A tale of two multi‐focal therapies for glioblastoma: An antibody targeting ELTD1 and nitrone‐based OKN‐007

Glioblastoma (GBM) is the most common primary malignant brain tumour in adults. Despite a multimodal treatment response, survival for GBM patients remains between 12 and 15 months. Anti‐ELTD1 antibody therapy is effective in decreasing tumour volumes and increasing animal survival in an orthotopic GBM xenograft. OKN‐007 is a promising chemotherapeutic agent that is effective in various GBM animal models and is currently in two clinical trials. In this study, we sought to compare anti‐ELTD1 and OKN‐007 therapies, as single agents and combined, against bevacizumab, a commonly used therapeutic agent against GBM, in a human G55 xenograft mouse model. MRI was used to monitor tumour growth, and immunohistochemistry (IHC) was used to assess tumour markers for angiogenesis, cell migration and proliferation in the various treatment groups. OKN and anti‐ELTD1 treatments significantly increased animal survival, reduced tumour volumes and normalized the vasculature. Additionally, anti‐ELTD1 was also shown to significantly affect other pro‐angiogenic factors such as Notch1 and VEGFR2. Unlike bevacizumab, anti‐ELTD1 and OKN treatments did not induce a pro‐migratory phenotype within the tumours. Anti‐ELTD1 treatment was shown to be as effective as OKN therapy. Both OKN and anti‐ELTD1 therapies show promise as potential single‐agent multi‐focal therapies for GBM patients.     

The Evolution of Tumour Composition During Fractionated Radiotherapy: Implications for Outcome

Current protocols for delivering radiotherapy are based primarily on tumour stage and nodal and metastases status, even though it is well known that tumours and their microenvironments are highly heterogeneous. It is well established that the local oxygen tension plays an important role in radiation-induced cell death, with hypoxic tumour regions responding poorly to irradiation. Therefore, to improve radiation response, it is important to understand more fully the spatiotemporal distribution of oxygen within a growing tumour before and during fractionated radiation. To this end, we have extended a spatially resolved mathematical model of tumour growth, first proposed by Greenspan (Stud Appl Math 51:317–340, 1972), to investigate the effects of oxygen heterogeneity on radiation-induced cell death. In more detail, cell death due to radiation at each location in the tumour, as determined by the well-known linear-quadratic model, is assumed also to depend on the local oxygen concentration. The oxygen concentration is governed by a reaction-diffusion equation that is coupled to an integro-differential equation that determines the size of the assumed spherically symmetric tumour. We combine numerical and analytical techniques to investigate radiation response of tumours with different intratumoral oxygen distribution profiles. Model simulations reveal a rapid transient increase in hypoxia upon regrowth of the tumour spheroid post-irradiation. We investigate the response to different radiation fractionation schedules and identify a tumour-specific relationship between inter-fraction time and dose per fraction to achieve cure. The rich dynamics exhibited by the model suggest that spatial heterogeneity may be important for predicting tumour response to radiotherapy for clinical applications.     

A mathematical model of brain tumour response to radiotherapy and chemotherapy considering radiobiological aspects

Glioblastoma is the most frequent and malignant brain tumour. For many years, the conventional treatment has been maximal surgical resection followed by radiotherapy (RT), with a median survival time of less than 10 months. Previously, the use of adjuvant chemotherapy (given after RT) has failed to demonstrate a statistically significant survival advantage. Recently, a randomized phase III trial has confirmed the benefit of temozolomide (TMZ) and has defined a new standard of care for the treatment of patients with high-grade brain tumours. The results showed an increase of 2.5 months in median survival, and 16.1% in 2 year survival, for patients receiving RT with TMZ compared with RT alone. It is not clear whether the major benefit of TMZ comes from either concomitant administration of TMZ with RT, or from six cycles of adjuvant TMZ, or both.The objectives were to develop our original model, which addressed survival after RT, to construct a new module to assess the potential role of TMZ from clinical data, and to explore its synergistic contribution in addition to radiation. The model has been extended to include radiobiological parameters. The addition of the linear quadratic equation to describe cellular response to treatment has enabled us to quantify the effects of radiation and TMZ in radiobiological terms.The results indicate that the model achieves an excellent fit to the clinical data, with the assumption that TMZ given concomitantly with RT synergistically increases radiosensitivity. The alternative, that the effect of TMZ is due only to direct cell killing, does not fit the clinical data so well. The addition of concomitant TMZ appears to change the radiobiological parameters. This aspect of our results suggests possible treatment developments.Our observations need further evaluations in real clinical trials, may suggest treatment strategies for new trials, and inform their design.     

Activation of executioner caspases is a predictor of progression-free survival in glioblastoma patients: a systems medicine approach

Glioblastoma (GBM) is the most common and aggressive primary brain tumor in adults. GBM cells are highly resistant to apoptosis induced by antitumor drugs and radiotherapy resulting in cancer progression. We assessed whether a systems medicine approach, analysing the ability of tumor cells to execute apoptosis could be utilized to predict the response of GBM patients to treatment. Concentrations of the key proapoptotic proteins procaspase-3, procaspase-9, Smac and Apaf-1 and the antiapopotic protein XIAP were determined in a panel of GBM cell lines and GBM patient tumor resections. These values were used as input for APOPTO-CELL, a systems biological based mathematical model built to predict cellular susceptibility to undergo caspase activation. The modeling was capable of accurately distinguishing between GBM cells that die or survive in response to treatment with temozolomide in 10 of the 11 lines analysed. Importantly the results obtained using GBM patient samples show that APOPTO-CELL was capable of stratifying patients according to their progression-free survival times and predicted the ability of tumor cells to support caspase activation in 16 of the 21 GBM patients analysed. Calculating the susceptibility to apoptosis execution may be a potent tool in predicting GBM patient therapy responsiveness and may allow for the use of APOPTO-CELL in a clinical setting.     

A risk signature with four autophagy-related genes for predicting survival of glioblastoma multiforme

Glioblastoma multiforme (GBM) is a devastating brain tumour without effective treatment. Recent studies have shown that autophagy is a promising therapeutic strategy for GBM. Therefore, it is necessary to identify novel biomarkers associated with autophagy in GBM. In this study, we downloaded autophagy-related genes from Human Autophagy Database (HADb) and Gene Set Enrichment Analysis (GSEA) website. Least absolute shrinkage and selection operator (LASSO) regression and multivariate Cox regression analysis were performed to identify genes for constructing a risk signature. A nomogram was developed by integrating the risk signature with clinicopathological factors. Time-dependent receiver operating characteristic (ROC) curve and calibration plot were used to evaluate the efficiency of the prognostic model. Finally, four autophagy-related genes (DIRAS3, LGALS8, MAPK8 and STAM) were identified and were used for constructing a risk signature, which proved to be an independent risk factor for GBM patients. Furthermore, a nomogram was developed based on the risk signature and clinicopathological factors (IDH1 status, age and history of radiotherapy or chemotherapy). ROC curve and calibration plot suggested the nomogram could accurately predict 1-, 3- and 5-year survival rate of GBM patients. For function analysis, the risk signature was associated with apoptosis, necrosis, immunity, inflammation response and MAPK signalling pathway. In conclusion, the risk signature with 4 autophagy-related genes could serve as an independent prognostic factor for GBM patients. Moreover, we developed a nomogram based on the risk signature and clinical traits which was validated to perform better for predicting 1-, 3- and 5-year survival rate of GBM.     

Effects of Anti-Angiogenesis on Glioblastoma Growth and Migration: Model to Clinical Predictions

Glioblastoma multiforme (GBM) causes significant neurological morbidity and short survival times. Brain invasion by GBM is associated with poor prognosis. Recent clinical trials of bevacizumab in newly-diagnosed GBM found no beneficial effects on overall survival times; however, the baseline health-related quality of life and performance status were maintained longer in the bevacizumab group and the glucocorticoid requirement was lower. Here, we construct a clinical-scale model of GBM whose predictions uncover a new pattern of recurrence in 11/70 bevacizumab-treated patients. The findings support an exception to the Folkman hypothesis: GBM grows in the absence of angiogenesis by a cycle of proliferation and brain invasion that expands necrosis. Furthermore, necrosis is positively correlated with brain invasion in 26 newly-diagnosed GBM. The unintuitive results explain the unusual clinical effects of bevacizumab and suggest new hypotheses on the dynamic clinical effects of migration by active transport, a mechanism of hypoxia-driven brain invasion.     

Characterization of brain cancer stem cells: a mathematical approach

Objective:  In recent years, support has increased for the notion that a subpopulation of brain tumour cells in possession of properties typically characteristic of stem cells is responsible for initiating and maintaining the tumour. Unravelling details of the brain tumour stem cell (BTSC) hierarchy, as well as interactions of these cells with various therapies, will be essential in the design of optimal treatment strategies.
Materials and methods:  Motivated by this, we have developed a mathematical model of the BTSC hypothesis that may aid in characterization of brain tumours, as well as in prediction of effective therapeutic strategies, which can be further validated in experimental and clinical studies. At the level of a small number of cells, the model developed herein is stochastic. For larger populations of cancer cells, the model is handled from a deterministic approach.
Results and conclusions:  In the stochastic regime, importance of a relationship between the likelihoods of two distinct types of symmetric BTSC divisions in determining BTSC survival rates becomes apparent, consequently emphasizing the need for a set of biomarkers that are able to better characterize the BTSC hierarchy. At the large scale, we predict the importance of the aforementioned symmetric division rates in dictating brain tumour composition. Furthermore, we demonstrate possible therapeutic benefits of considering combination treatments of radiotherapy and putative BTSC inhibitors, such as bone morphogenetic proteins, while reinforcing the importance of developing novel treatment strategies that specifically target the BTSC subpopulation.     

Csnk1a1 inhibition modulates the inflammatory secretome and enhances response to radiotherapy in glioma

Glioblastoma multiforme (GBM), a fatal brain tumour with no available targeted therapies, has a poor prognosis. At present, radiotherapy is one of the main methods to treat glioma, but it leads to an obvious increase in inflammatory factors in the tumour microenvironment, especially IL-6 and CXCL1, which plays a role in tumour to resistance radiotherapy and tumorigenesis. Casein kinase 1 alpha 1 (CK1α) (encoded on chromosome 5q by Csnk1a1) is considered an attractive target for Tp53 wild-type acute myeloid leukaemia (AML) treatment. In this study, we evaluated the anti-tumour effect of Csnk1a1 suppression in GBM cells in vitro and in vivo. We found that down-regulation of Csnk1a1 or inhibition by D4476, a Csnk1a1 inhibitor, reduced GBM cell proliferation efficiently in both Tp53 wild-type and Tp53-mutant GBM cells. On the contrary, overexpression of Csnk1a1 promoted cell proliferation and colony formation. Csnk1a1 inhibition improved the sensitivity to radiotherapy. Furthermore, down-regulation of Csnk1a1 reduced the production and secretion of pro-inflammatory factors. In the preclinical GBM model, treatment with D4476 significantly inhibited the increase in pro-inflammatory factors caused by radiotherapy and improved radiotherapy sensitivity, thus inhibiting tumour growth and prolonging animal survival time. These results suggest targeting Csnk1a1 exert an anti-tumour role as an inhibitor of inflammatory factors, providing a new strategy for the treatment of glioma.     

Control of the blood–brain barrier function in cancer cell metastasis

Cerebral metastases are the most common brain neoplasms seen clinically in the adults and comprise more than half of all brain tumours. Actual treatment options for brain metastases that include surgical resection, radiotherapy and chemotherapy are rarely curative, although palliative treatment improves survival and life quality of patients carrying brain‐metastatic tumours. Chemotherapy in particular has also shown limited or no activity in brain metastasis of most tumour types. Many chemotherapeutic agents used systemically do not cross the blood–brain barrier (BBB), whereas others may transiently weaken the BBB and allow extravasation of tumour cells from the circulation into the brain parenchyma. Increasing evidence points out that the interaction between the BBB and tumour cells plays a key role for implantation and growth of brain metastases in the central nervous system. The BBB, as the tightest endothelial barrier, prevents both early detection and treatment by creating a privileged microenvironment. Therefore, as observed in several in vivo studies, precise targetting the BBB by a specific transient opening of the structure making it permeable for therapeutic compounds, might potentially help to overcome this difficult clinical problem. Moreover, a better understanding of the molecular features of the BBB, its interrelation with metastatic tumour cells and the elucidation of cellular mechanisms responsible for establishing cerebral metastasis must be clearly outlined in order to promote treatment modalities that particularly involve chemotherapy. This in turn would substantially expand the survival and quality of life of patients with brain metastasis, and potentially increase the remission rate. Therefore, the focus of this review is to summarise the current knowledge on the role and function of the BBB in cancer metastasis.     

Correlation between initial chromatid damage and survival of various cell lines exposed to heavy charged particles

The biophysical characteristics of heavy ions make them a rational source of radiation for use in radiotherapy of malignant tumours. Prior to radiotherapy treatment, a therapeutic regimen must be precisely defined, and during this stage information on individual patient radiosensitivity would be of very great medical value. There are various methods to predict radiosensitivity, but some shortfalls are difficult to avoid. The present study investigated the induction of chromatid breaks in five different cell lines, including one normal liver cell line (L02), exposed to carbon ions accelerated by the heavy ion research facility in Lanzhou (HIRFL), using chemically induced premature chromosome condensation (PCC). Previous studies have reported the number of chromatid breaks to be linearly related to the radiation dose, but the relationship between cell survival and chromatid breaks is not clear. The major result of the present study is that cellular radiosensitivity, as measured by D₀, is linearly correlated with the frequency of chromatid breaks per Gy in these five cell lines. We propose that PCC may be applied to predict radiosensitivity of tumour cells exposed to heavy ions.     

Virtual glioblastoma: growth, migration and treatment in a three-dimensional mathematical model

Objectives:  Glioblastomas are aggressive primary brain cancers that are characterized by extensive infiltration into the brain and are highly resistant to treatment. Through mathematical modelling, we model the process of invasion and predict the relative importance of mechanisms contributing to malignant invasion. Clinically, we predict patterns of tumour recurrence following various modes of therapeutic intervention.
Materials and methods:  Our mathematical model uses a realistic three-dimensional brain geometry and considers migrating and proliferating cells as separate classes. Several mechanisms for infiltrative migration are considered. Methods are developed for simulating surgical resection, radiotherapy and chemotherapy.
Results:  The model provides clinically realistic predictions of tumour growth and recurrence following therapeutic intervention. Specific results include (i) invasiveness is governed largely by the ability of glioblastoma cells to degrade and migrate through the extracellular matrix and the ability of single migrating cells to form colonies; (ii) tumours originating deeper in the brain generally grow more quickly than those of superficial origin; (iii) upon surgery, the margins and geometry of resection significantly determine the extent and pattern of postoperative recurrence; (iv) radiotherapy works synergistically with greater resection margins to reduce recurrence; (v) simulations in both two- and three-dimensional geometries give qualitatively similar results; and (vi) in an actual clinical case comprising several surgical interventions, the model provides good qualitative agreement between the simulated and observed course of the disease.
Conclusions:  The model provides a useful initial framework by which biological mechanisms of invasion and efficacy of potential treatment regimens may be assessed.     

Genome-wide CRISPR/Cas9 screening identifies CARHSP1 responsible for radiation resistance in glioblastoma

Glioblastomas (GBM) is the most common primary malignant brain tumor, and radiotherapy plays a critical role in its therapeutic management. Unfortunately, the development of radioresistance is universal. Here, we identified calcium-regulated heat-stable protein 1 (CARHSP1) as a critical driver for radioresistance utilizing genome-wide CRISPR activation screening. This is a protein with a cold-shock domain (CSD)-containing that is highly similar to cold-shock proteins. CARHSP1 mRNA level was upregulated in irradiation-resistant GBM cells and knockdown of CARHSP1 sensitized GBM cells to radiotherapy. The high expression of CARHSP1 upon radiation might mediate radioresistance by activating the inflammatory signaling pathway. More importantly, patients with high levels of CARHSP1 had poorer survival when treated with radiotherapy. Collectively, our findings suggested that targeting the CARHSP1/TNF-α inflammatory signaling activation induced by radiotherapy might directly affect radioresistance and present an attractive therapeutic target for GBM, particularly for patients with high levels of CARHSP1.Subject terms: Tumour biomarkers, Glial biology     

Distinct cellular responses induced by saporin and a transferrin-saporin conjugate in two different human glioblastoma cell lines

Glioblastoma multiforme (GBM) is the most common primary brain tumour in adults, with a median survival of ~12-18 months post-diagnosis. GBM usually recurs within 12 months post-resection, with poor prognosis. Thus, novel therapeutic strategies to target and kill GBM cells are urgently needed. The marked difference of tumour cells with respect to normal brain cells renders glioblastoma a good candidate for selective targeted therapies. Recent experimental strategies focus on over expressed cell surface receptors. Targeted toxins represent a new class of selective molecules composed by a potent protein toxin and a carrier ligand. Targeted toxins approaches against glioblastoma were under investigation in phase I and II clinical trials with several immunotoxins (IT)/ligand toxins such as IL4-Pseudomonas aeruginosa exotoxin A (IL4-PE, NBI-3001), tumour growth factor fused to PE38, a shorter PE variant, (TGF)alpha-TP-38, IL13-PE38, and a transferrin-C diphtheriae toxin mutant (Tf-CRM107). In this work, we studied the effects of the plant ribosome-inactivating saporin and of its chimera transferrin-saporin against two different GBM cell lines. The data obtained here indicate that cell proliferation is affected by the toxin treatments but that different mechanisms are used, directly linked to the presence of an active or inactive p53. A model is proposed for these alternative intracellular pathways.     

Grand rounds at the National Institutes of Health: HDAC inhibitors as radiation modifiers, from bench to clinic

Glioblastoma multiforme (GBM) is the most common and aggressive malignant brain tumour. Patients afflicted with this disease unfortunately have a very poor prognosis, and fewer than 5% of patients survive for 5 years from the time of diagnosis. Therefore, improved therapies to treat this disease are sorely needed. One such class of drugs that have generated great enthusiasm for the treatment of numerous malignancies, including GBM, is histone deacetylase (HDAC) inhibitors. Pre-clinical data have demonstrated the efficacy of various HDAC inhibitors as anticancer agents, with the greatest effects shown when HDAC inhibitors are used in combination with other therapies. As a result of encouraging pre-clinical data, numerous HDAC inhibitors are under investigation in clinical trials, either as monotherapies or in conjunction with other treatments such as chemotherapy, biologic therapy or radiation therapy. In fact, two actively studied HDAC inhibitors, vorinostat and depsipeptide, were recently approved for the treatment of refractory cutaneous T cell lymphoma. In this review, we first present a patient with GBM, and then discuss the pathogenesis, epidemiology and current treatment options of GBM. Finally, we examine the translation of pre-clinical studies that have demonstrated HDAC inhibitors as potent radiosensitizers in in vitro and in vivo models, to a phase II clinical trial combining the HDAC inhibitor, valproic acid, along with temozolomide and radiation therapy for the treatment of GBM.     

A Fourier transform infrared microspectroscopic imaging investigation into an animal model exhibiting glioblastoma multiforme

Glioblastoma multiforme (GBM) is a highly malignant human brain tumour for which no cure is available at present. Numerous clinical studies as well as animal experiments are under way with the goal being to understand tumour biology and develop potential therapeutic approaches. C6 cell glioma in the adult rat is a frequently used and well accepted animal model for the malignant human glial tumour. By combining standard analytical methods such as histology and immunohistochemistry with Fourier Transform Infrared (FTIR) microspectroscopic imaging and multivariate statistical approaches, we are developing a novel approach to tumour diagnosis which allows us to obtain information about the structure and composition of tumour tissues that could not be obtained easily with either method alone. We have used a “Stingray” FTIR imaging spectrometer to analyse and compare the compositions of coronal brain tissue sections of a tumour-bearing animal and those from a healthy animal. We have found that the tumour tissue has a characteristic chemical signature, which distinguishes it from tumour-free brain tissue. The physical-chemical differences, determined by image and spectral comparison are consistent with changes in total protein absorbance, phosphodiester absorbance and physical dispersive artefacts. The results indicate that FTIR imaging analysis could become a valuable analytic method in brain tumour research and possibly in the diagnosis of human brain tumours.     

Prognostic value and immune cell infiltration of hypoxic phenotype‐related gene signatures in glioblastoma microenvironment

Glioblastoma (GBM) is a malignant intracranial tumour with the highest proportion and lethality. It is characterized by invasiveness and heterogeneity. However, the currently available therapies are not curative. As an essential environmental cue that maintains glioma stem cells, hypoxia is considered the cause of tumour resistance to chemotherapy and radiation. Growing evidence shows that immunotherapy focusing on the tumour microenvironment is an effective treatment for GBM; however, the current clinicopathological features cannot predict the response to immunotherapy and provide accurate guidance for immunotherapy. Based on the ESTIMATE algorithm, GBM cases of The Cancer Genome Atlas (TCGA) data set were classified into high‐ and low‐immune/stromal score groups, and a four‐gene tumour environment‐related model was constructed. This model exhibited good efficiency at forecasting short‐ and long‐term prognosis and could also act as an independent prognostic biomarker. Additionally, this model and four of its genes (CLECL5A, SERPING1, CHI3L1 and C1R) were found to be associated with immune cell infiltration, and further study demonstrated that these four genes might drive the hypoxic phenotype of perinecrotic GBM, which affects hypoxia‐induced glioma stemness. Therefore, these might be important candidates for immunotherapy of GBM and deserve further exploration.     

TRIM28 and β-Actin Identified via Nanobody-Based Reverse Proteomics Approach as Possible Human Glioblastoma Biomarkers

Malignant gliomas are among the rarest brain tumours, and they have the worst prognosis. Grade IV astrocytoma, known as glioblastoma multiforme (GBM), is a highly lethal disease where the standard therapies of surgery, followed by radiation and chemotherapy, cannot significantly prolong the life expectancy of the patients. Tumour recurrence shows more aggressive form compared to the primary tumour, and results in patient survival from 12 to 15 months only. Although still controversial, the cancer stem cell hypothesis postulates that cancer stem cells are responsible for early relapse of the disease after surgical intervention due to their high resistance to therapy. Alternative strategies for GBM therapy are thus urgently needed. Nanobodies are single-domain antigen-binding fragments of heavy-chain antibodies, and together with classical antibodies, they are part of the camelid immune system. Nanobodies are small and stable, and they share a high degree of sequence identity to the human heavy chain variable domain, and these characteristics offer them advantages over classical antibodies or antibody fragments. We first immunised an alpaca with a human GBM stem-like cell line prepared from primary GBM cultures. Next, a nanobody library was constructed in a phage-display vector. Using nanobody phage-display technology, we selected specific GBM stem-like cell binders through a number of affinity selections, using whole cell protein extracts and membrane protein-enriched extracts from eight different GBM patients, and membrane protein-enriched extracts from two established GBM stem-like cell lines (NCH644 and NCH421K cells). After the enrichment, periplasmic extract ELISA was used to screen for specific clones. These nanobody clones were recloned into the pHEN6 vector, expressed in Escherichia coli WK6, and purified using immobilised metal affinity chromatography and size-exclusion chromatography. Specific nanobody:antigen pairs were obtained and mass spectrometry analysis revealed two proteins, TRIM28 and β-actin, that were up-regulated in the GBM stem-like cells compared to the controls.     

A Fourier transform infrared microspectroscopic imaging investigation into an animal model exhibiting glioblastoma multiforme

Glioblastoma multiforme (GBM) is a highly malignant human brain tumour for which no cure is available at present. Numerous clinical studies as well as animal experiments are under way with the goal being to understand tumour biology and develop potential therapeutic approaches. C6 cell glioma in the adult rat is a frequently used and well accepted animal model for the malignant human glial tumour. By combining standard analytical methods such as histology and immunohistochemistry with Fourier Transform Infrared (FTIR) microspectroscopic imaging and multivariate statistical approaches, we are developing a novel approach to tumour diagnosis which allows us to obtain information about the structure and composition of tumour tissues that could not be obtained easily with either method alone. We have used a "Stingray" FTIR imaging spectrometer to analyse and compare the compositions of coronal brain tissue sections of a tumour-bearing animal and those from a healthy animal. We have found that the tumour tissue has a characteristic chemical signature, which distinguishes it from tumour-free brain tissue. The physical-chemical differences, determined by image and spectral comparison are consistent with changes in total protein absorbance, phosphodiester absorbance and physical dispersive artefacts. The results indicate that FTIR imaging analysis could become a valuable analytic method in brain tumour research and possibly in the diagnosis of human brain tumours.     

Heterogeneous phenotype of human glioblastoma: In vitro study

Glioblastomas (GBMs) are the most lethal primary brain tumours. Increasing evidence shows that brain tumours contain the population of stem cells, so‐called cancer stem cells (CSCs). Stem cell marker CD133 was reported to identify CSC population in GBM. Further studies have indicated that CD133 negative cells exhibiting similar properties and are able to initiate the tumour, self‐renew and undergo multilineage differentiation. GBM is a highly heterogeneous tumour and may contain different stem cell populations with different functional properties. We characterized five GBM cell lines, established from surgical samples, according to the marker expression, proliferation and differentiation potential. CD133 positive cell lines showed increased proliferation rate in neurosphere condition and marked differentiation potential towards neuronal lineages. Whereas two cell lines low‐expressing CD133 marker showed mesenchymal properties in vitro, that is high proliferation rate in serum condition and differentiation in mesenchymal cell types. Further, we compared therapy resistance capacity of GBM cell lines treated with hydroxyurea. Our results suggest that CSC concept is more complex than it was believed before, and CD133 could not define entire stem cell population within GBM. At least two different subtypes of GBM CSCs exist, which may have different biological characteristics and imply different therapeutic strategies. Copyright © 2013 John Wiley & Sons, Ltd.