CHIBCHA

SICA:

Collaborative Project (Small or medium-scale focused research project)

Research topic addressed:

HEALTH-2007-2.4.1-14: Studying cancer aetiology in Latin America

Project Co-ordinator:

Prof Ian Tomlinson, University of Oxford

Project Manager:

Dr Luis Carvajal-Carmona, University of Oxford

1. To collect well characterised series of CRC cases and controls in various LA countries. High-quality sample collection in large numbers is essential to any genetic project to identify and characterise common alleles that have individually modest, but collectively important, effects on disease risk. Our consortium is extremely well placed to undertake this collection, with considerable experience of large-scale sample collection in practice and of all the major relevant clinical disciplines (CRC Surgery, Gastroenterology, Clinical Genetics, Oncology). Standardised procedures for collection/processing of samples and data collection are already in place.
2. To identify high-risk (Mendelian) CRC genes in LA cohorts. The contribution of the major single-gene syndromes to CRC in LA is not well known. We shall undertake this testing locally. Expertise will be exported from Europe to LA should this be required.
3. To test European low-penetrance CRC SNPs in LA. Based on an initial GWAS comprising four phases in the UK, we have identified three SNPs associated with CRC in the UK population. Using a larger European collection, we have identified about three more such SNPs. These SNPs are highly likely to contribute to the risk of CRC in LA. Their contributions will be determined in CRC case-control series from different LA countries. This work will be undertaken locally, with advice on SNP typing provided from Europe if necessary.
4. To develop a new ancestry informative marker (AIM) panel. A new marker set incorporating whole-genome SNP typing will be developed.
5. To develop methods for GWASs in admixed populations. New methods to stratify for admixture in such screens will be developed, for example using principal components analysis.
6. To map new CRC genes using a GWAS: Our Europe-based GWAS is highly unlikely to have detected all variants associated with CRC. We shall use a multi-phase association study in LA populations, followed by phases in European populations, to screen for additional CRC-associated SNPs. Our European study did not analyse CNVs and these will be analysed in this project. The same analysis may also identify polymorphisms that influence the presenting features of the disease or patient outcome/survival.
7. To determine the locations of new CRC genes using admixture mapping. There are several very good reasons to undertake both a GWAS and AMS in LA populations, not least because the costs of performing an additional AMS are small once SNPs have been typed for a GWAS. Each method is, moreover, better suited to detecting disease alleles under different scenarios. We shall perform a phased AMS in the LA CRC series, followed by fine mapping. Any SNPs/CNVs that appear to have specific effects in the LA population, but not Europe, will form the basis of collaborative studies with other groups who have access to Amerindian and/or admixed Hispanic and/or Asian populations.
8. To identify functional polymorphisms associated with CRC in LA and Europe. The work above will lead to the identification of polymorphisms associated with CRC. The identification of the functional variation that causes increased CRC risk may lead to the development of new means of preventing the disease. However, the journey from the associated polymorphism to functional variant is not simple and is likely to require work following on from this project. In order to perform an initial search for readily identifiable functional variants ("low-hanging fruit"), we shall undertake some fine mapping in the regions close to each disease-associated SNP, enriching for variants in strong LD with that SNP and with possible functional effects. We shall also perform basic functional work, such as assessing the levels of mRNAs and proteins in each region in patients of known genotypes.
9. To develop a method for allele-specific copy number estimation. Both large-and small-scale genotyping platforms currently lack this important feature. We shall develop methods to do so.
10. To develop SNP sets for population-based predictive tests. The data derived from this study will be used to propose panels of SNPs/CNVs (not necessarily functional) that can be used in LA and Europe for population screening to predict CRC risk.
11. To provide the basis for testing CRC genes in other cancers. Some SNPs increase the risk of several cancers. We shall approach collaborators who have large collections of other cancer types, so that they can test any CRC SNPs in their data sets.
12. To train LA researchers and disseminate knowledge within LA and Europe. Exchange of knowledge by Europe-based laboratory attachments for LA researchers, expert lectures and training in methods or use of equipment is an important part of the programme.
13. To raise the profile of CRC in LA and thereby to improve diagnosis and outcome. In LA, the study will be used as a vehicle to raise awareness of CRC, including symptoms, importance of early diagnosis and increased familial risk. This will provide immediate benefits to the LA population.

1. Screening for Mendelian CRC syndromes in LA. Patients or families with CRC will be screened for germline mutations in the Mendelian CRC genes. Cases may be selected for screening based on clinic-pathological features and/or family history, or using results from ancillary molecular tests such as immunohistochemistry performed on tumours.
2. Role of existing low-penetrance CRC SNPs in LA. We shall examine the strengths of association between the 10 existing European CRC-associated SNPs and disease in the LA populations. We shall optionally investigate any differences between Europe and LA in terms of local LD and haplotype blocks, potentially deriving useful fine mapping data for identifying the causative SNP in each region.
3. A search for new CRC genes. The principal aim of the study is to undertake screens to identify new alleles/loci that are associated with CRC predisposition. This will take the form of a genome-wide association study in LA (including development of methods for stratifying by individual admixture) and an admixture mapping study in LA.
4. A new AIM panel. We shall develop an AIM panel based on the LA population using high-density SNP arrays.
5. New methods of CNV analysis. These will be developed by applying new algorithms to SNP typing output data.
6. Training and dissemination. The project will involve training of researchers and promoting awareness of CRC in the general population, healthcare professionals and government, with emphasis on LA.

1. Management.

   Leader:

   Ian Tomlinson, Luis Carvajal-Carmona, University of Oxford, UK

   Aims:

   1. Co-ordination of consortium activities
   2. Ensuring that agreed scientific directions are followed
   3. Monitoring and review of results and progress
   4. Resolution of disputes
   5. Optimisation of sample and data flows
   6. Reporting and financial management

2. Recruitment, sample collection and resource building.

   Leader:

   Maria Margdalena Echeverry, Universidad del Tolima, Colombia

   Aims:

   1. Identification and sampling of the agreed number of CRC cases (to include those with high-risk adenomas) and controls from each centre
   2. CRC and CRA samples from histopathology archives where possible
   3. DNA extracted, quantitated and QC'd for each case and control
   4. Arraying and storage of DNA resource for genotyping and future studies

3. Identification of CRC/CRA cases with known syndromes and effects of known CRC SNPs in LA populations.

   Leader:

   Augusto Rojas Martinez, Universidad de Nuevo Leon, Mexico

   Aims:

   1. Determination of the frequency of germline MMR, APC and MYH mutations as contributors to CRC susceptibility in LA.
   2. Transfer of medium-throughput SNP genotyping technologies to LA as required.
   3. Determination of the frequencies and effects of CRC predisposition SNPs identified in Europe in LA.

4. Genome-wide association study

   Leader:

   Angel Carracedo, Universidad de Santiago de Compostela, Spain

   Aims:

   1. Provision of high-quality genotypes (including copy number estimation) from all cases and controls in multi-phase design
   2. Analysis and assessment of data after each phase to test for strengths of association
   3. Identification of those SNP and CNVs that are associated with increased risk of CRC

5. Admixture mapping study

   Leader:

   Ian Tomlinson, Luis Carvajal-Carmona, University of Oxford, UK

   Aims:

   1. To undertake rapid, high-quality genotyping for an AMS screen, unless already done for the GWAS
   2. Analysis and assessment of data after each phase to test for strengths of admixture association
   3. Identification of those regions containing SNP and CNVs that are associated with increased risk of CRC

6. Database, sample reception, data exchange, data storage and statistics

   Leader:

   Chris Holmes, University of Oxford, UK

   Aims:

   1. Setting up database and computing facilities to receive and process data
   2. Analysis of data from known European CRC SNPs in LA series
   3. Establishment of updated AIM panel from Colombia
   4. New method for GWA analysis in admixed populations
   5. Analysis of AMS and GWAS data
   6. Testing for GGIs and risks associated with multi-locus genotypes

7. Functional studies

   Leader:

   Ian Tomlinson, Luis Carvajal-Carmona, University of Oxford, UK

   Aims:

   1. To determine whether different genotypes at SNPs/CNVs associated with CRC cause basic differences in gene function

8. Dissemination, training and education

   Leader:

   Ian Tomlinson, Luis Carvajal-Carmona, University of Oxford, UK

   Aims:

   1. To report the scientific results of the project in a high profile manner for clinicians, scientists and the public
   2. To utilise the project to raise the public, professional and political awareness of CRC in LA
   3. To provide training and experience in human molecular genetics for young LA scientists
   4. To flag and, where possible, exploit IP arising from the project
   5. To identify a panel of genetic variants that could be used in clinical practice for predicting CRC risk in Europe and LA

Dr Luis Carvajal-Carmona, CHIBCHA Project Manager
The Wellcome Trust Centre for Human Genetics
University of Oxford
luis@well.ox.ac.uk
http://www.well.ox.ac.uk/~luis

Professor Ian Tomlinson, CHIBCHA Project Co-ordinator
The Wellcome Trust Centre for Human Genetics
University of Oxford
iant@well.ox.ac.uk
http://www.well.ox.ac.uk/tomlinson