Research Group Molecular Tumor Pathology

Project description

Colon cancer (CC) is a major cause of death. Current treatment uses chemotherapy combined with anti-EGFR or VEGF drugs, and radiotherapy. Screening for biomarkers can indicate whether EGFR-inhibitors will be effective in patients but otherwise biomarkers for personalising patient treatment are scarce.

First-line combinations of chemotherapy and EGFR-inhibitors for stratified patients being RAS wild-type have led to an increase in overall survival to more than 30 months, but most patients develop RAS mutations under anti-EGFR therapy, or do not respond to EGFRi for unknown reasons. The majority of patients develop resistance and succumb to the disease.

We still have a poor understanding of how gene networks drive cancer, how they modulate response, and how they induce resistance to treatment. Good disease models that can provide insight are simply lacking. With the availability of large public data resources, our unique collection of patient samples and patient-derived models and with computational and experimental approaches becoming mature, the COLOSYS consortium aims to develop in silico therapy response predictors that allow precision medicine, based on patient-specific driver and resistance mechanisms. We will identify new CC tumor driver genes by integration of multiple data types from large public tumor and cell line repositories. A high quality, open repository of data and knowledge (knowledge commons) will be assembled and used to construct multiscale computer models of the molecular networks that underlie cancerous cell proliferation. Logical model simulations will predict the effect of drugs in cancer cell lines and patient tumors. We will test these predictions on cell lines as well as patient derived cell cultures, organoids and mouse xenografts, and perform preliminary testing in patients. The combined computational, experimental and clinical testing will provide understanding of resistance mechanisms, and allow personalised treatment of colon cancer.

Consortium

Coordinator:

Prof. Martin Kuiper
Norwegian University of Science and Technology
Trondheim, Norway

Co-Investigators:

1. Prof. Christine Sers
   Molecular Tumor Pathology, Charité–Universitätsmedizin Berlin
2. Dr. Emmanuel Barillot Institut Curie
   Paris, Frankreich
3. Prof. Alfonso Valencia
   Spanish National Center for Cancer Research (CNIO)
   Madrid, Spanien
4. Prof. Lodewyk Wessels
   The Netherlands Cancer Institute, Amsterdam 

Project description

Pancreatic NET (pNET) comprise the most prominent subgroup group of rare neuroendocrine tumors (NET) with distinct prognostic classes, and thus diverse therapeutic regimens. Available pNET treatments include somatostatin analogs, systemic chemotherapy, and novel molecular drugs targeting receptor tyrosine kinases (Sunitinib), or the mTOR pathway (Everolimus). However, tumor heterogeneity results in an unpredictable response to the therapy, and only a limited number of patients profits from either treatment. To date, no method for diagnostic stratification of patients exists.

A main hypothesis of the project is that improvement of therapy in pNETs can be achieved through a focussed systems biology approach, which considers the patient-individual mutation and expression profile. Patient-individual alterations such as mutations, expression changes and activation states will be represented within a combined MAPK-PI3K-mTOR mathematical model. This model will then be used to predict to what extend therapeutic interferences within these pathways are successful, or whether escape mechanisms are to be expected. We expect a deeper insight into the mechanisms that drive pNETs and furthermore the extraction of potential predictive biomarkers for indivualised therapy that can be used in routine diagnostics.

Consortium

Coordinator:

Prof. Dr. Christine Sers
Institute of Pathology (Charité – Universitätsmedizin Berlin)

Co-Investigators:

1. Prof. Dr. Kathrin Thedieck
   Institut für Biologie und Umweltwissenschaften, Neurogenetik
   Carl von Ossietzky Universität Oldenburg, D
   University Medical Center Groningen, Center for Liver, Digestive and Metabolic Diseases and Systems Biology Centre for Metabolism and Ageing, University of Groningen, NL
2. Prof. Dr. Nils Blüthgen
   Institute of Pathology, Systems Biology of Molecular Networks
   Charité, Universitätsmedizin Berlin
3. Prof. Dr. med. Marianne Pavel
   Medizinische Klinik m. S. Hepatologie und Gastroenterologie
   Charité, Universitätsmedizin Berlin
   Now:
   Universitätsklinikum Erlangen
   Leiterin Schwerpunkt Endokrinologie und Diabetologie
4. Dr. Katharina Detjen
   Medizinische Klinik m. S. Hepatologie und Gastroenterologie
   Charité, Universitätsmedizin Berlin
5. Prof. Dr. Ulf Leser
   Institute for Computer Science
   Humboldt-Universität zu Berlin

Project description:

The major objective of the project is to accomplish an in silico network of genetic, epigenetic and signaling processes to tailor diagnostics with predictive information for therapies targeting receptor tyrosine kinases (e.g. EGFR) in colorectal cancer. The network model is expected to provide diagnostic guidance for the improvement of therapy response prediction by achieving an in-depth functional understanding of receptor-mediated signaling and downstream processes. The core model will consist of a “static” colon-specific signaling network model that serves as a framework for integrating detailed kinetic models of MAPK and Wnt pathways, existing genetic/epigenetic information, high-throughput data, knowledge on predictive markers obtained by literature mining, and physiologically relevant data. The dynamics of MAPK and Wnt pathways, which are crucial for targeted therapies, will be subjected to detailed mathematical modeling based on results from focused experimental measurements. Once detailed models of pathways, key signaling nodes, feed-back loops and marker combinations are established, we will turn the static network into a dynamic model in an iterative process. Finally, this approach should generate a tool for the in silico evaluation of novel markers or combinations of them for an improved therapy prediction. We expect that our models will also be useful for other types of cancer and related predictive diagnostics.

Project description:

The overall aim of the OncoPATH Consortium is to analyze the qualitative and quantitative impact of frequent genetic and epigenetic alterations on the signaling pathways and metabolic networks in colorectal cancer cells. The changes in the metabolic networks and in signal transduction will be integrated into a mathematical model and functionally analyzed. Thereby, the theoretical model will be successively adapted and improved resembling single experimental models (e.g. a cell line) or even individual patients. The model can then be used to simulate therapeutic interference and propose new therapeutic options.

Coordinators:

Prof. Nils Blüthgen, Charité–Universitätsmedizin Berlin
Prof. Christine Sers, Charité – Universitätsmedizin

Project Partners:

1. Reinhold Schäfer, Charité – Universitätsmedizin
2. Markus Morkel, Charité – Universitätsmedizin
3. Ulf Leser, Humboldt Universität zu Berlin
4. Edda Klipp, Humboldt Universität zu Berlin
5. Tilmann Brummer, Institute of Molecular Medicine and Cell Research, University of Freiburg
6. Stefan Legewie, Institute of Molecular Biology, Mainz
7. Stefan Kempa, Max Delbrück Center for Molecular Medicine
8. Iduna Fichtner, EPO GmbH, Berlin

Supported by:

BMBF, e:BIO

ColoNet – A systems biology approach for integrating molecular diagnostics and targeted therapy in colorectal cancer more information about ColoNet

The major objective of the project is to accomplish an in silico network of genetic, epigenetic and signaling processes to tailor diagnostics with predictive information for therapies targeting receptor tyrosine kinases (e.g. EGFR) in colorectal cancer. The network model is expected to provide diagnostic guidance for the improvement of therapy response prediction by achieving an in-depth functional understanding of receptor-mediated signaling and downstream processes. The core model will consist of a “static” colon-specific signaling network model that serves as a framework for integrating detailed kinetic models of MAPK and Wnt pathways, existing genetic/epigenetic information, high-throughput data, knowledge on predictive markers obtained by literature mining, and physiologically relevant data. The dynamics of MAPK and Wnt pathways, which are crucial for targeted therapies, will be subjected to detailed mathematical modeling based on results from focused experimental measurements. Once detailed models of pathways, key signaling nodes, feed-back loops and marker combinations are established, we will turn the static network into a dynamic model in an iterative process. Finally, this approach should generate a tool for the in silico evaluation of novel markers or combinations of them for an improved therapy prediction. We expect that our model will also be useful for other types of cancer and related predictive diagnostics.

Subproject:  Molecular characterization of experimental models
Key features of the project are experimental models extensively characterized for para-meters known or expected to be involved in therapy efficacy. A significant proportion of information is available from the partners’ previous work and through public databases, while the generation of missing information will produce a high experimental standardization and ensure maximal clinical relevance.

Subproject: Assessing the dynamics of key pathways following specific intervention
This project will focus on two paradigmatic oncogenic pathways: the receptor-MAPK and the Wnt signaling systems, directly relevant for prediction and therapy and including the functionally associated receptors and underlying genetic and epigenetic alterations. For dynamic modeling at the systems level, multi-level data harboring functional and quantitative information will be gathered after defined perturbation of potential predictive markers. The focus will be on direct pathways intervention effects, the dynamics of such processes and effects on potentially associated “tumor” markers, e.g. transcriptional and DNA-methylation markers.

Supported by: BMBF, MedSYS

Systems biology of genetic diseases - Mutanom

Subproject: Effects of cancer gene mutations on the intracellular signalling network and transformed phenotype more information about Mutanom

The objective of this subproject is to analyze novel mutated cancer genes a) known to occur in solid human tumors, b) to be involved in cancer-related developmental disorders and c) identified/nominated by the consortium for their contribution to transformed phenotypes and for their impact on the signalling network. The experimental strategy depends on the nature of the mutations (deletion, duplication, insertion, nonsense, splice site alteration) and on the assignment of the affected gene to a functional group based on Gene Ontology (GO)-based molecular function or biochemical process groups. Depending on the previous characterization of the mutations, functional tests are chosen to either interrogate transforming or tumor-suppressing activities. More detailed investigations that exceed the proof of transforming or transformation-suppressing activity are done in collaboration with the partners. In view of already known common mechanisms of cancer initiation and progression, it is desirable to study the impact of mutated cancer genes, regardless of their putative tumor-promoting or -limiting function, on the handful of signalling pathways with oncogenic potential. Understanding how the perturbation of the signalling network ultimately determines the malignant features of cancer cells is a prerequisite for identifying the rate-limiting steps in cancer growth and to design rational therapies. 

Supported by  BMBF, NGFN Plus

Modelling of signalling cascades – the impact of RAS signal transduction on transcription

Special Collaborative Reserch Center (SFB 618), Theoretical Biology: Robustness, Modularity and Evolutionary Design of Living Systems, subproject A3 (jointly with Nils Blüthgen) more information about SFB618

Recently, the RAS pathway hit the headlines of translational cancer research again due to the finding that KRAS mutations have an impact on the clinical response toward receptor-targeted cancer therapies. Mutational activation of KRAS is primarily associated with therapy resistance, while wild-type KRAS activity is associated with a therapeutic benefit, however, only in a subset of cancer patients. At present, therapy responders and non-responders cannot be distinguished in molecular terms. Treatment allocation based on detailed knowledge of the molecular mechanisms is not possible. Therefore, signalling processes downstream of the RAS molecular switch are among the prime candidate mechanisms to be screened for clinically relevant alterations. Particularly, studying stimulus-response characteristics, dynamics and feedback regulation of the RAS/MAPK pathway and of other RAS-dependent pathways as well as assessing the integration of pathway activation with the transcriptional program are of utmost importance for defining functionally relevant nodes for therapeutic intervention and for identifying prognostic parameters.

During the past funding period, we have addressed several themes related to RAS/MAPK pathway dynamics. Using mathematical models, we investigated ultrasensitivity, bistability and feedback regulation in MAPK signaling. The models were experimentally tested in cell populations and in single cells stimulated by growth factors or by expression of conditional RAS genes. By expression profiling and the development of a new algorithm, which predicts biological functions regulated by a combinatorial interaction of transcription factors, we identified dual specificity phosphatases (DUSPs) as MAPK-dependent target genes involved in negative feedback regulation. Mathematical models were fitted to experimental time-series of MAPK activation and phosphatase expression. By mining microarray data and results from the literature we also confirmed that transcriptional negative feedback is a general design pattern in PI3K, cAMP, SMAD and STAT signalling pathways.

During the current funding period (until 2013), we (1) aim to investigate the noise/reliability in the expression of MAPK targets and downstream transcriptional networks identified before, to refine our predictions and to further underpin them with experimental data. Specific questions are as follows: How homogeneously do direct target genes respond, how does the noise in gene expression transmit into downstream transcriptional networks and does it rise in magnitude? (2) We have identified about 50 differentially regulated transcription factors in RAS-transformed epithelial cells compared to normal precursor cells. Seven factors have been selected for a series of RNAi-based pertubation experiments. Modular response analysis (MRA), a reverse engineering approach, is used to deduce transcription factor interactions. We plan to extend MRA and other computational methods to the entire set of transcription factors. Phenotypic read-outs of pertubation experiments established in the previous period will be used to study the biological consequences of network interference with respect to proliferation and neoplastic transformation. (3) The RAL pathway is gaining increased importance as the third major pathway downstream of RAS, particularly in model systems based on human cells as well as in tumours. Our previous expression profiling studies have identified functionally relevant target genes, grouped in a KRAS-dependent transcriptional module that responds neither to MAPK nor to PI3K pathway signals. We plan to investigate the role of the Ral pathway in the transcriptional and phenotypic response, using approaches developed for the MAPK pathway.

Supported by DFG/SFB 618

Ovarian carcinoma – Molecular and functional characterization of a KRAS oncogene-driven tumor model

We have established an ovarian cancer model based on ovarian epithelial cells. To mimick oncogenic pathway activation, which often is mediated by membrane-based receptors in the clinical setting, we introduced and expressed a mutated KRAS gene in those cells.  Transfected cells exhibit the typical features of transformation in vitro such as anchorage-independence, epithelial-mesenchymal transition (EMT) and motility. Progressively growing tumors are formed in nude mice following subcutaneous or intraperitoneal injection of cells. The malignant phenotype is associated with numerous alterations of gene expression. To understand the causal role of these alterations and the mechanisms of gene deregulation, we perform functional studies. For example, we investigate the role of the HMGA2 protein, an architectural transcription factor, in KRAS-transformed ovarian epithelial cells. HMGA2 expression is silenced by RNA interference and the effects on proliferation, EMT and anchorage independence are determined. For comparison, an HMAG2 expression vector is introduced into normal ovarian epithelial cells to find out, if HMGA2 may act as a oncogene itself. We measure the transcription of KRAS-responsive target genes by interrogating a customized oligonucleotide microarray (RASTA, RAS target array), which represents approx. 300 validated deregulated genes. Moreover, we search for transcription factor binding sites in the regulatory regions of HMGA2-regulated targets. This will help to elucidate the network of HMGA2, acting a as a potential master regulator, and interacting specific transcription factors driving malignancy in the ovarian epithelium.

Supported by: Berlin Cancer Society