KS01.5.A Allergic airway inflammation impacts tumor take and delays experimental glioblastoma progression

BACKGROUND

Numerous epidemiological studies have highlighted the protective role of immunoglobulin E-mediated allergic diseases on glioblastoma (GBM) susceptibility and prognosis. However, the mechanistic explanations behind these phenomena remain unexplored. Our objective was to set up a preclinical model and investigate the mechanisms underlying such protection to improve our understanding of the crosstalk between immune system and brain tumor development.

MATERIAL AND METHODS

A mouse model of allergic airway inflammation (AAI) induced by repeated nasal instillation of House Dust Mite extract was initiated before intracranial implantation of GL261 glioma cells, in both immunocompetent (C57BL/6) and immunodeficient (RAG-KO) mice. Tumor take and tumor growth were monitored by MRI. Central (microglia) and peripheral (spleen, bone marrow) immune cells were characterized by flow cytometry. The response of microglia was further assessed by RNA sequencing. Impact of candidate genes on patient survival was characterized by Cox regression analysis using data from TCGA and CGGA.

RESULTS

Following AAI induction in C57BL/6 mice, engraftment of GL261 cells in the brain was delayed and tumor growth rate was reduced. This correlated with an increase in survival of the mice and was accompanied by increased effector memory T-cells in the circulation. Of note, the survival benefit was lost in RAG-KO mice devoid of adaptive immunity. At the level of the brain, we observed enhanced secretion of TNFα and IL6 in microglia ex vivo. AAI induced a transcriptional reprogramming of microglia towards a pro-inflammatory-like state. We identified an allergy-related microglia gene signature that is associated with improved prognosis of glioma patients.

CONCLUSION

Our results demonstrate that AAI limits both tumor take and GBM progression in mice, providing a preclinical model to study the role of allergic inflammation in GBM susceptibility and prognosis, respectively. At the functional level, we identify a potentiation of microglial and adaptive anti-tumoral immunity. Further investigations are warranted to shed light on the reciprocal crosstalk between microglial reprogramming and peripheral immunity in the context of allergies and brain tumors.

ANGI-02. GENOME-WIDE shRNA SCREEN IDENTIFIES CANDIDATE GENES DRIVING GLIOBLASTOMA INVASION

BACKGROUND

A major hallmark of glioblastoma (GBM) is its invasive capacity, contributing to its aggressive behaviour. Invasive cells cannot be easily removed by surgery or irradiation and eventually result in lethal recurrence. A better understanding of the invasion process and the key molecular players underlying the invasive potential of GBM may lead to the identification of new therapeutic targets for GBM patients.

MATERIAL AND METHODS

To identify candidate genes responsible for invasion, a genome-wide shRNA screen was performed in patient-derived GBM cultures. The most promising candidate was validated in in vitro invasion assays, ex vivo brain slice cultures and in vivo orthotopic xenografts in mice. Gene knockdown in invasive GBM cells was compared with overexpression in non-invasive cells. RNAseq of knockdown cells, along with the generation of deletion constructs were applied to uncover the mechanisms regulating invasion.

RESULTS

A zinc-finger domain containing protein was identified as an invasion essential candidate gene. Knockdown of this gene confirmed a strong impact on invasion in highly invasive GBM cells. In contrast, gene overexpression switched non-invasive GBM cells to an invasive phenotype. Deletion of one or both zinc-finger motifs decreased invasion indicating that both are essential for regulating invasion. Mutation of the nuclear localisation signal resulted in retention of the protein in the cytoplasm and loss of the invasion phenotype demonstrating that the protein activity is required in the nucleus. Gene expression analyses revealed that invasion-related genes are significantly regulated by the candidate gene once it is localized in the nucleus.

CONCLUSION

We identified a zinc-finger containing protein as a novel driver of GBM invasion, presumably through transcription factor activity resulting in the induction of an invasive transcriptional program. This protein and its downstream pathway may represent novel promising targets to overcome invasive capacities in GBM.

GENE-02. ESTABLISHING PERSONALIZED TREATMENT OPTIONS FOR RECURRENT HIGH-GRADE GLIOMAS

BACKGROUND

High grade glioma (HGG) patients develop resistance to standard treatment leading to disease progression and limited life expectancy. Recent advances in the molecular characterisation of treatment-naïve HGGs based on next generation sequencing and DNA methylation analyses have led to a better delineation of HGG-subtypes and identification of distinct genomic abnormalities opening opportunities for personalized treatment strategies.

METHODS

We collected 300 fresh glioma specimen with approximately 100 longitudinal samples of initial and recurrent tumors from 43 matched patients. We succeeded in generating a live-biobank of HGG patient-derived orthotopic xenografts (PDOX) and 3D tumor organoids that neatly recapitulates the mutational spectrum including structural DNA variations and methylation-based subtypes of gliomas. A highlight is the generation of 19 PDOXs of paired initial and relapse HGGs from 9 glioma patients, enabling high-throughput drug screens. We performed comprehensive molecular profiling using arrayCGH, DNA-methylation and targeted DNA sequencing on patient specimen and their derivatives, 3D tumor organoids and PDOXs.

RESULTS

Detailed analysis of the paired longitudinal samples indicated that PDOXs closely recapitulate the evolutionary trajectory of the parental tumors. Furthermore, targeted genomic sequencing of paired HGGs suggests that relapse tumors also accumulate somatic mutations in epigenetic effectors. Based on patient-derived material we carried out drug response screening on 3D tumor organoids using a compound library matching the majority of genes that were assessed with targeted sequencing. Differential drug responses between initial and recurrent tumors were observed and the prevailing primary drug response profiles were essentially recapitulad in the relapse setting.

CONCLUSIONS

Response assessment of treatment-naïve gliomas and their recurrences provides crucial information on the differential sensitivity between initial and relapsed HGGs and offers novel personalized therapeutic options for the relapse setting. Furthermore, in depth correlation of the profiled somatic molecular landscape with drug response will enable pharmacogenomic predictions of potential inhibitors in the clinical setting.

P11.26 Genome-wide shRNA screen identifies candidate genes driving glioblastoma invasion

BACKGROUND

A major hallmark of glioblastoma (GBM) is its highly invasive capacity, contributing to its aggressive behaviour. Since invasive cells cannot be easily removed by surgery or irradiation, they are left behind and eventually result in lethal recurrence. Therefore, a better understanding of the invasion process and of the key molecular players underlying the invasive capacities of GBM may lead to the identification of new therapeutic targets for GBM patients.

MATERIAL AND METHODS

To identify candidate genes responsible for invasion, a genome-wide shRNA screen was performed in patient-derived GBM sphere cultures. The phenotype of the most promising candidate was validated in in vitro invasion assays, ex vivo brain slice cultures and in vivo orthotopic xenografts in mice. Gene knockdown in invasive GBM cell lines was compared with overexpression in non-invasive cells. RNA sequencing of knockdown cells, along with the generation of deletion constructs were applied to uncover the mechanisms regulating invasion.

RESULTS

Through a whole genome shRNA screen, a zinc-finger containing protein was identified as an invasion essential candidate gene. Knockdown of this gene confirmed a strong decrease in invasion capacity in two highly invasive GBM cell lines. In contrast, gene overexpression switched non-invasive GBM cells to an invasive phenotype. Deletion of either one or both zinc-finger motifs led to decreased invasion indicating that the two zinc-finger motifs are essential for regulating invasion. Mutation of the nuclear localisation signal resulted in retention of the protein in the cytoplasm and loss of the invasion phenotype demonstrating that the protein activity is required in the nucleus. Gene expression analyses revealed that invasion-related genes are significantly regulated by the candidate gene once it is localized in the nucleus.

CONCLUSION

We identified a zinc-finger containing protein as a novel driver of GBM invasion, presumably through a transcription factor activity resulting in the induction of an invasive transcriptional program. This protein and its downstream pathway may represent a novel promising target to overcome invasive capacities in GBM.

OS12.2 Targeting epigenetic pathways in the treatment of recurrent high-grade glioma

BACKGROUND

High grade glioma (HGG) patients develop resistance to standard treatment leading to disease progression and limited life expectancy. Advances in the molecular characterisation of treatment-naïve HGGs based on next-generation sequencing and DNA methylation analyses have led to a better delineation of HGG subtypes and the identification of distinct genomic abnormalities. Furthermore, using large patient cohorts of longitudinal tumor samples, comprehensive genomic profiling studies emerged to investigate therapy-associated evolution of gliomas. All together, those studies point out the need for personalised treatment strategies, where applied drugs will be adapted to the unique patient-specific genetic abnormalities.

MATERIAL AND METHODS

We collected fresh samples of more than 800 brain tumors containing almost 300 glioma specimen with approximately 100 longitudinal samples of initial and recurrent tumors from 43 matched patients. By now, we have successfully established 34 patient-derived orthotopic xenografts (PDOXs) in mice. We performed comprehensive molecular profiling using array comparative genomic hybridisation, DNA methylation analysis and targeted DNA sequencing on patient specimen and their derivatives such as 3D tumor organoids and PDOXs. The custom-design sequencing panel comprises 234 genes that reflect both established genetic identifiers for individual glioma subtype classification and novel genes encoding mainly epigenetic effector genes. Based on patient-derived material we carried out drug response screening on 3D tumor organoids using a compound library matching the majority of genes that were assessed by targeted sequencing.

RESULTS

We succeeded in generating a live biobank of HGG patient-derived xenografts and 3D organoids that neatly recapitulates the mutational spectrum including structural DNA variation and methylation-based subtypes of gliomas. A highlight is the generation of 19 PDOXs of paired initial and relapse HGGs from a total of 9 glioma patients. A detailed analysis of the paired longitudinal samples indicated that PDOX models closely recapitulate the evolutionary trajectory of the parental tumors. Targeted sequencing of longitudinal HGG PDOXs suggests that relapse tumors accumulate somatic mutations in epigenetic effectors compared with the Initial. Differential drug responses between initial and relapse tumors were observed after screening of in vitro 3D tumor organoids.

CONCLUSION

Response assessment of naïve initial gliomas and recurrences provides crucial information on the differential sensitivity between initial and relapsed HGGs and offers novel personalised therapeutic options in the relapse setting. Furthermore, in depth correlation of the profiled somatic molecular landscape with drug response will enable pharmacogenomic predictions of potential inhibitors in the clinical setting.

Estimating the Number of Clusters in DNA Microarray Data

Objectives: The main objective of the research is an application of the clustering and cluster validity methods to estimate the number of clusters in cancer tumor datasets. A weighed voting technique is going to be used to improve the prediction of the number of clusters based on different data mining techniques. These tools may be used for the identification of new tumour classes using DNA microarray datasets. This estimation approach may perform a useful tool to support biological and biomedical knowledge discovery.

Methods: Three clustering and two validation algorithms were applied to two cancer tumor datasets. Recent studies confirm that there is no universal pattern recognition and clustering model to predict molecular profiles across different datasets. Thus, it is useful not to rely on one single clustering or validation method, but to apply a variety of approaches. Therefore, combination of these methods may be successfully used for the estimation of the number of clusters.

Results: The methods implemented in this research may contribute to the validation of clustering results and the estimation of the number of clusters. The results show that this estimation approach may represent an effective tool to support biomedical knowledge discovery and healthcare applications.

Conclusion: The methods implemented in this research may be successfully used for the estimation of the number of clusters. The methods implemented in this research may contribute to the validation of clustering results and the estimation of the number of clusters. These tools may be used for the identification of new tumour classes using gene expression profiles.

Discovering relevance knowledge in data: a growing cell structures approach

Both information retrieval and case-based reasoning systems rely on effective and efficient selection of relevant data. Typically, relevance in such systems is approximated by similarity or indexing models. However, the definition of what makes data items similar or how they should be indexed is often nontrivial and time-consuming. Based on growing cell structure artificial neural networks, this paper presents a method that automatically constructs a case retrieval model from existing data. Within the case-based reasoning (CBR) framework, the method is evaluated for two medical prognosis tasks, namely, colorectal cancer survival and coronary heart disease risk prognosis. The results of the experiments suggest that the proposed method is effective and robust. To gain a deeper insight and understanding of the underlying mechanisms of the proposed model, a detailed empirical analysis of the models structural and behavioral properties is also provided.

Special section on machine intelligence approaches to systems biology

The three papers in this special section focus on machine intelligence approaches to systems biology.

Estimating the number of clusters in DNA microarray data

OBJECTIVES

The main objective of the research is an application of the clustering and cluster validity methods to estimate the number of clusters in cancer tumor datasets. A weighed voting technique is going to be used to improve the prediction of the number of clusters based on different data mining techniques. These tools may be used for the identification of new tumour classes using DNA microarray datasets. This estimation approach may perform a useful tool to support biological and biomedical knowledge discovery.

METHODS

Three clustering and two validation algorithms were applied to two cancer tumor datasets. Recent studies confirm that there is no universal pattern recognition and clustering model to predict molecular profiles across different datasets. Thus, it is useful not to rely on one single clustering or validation method, but to apply a variety of approaches. Therefore, combination of these methods may be successfully used for the estimation of the number of clusters.

RESULTS

The methods implemented in this research may contribute to the validation of clustering results and the estimation of the number of clusters. The results show that this estimation approach may represent an effective tool to support biomedical knowledge discovery and healthcare applications.

CONCLUSION

The methods implemented in this research may be successfully used for the estimation of the number of clusters. The methods implemented in this research may contribute to the validation of clustering results and the estimation of the number of clusters. These tools may be used for the identification of new tumour classes using gene expression profiles.

Integrative data analysis for functional prediction: a multi-objective optimization approach

An integrative classification system for functional genomics is introduced. A comparison with a previous study of the yeast mitochondrial proteome is presented.

AVAILABILITY

A demonstration prototype, interSearch, is available on request.

SUPPLEMENTARY INFORMATION

http://ijsr32.infj.ulster.ac.uk/~e10110731/interSearch.

A cluster validity framework for genome expression data

This paper presents a method for the assessment of expression cluster validity.

A computational neural approach to support the discovery of gene function and classes of cancer

Advances in molecular classification of tumours may play a central role in cancer treatment. Here, a novel approach to genome expression pattern interpretation is described and applied to the recognition of B-cell malignancies as a test set. Using cDNA microarrays data generated by a previous study, a neural network model known as simplified fuzzy ARTMAP is able to identify normal and diffuse large B-cell lymphoma (DLBCL) patients. Furthermore, it discovers the distinction between patients with molecularly distinct forms of DLBCL without previous knowledge of those subtypes.

Improving clinical decision support through case-based data fusion

This paper presents an information fusion technique based on a knowledge discovery model, and the case-based reasoning decision framework. Using signal data and database records from the heart disease risk estimation domain, three data fusion methods are discussed. Two of these methods combine information at the retrieval-outcome level, and one method merges data at the discovery-input level. The result of these three models are compared and evaluated against the performance of single-source models. It is shown that the methods that fuse information at the retrieval-outcome level are significantly superior.

Predicting coronary disease risk based on short-term RR interval measurements: a neural network approach

Coronary heart disease is a multifactorial disease and it remains the most common cause of death in many countries. Heart rate variability has been used for non-invasive measurement of parasympathetic activity and prediction of cardiac death. Patterns of heart rate variability associated with respiratory sinus arrhythmia have recently been considered as possible indicators of coronary heart disease risk in asymptomatic subjects. The aim of this work is to detect individuals at varying risk of coronary heart disease based on short-term heart rate variability measurements under controlled respiration. Artificial neural networks are used to recognise Poincaré-plot-encoded heart rate variability patterns related to coronary heart disease risk. The results indicate a relatively coarse binary representation of Poincaré plots could be superior to an analogue encoding which, in principle, carries more information.