Final Report Summary - GENINCA (Genomic instability and genomic alterations in pre-cancerous lesions and/or cancer) Project context and objectives: The overall objectives of the GENINCA consortium included to provide novel information on cancer initiating stem cells, on early stage lesions and exploration of underlying mechanisms of genomic alterations, investigation of the microenvironment and immune response. As GENINCA was a translational research effort, an especial focus was also on the discovery and development of novel biomarkers for tumour diagnosis, prognosis, and disease monitoring. In the following we summarise the most important achievements. Novel information on cancer initiating stem cells Beneficiary four developed protocols for the isolation and characterisation of cancer stem cells. Beneficiary four succeeded in the development of cancer stem cell based murine models of colorectal cancer disease and its use in pharmacological testing. These efforts were accompanied by functional studies. Beneficiary 12 identified p53 dependent checkpoints that have a role in preventing the evolution of chromosomal instability in stem cells with dysfunctional telomeres. Together with Beneficiary one Beneficiary 12 investigated the consequences of telomere dysfunction in stem cells in detail using sophisticated single-cell technologies. Thus, both Beneficiaries characterised the role of telomere dysfunction and telomerase in cancer initiation and progression and analysed functional consequences of conditional intestinal p53 deletion of mice with functional or dysfunctional telomeres. These efforts allowed the identification of p53 dependent checkpoints that have a role in preventing the evolution of chromosomal instability in stem cells in tissues with dysfunctional telomeres. The deletion of p53 impaired the depletion of chromosomal instable intestinal stem cells of telomere dysfunctional mice. These instable stem cells contributed to epithelial regeneration, thereby leading to an accumulation of chromosomal instability, increased apoptosis, altered epithelial cell differentiation, and premature intestinal failure. The GENINCA consortium also defined the role of p53 in liver stem cells. p53 can limit the self-renewal of stem cells from various tissues. Experimental evidence has suggested that deletion of p53 can cooperate with other oncogenic events to induce aberrant self-renewal and transformation of progenitor cells. It was unknown whether p53 deletion alone can lead to liver tumour formation. The GENINCA consortium showed that loss of p53 alone is sufficient to induce primary, bi-lineal liver cancer originating from chromosomal instable liver progenitor cells or dedifferentiated hepatocytes. Thus, the GENINCA consortium has contributed to an improved understanding of the role of stem cells in colorectal and liver cancer. Early stage lesions and exploration of underlying mechanisms of genomic alterations Beneficiary one developed and applied novel technologies for the analysis of genomic alterations in single cells and precursor lesions. This was instrumental for the analysis of circulating tumour cells (CTCs) as detailed below. In addition, Beneficiary one investigated together with Beneficiary three the genomic consequence of oncogene induction, especially of Ras oncogene induction. The GENINCA-consortium studied hyper-replication after Ras oncogene induction as an important mechanism for pathogenic genome-wide alterations, which occurred in a random, non-clonal fashion. As the alterations were associated with losses of tumour-suppressor genes and gains of oncogenes, this mechanism is likely important in human tumourigenesis. The exploration of mechanisms resulting in cancer is an ultimate prerequisite for cancer preventive means and the development of targeted therapies. Beneficiaries one and seven especially focused on genomic instability as reflected in CTCs. The GENINCA consortium compared the single cell CTC data with those obtained from the respective primary tumours and metastases. The results revealed a surprising heterogeneity between the CTCs. Some CTCs showed a combination of chromosomal gains and losses which significantly differed from those observed in the primary tumour and/or metastases. However, our results clearly indicate that with the spectrum of technologies applied by the GENINCA consortium, CTCs have become amenable to be used as predictive biomarkers. The single cell technologies were further extended to the application of novel next-generation sequencing technologies. This allowed the simultaneous sequencing of up to 68 cancer related genes. Thus, genomic instability was also explored on the base pair level. Due the results obtained by the GENINCA consortium, CTC analysis will now evolve to a novel tool for disease monitoring. Such non-invasive methods for disease monitoring from peripheral blood have also been referred to as 'liquid biopsy'. Furthermore, the CTC analysis will likely provide novel insights into metastasis. Their analysis can serve as predictive biomarker, e.g. by monitoring the KRAS mutation status in patients with colorectal cancer. The analyses regarding genomic instability were extended by Beneficiary 13 to colon adenomas, which surprisingly revealed that a large number of adenomas have already genomic instability to a large extent. In summary, the GENINCA consortium characterised in detail precursor lesions and early stage cancer lesions, for which prior to the project only limited information was available. A surprising finding was that genomic instability is apparently a very early event during tumourigenesis. The consortium successfully identified some mechanisms which may give rise to the instability and which may evolve to novel therapeutic targets. Microenvironment and immune response The GENINCA consortium analysed the role of the immune system in cancer and established classification systems based on immune response, which allow predicting the patient's prognosis. In fact, we improved the current understanding of the microenvironment, especially the function of T cells in this process. Tumours interact with their microenvironment. Beneficiary eight explored the distribution of intra- and peri-tumour T lymphocytes in colorectal cancer (CRC) and their importance. The GENINCA-consortium established the immune landscape in human colorectal cancer and the major hallmarks of the tumour microenvironment, which are associated with tumour progression and recurrence. A surprising finding was that the immune landscape is independent of genomic instability. To this end, Beneficiary eight identified tumour micro environment markers associated with tumour progression and characterised immune cells associated with metastatic dissemination. The definition of an immune classification with prognostic value was compared with current classification schemes and found to outperform other prognostication systems. As such, Beneficiary eight identified patients with high risk of tumour recurrence using novel immuno-staging. Furthermore, Beneficiary 14 elucidated the role of vascular endothelial growth factor (VEGF) signalling from tumour environment to tumour cells. In sum, the GENINCA consortium provides novel insights into the role of the immune system in cancer development. We established the detailed immune landscape, which has important implications for prognosis and treatment options. Furthermore, several mechanisms involved in invoking an immune response were identified. Discovery and development of novel biomarkers for tumourdiagnosis, prognosis, and disease monitoring The GENINCA consortium developed a number of novel biomarkers, which are of importance for the clinician. In brief, Beneficiary 14 employed novel approaches for the molecular imaging of tumours, using EGF-Receptor and VEGF. Beneficiary 12 developed novel biomarkers of telomere dysfunction and deoxyribonucleic acid (DNA) damage. Beneficiary 10 generated novel approaches for the proteomic identification of secreted RAS-induced senescence biomarkers. In addition Beneficiary 10 identified novel diagnostic cholangiocellular carcinomas-specific tumour markers in bile juice, which will improve diagnostic of this liver cancer entity. Beneficiaries seven and one developed and applied technologies for the detection and molecular characterisation of CTC from CRC in HCC patients and mouse models so these cells may now be amenable as predictive biomarker. Most cancer-related deaths are not caused by the primary tumour itself but by metastases. Metastases can be initiated by cancer cells, which exit from the primary tumour, intravasate into the circulation, and travel to distant organs to establish distant lesions in the lung, brain, liver, or bone. Circulating tumour cells (CTCs) are tumour cells identified by various means in transit within the blood stream. Analysis of CTCs is a challenge, as these cells are extraordinarily rare. According to estimates there is one CTC per billion normal blood cells in the circulation of patients with advanced cancer. Up to date technologies for the isolation of these rare events and their subsequent detailed analysis are limited. Despite these difficulties the GENINCA consortium performed detailed characterisations of CTC in patients with both colon and liver cancer. CTCs were enumerated using a special, FDA approved device (i.e. the Veridex or CellSearch system). Furthermore, in several patients with colorectal cancer CTCs were isolated, the DNA was amplified using special amplification protocols developed by the GENINCA-consortium, and then subjected to detailed analyses to establish copy number changes and the mutation status in a panel of genes. To this end the GENINCA consortium standardised whole-genome amplification protocols to multiply the DNA of single cells in an unbiased fashion, developed novel evaluation algorithms for the evaluation of single-cell amplification protocols for copy number alterations, and combined these approaches with next-generation sequencing technologies. Finally, Beneficiary nine developed and tested DNA-based markers for patient stratification and monitoring by blood analysis. In summary, we identified a number of predictive and prognostic biomarkers and furthermore some markers, indicating specific biological processes. Project results: The main scientific and technical (S&T) results relate to the following areas: 1. 1. cancer stem cells 2. genomic instability in precancerous lesions and exploration of underlying mechanisms of genomic alterations 3. inflammation, microenvironment and cancer 4. development of prognostic and predictive biomarkers Cancer stem cells In the Geninca project framework, Beneficiary four has gained further knowledge on Cancer Stem Cells in the initiation of tumour and, through the project collaboration scheme, has indirectly contributed to the study of genomic instability in neoplastic lesions by providing the other partners in the consortium with the biological material that was essential for their analyses to be performed. Although the role of CSC in the initiation of tumours was still debated at the beginning of the consortium activities, the experimental work by beneficiary four has contributed to the notion that CSC cultures represent one of the more advanced models for the study of colorectal cancer in a preclinical phase. Beneficiary four was able to purify and expand CSC from surgical samples derived from colorectal cancer (CRC) resections, so as to obtain a semi stable derivative that grows indefinitely in culture. This, compared to other laboratory practices, where a limited number of CSC is purified from the lesion and exhausted in the use for tests, represents a clear opportunity for more thorough analysis, and the effort undertaken by beneficiary four represents a technical advancement for the whole scientific community in that it demonstrates that CSC cultures can be obtained by CRC specimens and kept in culture. One clear objective stated in the grant agreement was to identify molecular determinants on the surface of CRC-derived CSCs, and their use as tools in oncology research. To identify the molecular determinants of a functional property, an unbiased search of the molecules that are selectively expressed in these cells but not in other, CRC-derived cell lines is performed on data gathered on the transcriptome. This is currently the widest and more reproducible amount of data obtained from biological samples such as cell cultures. We have taken advantage of the CSC cultures and compared their transcriptome to conventional CRC cells lines, and to t derivative of CSCs that has undergone differentiation and is therefore significantly depleted of stem CSC. The analysis identified very few transcripts that are expressed at high levels in CSC but not in CRC lines and CSC cultures induced to differentiate. That the molecular determinants of stemness be just a few is not unexpected, as the distinction between the stem cell and its derivatives might be very tight, and some models hypothesise that the cells of the immediate progeny are capable of switching their fate from terminal differentiation commitment to stemness. For the same reason, the amplitude of the molecular changes between the stem cell and the committed progeny might fall below the resolution of the current separation techniques. In model organisms, this issue was tackled by elegant solutions which are too elaborate to be adopted in humans, yet beneficiary four could take advantage from the advancement in the research on mouse intestinal stem cells (mISC) over the recent past. The data gathered by beneficiary four were evaluated taking into account the mISC signature, consisting of highly expressed transcripts, and used it as a criterion to identify other transcripts which behaved similarly in the human CSC-derived lines. This strategy provides by itself a valid proof of principle for the results obtained, but would not be accepted as sufficient for the achievement of the goal. Beneficiary four undertook the functional characterisation of cells isolated on the basis of putative CSC determinants, but was limited in the outcome of such analysis by the fact that the reagents (antibodies) commercially available for the recognition of putative CSC antigens were ineffective. Beneficiary four then proceeded with a parallel effort to autonomously raise antibodies against CSC and isolated several independent antibodies that bind selectively CSC but not other cells. The analysis will be extended to frozen tissue sections from matched samples of normal and tumoural tissue from CRC patients. This effort has also generated protocols for the isolation of antibodies that will be shared at the moment results are published. In cooperation with beneficiary one and 3, beneficiary four has analysed the genomic instability (GI) in CSC, coming to the conclusion that genomic instability is present in CSC, that copy number variations occur over time in CSC cultures and that chromosomal aberrations can occur independently of the GI status of the CSC, be them microsatellite instable or chromosome instable. These results are highly informative, because the GI status is often tested as a predictive factor in clinical studies, and if any differences occurred between CSC in a given tumour and the bulk tumour tissue, this might have led to an adjustment in the focus of the molecular analysis. In cooperation with beneficiary 3, beneficiary four has investigated the proficiency of CSC to mount an adequate response to DNA damage (DDR), and since this was shown to be the case, beneficiary four has used this indication to translate it into a therapeutic opportunity, by investigating whether interference with the DDR mechanisms that might be responsible for protecting CSC from the effect of chemotherapeutic regimens would lead to sensibilisation. The results obtained with CSC derived from CRC haven't been conclusive, however, beneficiary four has established that CSC derived from breast cancer are clearly impaired in their resistance to genotoxic agents by the treatment with DDR inhibitors and this finding was also confirmed in preclinical murine models of CSC-based disease. Beneficiary four has succeeded in the establishment of murine models of CSC-based disease as indicated in the Geninca agreement. The suitability of such models, obtained by injection of CSC into mice, has been assessed by the histochemical profile that closely resembles that of parental samples, histochemistry being the primary criterion of classification for tumours in current pathology procedures. This represents a clear advancement compared to tumours derived from conventional CRC lines, that form tumours with an architecture that lacks any resemblance to CRC in beneficiary four practice. For more convenient use of such tumours, CSC have been engineered genetically with minimal impact on their behavior, so as to be easily monitored without the use of invasive techniques in live animals. As the course of deadly disease in CRC mostly involves the dissemination and creation of metastasis in distant organs, and primarily in the liver, beneficiary four has endeavored with success to replicate such phenomenon and both local and metastatic models of disease were subjected to molecular analysis to identify the prominent features that distinguish these models from the in vitro CSC cultures. Reverse phosphoroteomic analysis (RPPA) of these samples has revealed that also at the molecular level the tumours obtained with CSC are more closely related to the patient CRC samples than tumours obtained from conventional CRC lines. To which extent this can be exploitable for predicting the weaknesses of the tumour has not been clarified, despite a massive effort of beneficiary four is devoted to the analysis and the verification of potentially critical mechanisms for CSC survival, proliferation and chemoresistance. Genomic instability in precancerous lesions and exploration of underlying mechanisms of genomic alterations Analysing the role of p53 dependent checkpoint response in the evolution chromosomal instability in tissue stem cells (UULM) There is growing evidence that the function of adult stem cells declines during aging. This decline is associated with an accumulation of DNA damage in stem cells. Studies in telomere dysfunctional mice have provided a 'proof of principle' that increasing DNA damage can impair the function of adult stem cells in tissue maintenance. Within the GENINCA project we have analysed the role of p53 dependent checkpoint responses to restrict stem cell maintenance and the evolution of chromosomal instabilities in response to telomere dysfunction. Our studies provided the first experimental evidence that telomere dysfunction cooperates with the loss of p53 function to induce chromosomal instability in somatic stem cells (intestinal stem cells). It is conceivable that this evolution of chromosomal instability could contribute to impairments in tissue homeostasis and an increased cancer risk in aging tissues. In a follow up project we are analysing the role of p53 upregulated modulator of apoptosis (PUMA) dependent apoptosis in stem cell and tissue aging of telomere dysfunctional mice. Our studies show that PUMA dependent apoptosis contributes to the depletion of stem cells in telomere dysfunctional tissues. PUMA deletion rescues stem and progenitor cell function and tissue homeostasis but does not lead to an instability phenotype at stem cell level. These data suggest that PUMA independent pathways contribute to the maintenance of chromosomal stability at stem cell level. In addition, our current studies revealed that p53 deletion as a single genetic element can induce liver cancer formation but not intestinal cancer formation (Rudolph et al. unpublished). The studies on liver can show that p53 deletion enhances self-renewal of bipotent liver progenitor cells. Moreover, it facilitates the de-differentiation of hepatocytes into tumour forming progenitor cells. Genomic instability in circulating and disseminated tumour cells (UKE) For the detection of epidermal growth factor receptor (EGFR) expression with the CellSearch system, a fluorescein-labeled anti-EGFR-antibody was used by Veridex, Raritan, New Jersey, United States of America (USA). With the help of cell line cells with known EGFR gene status Beneficiary seven established an immunoscoring for EGFR expression. Using this immunostaining, CTC could be categorised as EGFR-negative, weakly, moderately or strongly EGFR-positive, however, only a minority of CTC presented with a strong EGFR expression. This result has to be validated in further large clinical trials to find out whether EGFR expression levels of CTC are somehow related to the response of CRC patients to EGFR-directed therapies. To find out more about the molecular features of the detected CTC and whether different individual CTC of one patient show common or varying genetic changes such as gene mutations or chromosomal aberrations, a technique allowing the analysis of whole genome amplified DNA obtained from a single cell was established. Using this technique, single CTC isolated from colon cancer patients were screened for mutations in a panel of four genes (p53, KRAS, BRAF, PIK3CA) frequently aberrant in primary tumours of CRC patients. While the p53 status is interesting to confirm the malignant origin of CTC, the presence or absence of mutations in the KRAS, BRAF and PIK3CA gene is highly relevant for therapeutical decisions. Only patients whose primary tumours posses normally functioning forms of these three genes, can profit from an anti-EGFR immunotherapy. Currently, treatment decisions only rely on the analysis of the primary tumours for KRAS mutations, but differences concerning the KRAS status between primary tumours and corresponding metastases have been described. Since metastases are rarely surgically removed, CTC have the potential to become useful as 'liquid biopsy'. Moreover, it is of high interest whether there is heterogeneity between individual CTC from the same patient regarding these genetic alterations. Within the GENINCA project, Beneficiary seven detected several mutations in all analysed genes on single CTC from CRC patients, revealing a quite significant heterogeneity between CTC of the same patient. The comparison of these results to the mutational analysis of the corresponding primary tumours performed by Beneficiary one will provide valuable information, which will be especially useful in view of the development of more individualised therapies. As Beneficiary one demonstrated, arrayCGH profiles from individual CTC of the same patients show both common genomic aberrations as well as cell-specific differences in certain genomic regions that will be further investigated in detail. Losses of 5q (probably involving at least parts of the APC gene) or gains in 8q (c-Myc) were frequently observed alterations. Chromosomal and genomic instability in liver cancer (MUG) Hepatocellular carcinoma (HCC) is the fifth most common cancer in men and the eighth most common in women worldwide. The incidence of HCC is increasing in developed countries and is one of the growing major causes of cancer-related death. HCC is a tumour with poor prognosis and with only few treatment options. The development of HCC is a multistep process. HCC arises most frequently in the setting of chronic liver inflammation and fibrosis due to viral infection, metabolic injury, toxic insults, or autoimmune reactions. Knowledge about molecular events in early stage HCC development is limited because of difficulties in the histomorphologic distinction between non-malignant nodular lesions from early HCC. As it is more or less impossible to obtain sufficient amounts of early stage liver tumours or even precancerous lesions from humans, animal models facilitate the study of different stages of hepatocarcinogenesis. One of the most frequently used mouse HCC models is the diethylnitrosamine (DEN) treatment of mice. The molecular signature found in tumours resulting from DEN exposure reflects the situation of human HCC with poor prognosis. Mdr2-knockout (Mdr2-KO) mice represent a model of inflammation-associated HCC. Homozygous disruption of the multidrug resistance two (Mdr2) gene, which encodes for a P-glycoprotein and is responsible for the transport of phosphatidylcholine into the bile, leads to the development of cholestatic hepatitis followed by dysplasia, dysplastic nodules, and carcinoma. Thus, this model represents a prototype of inflammation-associated cancer which is also common in humans. Therefore, both mouse HCC models have respective counterparts in humans so that results obtained with these mouse models can be transferred to the human situation. The GENINCA consortium has succeeded in establishing a detailed order of genetic changes within the HCC tumourigenesis from very early to advanced stages. Such a detailed knowledge about events especially in the early HCC development has not existed previously. Specifically we elucidated the role of two genes in HCC carcinogenesis. The gene b-catenin (CTNNB1) was thought to be critically involved in causing chromosomal instability. The GENINCA consortium could confirm frequent b-catenin mutations during HCC tumourigenesis, however, these were unrelated to the chromosomal stability. Thus, other mechanisms must be the driving force behind the chromosomal instability in this tumour entity. Furthermore, we identified a gene which is frequently up-regulated already in early stage disease and which often continues to be highly expressed throughout the disease course. We elucidated the function of this gene as critical for increasing the proliferation rate of hepatocytes. This gene is not only up-regulated in the mouse models studied by the GENINCA-consortium but also in humans. This may evolve to novel strategies for therapeutic interventions, as an inhibition of this gene should diminish the growth of HCCs. Mechanisms for chromosomal instability in colon precancerous lesions (MUG) Genomic instability is a hallmark of cancer. Yet, the mechanisms giving rise to pathogenic genomic alterations and their evolution are not fully understood and have not been explored in detail. The GENINCA-consortium studied the consequences of hyper-replication after Ras oncogene induction in normal human cells. Without enforcement of a DNA-damage checkpoint response oncogene induced replication stress resulted in genome-wide alterations in a random, non-clonal fashion associated with losses of tumour-suppressor genes and gains of oncogenes. The GENINCA consortium established detailed maps of sites in the human genome which are preferentially affected by re-replication stress. Inflammation, microenvironment and cancer How the inflammation can support tumour growth (UKER) The immune system is generally considered to be important to prevent tumour development. Individual tumour cells are detected and removed by the host immune system. In contrast, some cancers develop on an inflammatory background, suggesting that under certain circumstances, immune cells or immune cell derived molecules can promote rather than prevent tumour development. One such example is colon cancer which usually develops sporadically. However, patients suffering from ulcerative colitis, a chronic inflammatory bowel disease, have a high risk to develop colon cancer. The risk increases with the duration of the inflammatory disorder and the degree of inflammation in the colon. The molecular mechanisms driving colitis-associated colon cancer are currently unknown. Knowledge of these mechanisms would increase our understanding of the underlying causes of cancer and would allow developing novel therapeutic strategies. Beneficiary 14 reasoned that comparing the gene expression between colitis-associated and sporadic colorectal cancer would allow identifying inflammation related factors driving tumour development in the gut. Beneficiary 14 has profiled gene expression patterns from the APCmin and AOM/Dextran sodium sulphate tumour models, two widely accepted mouse models of sporadic or respectively colitis-associated colon cancer. Experimental tumour growth was monitored by mini-endoscopy, and tumours and tumour-free colon were probed using Affymetrix gene chip analysis. Beneficiary 14 compared the genes differentially expressed in colitis associated colon cancer with those differentially expressed in sporadic colon cancer. The data demonstrated partly overlapping gene expression patterns, indicating that many cellular pathways were activated in both tumour entities. Markers of cell proliferation and tissue remodeling ranked among the most upregulated transcripts in both tumour types. In agreement with the general dedifferentiation associated with tumour development, many genes involved in cell differentiation were markedly reduced in both models. Interestingly, Beneficiary 14 discovered more than 600 genes that were differentially expressed in colitis associated, but not sporadic tumours arguing for important molecular differences between colitis associated and sporadic colorectal cancer. Among these genes, the gene coding for the protein Epiregulin (Ereg), a member of the epidermal growth factor (EGF) family was highly expressed especially in colon tumours developing in mice with colitis. The members of the EGF family are best known for their ability to stimulate cell growth and proliferation and are important for many developmental processes and in cancer development. Beneficiary 14 characterised Epiregulin expression and found a strong correlation with colonic inflammation. Interestingly, mice genetically deficient for EREG displayed increased susceptibility to experimental colitis, but were highly protected from tumour development in the AOM/DSS model of colitis-associated cancer as evaluated by endoscopy. The number of tumours developing in epiregulin deficient mice was not only smaller, also the sise of tumours was significantly reduced, suggesting that epiregulin may promote tumour cell growth. The epiregulin effect was inhibited by blocking activation of the transcription factor signal transducers and activators of transcription three (STAT3). Thus, beneficiary 14 discovered a novel mechanism promoting the growth of colitis associated tumours via Epiregulin and STAT activation. Beneficiary 14 has identified further cellular pathways of communication between the microenvironment and colon tumour cells. Alterations of the expression of several cytokines and growth factor and their receptors including TGF-beta, IL-6, IL-17, IL-22 and VEGFR2 were detected in the course of tumour development and in precancerous lesions. Beneficiary 14 made a surprising discovery involving vascular endothelial growth factor (VEGF), a signalling protein that stimulates vasculogenesis and angiogenesis. Solid cancers can only grow beyond a certain sise if adequately supplied by blood vessels. VEGF is therefore frequently overexpressed in cancers. Beneficiary 14 could demonstrate that also epithelial cells in tumour tissue expressed the receptor for VEGF (VEGFR2) and responded to VEGF stimulation with augmented proliferation. Blockade of VEGF function via soluble decoy receptors suppressed tumour development, inhibited tumour angiogenesis, and blocked tumour cell proliferation. Functional studies revealed that chronic inflammation leads to an up-regulation of the expression of VEGFR2 on intestinal epithelial cells. Studies in STAT3 mutant mice showed that VEGFR signalling requires STAT3 to promote epithelial cell proliferation and tumour growth in vivo. Thus, VEGFR-signalling acts as a direct growth factor for tumour cells in colitis-associated cancer, providing a molecular link between inflammation and the development of colon cancer. Beneficiary 14 also demonstrated that colonic inflammation is associated with the induction of STAT3 activity in intestinal epithelial cells (IECs) and high constitutive activity of STAT3 was detected in tumours arising from inflammation. In these studies, genetically engineered mice showed epithelial STAT3 activation in acute DSS colitis dependent on the cytokine interleukin-22 (IL-22). Mice with an IEC-specific deletion of STAT3 activity were highly susceptible to experimental DSS-induced colitis, indicating that epithelial STAT3 regulates gut homeostasis already before the development of colon tumours. STAT3 deficient mice, upon treatment with DSS, showed a striking defect of epithelial restitution. Gene chip analysis indicated that STAT3 regulates the cellular stress response, apoptosis, and pathways associated with wound healing in IECs. In summary, the data of beneficiary 14 demonstrate that STAT3 activation can be induced in the microenvironment by multiple inflammation-associated molecules and that STAT3 is a major mediator between immune cells and tumour cells. The immunologic landscape of cancer Immunohistochemical analyses was then performed for a subset of adaptive immune markers (CD3, CD8 granzyme B, and CD45RO) using tissue microarrays prepared from cores of the center of the tumour and the invasive margin of more than 500 colorectal cancer patients (three independent series). A statistically significant correlation between immune cell density and patient outcome was observed. We showed that the natural adaptive immune reaction in each tumour region influenced clinical outcome at all stages of the disease, and that the immune reaction at the tumour site determined the cancer evolution and clinical outcome regardless of the local extent and spread of the tumour. Thus, the adaptive immune reaction and the intratumoural T-cell subpopulations were a better predictor of survival than the traditional staging based on the cancer sise and spread. We analysed the functional immune coordination, and the cytotoxic T cell markers in relation to T helper subpopulations (Th1, Th2, Th17 and Treg). Four major groups of patients according to immune status and outcome were defined. Patients with high expression of the Th17 cluster had a poor prognosis whereas patients with high expression of the Th1 cluster had a prolonged disease-free survival. The combined analysis of cytotoxic and Th17 clusters gave a better discrimination for relapse. We demonstrated the prognostic value of using immune criteria in patients with early stage CRC (UICC TNM stage I-II CRC) to predict survival and relapse. These results underline the clinical relevance of this investigation, which could profoundly improve the accuracy of prognosis and guide therapeutic decision making. The Journal of Clinical Oncology has published a two-page editorial entitled 'TNM Staging in Colorectal Cancer: T is for T cell and M is for Memory' on our last publication to comment on the major prognostic significance of the immune score that we propose. Thus, our data revealed: 1. identification of tumour micro environment markers associated with tumour progression 2. characterisation of immune cells associated with metastatic dissemination 3. definition of an immune classification with prognostic value above current classification 4. identification of patients with high risk of tumour recurrence using novel immuno-staging. Development of prognostic and predictive biomarkers Development of novel markers for prognosis and therapeutic monitoring in colon cancer using confocal laser endomicroscopy (UKER) Confocal laser endomicroscopy (CLE) is a newly introduced endoscopic tool allowing to carry out confocal microscopic examination of the mucosal layer during ongoing endoscopy. CLE has revolutionised endoscopy. Various types of tissue and several diseases can be diagnosed immediately, facilitating early diagnosis of colon cancer. In addition, analysis of the tissue architecture is useful in targeting biopsies to relevant areas. While currently, unspecific fluorescent dyes are used to stain the surface of the colon and improve the contrast, research groups including beneficiary 14 envision molecular imaging of specific tumour markers to either discriminate tumour tissue from non-tumourous tissue, or to predict the response to biological therapies. Experiments performed by Beneficiary 14 provided promising results concerning in vivo molecular imaging of VEGF and EGFR in surgical specimens of patients with CRC and in tumours in mice. Beneficiary 14 evaluated the ability to use CLE in vivo for instantaneous molecular imaging of EGFR in experimental colorectal cancer models. Tumours were visualised in vivo with a handheld CLE probe after injection of fluorescently labeled EGFR antibodies. Neoplastic and non-neoplastic specimens from human colorectal mucosa were examined. In vivo findings were correlated with histopathology, immunohistochemistry, and fluorescence microscopy analyses. Importantly, CLE analysis of EGFR expression in human specimens allowed distinction of neoplastic from non-neoplastic tissues. These findings confirmed that CLE can be used for in vivo, molecular analysis of CRC and to differentiate EGFR expression patterns in human tissue samples. Because CLE can be performed during endoscopy, in vivo molecular imaging might be used in diagnosis of CRC and to predict response to targeted therapies. In addition to EGFR, VEGF is a therapeutic target in colorectal cancer. Molecular imaging of VEGF on tumours in experimental colon cancer models and in surgical specimens of patients with colorectal cancer was achieved after application of fluorescence labelled antibodies detecting VEGF. CLE enabled the cytoplasmatic distribution of VEGF to be displayed due to its subcellular resolution. In human tissue, a VEGF-specific signal was observed in 12/13 malignant specimens and in 10/11 samples from healthy mucosa from the patients. CLE findings correlated well with ex vivo microscopy. Collectively, the findings of Beneficiary 14 demonstrate that confocal endomicroscopy allows specific targeting of markers enriched in tumours and can contribute to the identification of lesions at risk and potentially predict response to targeted treatment. Identification and clinical analysis of protein markers of telomere dysfunction The accumulation of DNA damage represents one of the hallmarks of human aging and it is associated with an increased cancer risk. Telomere shortening represents one cell-intrinsic mechanism contributing to the accumulation of DNA damage in aging cells. Within the GENINCA consortium we have identified a set of biomarkers of telomere dysfunction and DNA damage that can be measured in human blood serum. Interestingly, the biomarker levels increase during human aging and in disease stages that are associated with an increased cancer risk such as liver cirrhosis and myelodysplastic syndromes (MDS). Moreover, lifestyle factors that are associated with an increased cancer risk also associate with increased levels of the biomarkers. Together our studies indicate that biomarkers of telomere dysfunction and DNA damage are associated with human aging and an increased cancer risk. We have initiated clinical studies to analyse the expression level of biomarkers of telomere dysfunction and DNA damage in cancer, specifically focusing on the question whether increased biomarkers can predict an increased cancer risk. Within the GENINCA project we have also analysed whether telomerase mutations would associate with the development of cirrhosis - a precancerous disease stages - in patients with chronic liver disease. The telomerase enzyme complex can synthesise telomere de novo. In our studies we sequenced two essential genetic components of the telomerase enzyme (TERT and TERC) in 521 cirrhosis patients and 600 control patients. The study revealed the first experimental evidence that a subset of cirrhosis patients carries mutations in the telomerase enzyme that lead to a reduction in telomere synthesising activity. These results support the concept that defects in telomere maintenance can contribute to the progression of chronic liver disease towards cirrhosis development and an increased cancer risk. Proteomic identification of secreted RAS-induced senescence biomarkers Early malignant processes in tumour formation are associated with accumulation of DNA damages. As a consequence, DNA repair processes are activated resulting in replicative aging of cells (cellular senescence) and loss of replication. Ultimately, this applies a formidable pressure on cells to inactivate DNA damage checkpoint genes and loss of replication has been identified as a cancer promoting factor. Once cells undergo senescence they significantly alter secretion of proteins. To identify novel secreted senescence indicating proteins (biomarkers) we cultivated human cells in cell culture. Cells were forced to initiate senescence. Then the secreted proteins were isolated from the cell culture supernatant and analysed utilising high resolution proteome analysis. This analysis identified fragments of Vimentin, Collagen alpha-2(I) and Lysyl oxidase as novel senescence biomarkers. Lysyl oxidase is a secreted enzyme that catalyses formation of cross links in collagen and elastin, major components of the extracellular part of tissue. These two proteins provide structural support to the cells (extracellular matrix). Lysyl oxidase has reported links to cellular senescence and tumour suppression. Interestingly, we identified a fragment of lysyl oxidase and of its substrate collagen as biomarker candidates. Vimentin is another structural protein. In essence Vimentin is responsible for maintaining cell shape, integrity of the cytoplasm, and stabilising cytoskeletal interactions. As such Vimentin is used as a marker for tissue tumours, so called sarcomas. Identification of novel diagnostic CC-specific tumour markers in bile juice Tumours of the bile duct (cholangiocellular carcinomas, CC) show dramatically increasing rate of new cases per year (incidence) and are particularly associated with poor survival prognosis. Usually patients die six to 12 months after diagnosis. The early differentiation between malignant and benign biliary diseases, in particular CC in patients with chronic gall bladder inflammation (primary sclerosing cholangitis, PSC), is very difficult and may come too late to give the patient a therapeutic benefit. Therefore, we applied proteomic analysis of bile to distinguish cholangiocarcinoma from non-malignant lesions, such as PSC and bile stones (choledocholithiasis). To define novel CC specific proteins, we in total analysed 50 samples from patients suffering from choledocholithiasis (n=16), PSC (n=18) and CC (n=16) patients. This resulted in two different biomarker profiles. A first profile allowed differentiation of PSC and CC from bile stones and a second profile the subsequent differentiation of PSC from CC. For biomarker validation the two profiles were applied to an independent test cohort (choledocholithiasis n=8, PSC n=10 and CC n=13). Validation resulted in correct identification of 7/8 patients with choledocholithiasis (88%) and 20/23 patients with PSC or CCC (87%). Subsequently, 12/13 patients with CC (92%) were correctly distinguished from 7/10 patients with PSC (70%). The majority of sequence-identified biomarkers were fragments of hemoglobin alpha and beta chains, followed by peptides of serum albumin, pancreatic triacylglycerol lipase and cytoplasmic actin 1. Other peptides were assigned to structural proteins, i.e. keratins, histones, proteins involved in proteolysis and degradation of lipids and polysaccharides, i.e. proteasome subunits, carboxypeptidases, trypsin, alpha-amylase, bile salt-activated lipase, as well as proteins involved in immune responses, i.e. complement factors, immunoglobulins. Circulating and Disseminated Tumour Cells as Biomarker Tumour cells are able to detach from the primary tumour early in cancer progression and to get access to the lymphatic and/or blood system. A small proportion of these disseminated tumour cells (DTC) detectable in bone marrow aspirations or CTC determined in the peripheral blood of cancer patients seem to be responsible for the development of distant metastasis in cancer patients. The incidence and also the number of CTC found in a distinct blood volume of 7.5 mL represent a valuable indicator for the prognosis of patients with metastatic cancer. Furthermore, detailed analysis of the molecular characteristics of CTC not only contributes to a better understanding of the mechanisms underlying metastasis in general, but also might offer new options for therapeutical interventions. The detection of these extremely rare cells, with as few as one CTC in millions of blood cells, is challenging and requires highly sensitive methods. Currently, there are different techniques for the enrichment and detection of CTC available, with some of which validated in clinical trials. Thus far, only one standardised automated device is commercially available, the CellSearch system that has been cleared by the US Food and Drug Administration for CTC analysis of patients with metastatic breast, prostate and colorectal cancer. This device captures CTC with high sensitivity by an antibody that binds to EpCAM - an epithelial adhesion molecule that is present on the surface of most carcinoma cells, but absent on blood cells. This immunomagnetic capturing sometimes lacks specificity, but specific detection of CTC is subsequently ensured by immunostaining with anti-cytokeratin antibodies and to exclude leukocytes with anti-CD45 immunostaining. For patients with metastatic colorectal cancer (mCRC), prognostic relevance of three and more CTC in 7.5 mL blood was shown for overall and progression-free survival, however less is known for patients with primary non-metastatic CRC. Results of Beneficiary seven within the GENINCA project demonstrate that CTC also are detectable in a considerable number of patients with non-metastatic CRC (20/84, 24%) and in 61% of 66 patients with mCRC. Whether CTC in non-metastatic patients have prognostic relevance as well will be investigated by long term follow up observation of the patients. Analysing blood samples from the mesenteric vein resulted in a significantly higher frequency of CTC-positivity and higher CTC numbers per patient compared to peripheral blood in non-metastatic CRC patients. Collection of follow up data of these patients will show whether CTC detected in the mesenteric vein are related to the high frequency of liver metastases in CRC patients. Within the GENINCA project, Beneficiary seven established different methods to detect CTC in blood and bone marrow samples from immunosuppressed and transgenic mouse models. Characterising CTC in mouse models is a promising tool for testing genomic instability of cancer cells and new tumour cell-targeting therapies. To analyse whether there is a relation of intratumoural and peripheral immune response to tumour cell dissemination, Beneficiary eight analysed paraffin-embedded CRC tissue from CRC patients for the number of CD45RO-positive memory T cells as well as for the intensity of immunostaining. The infiltration of the tumour with these immune cells is connected with a good prognosis for the patients as could be shown by Beneficiary eight in former experiments. CD45RO-positive cells were counted intratumourally and at the invasive margins. Evaluation of immunostainings revealed considerable differences for both the T cell counts as well as for the intensity of immunostainings between individual cases; however, no clear correlation to the presence of CTC was obtained thus far, probably due to the still low number of cases. Further investigation of this interesting topic is planned. Serum-DNA as Biomarker Diagnosis of cancer and diagnosis of recurrence has a major impact on therapy outcome. Based on clinical observations we know that the earlier a disease is diagnosed the more successful are subsequent medical interventions. Therefore it is of utmost importance to identify biomarkers which allow disease diagnosis as early as possible. Second in clinical day to day practice as well as during drug development we would like know to as early as possible if a drug is effective. Meaning that tumour growth is blocked or even more desirable the tumour shrinks and gets killed by the drug. Very likely biomarkers that monitor tumour response will have the power to improve patient treatment by enabling the oncologist to switch earlier to effective treatment options and improve patient survival. Besides patient management, the analysis of tumour response and prediction of treatment efficacy are important information during clinical development of new drugs. Drug efficacy and dose-response correlations are very critical for successful drug development. Therefore so called surrogate markers are of urgent need to further improve clinical drug development. There are multiple attempts to address and improve tumour diagnosis and tumour response detection. One approach implies the identification of tumour derived DNA in the blood circulation. Normally DNA is covered inside cells, mainly in the nucleus of the cell and, to a lesser extent, in the mitochondria. The hallmark of cancer is its genetic instability, which is characterised by a huge number of DNA aberrations, as for example DNA mutation, gene amplification, gene loss, and gene translocation and fusion. The observation that cell-free DNA can be found in blood circulation is known for decades. More recently several technological improvements made it possible to identify genetic aberrations in the very limited amount of cell-free DNA typically present in blood. Many but not all tumour associated DNA aberrations or DNA modifications are tumour specific and allow tumour diagnosis in blood samples. However, tumour derived DNA is an indicator for tumour presence and furthermore its amount correlates with tumour mass, meaning a further increase or decrease is expected to mirror resistance or response to treatment. Potential impact: Cancer causes a substantial number of deaths in the EU and it is a complex socio-economic issue. The societal aspects are highlighted by the fact that incidence and mortality rates are higher in the lower socio-economical classes of the European society. The economic aspect centres on the fact that the population is aging in the Western world, concomitant with increasing rates of cancer. Our studies have received great attention in the scientific community, specifically in the areas of age-related cancer, stem cell aging, stem cells as cell-type of origin of cancer, checkpoints in cancer. The studies were published in highly recognised international journals including Nature Genetics and Hepatology. In addition, the study results were presented at renowned international conferences including the Cold Spring Harbor meetings on Aging and Stem Cells and the Cologne Spring Meeting, among others. Dissemination beyond the Consortium At the beginning of the project a logo and an overall project design has been created, which is used together with the Grant Agreement number HEALTH-F2-2007-202230 and the Seventh Framework Programme (FP7) logo/EU flag, on any printed and electronic issues for the public and for any other official contact. In general, GENINCA dissemination aimed at putting the project topics in the bigger pictures of cancer in Europe. The main goal was to show how little alterations in our genes might be a small problem on the molecular level, but a huge burden for the society when viewed on a European level. All dissemination activities therefore invite the audience to follow the track of cancer impact from the level of genes to Europe. Public awareness of GENINCA has been addressed by the website http://www.geninca.eu The content is regularly checked and updated and includes key background information about the project, its partners and the increasing burden cancer poses for the European society. The website is available in two languages (English, German), while basic project information is available in multilingual fact sheets providing project facts at a glance. Special emphasis has been placed on media-rich presentation of the content. Wide usage has been made of animation, audio and video material as well as picture galleries. Video material is also available on YouTube (see http://www.youtube.com/user/GENINCA online) and vimeo. The news-section provides the visitor with the latest publications, press releases and other dissemination activities of GENINCA. Additional audience was targeted at special events. GENINCA was presented to the public at large during the European Researchers' Night 2009 in Vienna. A scanimated poster visualised the project content, but also gave information about liver and colon cancer in general. In June 2011, GENINCA launched a dissemination activity unique for research projects: a jump&run browser game (http://game.geninca.eu/). As of 2008, more than 100 million people in the EU consider themselves as active computer gamers. While these people spend several hours per week gaming, they are difficult to reach by other means of communication. The GENINCA office tried to fill this gap in communication by creating this free online game. Following the concept of tracking the impact of cancer on different levels, the game takes the player from a European level to the genes. An intro video makes the player familiar with this concept. Sticky notes the player passes in the different stages tell him/her more about the project, the impact of cancer as well as the burden it poses for the European society. The Consortium also takes measures to provide GENINCA exposure to the public through conventional public relations (PR) media work, which will complement the public access website, particularly for those that do not have internet access. Posters and folders giving striking facts about the impact of cancer were created to be distributed by the partners. When a foil attached to this folder is moved over the images they are animated. This effect is called 'scanimation' and aims at filling the gap between animated and printed material. The scanimated GENINCA poster was presented by Beneficiary 14 /UKER during the Inflammation and Cancer Summer School in Crete, which was organised by the PF7-project INFLA-CARE. Beneficiary one (Medical University of Graz) published two press releases during the funding period, one at the beginning of the project and another at month 18. As the team of Beneficiary one consists mainly of human geneticists, Beneficiary one participated regularly in educational workshops for the public, e.g. to teach about hereditary forms of cancer. In these workshops Beneficiary one also referred to the on-going GENINCA research efforts. Two more press releases were published by Beneficiary10 (mosaiques) and Beneficiary 14 (Universitätsklinikum Erlangen). The scientific community has a vested interest in keeping up with GENINCA progress and results, to build on the body of knowledge that currently exists. This was achieved through conventional scientific dissemination means, by publishing more than 30 publications in international journals and by attending more than 20 scientific conferences. Communication within the project Consortium To foster communication and interactions within the Consortium, Beneficiary 11 (biolution) established a web-based internal communication platform. Constant scientific exchange was also encouraged by organising bi-annual project meetings for the partners. Project website: http://www.geninca.eu Related documents Final Report - GENINCA (Genomic instability and genomic alterations in pre-cancerous lesions and/or cancer)
