Harmonising Biobank Reseach: maximising value - maximising use
25-27 March 2009
Brussels, Belgium

Organisers:

Jennifer Harris, The Norwegian Institute of Public Health, Oslo, Norway

Draft Report

Summary

Many countries have made substantial investments in biobanking and intense activity has been targeted at building a harmonised network of biobanks to support large-scale genomics. Several of these countries envision that biobanks will become part of their medical research and healthcare infrastructure and biomedicine will benefit greatly from building synergies between diverse sources of health information. A range of funded initiatives are serving to create a biobanking infrastructure, the scope of these projects spans a wide array of designs, including population-based and special populations, disease-specific, tissue banks and those with specific ELSI focus. The collective output
from these projects includes harmonisation tools and technologies related to the design and management of biobanks, development of standard operating procedures (SOPs) for sample handling, cataloguing and comparing information, coordinating development of compatible bioinformatics, and mapping ethico-legal frameworks. An important consequence of harmonisation activities is that a new reservoir of knowledge, experience, and expertise is emerging that is crucial to share with the biobanking community at large in order to maximise scientific value and use of biomolecular resources.

This international conference, ‘Harmonising Biobank Research: Maximising Value-Maximising Use’ (1) provided a forum for much-needed information exchange centred on the grand theme of biobank harmonisation to maximise use. The conference was held on March 25th-27th in Brussels, Belgium, and was co-organised by three interrelated initiatives working in tandem to further biobank science and research: Promoting Harmonisation of Epidemiological Biobanks in Europe (PHOEBE) (2), Public Population Project in Genomics, (P3G) (3) and Biobanking and Biomolecular Resources Research
Infrastructure (BBMRI) (4). Additional funding was provided by the European Science Foundation, PreAnalytiX, Qiagen, and GenVault.

The conference was organised around four keynote presentations and eight sessions which focused on topics necessary for moving the international agenda forward. The final session, structured around an expert panel discussion, identified critical next steps for helping to realise this. Approximately 250 people from over 35 countries attended.

1. Presentations available at http://www.phoebe-eu.org

2. PHOEBE is funded by the European Commission under the 6th Framework
Programme

3. P3G is funded by Genome Canada and Genome Quebec

4. BBMRI is funded by the European Commission under the 7th Framework Programme

Scientific Content

The scientific content and discussions centred around 8 main thematic areas: 1) phentoype harmonisation, 2) making the most of biobank data via the internet, 3) isolated populations, 4) statistical analyses, 5) ethics, recontact and reconsent, 6) models of data sharing and how they affect science, 7) practical experiences with clinical biobanks and 8) building an international biobanking community.
Presentations from these sessions are available at http://www.phoebe-eu.org.

Phenotype Harmonisation
There is good consensus that harmonisation is among the most crucial yet most daunting undertakings facing international biobanking because it pertains to targets across the key activities upon which biobank-based science relies. Even if there is agreement about the information to collect, there is significant heterogeneity between biobanks in their purpose (e.g., epidemiological or population versus clinical), design, population, selection criteria, nomenclature, biases, etc., as well as increasing complexity in the information collected and a lack of common standards for collecting and disseminating that information. Therefore, multiple harmonisation efforts must occur in concert in order to achieve interoperability, maximise the sharing and exchange of information so science can reach its full potential. Specific efforts in phenotype harmonisation (e.g. DataSHaPER, PhenX), were presented that will promote meta-analysis and sharing of data. Much of this work is already embedded in ongoing projects that are taking advantage of the harmonization initiatives. Additional efforts related to Web technologies are also being used to promote semantic integration of biological resources and to better connect biobanks and biospemens to the broader scientific enterprise.

Making the Most of Biobank Data via the Internet
Biobanking has matured substantially in recent years, not least in terms of the sophisticated solutions employed to run operations and record research findings. In parallel, genotype-phenotype databasing on the Internet is becoming a major field in its own right. Progress in these two areas will require bridging the gap between intrabiobank data handling and Internet-based data sharing and integration. The scope of this effort goes far beyond simply building a database atop each biobank, and issues
raised in this new domain of “Internet biobanking” were addressed.

To date most biobanks have devoted considerable attention to DNA interrogation, some attention has been directed towards sample processing and phenotyping but relatively little attention has been directed to data handling (analysis/databasing), even though this poses the biggest challenge to come. An attractive vision for the future is an Internet-based “knowledge environment” for gene-to-phenotype (G2P) information, supporting direct data submission, seamless data integration, and holistic searching. Implementing a G2P database network requires consideration of design and utilisation. In terms of design, there appears to be a desire to move from the disparate silos that have traditionally existed to a federated network with nodes and some centralised coordination; i.e., a mixed model. A number of efforts are already underway to coordinate biological data resources throughout Europe that provide useful lessons for moving forward. Committing to Internet biobanking requires changes to the incentive structure to reward and promote contributions along the entire pipeline from biobank conception to the sharing of samples and data. Meeting participants put forward a number of countermeasures to address the issue of incentives and proper credit.

Isolated Populations
Advantages of using isolated populations for understanding genetic influences on complex disorders were presented. Importantly, methodological advances allow increased bridging between studies that are population-based and those using genetic isolates. Within Europe substantial research has been conducted on population isolates and this work faces many of the same challenges in harmonisation and data sharing that characterise the population and clinical based biobanks. Although decisions about study design in isolated populations will be study specific, striving to increase interoperability will facilitate comparisons between populations.

Studies of isolated populations also raise some unique consenting issues because they typically involve a narrow research environment, with mostly healthy volunteers, but also include people who are often related to each other. Therefore, an individual’s decision to participate may reveal information about extended family members who may not have agreed to participate. Consenting may involve multiple levels: Individual and social levels involve the supra-individual dimension by promoting discussion in the community to foster a wider social understanding; the legal and participative levels include informed consent, feedback, and a deliberative phase.

From Data to Statistical Analyses
Biobank data harmonisation is focusing increased attention on higher level computation and modeling. Information structures are becoming more complicated as the handling of biomedical research data requires integration of diverse types of data, harmonisation, storage, data curation, access, tools, web services and high capacity networks. This session illustrated how an analysis pipeline will enhance our ability to address the growing complexity surrounding the phenotypic data, analytic strategies and the larger quantities of data being generated. This pipeline would span from raw data via low level filtering to complex statistical analyses and educational aspects of future challenges. Substantial infrastructure and integration of technologies is required to achieve this and considerable investments in the biomedical and life sciences are aimed at generating important pieces of this. e-Infrastructures allow researchers to share resources, strengthen collaborations and will rapidly change the landscape of science by elevating existing applications to higher levels of usability. Successes of genome-wide association studies (GWAS) have been built on the harmonisation of genotypes, simple phenotypes and statistical analyses. Challenges in the next phase include harmonisation of more complicated phenotypes, harmonisation of analytical approaches and integration of increasing amounts of information produced via high-throughput data domains.

In tandem with these developments, the authoring of educational content is gradually shifting from a traditional approach (lectures with text books) to a more complex blended learning approach, in which publishers, lecturers, and students play complementary roles.

Recontact and Reconsent
Recontact (contacting participants who are already part of a study/project) and reconsent (Providing additional consent for further aspects of a study) issues are becoming more urgent for several reasons. New requests may involve the use of data or samples for studies not foreseen at the time of consenting, and as our genetic
knowledge base grows return of results or informing participants of incidental findings will become more relevant. This session provided snapshots of the policies related to reconsent and recontact in a number of European countries. While some studies recontact source persons in cases of abnormal results, most consent forms state that participants will not be provided with specific information. When recontact has been agreed to, there is so far no consensus about whether communication should be initiated by the physician, the researcher, or the biobanker. Qualitative studies have found that consents are used to protect the institution as much as the donor, and most participants do not fully comprehend them. An overwhelming majority of particpants desire feedback about research results, particularly if personally relevant and if treatment is available.

Currently within Europe, there is no specific, binding legal instrument that applies to biobanks as legal instruments that apply to data protection, medical research, and tissue regulation were not necessarily designed with biobanks in mind. There are essentially different legal requirements for tissue and data; in practice, it is very difficult to separate these two entities. To further complicate matters, the requirements can vary between the different jurisdictions across Europe.

Data Sharing: Various Models—Same Science?
Two keynote addresses illustrated that striking advances in GWAS underscore the power of consortia and data sharing and characteristics of two data access models (one European and one American) were presented for comparison: The European Genotype Archive (EGA) and dbGaP. Notable differences exist between the ways in which data access are granted. This is attributable in large part to differences in operating context. In Europe, ethical oversight is mostly at the study level using individual consents and sample collection. In the United States, a main ethical checkpoint is at the protocol level; e.g., NIH requirements. U.S. institutional review boards (IRBs) provide scrutiny and strict interpretation of “consent to specific end use” while in Europe access decisions are left to the investigator or data-generating body.

Presenters highlighted practical ways in which data sharing has contributed to their work but also the challenges they have confronted, potential solutions employed, and unintended consequences.

The discussions helped to delineate emerging issues with which many projects are grappling as new methods have become available. A constant consideration is the balancing of science, scientist, and patient interests and practicable solutions must take into account whether oversight and access will be administrated through local, federated, or centralised models. Participants expressed concern about the potential for “bad science” due to misunderstanding or incorrect analysis that can spread misinformation or reflect poorly on the parent study. They noted that data access committees typically give primary consideration to ethical/legal issues and defer questions about science quality or significance to peer review that is part of the grant funding and publication process.

Practical Experiences With Clinical Biobanks
Clinical biobanks exhibit great diversity of materials, much more than population-based biobanks. Nevertheless, clinical biobanks must address many of the same scientific issues as population biobanks. Many countries clearly recognize that national and international networks can bring discipline and credibility to biobanking, helping projects to operate under a similar framework so as not to isolate precious biospecimens. Presenters from Sweden, Canada, Australia, Belgium, and The Netherlands shared practical experiences in building clinical-based biobanks in their respective countries. These presentations highlighted a number of cross-cutting themes for clinical biobanks including the need for harmonisation of sample handling routines between biobanks and clinical practice, a broad biobanking consent system in routine medical care, education and networking.

Assessment of the results & impact of the event on the future direction of the field

All biobanks are facing similar strategic issues and there was broad consensus at the meeting to develop an overarching strategy so that biobanking could develop globally and in a harmonised fashion. The final session on Building an International Biobanking Community discussed realistic next steps. The most productive approach is to move forward together to address the challenges ahead, with an eye toward avoiding duplication, for the benefit of all. It is crucial to continue to interact as science advances—to share insights, to continually raise the quality of biobanking science, and to speak as one voice. A number of coordination efforts have contributed to building an international biobanking community to address many of these overarching issues. In addition to building on these current harmonisation and sustainability efforts, participants identified two primary next steps: a coordinated strategy and continued interactions.

A coordinated strategy would outline which effort(s) are being undertaken by which
organisations and how action items will be addressed. It is important to be able to communicate among different disciplines and to create an environment where basic, clinical, and epidemiological researchers can come together to provide a more unified voice with respect to biobanking science. To encourage international collaboration across biobanks, brainstorming sessions could be organised in the context of professional meetings where people can put together ideas for study concepts that can be taken forward. Specific strategies for achieving part of these goals could be taken on by already ongoing initiatives such as P3G and BBMRI.

There was considerable interest in the biobanking community bringing coherence to biobanking issues. We decided that a good way to do this is through the issuance of a white paper that provides a blueprint or common strategy for the way forward. Such a white paper should reflect international consensus to clarify the issues that are of priority concern, to mitigate misunderstanding and misimpression, and to give direction about funding strategies and policy in the biobanking field. A small working group has been tasked to outline next steps and take this forward and it is essential to engage the broader community in this endeavour.

Programme
Wednesday 25th of March 2009
18:30 – 19:00 Conference Opening with Refreshments
19:00 – 19:45 Opening Session with Key-Note Speaker
Harmonisation: What Does it Mean in Ethics?
Dr. Ruth Chadwick, Director of the ESRC, Cesagen, Cardiff University, UK
19:45 – 21:30 Reception with Poster Session

Thursday 26th of March 2009
08:45 – 09:00 Welcome
Dr. Jennifer Harris, PHOEBE Coordinator, Norwegian Institute of Public Health, Norway
Professor Kurt Zatloukal, BBMRI Coordinator, Institute for Pathology, Medical University of Graz, Austria
Professor Bartha Maria Knoppers, P3G, University of Montreal, Canada

09:00 – 10:30 Session I – Phenotype Harmonisation
Chair: Professor and Chairman Julian Little, University of Ottawa, Canada Research Chair in Human Genome Epidemiology, Canada
From Harmonisation to Standardisation, Just a Step?
Dr. Isabel Fortier, Director Research and Development, P3G, Canada
Accelerating and Improving Research through Standardisation
Peter Geary, Chief Executive Officer, Canadian Tumour Repository Network and Chair of the Marble Arch International Working Group on Human Specimen Biobanking for Research Purposes, Canada
PhenX Measures
Dr. Carol M. Hamilton, Director of Bioinformatics, RTI International, PhenX Principal Investigator, USA
Concept Web Technologies to Harmonise Dispersed and Ambiguous Information
Dr. Barend Mons, Biosemantics Group, Department of Medical Informatics, Erasmus Medical Center, Rotterdam and Department of Human Genetics, Leiden University Medical Center, The Netherlands

10.30 – 11.00 Break

11.00 – 12.30 Session II – Making the Most of Biobank Data via the Internet
Chair: Professor Rex Chisholm, Dean for Research, Feinberg School of Medicine, Northwestern University, USA
Bridging the 'Internet-Biobanking' Gap: Modular Resources and Researcher IDs
Professor Anthony J. Brookes, Department of Genetics, University of Leicester, UK
Anonymous Bosh: Identity, Authority and Reputation in a Mashed-Up World
Geoffrey Bilder, Director of Strategic Initiatives, CrossRef, UK
Tools for Multi-Center Data Harmonisation and Analyses
Dr. Maria Krestyaninova, European Bioinformatics Institute, EMBL-EBI, Hinxton, UK Tracing Biological Collections: an Incentive for Collaboration
Dr. Francine Kauffmann, Inserm U780, Epidemiology and Biostatistics, France

12-30 – 13.30 Lunch – Cap Nord Restaurant

13:30 – 14:15 Key-Note Speaker
Pooling Data of Isolated Populations: Prospects and Limitations
Professor Cornelia van Duijn, Department of Epidemiology & Biostatistics, Erasmus University Medical School, The Netherlands

14:15 – 15:15 Session III – Isolated Populations
Chair: Professor Thomas Meitinger, Head of the Institute of Human Genetics, Helmholtz Zentrum Munich and Head of the Institute of Human Genetics, Technical University Munich, Germany
Complex and Quantitative Traits: Definition of a General Picture from The Study of a Network of Isolated Populations
Professor Paolo Gasparini, University of Trieste/IRCCS-Burlo Garofolo, Italy
Consenting in Isolated Population Biobanks: Individual Autonomy vs. Intersubjective
Responsibility
Deborah Mascalzoni, EURAC Bolzano, Italy
Genetic Structure of Some of the Isolated and Non-Isolated Populations in Europe
Professor Andres Metspalu, Director of the Estonian Genome Project of the University of Tartu, Estonia

15.15 – 15.45 Break

15.45 – 16.30 Key-Note Speaker
Sequencing 1000 Genomes to Provide a Deep Catalogue of Human Genetic Variation
Dr. Richard Durbin, Principal Investigator, Wellcome Trust Sanger Institute, UK

16.30 – 18.30 Parallel Session IV:From Data to Statistical Analysis
Chair: Professor Paul Burton, Department of Health Sciences and Department
of Genetics, University of Leicester, UK
eInfrastructures in the Updated ESFRI Roadmap
Professor Juni Palmgren, Department of Medical Epidemiology and Biostatistics, Karolinska Institutet and Department of Mathematical Statistics, Stockholm University, Sweden
From Data to Statistical Analyses: A Lesson and a Vision from the Domain of Lipids and
CHD in ENGAGE
Dr. Samuli Ripatti, Institute for Molecular Medicine, University of Helsinki, Finland
Statistical and Bioinformatical Challenges in the Analysis of Genome-Wide Association Data
Dr. Yurii Aulchenko, Department of Epidemiology and Biostatistics, Erasmus MC, University Medical Center, Rotterdam, The Netherlands
From Data to Statistical Analysis: Explaining Genetic Heritability by Interaction Analysis?
Dr. Kristel Van Steen, Montefiore Institute, Bioinformatics, Biometry and Statistics, University of Liège, Belgium
EUROGENE – The First Pan-European Learning Service Dedicated to Genetic Medicine
Professor Heike Bickeböller, Department of Genetic Epidemiology, University of Göttingen, Medical School, Germany and Petr Knoth, Researcher and Doctoral Student, Knowledge Media Institute, The Open University, UK

Parallel Session V:Recontact and Reconsent
Chair: Dr. Ruth Chadwick, Director of the ESRC, Centre for Economic and Social Aspects of Genomics (Cesagen), Cardiff University, UK
Deconstructing and Reconstructing a Terminology
Dr. Anne Cambon-Thomsen, Director of Research, CNRS, INSERM U558, France and
Professor Bartha Maria Knoppers, University of Montreal, Canada
UK Biobank’s Recontact and Reconsent Policy
Dr. Jane Kaye, Wellcome Trust Fellow, Ethox Centre, University of Oxford, UK
Recontact and Reconsent in Estonia
Professor Andres Metspalu, Director of the Estonian Genome Project of the University of Tartu, Estonia
French Policies and Examples Regarding Re-Contact and Re-Consent
Mrs. Emanuelle Rial-Sebbag, INSERM, France
Recontact and Reconsent in CARTaGENE
Ms. Karine Bédard, CaRTaGENE, Canada
Public Perspectives on Recontact and Reconsent
Dr. Klaus Hoeyer, Institute of Public Health, University of Copenhagen, Denmark

20.00 Dinner – Cap Nord Restaurant

Friday 27th of March 2009
09.00 – 09.45 Key-Note Speaker
Thinking Big: Collaboration and Data-Sharing, The New Genetics, and Genes Influencing Diabetes and Obesity
Professor Mark McCarthy, Oxford Centre for Diabetes, Endocrinology and Metabolism, UK

09.45 – 10.45 Session VI – Data Sharing: Various Models - Same Science?
Chair: Professor Paul Burton, Department of Health Sciences and Department of Genetics, University of Leicester, UK
The European Genotype Archive Data Access Model
Dr. Mario Caccamo, European Genotype Archive, EMBL-EBI, UK
dbGaP as a Model for Sharing GWAS Data
Dr. Rebekah Rasooly, Program Director for the Genetics and Genomics Program, National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK), National Institutes of Health, USA
Data Sharing: Can We Align The Needs of Science and the Needs of Individual Scientists?
Professor Paul Burton, University of Leicester, UK

Panel discussion:
Professor Anthony J. Brookes, University of Leicester, UK
Professor Paul Burton, University of Leicester, UK
Dr. Mario Caccamo, European Genotype Archive, UK
Dr. Jennifer Harris, Norwegian Institute of Public Health, Norway
Professor Mark McCarthy, Oxford Centre for Diabetes, UK
Dr. Rebekah Rasooly, National Institute of Health, USA
Professor Cornelia van Duijn, ERASMUS MC, The Netherlands

10.45 – 11.15 Break

11.15 – 13.15 Session VII – Practical Experiences with Clinical Biobanks
Chair: Dr. Peter Riegman, ERASMUS MC, The Netherlands
Clinical Population-Based Biobanks: Experiences of Building and Scientific Use
Professor Joakim Dillner, Department of Laboratory Medicine, Lund University, Sweden Building a Tissue Bank Network
Peter Geary, Chief Executive Officer, Canadian Tumour Repository Network and Chair of the Marble Arch International Working Group on Human Specimen Biobanking for Research Purposes, Canada
Practical Experiences with Clinical Biobanks: Lessons from Down Under
Professor Lyle Palmer, Foundation Chair in Genetic Epidemiology and Director of the Centre for Genetic Epidemiology and Biostatistics, University of Western Australia, Australia
Cervical Cytology Biobanks as a Resource for Molecular Epidemiology
Dr. Marc Arbyn, National Centre for Cancer Control, Scientific Institute of Public Health, Belgium
Bridging Population Biobanks and Disease Based Post-Mortem Biobanks - Yes, We Can Dr. Rivka Ravid, Netherlands Institute for Neurosciences, Dutch Royal Academy of Sciences, The Netherlands

13.15 – 14.15 Lunch

14.15 – 16.00 Session VIII – Building an International Biobanking Community
Chair: Professor Lyle Palmer, University of Western Australia, Australia
This structured discussion aims to stimulate our thinking about the reasons and approaches for building an international biobanking community. The panellists will identify and discuss important next steps that could help move forward this agenda forward through international biobanking initiatives.

Panel Discussion:
Professor Paul Burton, University of Leicester, UK
Professor Rex Chisholm, Dean for Research, Feinberg School of Medicine, Northwestern University, USA
Mrs. Mylène Deschênes, Executive Director, P3G Consortium, Canada
Dr. Jennifer Harris, PHOEBE Coordinator, Norwegian Institute of Public Health, Norway
Dr. Robert Hewitt, President ISBER, Director, National University Hospital, Singapore
Professor Gerardo Jimenez-Sanchez, Director General, National Institute of Genomic Medicine, Mexico
Dr. Jane Kaye, Wellcome Trust Research Fellow, Ethox Centre, University of Oxford, UK
Dr. Jan-Eric Litton, Department of Medical Epidemiology and Biostatistics, Stockholm, Sweden
Professor Lyle Palmer, University of Western Australia, Australia
Mr. Markus Pasterk, Scientific Coordinator, International Agency for Research on Cancer, France
Dr. Camilla Stoltenberg, Deputy Director, Norwegian Institute of Public Health, Norway
Professor Gert Jan B. van Ommen, Head of Department of Human Genetics, Leiden University Medical Center and Director, Center For Medical Systems Biology, The Netherlands
Professor Kurt Zatloukal, BBMRI Coordinator, Institute for Pathology, Medical University of Graz, Austria

16.00 Adjourn