Search

Despite major advances in the molecular exploration of pediatric cancers, approximately 50 % of children with high-risk neuroblastoma lack effective treatment. To identify new therapeutic options for this group of high-risk patients, we have combined integrative data analysis with experimental evaluation in patient-derived xenograft cells. We propose a new algorithm, TargetTranslator, which combines data from tumor biobanks, pharmacological databases, and cellular networks, to predict how particular targeted interventions will affect mRNA signatures associated with high patient risk. We find more than 80 known and novel targets to be associated with neuroblastoma risk and differentiation signatures. To evaluate these predictions, we performed RNA sequencing of drug-treated cell lines derived from high-risk patients and show that predicted compounds suppress risk signatures and malignant phenotypes. Using a xenograft model, we establish CNR2 and MAPK8 as promising candidates for the treatment of high-risk neuroblastoma. We expect that our method (made available as a public tool, targettranslator.org) will enhance and expedite the discovery of risk-associated targets for pediatric and adult cancers.

Glioblastoma (GBM) is the most common malignant brain tumor in adults with a 15 month median survival after diagnosis. Patients receive surgical resection of the tumor, followed by aggressive radio- and chemotherapy with temozolomide (TMZ). Many human cancers including GBM demonstrate addiction to MYC transcription factor signaling and thus become susceptible to inhibition of MYC downstream genes. JQ1 is an effective inhibitor of BET Bromodomains, a class of epigenetic readers regulating expression of MYC genes and their downstream targets. In a panel of 19 patient-derived GBM cell lines, we showed that BET inhibition alone and in combination with TMZ decreases viability by inducing cell cycle arrest, apoptosis and senescence. We propose a distinct expression signature of GBM cells that correlates with a sensitivity to BET inhibition. Cell lines most sensitive to BET inhibition exhibited ten-fold upregulation of MYCN, as compared to resistant cell lines, while we could not find a correlation between MYC expression and sensitivity to the inhibition. Following BET inhibition, JQ1-sensitive cells downregulated MYCN both on mRNA and protein level, while the inhibition of JQ1-resistant cells had no effect on MYCN expression. In JQ1-sensitive cells, we found enrichment of pathways regulating cell cycle, DNA replication, DNA damage response and repair. As DNA repair leads to an acquired chemoresistance to TMZ, JQ1 treatment in combination with TMZ can further inhibit proliferation of TMZ-resistant cells. Intriguingly, JQ1-sensitive cells expressed high levels of OLIG2 and LGR5, suggesting that BET inhibition targets a subset of proneural GBM progenitors. Taken together, we suggest that BET inhibition can increase the potency of TMZ therapy in particular for GBM patients with a MYCN-positive signature. Ongoing in vivo evaluation of JQ1 and TMZ aims to determine the efficacy of the drug combination in inhibiting tumor growth.

We have recently identified LGR5 as a marker of an immature, neural stem cell-like cell of origin for glioblastoma (GBM). Here we have further investigated the role of LGR5 in GBM. We have found that in a panel of 61 patient-derived GBM stem cell lines, LGR5 was differentially expressed and significantly enriched in those of the proneural subtype. High LGR5 expression in proneural GBM lines (LGR5high-PN) was associated with increased self-renewal and tumorigenicity compared to low expressing (LGR5low) and high expressing mesenchymal (LGR5high-MS) GBM cell lines. Knock-down of LGR5 gene expression in LGR5high-PN lines resulted in decreased self-renewal, while over-expression in both LGR5high-PN and LGR5low lines produced a higher self-renewal capacity of the cells. Furthermore, by comparing LGR5high-PN cells with and without si/shLGR5 we found that LGR5 also could regulate tumor cell invasion in vitro and in vivo. When analyzing transcriptome changes upon LGR5 depletion we detected a transition from the proneural to the mesenchymal subtype in LGR5high-PN GBM cells treated with shLGR5 that was coupled to an upregulation of genes of the TNFA/NFKB pathway. Taken together our data suggests that LGR5 could be an essential regulator of a subset of GBMs. To stratify patients carrying LGR5-dependent GBMs we have identified a number of additional candidate biomarkers, which together with LGR5 could help defining LGR5-driven GBM cells.

Glioblastoma Multiforme (GBM) is the most frequent adult primary malignant brain tumor that remains incurable despite aggressive treatment. The cell of origin (COO) for GBM is unknown but assumed to be a glial stem or progenitor cell. GBM harbours hierarchical tumor cells called glioma stem cells (GSCs) that maintain tumor growth, drive tumor progression and cause tumor relapse due to their increased resistance to therapy. We have analyzed the significance of cellular origin for GBM development and GSC properties by comparing mouse GBMs and GSCs derived thereof induced in neural stem cells (NSCs), glial-restricted precursor cells (GPCs) or oligodendrocyte precursor cells (OPCs) by identical mutations. There were striking differences in GBM development and the phenotypes of GSCs and their response to drugs owing to the COO. Global gene expression analysis of mouse GSC lines displayed a clear separation due to COO and differential gene expression analysis identified a COO gene signature of 175 genes. Cross-species bioinformatics analyses were performed. First we analyzed the human cancer genome atlas (TCGA) GBM tissue samples and the mouse GSC expression data for a collection of TCGA GBM subtype signature genes. This showed that we could model both Proneural and Mesenchymal GBMs in mice by merely switching the COO. Next, we used the mouse COO gene signature to stratify a large number of newly established human glioma stem cell lines. This produced two groups of human GSCs; the NSC origin group and the progenitor cell (PC) origin group in which the mouse GPC- and OPC-derived genes were combined. Importantly, patient survival was significantly different between the NSC and PC COO groups with a better prognosis for the PC group patients. Thus, the cell of origin is essential for GBM biology and needs to be considered for more accurate patient stratification, target identification and drug discovery.

Glioblastoma multiforme (GBM) is the most aggressive brain tumor in adults, and despite state-of-the-art treatment, survival remains poor and novel therapeutics are sorely needed. The aim of the present study was to identify new synergistic drug pairs for GBM. In addition, we aimed to explore differences in drug-drug interactions across multiple GBM-derived cell cultures and predict such differences by use of transcriptional biomarkers. We performed a screen in which we quantified drug-drug interactions for 465 drug pairs in each of the 5 GBM cell lines U87MG, U343MG, U373MG, A172, and T98G. Selected interactions were further tested using isobole-based analysis and validated in 5 glioma-initiating cell cultures. Furthermore, drug interactions were predicted using microarray-based transcriptional profiling in combination with statistical modeling. Of the 5 465 drug pairs, we could define a subset of drug pairs with strong interaction in both standard cell lines and glioma-initiating cell cultures. In particular, a subset of pairs involving the pharmaceutical compounds rimcazole, sertraline, pterostilbene, and gefitinib showed a strong interaction in a majority of the cell cultures tested. Statistical modeling of microarray and interaction data using sparse canonical correlation analysis revealed several predictive biomarkers, which we propose could be of importance in regulating drug pair responses. We identify novel candidate drug pairs for GBM and suggest possibilities to prospectively use transcriptional biomarkers to predict drug interactions in individual cases.