Bioresources access and sharing policies...by Mylène Deschênes, P3G, Montreal, Canada

Embedded Scribd iPaper - Requires Javascript and Flash Player
Bioresources data access and sharing policies: considerations for improving bioresource impact factors
MYLÈNE DESCHÊNES, B.C.L., LL.B, LL.M. P3G Executive Director with special acknowledgment to Saminda Pathmasiri for the preparation of this presentation
BRIF Workshop Toulouse, January 18th, 2011
P3G Funders
• (1) the guidelines/policies promoting datasharing • (2) the policies of the biobanks/cohorts
1. ACCESS GUIDELINES/POLICIES FOR PUBLICLY FUNDED ‘RESOURCE’ PROJECTS
An era of sharing?
• ‘Recent advances in information technology have revolutionised science. Providing new opportunities for researchers to share data and build on one another’s work. Informatics and the ability to mine large datasets and combine them with information from many other sources present a huge potential to advance developments in publics health. The importance of data sharing in advancing health is becoming increasingly widely recognized (…)’
• Wellcome Trust 2010 Joint statement
An era of sharing?
• ‘Much of data collection that could improve public health research is expensive and time consuming. As public and charitable funders of this research, we believe that making research data sets available to investigators beyond the original research team (..) – Faster progress in improving health – Better value for money – Higher quality science’
• Wellcome Trust 2010 Joint statement
An era of sharing?
• 1996 Bermuda Principles provided guidelines on the rapid release of data from large-scale sequencing projects – sequence assemblies of 1-2 kb or greater in the public domain;
An era of sharing?
• 2003 Fort Lauderdale : reaffirm and extend the rapid data release policies developed to implement the 1996 Bermuda Principles. – ‘Require, as a condition of funding, free and unrestricted data release from community resource projects to appropriate central and searchable public databases, and vigourously ensure that this occurs ‘ (A.2) – ‘Make data generated by the resource immediately and freely available without restriction’ (B.3) – Extended to all types of sequence data – Extended to ‘community resource projects’ - conceived to develop a set of data, reagents, or other material whose primary utility will be as a resource for the broad scientific community
An era of sharing?
• OECD Principles and Guidelines for Access to Research Data from Public Funding 2007 • Toronto International Data Release Workshop (2009) • Reaffirmed the data sharing principle with a wider group of stakeholders : Large reference datasets in biology and medicine that have broad utility. ‘Should go beyond genomics and proteomics studies to other datasets including chemical structure, metabolomic, TNA and annotated clinical resources (cohorts, tissue bank, and case control studies)’ Nature 461, 168-170 (2009) ‘Require the sharing plan to be presented as part of the grant application and (…) be subject to peer review. (…) recognizing that largescale data generation projects need not necessarily lead to traditional publications.
An era of sharing?
• 2010 Joint statement by funders of health research (Wellcome Trust) • ‘Expectation that all our funded researchers should maximise access to their research data with as few restrictions as possible’. Need to submit a sharing plan in the following types of application: - Primary goal is to create a database resource - Might generate a community resource as defined in the 2003 FL Statement - Proposal generating large scale or other high value data outputs with clear utility to research questions beyond those the data generator are seeking to address - **Exception (footnote) ‘does not currently cover routinely collected clinical data, or public health data to which the signatories of this statement contribute no funding’
What is the link with BRIF?
• There is a clear push for sharing/providing access • While open access might be presumed to be a good thing to increase the impact of a biorepository : open access does not necessarily mean measurable impact. • Open access shouldn’t mean ‘open bar’!
What is the link with BRIF?
• Access arrangements are the most appropriate vehicule/opportunity to insert :
– Control points – Tools/Markers that can be measured in an impact assessment in a way that is clear, flexible and enforceable.
What is the link with BRIF?
• needs to balance the obligation to open access with appropriate measures that are compatible with the interest of the biorespository itself: proper recognition of scientific contribution and sustainability through the capacity of measuring their own impact.
An equilibrium between access and capacity to measure impact
Funders agree to promote greater access to and use of data in ways that are: (1) Equitable: In any approach to the sharing of data should recognize and balance the needs of researchers who generate and use data, other analysts who might want to reuse those data , and communities and funders who expect health benefits to arise from research (...). (WT 2010)
An equilibrium between access and capacity to measure impact
SPECIFIC MEASURES: 1. Citation of sources • C.1 Resources users: appropriately cite the source of the data analysed and acknowledge the resource producers. (FL 2003) (similar wording in WT 2010) • In the OECD 2007 guidelines: it is a concern of quality/security that requires ‘traceability’ • C.1 The early publication of a Project Description (...) would provide users with an appropriate reference to cite before the data are formally published’ (FL 2003) (similar wording in Toronto statement)
An equilibrium between access and capacity to measure impact
SPECIFIC MEASURES: 2. Track of usage and publication ‘To encourage compliance, funding agencies and academic institutions should give credit to investigators who adopt prepublication data-release practices. One option would be to recognize good data release behaviour during grant renewals and promotion, another would be to track the usage and citation of data sets using electronic systems similar to those used for traditional publication’ Toronto 2009 statement
An equilibrium between access and capacity to measure impact
3. Cataloguing of resources • OECD: Principle of transparency : ‘Lack of visibility of existing research data resources and future data collection poses serious obstacles to access.’ • FL 2003: A.4 Funding agencies: ‘ensure that a centralized view of existing community resource projects is available as an information source for the community;
Cataloguing System
• Characteristics:
– uniform set of information that bioresources will make publicly available. This is important for: • Investigators to enable a quick reference • Avoid methodological repetition in scientific article using bioresources • Enable peer reviewers easy access to bioresource information – Publicly accessible through the internet – Independent and reputed management of system – Ability for each bioresource to update its information – Shall contain not only the publications but the uses
An equilibrium between access and capacity to measure impact
Conclusion part 1 • While biorep are expected to open access now, the tools are not in place to reach the equilibrium necessary for them to measure impact. • This work is URGENT!
2. BIOREPOSITORY ACCESS ARRANGEMENTS
Access Arrangements
• Biorepositories who are ‘open for business’ and want to give access will usually have: • a) an access policy • b) a data and sample access committee (SDAC) • c) a generic material/data transfer agreement contract for future users
Empowering biorepository with tools to measure their impact
• Access agreement = appropriate vehicle to implement impact factor tools
– Clear/specific/detailed content – Flexible (can be modulated to suit each case scenario) – Enforceable (contract)
Ideally a harmonized mechanism among bioresources through which bioresources will be able to monitor its usefulness within the scientific community.
Contractual clauses/ policy requirement
1. Mandatory reference of the bioresource(s) which provided the data/samples for the research project.
Consider providing a precise phrase that the investigator ought to include in the publication.
E.g. All publications and presentations must contain the following sentence: “The/Certain data and/or samples used for this research are from NAME OF BIORESOURCE (www.name....)”
Co-Authorship or acknowledgement?
Contractual clauses/ policy requirement
2.Request a copy (or link) of all published articles which uses bioresource materials
Consider providing a precise time frame within which this should be done.
E.g. Approved investigators shall return to the bioresource a copy of or an internet link to all published articles.
Contractual clauses/ policy requirement
1+2: Enables the biobank and others to measure impact through traditional ways of measuring success in science field: publications. ** but is publication the only ‘endpoint’.. Wouldn’t mere use be interesting?: lots of scientific activities might never be published. Other types of scientific/social contributions of biobanks
Contractual clauses/ policy requirement
3. Get permission from users to disseminate a short summary of their research activity.
This is crucial to be able to show the uses and not just the publications
E.g. Executive Summaries submitted to the Data and Access Committee for review shall be made public on the biobank website and on any other dissemination tools available to the public and the scientific community for the purpose of identifying the type of research being conducted with the material/data
Should we ask for annual activity reports? How much monitoring can biorepository impose without becoming too cumbersome to users
3G P
Access Catalogue
45 studies/cohort/biobanks surveyed: • Periodical reports have to be submitted to the study – 15 yes; 21 no; • ‘Presentations/Publications have to be reviewed by study representative to ensure compliance prior to public disclosure’ – 24 yes; 12 no; • ‘other important conditions of use’ – Co-authorship: 6 – Acknowledgement: 6 – Receive copies of all published material : 2
Contractual clauses/ policy requirement
4. Prevent any third party use of bioresource data/samples (i.e. by investigator other than the signatories of the access agreement).
This is crucial to be able to adequately monitor the use of the material provided to an investigator.
E.g. Transferred Materials, including any copies thereof, may not be
disclosed, transmitted or shipped to anyone except employees working with the Approved User or co-investigators indicated in the Study who have a need to know such information for the Approved Study and who agree to be bound by the terms of this Agreement.
However: serious impediment to the principle of ‘free/open access’
Combined effect of these 4 clauses:
Enables the bioresource to monitor the scientific output directly resulting from its use Measure and advertise the scientific utility of the resource to the scientific community (as well the general public and the funders)
For discussion:
• What else is needed to empower biorepositories in the contractual access arrangements?
Some limits...
• Enforceability
– How will we know a breach occurred? – Who will sue for breach of contract? – More realistic: sanctions/deterrents
• be excluded from future funding opportunity
– Funding agencies need to clearly support the respect of tools that allow biobank to make their impact measurable.
• be excluded from ‘a club’ (consortia, hospital based research group)
– Code of conduct for researchers (users) – Federated model of access
Some Limits...
• Vision based on a linear model ...
Biorepository
Scientist
Research hypothesis
Breakthrough leading to a publication
Organic model
Research is a much more ‘organic’ enterprise.. And so are biorepositories = Resulting in lots of contractual arrangements
NETWORK/ ALLIANCE
BIOBANK
BIOBANK BIOBANK
BIOBANK BIOBANK BIOBANK BIOBANK
BIOBANK BIOBANK BIOBANK
32
BIOBANK
Organic model
Will the users/researchers be able to keep track of it all?
BIOBANK BIOBANK BIOBANK BIOBANK BIOBANK BIOBANK BIOBANK
NETWORK/ ALLIANCE
BIOBANK BIOBANK BIOBANK
33
BIOBANK
Different models of sharing the material/data
• Example 1: NIH - GWAS
– Expectation that data from NIH Supported GWAS be deposited into the NIH GWAS data repository dbGAP: it is de-identified (but the investigator retains code). Make them available to other researchers. – Data is ‘tagged’ but the users don’t know who contributed.
Different models of sharing the material/data
• Example 1: NIH – GWAS • Core Elements for National Institutes of Health (NIH) Data Use Certification Agreements:
– No distribution (to other users by the initial one) – Public listing of approved research uses – Progress report on the uses – Recognition of the IP policy
Different models of sharing the material/data Example 2:
• Create a ‘community’ of interest whereby material is federated in a ‘club’ (*Enforceability):
– ICGC consortium
• • • •
Material/Data is shared within the group Common adhesion to governance framework 2 datasets: open access/controled access) The access procedure are centralized: not the data/material : easy to trace the ‘contributing biobanks’
Access : Option 2 – Federated model: International Cancer Genome Consortium (ICGC)
Governance/Pac kage PreApproved by Institutions and REBs
Open Access Datasets (OAD) Controlled Access Datasets (CAD)
International Data Access Committee
Franchise Database Franchise Database Franchise Database
Local Ethics Approval of Project
Access Form (CAD)
Researcher
Access Agreement + Cost Decision = Access Number
Data Access Officer and Data Access Compliance Office
Data Coordination Centre
Access Number
Data
Biorepository access arrangements
• Through contractual agreement we can impose some tools that will allow to measure impact. • However:
– challenge of ‘organic’ research model – Devil’s in the details ; need to think about the technical aspects of traceability of source in ‘natural flow’ of research activities (food?) – we need commonly recognized tools (ex.: ID numbers)
• Tools to measure impact need to be recognized in it’s own value.. (not just for quality/security/IP.. )
Need common tools
Thank you...