonstrated an additive cytoxicity in vitro and improved tumour growth delay in a preclinical mouse model. Therefore, targeting HOX proteins shows potential as an adjuvant therapy with current standard of care chemoradiotherapy for glioblastoma.