Abnormal expression of tyrosine kinase (tk) genes is a common phenomenon in papillary thyroid carcinomas (PTCs), where it is believed to alter cell growth and response to external signals such as growth factors, hormones etc. While the pathogenesis of radiation-induced PTC remains unclear, there is evidence that tk genes such as the receptor tyrosine kinases ret and NTRK-1 are abnormally expressed, and that the overexpression of some tk genes due to gene amplification or changes in gene regulation in the absence of structural alterations may lead to oncogenic transformation of cells. Using a DNA microarray based technique, we have identified several tk genes with abnormal expression in human tumor cell lines. We now propose to apply the technique to measure the relative expression levels of more than 50 tk genes in the PTC's that arose after the Chernobyl nuclear accident and to compare these expression profiles with the gene expression pattern found in sporadic PTC cases and tumors that arose following low level therapeutic irradiation of the thyroid. Results from this study may allow the identification of molecular markers that can be used to facilitate tumor diagnosis and staging, and, eventually, provide targets for therapeutic intervention.

The objective of this study is to investigate the link between exposure of children to radiation, the subsequent development of tumors and how their morphology, molecular and cell biology influence clinical outcome. The project is an integrated approach involving 5 leading European centers. Samples of the same tumors will be studied by the 5 different centers to determibe tumor morphology and type; the degree of variation within the tumor, including the variation of the proportion of cells in cycle using antibodies to novel DNA replication associated peptides; the gene involved in the carcinogenic process, using DNA chip technology; specific studies of the pathways associated with one oncogene (ret) known to be linked to the tumor type involved; and studies of novel gene rearrangements using FISH technology. By using the same tumor/normal pairs in these studies, integrating the results from the different centers and studies and correlating these with detailed morphological analysis and patient details including evidence on tumor aggressiveness and recurrence, we will increase our understanding of the link between radiation exposure and cancer development, and provide evidence which will inform decisions on radiation protection and on clinical management of patients with radiation associated cancers.

The papillary thyroid carcinoma oncogene (RET/PTC) is a rearranged version of the tyrosine kinase RET. It is know that the incidence of RET/PTC activation is increased in radiation-induced papillary thyroid carcinomas compared to papillary thyroid carcinomas without a radiation history. The prognostic value of RET/PTC rearrangements and its importance as a radiation-specific marker is still unclear. It is proposed to screen 70 childhood cases from Belarus and Ukraine with interphase FISH in conjunction with RT-PCR to confirm the presence and type of the chimeric transcripts. Cases with indications for atypical RET/PTC rearrangements will be further investigated using 5'RACE for the presence of novel types of alteration. For this part of the proposal RNA samples as well as a limited number of paraffin-embedded sections would be needed. The RNA samples will be further investigated for expression profiles of other tyrosine kinase genes to identify other gene rearrangements which may occur as a sole abnormality or in addition to RET/PTC rearrangements.

Mittochondrial DNA (mtDNA) is a likely hotspot for mutation in cancer as it is preferentially modified by many carcinogens. We have previously shown that there is a specific association between sequence variants of Complex I genes and ATPase6, one of the two mitochondrial genes of Complex V, and the occurrence of malignancy and of oxyphil features in thyroid tumors. We have also found a significant association between mitochondrial sequence variants and the occurrence (and degree) of the so-called mitochondrial common deletion. In an attempt to elucidate the role of post-Chernobyl irradiation in mtDNA alterations and to find out whether or not such alterations are involved in the etiopathogenesis of thyroid tumors we will search for the mtDNA common deletion and for somatic mutations and sequence variants in the D-loop region, in the 13 coding genes and in the 22 tRNAs genes of cases from which there are RNA and DNA samples extracted from blood (set without radiation), normal thyroid (set irradiation) and tumors (set irradiation and tumorigenesis). In a first step, we will study exhaustively 20 cases. The results obtained in this first part will be used together with the data we and others have previously obtained to decide the most appropriate targets for the second part of the study. If possible we would like to correlate the results of our study on mtDNA deletions and mutations with those on ret oncogene. In case there are also clinico-pathyological data available, we would like to collaborate with the pathologists who have studied the cases in order to clarify the putative clinical significance of the mtDNA alterations.

Genetic alterations of the ret proto-oncogene play a critical role in the pathogenesis of papillary thyroid carcinomas both naturally occurring and radiation induced. We have recently found that classical RET/PTC rearrangements are present also in benign thyroid nodular disease. Furthermore, in several cases the independent expression of tyrosine kinase (TK) and extracellular (EC) domains of RET was found. This finding may be interpreted as RET wild type gene expression. In other cases, especially radiation exposed, the TK domain in the absence of EC was found and was interpreted as unknown RET/PTC rearrangements (PTCX). Aim of this project is to clarify the modality of expression of RET protooncogene in nodular thyroid diseases. In cases with EC and TK expression, we will search for RET wild type by an extralong PCR encompassing the TK and EC domains, followed by sequencing of the PCR product. In cases of TK positive expression only (but not classical PTC1, PTC2 and PTC3) we will identify the 5' domain rearranged with the TK domain. As for other unknown RET/PTC rearrangements will use the 5'race approach.

All samples positive for TK expression will be submitted to quantitative PCR (ABI Prism 7700 sequence detector, Perkin Elmer) for TK mRNA. The TK mRNA expression will be correlated with the histological characteristics of the analysed tissues, and the immunohistochemical pattern of RET protein expression, using an antibody to recognize the TK domain.

The expression of RET/PTC oncogenes in thyroid PC Cl 3 cells induces a complex phenotype with a block of the differentiation program, hormone-independent proliferation and increased apoptotic rate. Oligonucleotide GeneChips were used to analyse gene expression profiles of PC Cl 3 cells expressing either RET/PTC1 or RET/PTC3 oncogenes in comparison to parental cells. About 2,000 genes showing at least two-fold increase and 2,000 genes showing at least two-fold decrease were identified in RET/PTC-expressing thyrocytes. Virtually all the genes up-regulated by more than 5-folds (about 100) were confirmed by RT-PCR and some of them by immunoblot. Genes upregulated by RET/PTC could be functionally divided in genes involved in proliferation (such as D-type cyclins), apoptosis, proteolysis, inflammation, and metabolism. We plan to extend these studies on the identified genes up and on the downregulated by analyzing their expression in human thyroid tumors of different histotypes. We propose to study their expression by semiquantitative reverse transcriptase PCR and for selected genes by real time quantitative PCR. Immunohistochemical analysis will be use to verify protein expression.

The findings of these studies can reveal clues to the molecular pathways involved in papillary thyroid carcinoma and may provide biomarkers for clinical use.

Papillary thyroid carcinoma (PTC) accounts for 80% of all thyroid malignancies. It has a variable disease course and to date no pathways or specific genes have been implicated as causative in this tumor. RET activation, through translocations involving several genes, have been noted in a high incidence of PTCs. However, the activation of this oncogene is found at all stages from benign through well-differentiated to undifferentiated carcinoma. This suggests that it represents an early event and that this defect is not in itself sufficient for carcinogenesis. It may also suggest that the classification of PTC covers more than one tumor subtype. The relationship between radiation exposure and PTC is well established. We plan to perform a genome wide scan using microarray analysis for alterations in tumors from children exposed to radiation and compare them to sporadic adult tumors to identify which genes are commonly altered and which genes are display differential alteration expression patterns. We plan to extend this study using high-resolution BAC-CGH to define a molecular pattern for these tumors and evaluate this approach for diagnostic applications. The study of molecular alterations which cause thyroid carcinoma is of importance since the identification of causative factors could lead to new approaches for treatment.

Development of papillary thyroid cancer (PTC), similarly to that of most of other human malignancies, is likely to comprise a multistep and multihit process. It is quite probable that mutational events initiating, promoting and/or driving the tumor progression are quite similar in the sporadic and radiation-induced PTC. Along with this, one may expect there may be unidentified to date molecular distinctive features peculiar to thyroid cancers of different etiology. Thus, a comparative study of various molecular characteristics in the two groups of PTC may provide additional information for the determination of the molecular signature of radiation-induced thyroid cancerogenesis.

In the proposed project we intend to study the following molecular characteristics of the DNA extractedfrom normal and tumor tissue of radiation-induced PTCs: i) relative content of mtDNA and number of large-scale deletions in mtDNA; ii) prevalence of gene mutations of MAPK signal molecules, including the Ras,BRAF, Raf-1 and MEK genes; and iii) distribution of the codon 72 allelic variants of theTP53.

After the data are obtained, we will perform a comprehensive univariate and multivariate statistical analysis against already available results of examination of sporadic PTC in order to identify molecular parameter(s) specific to radiation-induced PTC.

This study investigates the occurrence and molecular characteristics of thyroid cancer in residents of the Bryansk Oblast of the Russian Federation, who were 0-50 years of age at the time of exposure to radiation from the Chernobyl Power Station accident (ATA) on April 26, 1986. The study has three primary purposes: 1) to characterize cases of thyroid cancer according to specific molecular markers of genetic change, and investigate whether the presence of such markers is associated with individual thyroid radiation dose from the Chernobyl accident; 2) to investigate whether age-at-exposure dependent radiation dose response for thyroid cancer differs between cancers that are positive versus negative for the molecular markers investigated; and 3) to investigate whether the presence of these same molecular markers is associated with clinical outcomes. Included will be thyroid cancer cases diagnosed between April 1, 2001 and March 31, 2006 and confirmed by a panel of expert thyroid pathologists. An equal number of controls will be individually matched to cases by sex, age, type of settlement and raion of residence on April 26, 1986. Data collected will include in-person interviews for all participants, and for cases only, paraffin embedded tissue or fresh frozen tissue, clinical history and outcome information.

Thyroid cancers are the most common endocrine malignancies and the vast majority of them are papillary thyroid cancers (PTC). Several genetic abnormalities, including Ras mutations and RET/PTC rearrangements have been well characterized in these cancers. Recently, we and several other groups have reported the BRAF mutation in PTC with a high prevalence. We have also characterized aberrant DNA methylation in several genes in thyroid cancers, including novel tumor suppressor genes and some thyroid-specific genes. Except for the RET/PCT rearrangements, these genetic and epigenetic abnormalities have been studied mainly in sporadic thyroid cancers, and their role is unknown in the special group of thyroid cancers induced by radiation, the most common and well-established environmental risk factor for thyroid cancer. Chernobyl nuclear accident has been associated with a significant increase in the incidence of PTC, which represent an ideal thyroid tumor model for the study of radiation-induced thyroid tumorigenesis. In the present project, we propose to use such special thyroid cancer samples to study novel genetic and epigenetic abnormalities, their relationship, and their effects on the expression of key thyroid genes. Well-established experimental protocols and techniques, including RT-PCR, methylation-specific PCR, real-time quantitative PCR, and a recently established colorimetric mutation detection method will be used. The study is expected to result in important insights into radiation-induced thyroid tumorigenesis and provide novel clinical implications for this special group of thyroid cancers.

Thyroid follicular carcinomas frequently exhibit RAS mutation, and closely resemble benign follicular adenomas with respect to morphology and differentiation. Cell culture studies suggest that at least one requirement for progression from an adenoma to a carcinoma is failure of an intrinsic mechanism that normally limits the proliferative lifespan of RAS-induced cell clones. One current candidate for over-riding the mechanism is the tumor suppressor gene p16 INK4a.

There is great clinical interest in this area, as currently there is no marker to distinguish between thyroid follicular adenoma and carcinoma when evaluating fine needle aspirates and biopsies of thyroid glands. This means that many people have unnecessary operations on the basis of presumed malignancy. We now therefore wish to carry out a comprehensive comparison between thyroid carcinoma and adenoma cells to identify differences that may confer an extended proliferative lifespan on carcinoma cells.

Chernobyl thyroid cancers represent a unique resource in oncology and radiation biology. They appear in a cohort of patients irradiated at the same time and in which there is no doubt that the cancer originated from radiation exposure. The precise timing of the course allows to follow precisely the kinetics of appearance of the cancers. On the other hand study of the gene expression pattern of cancers by microarrays allows a precise molecular definition of each cancer. Using this methodology we were able to show that the clustering of gene patterns of Chernobyl cancers of the first wave and European and US sporadic cases could not separate them. This work should now be extended to a larger series of cases.

1. to try to distinguish subtypes of Chernobyl and sporadic papillary carcinomas and their signature.
2. to relate patterns of gene expression with clinical variables such as the duration of the incubation period and with the genetic diagnosis.
3. to validate data at the RNA level by PCR and at the protein level by Western and to define potential diagnostic markers and therapeutic targets.

High copy-number gene amplification is known to take place at a few genomic loci in numerous human cancers, but widespread low-level copy-number changes in genomic DNA have not been described. cDNA microarray-based comparative genome hybridization yields high-resolution copy-number profiles that enable the detection of low-level amplification events at individual gene loci. We have shown in a small pilot study that pediatric thyroid carcinoma in residents of a region contaminated by 131I from the Chernobyl accident (the Bryansk Oblast) exhibits gene amplification at a higher frequency than that seen in pediatric thyroid carcinoma in US children with no history of radiation exposure. This result suggests that exposure to ionizing radiation from the environment may be associated with an increased rate of gene amplification in a human cancer. The consistent amplification of many genes among cases of post-Chernobyl thyroid carcinoma from Bryansk suggests the existence of a target pool of radiation-sensitive genomic loci that respond to exposure by initiating local amplification events. The pattern of gene amplification may represent a radiation signature that could be used to map amplicons likely to harbor participating oncogenes.

Based upon these differences observed in the pilot study in apparent gene amplification between post-Chernobyl and spontaneous pediatric papillary thyroid carcinoma (PTC), we hypothesize that radiation exposure leaves a measurable genomic signature in the form of stable changes in gene amplification. Some chromosomal regions identified by this method are likely to harbor participating oncogenes, but it is unreasonable to expect that so many genes be directly involved in oncogenesis. We hypothesize the presence of genomic hot spots in human DNA that are susceptible to radiation-induced amplification. These genomic targets are unlikely to be saturated by the doses of radiation delivered to these patients. The number of targets hit in each case should therefore be directly proportional to radiation dose.

Genetic damage following radiation exposure is subject to correction by the caretaker systems of DNA repair. Our interest is in the role that these systems may play in the molecular pathogenesis of cancer. We propose a pilot project to assess the influence of genetic variation on cancer risk following exposure to ionising radiation. This will be achieved by investigation of variation in the genes involved in the DNA repair pathways, in DNA derived from blood samples or normal tissue samples from patients with thyroid tumors of radiation-associated and non-radiation-associated etiology. There are two main justifications for such a study, first to identify SNPs which indicate possession of an at risk genotype, and secondly, to identify the genes in which genetic variation is a significant modulator of cancer risk. This has particular and wider relevance to the involvement of DNA repair-associated factors in the pathology of other cancers as well, such as breast cancer.

Genotyping of genes involved in double-strand break repair will be performed using a mass spectrometry-based SNPing platform which has been developed by our group at the University of Wales, Swansea. Data will be provided to the Chernobyl Tissue Bank for correlation with pathology and expression of oncogenes such as ret and BRAF.

It is proposed to study chromosomal imbalances in post-Chernobyl papillary thyroid carcinomas (PTC) by means of array CGH using 1Mb BAC arrays. As derived from interphase FISH experiments RET/PTC rearrangements are heterogeneously distributed within tumor tissues leading to the assumption that additional gene alterations may play an important role in these tumors. To address this question post-Chernobyl PTC, with and without RET/PTC rearrangements, will be analysed by array CGH. Altered candidate genes will be derived from recurrent regions of amplifications and deletions and will be confirmed by interphase FISH on paraffin sections and further studies by PCR-based approaches to investigate expression of these genes. In a first pilot study it is intended to compare 10 RET/PTC-positive (RET/PTC3) and 10 RET/PTC-negative cases with similar histological features, a comparable age range of patients at time of exposure and a similar latency after exposure is a first study. The pilot study will use cases from the age-matched series with known RET/PTC status. If successful this study will be extended to a larger series of cases linked to the GENRISK-T project.

The aim of this study is to investigate whether different transcriptomic profiles can be related to exposure to radioiodine in fallout from the Chernobyl accident and to lower level radiocaesium exposure present in the contaminated environment. We will use Affymetrix microarray technology to define transcriptomic profiles in two cohorts of children, matched on age, oblast and pathological type of tumor. The research will be carried out in two separate laboratories and cross-validated. This project is one of a series of projects that will study the transcriptomic and genetic profile of two well-defined cohorts to investigate the relative effects of radioiodine and radiocaesium exposure on the development of thyroid cancer.

Gene expression has received less attention than the role of germline polymorphisms or somatic mutations in studies of radiation and thyroid cancer. The increase in papillary thyroid cancers (PTC) in exposed children following the Chernobyl nuclear accident presents an opportunity to pursue the role of gene expression further. Recently, we reported on expression of seven genes, each of which was able to distinguish post-Chernobyl PTCs from sporadic PTCs (Port et al 2007). Our approach involved (i) a whole genome microarray used for screening purposes; and (ii) quantitative examination of the 92 target genes with a high throughput RTQ-PCR technique (LDA). The study had some limitations such as the origin of sporadic PTCs (Eastern Germany), their different age at diagnosis and a small number of cases (n=11). Subsequently another group has reported on expression of thirteen genes involved in homologous recombination suggesting a distinct radiation pattern of post-Chernobyl PTCs (Detours et al 2007). Using the already established 2-stage design, the purpose of the present application is to overcome limitations in the previous study and extend the findings (Port et al 1007) by evaluating a dose-dependent gene expression pattern in 74 post-Chernobyl PTCs with individual I-131 dose estimates.

Radiation exposure is a well established risk factor for thyroid cancer. Ionising radiation is known to cause extensive DNA damage including double strand breaks, which may lead to the generation of somatic mutations in thyroid cells and cancer initiation. However, environmental triggers cannot fully explain the inter-patient heterogeneity in the individual response to exposure to radiation, which points to the existence of genetic variations that define the individual susceptibility to radiation-related cancer. We propose to analyse several candidate DNA repair genes and perform a genome-wide analysis of single nucleotide polymorphisms (SNPs) in Ukrainian patients who developed thyroid cancer after Chernobyl and in control cancer-free individuals to identify mutations and SNPs that are involved in genetic predisposition and to identify the genes that are affected. We will then perform functional analysis to find whether these genetic variations alter gene function. We will also study the link between specific SNPs and known or new mutations found in these tumors. The overall goal of this study is to identify genes involved in genetic predisposition to radiation-associated thyroid carcinogenesis and a pattern of SNPs that can detect it.

Cancer of the non-medullary (follicular epithelium) component of the thyroid is induced by external irradiation and by radionuclides deposited within the thyroid tissue. Estimates of the radiological risk of developing thyroid cancer are derived from epidemiological studies performed in. Populations receiving high doses where, according to Ron et al, the threshold in 100 mSv. Extrapolation of this risk to exposures at much lower doses is compromised by the lack of an accurate model of the dose response curve for thyroid cancer at low doses. Moreover, such population based estimates fail to take into account the contribution of individual genetic variability to the risk estimate. Individuals with an increased genetic predisposition to develop thyroid cancer are not identified, and it is precisely these individuals who will be at greatest risk at low doses. The GENRISK-T consortium is composed of thyroid cancer experts with experience in the fields of radiation biology, animal models of radiation-induced cancer, tumor banking, cancer biology, molecular genetics, histopathology, cyto genetics and risk modelling. We will use this interdisciplinary knowledge to define the genetic component influencing the risk of radiation-induced thyroid cancer. This will be achieved through a combination of studies using animal models and in human radiation-induced thyroid tumors. This new understanding of the genetic risk modifiers will be used to develop an animal model of thyroid cancer that is responsive to low dose radiation in the cGy range, thereby providing an experimental solution to resolving the uncertainties of the low dose-response curve. This EC collaborative project (PI Professor M Atkinson, Helmholtz Zentrum, Munich) combines the use of animal models and human studies. This particular application is to support the investigation of human germline SNPs that may predispose to radioiodine induced thyroid cancer in those exposed as children and adolescents.

The aim of the study is to investigate transcriptomic profiles of follicular thyroid tumors (malignant and benign) that arose after expose to radioiodine fallout from Chernobyl power station and to low level radiocesium exposure present in the contaminated environment. Affimetrix microarray technology will be employed to define transcriptomic profiles in two cohorts of children matched on age, oblast and pathological type of tumor. This project is the continuation of a series of projects that are carry on to investigate the transcriptomic and genetic profile of radiation induced thyroid cancer.

Follicular thyroid cancers are less frequent than papillary thyroid carcinomas (PTC), however, they are associated with a poorer survival outcome than PTC. Although several genetic changes have been identified so far, the molecular genetic mechanisms of tumor development in follicular thyroid neoplasms are still unclear.

To investigate novel gene alterations and potential radiation signatures in follicular thyroid adenomas (FA) and follicular thyroid carcinomas (FTC) it is proposed to investigate genome-wide copy number changes of 100 thyroid tissue samples by array CGH using 1 Mb BAC arrays. For this purpose we want to compare genomic profiles of tumors (FA and FTC) developed pre- and post-fallout of the Chernobyl accident. The proposed study aims to identify gene alterations in follicular thyroid neoplasms from altered genomic regions and to determine aberrations patterns that correlate with the radiation history of patients as well as with any of the clinical phenotypes of the tumors.

The aim of the study is to seek for differences in transcriptomic profiles of childhood papillary thyroid cancers that arose after radiation exposure from Chernobyl power station fallout and sporadic cancers. As the first step Affimetrix microarray technology was employed to define transcriptomic profiles in two cohorts of children matched on age, oblast and pathological type of tumor. The second step of the study is to validate microarray results with Q-PCR analysis. We also plan to extent the analysis to exon microarray study which allow to detect transcript isoforms, chromosomal deletions and amplifications.

Post-Chernobyl pediatric thyroid cancers are associated with a high frequency of recombination events leading to the generation of fusion oncogenes, resulting in aberrant expression and activation of RET, and less frequently of NTRK and BRAF. Altogether, ~60% of PTC arising in this patient population harbor one of these abnormalities. The discovery of novel somatic rearrangements using conventional methods has low sensitivity and/or resolution. We propose to use genome-wide parallel paired-end sequencing to identify somatic rearrangements in childhood thyroid cancers induced by radiation, as compared to age-matched thyroid cancers without radiation exposure. Altogether we will select 6 samples of each population: 2 with a known rearrangement in RET, NTRK or BRAF (as positive controls), and 4 without. We will construct 3kb insert Illumina libraries from each tumor DNA sample, and ~35bp of sequence from both ends of each fragment will be obtained. Each end will be mapped back to the reference genome. Fragments for which the ends do not map back within 3kb of each other and/or are in inappropriate orientation will be further studied as they may represent rearrangements present in the thyroid cancer genome. We will acquire ~1 fold genome coverage (~3Gb) from each sample. This will approximate to 30-fold physical coverage, allowing detection of essentially all rearrangements present in the dominant clone of the cancer. Rearrangements will be processed using a suite of informatics tools to predict which may generate an in-frame fusion gene. RNA from each of the samples will then be tested by exon-exon PCR to determine whether the fusion is expressed, and based on its predicted function, whether it may correspond to a driver mutation. The number of rearrangements in each cancer and their architecture will then be compared between the two classes. Sequences at the rearrangement junctions will also be compared particularly to examine the complexity of the rearrangement, sequence contexts of breaks, presence of repeats and overlapping microhomology of the rearrangement. Each of these indices may provide clues to the way large radiation doses induce DNA double strand breaks and how they are repaired.

MicroRNAs (miRNAs) are 21-23 nucleotide long non-coding RNA molecules that have been shown to regulate the stability or translational efficiency of target messenger RNAs. Dysregulation of miRNAs has been implicated in a variety of cancers, and in the thyroid germline SNPs in miRNA binding sites and in the coding sequence for miRNAs themselves, have been implicated in Papillary Thyroid carcinogenesis. The Human Cancer Studies Group is part of an EC sponsored consortium (Genrisk-T) that has intensively studied papillary carcinomas from a series of 100 patients, and is currently extending this approach to follicular tumors. Half of this group were exposed to radiation and half were born after 1/1/87 and not exposed to radioiodine in fallout. The cohort is carefully age and sex matched. The Genrisk-T project has provided data on RNA expression using Affymetrix technology, bac array CGH data on copy number variation in the tumors and SNP array data from normal tissue from these patients. We now seek to add miRNA data from this cohort and to correlate data on SNPs and copy number variation in the tumor, with changes in miRNA level, and miRNA expression levels with changes in RNA expression levels. The combination of this data will give us a thorough understanding of the regulation of a number of different growth control pathways involved in carcinogenesis of the thyroid follicular cell and their relationship to radiation exposure.

We are a small biotech company developing statistical and bioinformatics methodology that would allow for the direct integration and co-analyses of RNA-seq and microarray data. As quantitative RNA-seq data is becoming the method of choice for gene expression analyses, we are developing the statistical and bioinformatics technology which that allow all previous published microarray data to be incorporated with new gene expression platform data, to ultimately provide more comprehensive data-sets for downstream applications/analyses.

We have been working with Dr C Maenhauts group to develop the methodology to integrate her published Affymetrix gene expression data with our own Illumina RNA-seq data from papillary thyroid cancers. We propose now to subject surplus material from the samples that Dr Maenhaut received from the CTB (project 002/2007) to RNA-seq in order to not only obtain additional expression data over and above that obtained by microarray but also to further validate our methodologies.

The EU funded project EpiRadBio project seeks to model the cancer of the lung, breast and the thyroid after exposure to radiation in the low-dose range (cumulative dose < 100 mGy). The formalin-fixed paraffin-embedded (FFPE) papillary thyroid cancer tissue sections we apply for will be used in a work package of EpiRadBio that focuses on the cancer risk of papillary thyroid cancer. In another, recently finished EU funded project on young onset childhood thyroid carcinomas that also used CTB material we found that a gain of the chromosome band 7q11 was associated with exposure to low-dose ionising radiation (He et al., PNAS, in press). The FFPE tumor sections from patients, which are part of the UkrAm cohort and which come with estimates on radiation dose the patients have received will be used for validation and further characterisation of this marker. The FFPE sections will be used for in situ hybridisation (FISH) using 7q11 specific probes. DNA and total RNA will also be extracted from the FFPE sections. The DNA will be used for high-resolution array CGH and the total RNA for qRT-PCR mRNA expression analysis of candidate genes from the gained region. The copy number data and qRT-PCR data will be integrated with the dose estimates in order to identify a potential relationship between the radiation-associated biomarker and radiation dose. The resulting data will be provided to the modellers of the project who will use this information to build and refine their models on radiation risk of papillary thyroid cancer after exposure to low-dose ionising radiation.

The constituted network to realise this program involved two teams well known in diagnosis and treatment of thyroid tumors and one team which were already implicated in the search of radiation-induced signatures in the thyroid and in the identification of the molecular mechanisms of radiation-induced tumorigenesis. To date, we focused our approach at the transcriptomic level. We wish now to analyse miRNA and mRNA deregulations in a series of post-Chernobyl thyroid tumors by microarray analysis: 1) to identify a radiation-induced miRNA and mRNA signatures. We will assess the robustness of these signatures for a potential use as a diagnostic tool alone or in combination. 2) to obtain an integrated miRNA/mRNA overview of radiation-induced tumorigenesis by taking advantage of the deregulated pathway identified by the transcriptomic approach. To date, no such integrated analysis has been performed as most of the studies focused on either transcriptome analysis or miRNA analysis separately. Ultimately, we will cross the obtained data with others results from analysis of post-radiotherapy tumors (Ory et al. 2011; Ugolin et al. PLosONE under revision).

The project using the applied biomaterial is an extension of the EU-funded GENRISK-T project and aims to validate a gain on chromosome band 7q11 that was found to be associated with papillary thyroid carcinomas from young patients (< 19 years) that were exposed to radiation from the Chernobyl fallout at very young age (median: 2 years). The validation set containing exposed and unexposed cases will be matched for age, morphology and residence - these criteria were already used for case selection of the GENRISK-T set. A high-definition array CGH platform will be used to validate the gain and to type for additional radiation-markers that are smaller in size and were therefore not detectable by 1Mb BAC array CGH that was used in GENRISK-T. Further, we will use RNA samples and paraffin sections for characterisation of expression candidate genes and proteins by qRT-PCR and immunohistochemistry. Fresh-frozen tissue from selected samples expressing the candidate gene CLIP2 from the gained region on 7q11 and those that do not express the gene will be analysed using a whole proteomics approach (liquid chromatography-tandem mass spectrometry, LC-MS/MS). The results will be used to identify the dysregulation networks in which CLIP2 is specifically involved. The projects aims to validate the radiation marker on chromosome band 7q11, to identify the networks that are dysregulated in tumors harbouring the marker and to find new markers that are associated with exposure to radiation.

Cancer is a genetic disease, a concept which has been consistently observed for all tumors. Our approach is to survey the entire genome of these tumors to look for mutations and other perturbations that are involved in radiation-induced papillary thyroid carcinoma (RI-PTC) and create a genomic profile of this tumor. A new technology, termed Next-generation sequencing allows sequencing of the entire human genome in 8 days. We will begin by sequencing the DNA of 50 individual RI-PTC tumors from Chernobyl pediatric patients to look for mutations in genes. We will correlate this data with RNA sequencing from the same patient samples. This data will provide information about events that are taking place at the level of gene expression, providing information about over and under expressed genes and the expression of aberrant genes, and gene fusion products. This data will be compared to RIP-PTC in age matched patients who were not exposed to radiation to develop a radiation-specific profile. We have bioinformatic specialists in the group who will integrate these various kinds of data. This study will provide a comprehensive profile of RIP-PTC.

Within the scope of the EpiRadBio-project, Mark van de Wiel, Carel Peeters and Wessel van Wieringen from the department of Epidemiology & Biostatistics of the VU University Medical Center are responsible for integrative analysis of the molecular data from (radiation) exposed and non-exposed thyroid cancer samples. Integrative analysis combines heterogeneous biological data, be it experimental (e.g., copy number, gene expression, microRNA expression) or from the biological literature (e.g., gene annotation, pathway information). Integration of data from multiple sources is imperative for a mechanistic understanding of cancer. By putting together partial views of a complex process like tumorgenesis, we may obtain a more accurate and complete picture of the molecular mechanisms underlying it. No off-the-shelf methodology for the statistical analysis of the data is available. We therefore aim to develop statistical models from the experimental data of these biological processes. Such models will enhance our ability to identify biomarkers and therapeutic targets more unambiguously and to interpret genotype information. This should enhance the understanding of what distinguishes the exposed from the non-exposed thyroid cancers. In addition, the biological insight these models provide is likely to result in more targeted follow-up experiments and efficient use of available resources.

The aim of the study is to investigate transcriptomic profiles of papillary thyroid carcinomas that arose after expose to radioiodine fallout from Chernobyl power station and to low level radiocesium exposure present in the contaminated environment. This project is the continuation of a series of projects that are carry on to investigate the transcriptomic and genetic profile of radiation induced thyroid cancer. As a first step of the study Affimetrix microarray technology was employed to define transcriptomic profiles in two cohorts of children matched on age, oblast and pathological type of tumor. At present we want to validate the results with qPCR.

It has long been accepted that cancers arise by the sequential accumulation of errors in the DNA of a clone of cells within a given tissue. This in part explains their long latency. However, recent studies using whole genome sequencing have suggested a different mechanism for some cancers, particularly those that arise in children. This mechanism is called chromothripsis and involves shattering of chromosomes and then restitching them together. This results in multiple small changes in the DNA that occur simultaneously rather than sequentially. The aim of this study is to use whole genome sequencing to identify novel mutations in radiation induced thyroid cancer, to identify changes in the mitochondrial genome and to investigate the frequency of chromothripsis in childhood thyroid cancer, especially with respect to radiation exposure.

Convincing epidemiologic data has established a strong causative association between ionizing radiation and the development of papillary thyroid cancer (PTC). This finding was due, in large part, to studying those affected by the Chernobyl accident. More recently other solid tumors, such as lung and breast, have been associated with radiation. Medullary thyroid cancer (MTC), a rarer but more lethal thyroid cancer than PTC, has not been associated with radiation exposure to date. However, there appears to be a higher number of MTC cases than would be expected in the Chernobyl population. We would like to determine if there is a link between radiation and MTC by studying this population to look for mutations that are associated with radiation exposure. Additionally, because MTC is a hereditary condition in 25% of cases, we would also like to assess whether there could be a hereditary cause to MTC in this small geographic region. Such data will give us an insight into the mechanisms in which tumors are formed in MTC and potentially help us target ways to stop these tumors.

Previous analysis of data derived from the Chernobyl accident has shown strong correlation between absorbed doses of IR and the induction of papillary thyroid cancer (PTC). Specific genetic alterations, such as rearrangements of the RET oncogene, are linked to previous exposure to radioiodine. Recently, rearrangements of the anaplastic lymphoma kinase (ALK) gene have been found to be selectively expressed in papillary thyroid cancer (PTC) amongst atomic bomb survivors (ABS), but not in PTC patients lacking radiation exposure. In PTC, ALK rearrangements have been shown to be associated with tumor aggressiveness and, importantly, represent a possible pharmacologicaltarget for the compound Crizotinib. Interestingly, radiation-induced PTCs that show ALK rearrangements lack additional genetic alterations that are frequently found in sporadic thyroid cancer, such as RET, NTRK1, BRAF, or RAS; these findings underline the oncogenic potential of ALK rearrangements in radiation-induced PTC. We plan to investigate a possible correlation between radiation exposure and ALK rearrangements in PTC biospecimens from patients exposed to radioiodine in the context of the Chernobyl accident as well as in a control cohort of sporadic PTCs, performing fluorescence in situ hybridization (FISH) analyses. Furthermore, we intend to assess the mutation status of commonly altered oncogenes in PTC, such as BRAF or RAS, using pyrosequencing.

New advances in genomic characterization technologies afford new opportunities to comprehensively investigate the genetic basis of the established association between childhood exposure to iodine-131 (I-131) from the Chernobyl Nuclear Power Plant accident and risk of thyroid cancer. The opportunity to analyze a large set of thyroid cancers and normal tissue/blood from the Chernobyl Tissue Bank (CTB) could further our understanding of the molecular mechanisms of radiation carcinogenesis in humans based on careful evaluation of epidemiological risk factors and genomic alterations. Previously, we had reported a preliminary study of dose-related alterations in gene expression and chromosomal translocations in approximately 70 cases drawn from the Ukrainian-American (UkrAm) cohort component of the CTB biorepository. Recently, we have completed a pilot study in which we have conducted a comprehensive genomic characterization of 12 UkrAm cases, including paired fresh frozen and formalin-fixed paraffin-embedded tissue samples using the infrastructure and analytical pipeline of the Cancer Genome Atlas (TCGA) of the U.S.A. National Cancer Institute (NCI). This includes whole genome sequence analysis of RNA, DNA and micro-RNAs as well SNP and methylation microarrays of the tumor tissue. SNP microarray and DNA analysis was performed on peripheral blood for comparison of germline to somatic alterations. The success of the feasibility study establishes a strong scientific basis for extending this approach to a substantially larger study of thyroid cancer cases drawn from the CTB. Here, we propose conducting a comprehensive genomic characterization study of 500 papillary thyroid cancers (PTC) from the Ukraine using biological samples, information on radiation exposure, and demographic characteristics. The main objective of the study is to provide comprehensive, integrated characterization of the genomic, transcriptomic, and epigenomic landscapes of radiation-related PTC for comparison across a spectrum of I-131 exposure as well as a comparison with approximately 500 sporadic PTCs available from The Cancer Genome Atlas recently published in Cell. The proposed study has the potential to provide unique insights into the mechanisms of radiation carcinogenesis and to generate a rich data resource accessible to other investigators.

In this study we plan to analyze samples from patients with papillary thyroid cancer. We will examine primary tumors and serum samples in 40 pediatric patients (aged ≤ 18 years in 1986), and in 40 young adults (18-25 years in 1986). To determine a possible correlation between the presence of mutations/fusion in liquid biopsy and the extent of metastatic spread we plan to examine samples from patients presenting with lymph node metastases only and from patients with distant metastases.

We will first identify driver mutations in the primary tumor by performing Ion Torrent Oncomine Pan-Cancer assay analysis using DNA/RNA extracted from tumor. Next, we will employ the same approach for analysis of DNA/RNA extracted from the serum sample of the same patient. We expect to show that in patients with metastatic thyroid cancers, results of NGS analysis in serum will recapitulate results of NGS analysis in primary tumors. We also plan to perform quantitative assessment of genomic abnormalities using the digital PCR technique. To do so, we will examine the presence of specific driver mutation(s) in primary tumor by dPCR, and determine the mutant/wild copy numbers in DNA/RNA samples extracted from the whole tumor. We also plan to complete the assessment of primary tumors through the detection of oncogenic proteins by immunohistochemistry, and by evaluation of mutant/wild copy numbers after performing laser captured microdissection. Serum from the same patient will be then examined by dPCR.