/Always_expect_the_unexpected by CSG Centre for Society and the Life Sciences

Always expect the unexpected Legal and social aspects of reporting biobank research results to individual research participants

Jasper Bovenberg Tineke Meulenkamp Ellen Smets Sjef Gevers

Jasper Bovenberg Jasper Bovenberg is practicing attorney in The Netherlands, specializing in life sciences. Previously, he acted as attorney and the US. From 2005-2008, he was part-time research fellow at the Amsterdam Academic Medical Centre (AMC), where he conducted the research for this report. In 2005, at Leiden University, he received his doctorate thesis “Property Rights in Blood, Genes & Data: Naturally Yours?”. Bovenberg is founder and director of Legal Pathways Institute for Health and Bio-law (2007). He has been a member of several national and international advisory and review boards in the area of health care, ownership and biobanks.

Tineke Meulenkamp Tineke Meulenkamp ﬁnished her studies in psychology at Utrecht University in 1996, after which she worked as a researcher and management consultant at Prismant, the service provider for Dutch healthcare. Since 2004, she has been working at the department of Medical Psychology of the Academic Medical Centre (AMC) in Amsterdam, where she performed the social research of the study underlying this report. Currently she is employed by the Netherlands Institute for Health Services research (NIVEL).

Ellen Smets Ellen Smets earned a masters degree (1987) as a social psychologist. Her doctorate thesis at the University of Amsterdam (1997) addressed cancer patients’ radiotherapy related fatigue. From that time, she has been active in research in the ﬁeld of medical communication. As Principal Investigator she presently coordinates the research line Medical Communication at the department of Medical Psychology of the Academic Medical Centre, University of Amsterdam. Her current interests lie in patients’ wish for and use of medical information and physician’s behavior in this regard, with a focus on oncology and genetics. Smets is a member of the Dutch Health Council.

Sjef Gevers Sjef Gevers studied law and sociology at the University of Amsterdam; since 1989 he is professor of health law at the same university, both in the Faculty of Law and at the Academic Medical Centre. He has published widely on the legal aspects of health protection and health care. His main research interests include the response of the law to new technologies in health care, the evaluation of health legislation and ethical and legal issues in the ﬁeld of medical research. He is editor in chief of the Dutch Health Law Journal and one of the editors of the European Journal of Health Law.

Always expect the unexpected Legal and social aspects of reporting biobank research results to individual research participants

Jasper Bovenberg Tineke Meulenkamp Ellen Smets Sjef Gevers

Centre for Society and Genomics, November 2009

Table of contents

Preface

Summary

1. Introduction

1.1 Aim 1.2 Background 1.3 Research question 1.4 Methods 1.5 Deﬁnitional issues 1.6 Hypothesis: Always Expect the Unexpected 1.7 Literature overview: bioethical discourse 2. Preferences of researchers and (potential) participants (empirical research)

2.1 Introduction 2.2 Methods 2.3 Information preferences of citizens and patients 2.4 Attitudes of citizens, patients and researchers towards the communication of research results 2.5 Discussion 2.6 Interim conclusion empirical research regarding Guiding Principles 3. Rights and obligations of researchers and participants (legal research)

3.1 Case study 3.2 Whom to sue? 3.3 X’s brief 3.4 Biobankers’ defence 3.5 Expected outcome of case study

13 14 14 15 16 17 19 21

23 24 24 25 28 29 31 33 34 36 37 50 57

4. General discussion

5. Proposed guiding principles

67 69 71

5.1 Structural arrangements 5.2 Disclosure decisions Samenvatting

Sources

Appendix A Factors associated with information preference for individual results and attitudes on researchers’ duties to communicate research results Appendix B Supervisory Board Appendix C List of experts

87 88 88

Preface

Imagine you have been asked to participate in a research project on a rare disorder. You have given blood, medical data and some additional information about yourself. For years, you know you have some of the symptoms of this disorder but this never really made your life difficult (or more difficult than it is anyway). Years later, you receive a letter from your consultant, telling you that – on the basis of their research – they have discovered something about your future health that you may want to know about. Would you be pleased or perhaps astonished to receive this letter? And what would you do? The case described above no doubt is imaginary. But it is not one that is entirely unthinkable. Developments in the field of genomics are moving fast, often creating new and sometimes unforeseen societal issues. One of these issues is the question what we should do with findings of health-related information that is the result of DNA sequencing. If this DNA has been stored in a biobank, it may indeed be the case that people are contacted several years after they have donated a blood sample and gave consent to the use of this sample for research purposes. The future use of biobanks is a topic that is typical for the work and approach of the Centre for Society and Genomics (CSG). It is an urgent issue, it emerges at the intersection of disciplines (healthcare, legal issues, ethics), it touches upon the lives of different people (patients, citizens, scientists, policy-makers, lawyers and professionals) and it requires early anticipation in order to prepare both science and society for the future. This report concerns the communication about findings generated from the material stored in biobanks. Communicating about genetic predispositions and increased risks for multifactorial diseases is not an easy undertaking. What is legally allowed, obliged or reasonable? What information do participants consider to be useful? Do experts think the same about this question? Who should communicate research findings? What should be communicated? And when should researchers refrain from communicating their findings? The authors of this report are affiliated to the Amsterdam Medical Center (AMC) of the University of Amsterdam (UvA). They represent legal, medical and social science expertise and have provided a comprehensive analysis of the aforementioned questions. The authors have consulted scientific literature, law texts and relevant groups: patients, citizens and researchers. The main conclusion of this report is that biobanks are under increasing pressure to give more and better information regarding results to individuals who have participated in biobanks. The report concludes with a large set of guiding principles for the future; improvements for setting up biobanks and recommendations for decision-making about the disclosure of research findings. I do belief that these guiding principles will be of benefit for policy-makers, officials and professionals involved in the communication and organisation of biobanks, both now and in the future. Annemiek Nelis General Director CSG

Summary

The research of common complex disorders requires data-driven, multi-purpose and multidisciplinary research using large-scale DNA-banks or biobanks. Complex disorders are caused by a large number of small, often additive effects, representing the outcome of the interplay, at various levels, of genes, lifestyle and the environment. Revealing these complex interactions will depend critically on the study of large sets of well-documented, up-to-date epidemiological, clinical, biological and molecular information and corresponding material from large numbers of patients and healthy persons, collected and made available by biobanks. Over the last decade, in many countries such DNA research databases have developed, sometimes in the form of national, population based biobanks. Examples of large scale biobank initiatives in the Netherlands are the UMCG Groningen initiative LifeLines, and the initiative of the academic-medical centres String of Pearls. Our research question was whether and, if so, how any results to be derived from (studies using) these biobanks should be communicated to the individual research participants. Specifically, our research intended: 1. to examine whether there are any legal obligations of researchers to report results from biobank research to individual research participants, and 2. to assess the attitudes and information preferences of major stakeholders. This research question is not only related to the emergence of large scale biobanks but also to the recent debate in the international literature about whether research participants should be informed about the results of the research in which they are involved. In fact, a trend in international legal instruments towards the recognition of right to feedback of research results has been reported in ethical and legal commentaries. Traditionally, research results are, in principle, not communicated to individual participants, except for incidental or unexpected findings. Unexpected findings are incidental (diagnostic) findings which pertain exclusively to one individual and which were not foreseen. New findings are generalizable results of scientific research which have a predictive value for present or future health of a group of people. In our research proposal, we hypothesized, that the distinction between ‘expected’ and ‘unexpected’, which underlies the traditional standard of non-disclosure, would be challenged by fundamental shifts in both the specifics of biobank research and the perspectives and expectations of participants. We hypothesized that these expectations would put additional pressure on biobank operators and investigators to translate and communicate results to participants faster and more routinely. We submit that our hypotheses have been verified by the following developments: • The proliferation of biological analyses, in a standardized format, in combination with the availability of a broad range of personal phenotype and environmental data in a biobank, over a prolonged period of time, could facilitate and accelerate the translation of the aggregate results into clinically relevant information at the level of the individual participant.

• The application of comprehensive genotype and functional genomic measurements across the general population is likely to yield unexpected, incidental findings for nearly everyone. • The widely advertized availability of both personal genotype and biomarker information, in unprecedented detail and breadth and direct to consumer, is likely to raise the level of expectations of those who participate in a biobank. • The participants’ long term, indeed life-time, involvement, coupled with the breadth and the sensitivity of the information they might provide, may raise these expectations of receiving benefits in the form of individual feedback of results, regardless of whether these were “expected” or “unexpected”. As a result of these developments, the problem of individual feedback in the biobank context relates now to the more comprehensive issue of informing individual research participants about research results in a broad sense, including both incidental and intended findings, as well as study outcomes at the individual level and at the aggregate level. In the years 2006-2008 we have conducted an extensive study of this issue that included both empirical research to assess the attitudes and information preferences of major stakeholders (i.e. (potential) participants and researchers) and legal research to examine whether there are any legal obligations of researchers to provide feedback to individual participants. Our aim was – also bearing in mind the bioethical literature and relevant international policy documents - to elaborate a general normative framework that could provide (further) guidance in this matter and be taken into account in the establishment and operation of biobanks. The empirical study entailed a cross-sectional questionnaire study among citizens (as potential participants), patients (as participants, as they had already donated blood or saliva to biobanks) and researchers. The conclusion of the empirical research is that the majority of (potential) participants showed a keen interest in feedback, not only in individual findings, but also in aggregate study results that could be relevant to their health, including information on small, incremental risks, and largely irrespective of the nature of the disease and the possibilities for prevention or cure. However, a sizeable minority preferred not to receive any feedback. Contrary to the opinion of researchers, (potential) participants also thought that researchers have the duty to inform participants about all kind of results. Biobanks should consider the feedback issue. They must give attention to the management of expectations of participants (which may be high), the diversity of information preferences (i.e. a small group does not have any interest in receiving (specific) results) and the gap between researchers and participants in expectations. The legal analysis involved a survey of international legal instruments, guidelines and policies, policies in use with existing biobanks, analysis of case law and statutes

governing analogous situations. In the report we present the results of this analysis in the form of a case study featuring a recruit in a population based biobank who, after developing several diseases, requests the biobank to inform her about any risk factor that it may know about her, together with advice on prevention/treatment. After briefly discussing the question to whom this participant may bring legal action, we elaborate the legal arguments she could forward to ground her claim, as well as the possible counter arguments of the defending biobank. The conclusion of the legal research is that under the current legal regime, a series of scientific, statutory, medical-legal arguments and practical barriers (complexity, competence, costs and consent) militates against the imposition of an (unconditional) obligation to report biobank research results to individual participants, with the exception of abnormal measurements which pose a present and imminent risk of a serious, treatable condition. However, the above scientific, technological and societal developments and demands are likely to tip the scale towards a more pro-active stance for reporting results and data in the near future. These developments might also call for (a mandatory offer of) disclosure of results and data pertaining to health risks that are not or less imminent, not life-threatening and not instantly treatable. While they may not be under a specific legal obligation to assess and report each and every finding, biobanks might be found negligent in the future, if they fail to adequately address the issue, fail to have a clear policy and/or fail to communicate such policy to participants and other stakeholders. Taking account of the bioethical debate and of our interim conclusions in respect of the legal and empirical aspects and referring to the 2008 OECD Draft Guidelines on Human Biobanks and Genetic Research Databases (which recommend that biobanks should elaborate an adequate feedback policy), we propose a set of Guiding Principles for a feedback or reporting policy to be used in the establishment and operation of general, national and international large-scale biobanks. The Guiding Principles are addressed to those involved in the funding, establishment, operation, use, governance and oversight of such banks. The principles relate on the one hand to ‘structural arrangements’, on the other to ‘disclosure decisions’. The first group of (13 structural or organizational) principles includes inter alia: • that in establishing a biobank the feedback issue should be addressed and a comprehensive policy should be developed; • that the most important stakeholders should be involved in this; • that the policy should provide a framework for disclosure/non-disclosure, to be elaborated at the level of individual research projects; • that upon enrolment participants should be fully informed and be enabled to express their wish to be informed or not about findings relevant to their health; • that the policy should be communicated to the public so that future participants may know what they can expect.

The second group of (10 disclosure decision) principles includes inter alia: • that individual feedback should be given where the information available from research relates to a serious health problem and where the possibility of an individual health benefit is realistic; • hat in other cases, validated research results should be made available on a collective basis to interested participants; • that research results should be disclosed in ways designated by the biobank; external researchers should not directly contact research participants; • that disclosure of research results should be offered with appropriate supplemental information and with counselling when needed; • that in case of doubt, decisions to give or to withhold individual feedback on research results should be submitted for prior advice to an independent ethics committee. The outcome of the empirical and legal research as well as the draft Guiding Principles have been presented and discussed at an expert meeting organized at the Academic Medical Centre of the University of Amsterdam on September 12, 2008. The purpose of this expert meeting was to provide a number of experts drawn from relevant disciplines with an opportunity to challenge our assumptions, submissions, arguments and empirical findings and to review and comment on the Guiding Principles we propose. In addition, the outcome of the empirical research and the legal research have been presented and commented upon by fellow ELSI scientists from various backgrounds in an interdisciplinary peer and non peer review session organized by the Centre for Society and Genomics. The outcomes of these reviews and discussions have been incorporated in this final research report. Final commentary on the legal part of the research report has been provided by a foreign expert, prof dr B.M. Knoppers of the University of Montreal. All findings and Guiding Principles presented in this research report remain the sole responsibility of the authors.

1 Introduction 13

1.1 Aim 1.2 Background 1.3 Research question 1.4 Methods 1.5 DeďŹ nitional issues 1.6 Hypothesis: Always Expect the Unexpected 1.7 Literature overview: bioethical discourse

1. Introduction

“To wrest from nature the secrets which have perplexed philosophers in all ages, to track to their sources the causes of disease, to correlate the vast stores of knowledge, that they are quickly available for the prevention and cure of disease -- these are our ambitions.” Sir William Osler, 1906

1.1 Aim

Our aim is to develop Guiding Principles for the reporting of results derived from research on biobanks to individual participants (Gevers et al., 2006).

1.2 Background

Biobanks One of the challenges in the post-genomics era is the research of common complex disorders (Collins, 2003). Common complex disorders are caused by a large number of small, often additive effects, representing the outcome of the interplay, at various levels, of genes, lifestyle and the environment. Revealing these complex interactions will depend critically on the study of large sets of well-documented, up-to-date epidemiological, clinical, biological and molecular information and corresponding biological specimen from large numbers of patients and healthy persons, collected and made available by biobanks (Kaiser, 2002; Collins et al., 2003; Khoury, 2004; KNAW, 2006; Bouchie, 2004; Hagen and Carlstedt-Duke, 2004; Manolio, 2006). Linking genotype and phenotype Research on biobanks could provide key-links between abstract genomic data and concrete patient medical records. Information from a Biobank may ‘help specify meaningful subgroups of illness and improve the specificity, effectiveness of all kinds of healthcare’ (UK Biobank). Their ultimate objective is to enhance the development of more effective and personalized treatment, reduce undesired side effects of new treatments, and to generate more personalized forms of disease prevention. Oceans of data The data-driven, multi-purpose and multi-disciplinary research on biobanks will yield new results and new data. Specifically, the research is expected to identify risk factors related to more than one disease (common denominators) and the occurrence of various diseases within one individual. As the research findings and updated data are fed back into the DNA-Bank and the data is continuously updated, we expected the bank to become a “revolving results and data facility”.

1.3 Research question

Reporting research results to individual participants? Given the expected flow of results and data, our research question is whether, when, how and by whom/to whom the biobank research results would have to be reported to the participants individually. Traditional standard: no reporting of research results to participants Traditionally, research results are, in principle, not communicated to individual participants, except for the so-called incidental or unexpected findings of clear clinical and actionable utility. For example, according to the Code Goed Gebruik, a distinction can be made between new findings and unexpected findings (Code Goed Gebruik; KNAW, 2006). New findings are generalizable results of scientific research which have a predictive value for present or future health of a group of people (Code Goed Gebruik). New findings are typically relevant at population level but not so much for the individual participant (Code Goed Gebruik; KNAW, 2006). New findings are typically foreseen as the outcome of the verification of the hypothesis underlying the research (Code Goed Gebruik). Unexpected findings are incidental (diagnostic) findings which pertain exclusively to one individual and which were not foreseen. Unexpected or incidental findings may have direct consequences for the health of a participant and treatment or intervention may be possible (Code Goed Gebruik; KNAW, 2006). The chances of making incidental or unexpected findings are considered to be minimal (Code Goed Gebruik; KNAW, 2006). The Code Goed Gebruik even prescribes that the research must be designed so as to minimize the chances of an incidental finding. Besides the incidental or unexpected findings, there seems consensus in the (bioethical) literature that the general research findings should be communicated to the research participants. Also measurements taken at the enrolment of biobank participation (e.g. blood pressure or cholesterol levels) are being communicated in most larger (national) biobanks. Rationale for traditional non-disclosure The traditional policy of non-disclosure rests on a number of considerations which include, but are not limited to, (i) the fact that most research findings in epidemiology or cohort studies are aggregate findings of an exploratory nature, with little or no analytical validity or clinical utility for the individual concerned, (ii) the costs, competence and complexity involved in proper reporting to individuals and (iii) the fact that these findings ordinarily cannot be linked to identifiable participants (Beskow et al., 2001; Code Goed Gebruik, 2002; Knoppers et al., 2006; Boggio, 2008). Shifting standards? In our research proposal, however, we hypothesized that the rationale underlying the traditional non-disclosure policies (specifically the distinction between ‘expected’ and ‘unexpected’ findings) might no longer hold for biobanks and biobanks research

results for the following reasons: 1 Biobank research is expected to identify risk factors related to more than one disease (common denominators) and the occurrence of various diseases within one individual. 2 Biobanks are typically designed so as to form a resource that will enable the conduct of hundreds of different research projects into all kinds of diseases, rather than focusing on a specific disease. 3 New technologies will allow multiple research projects using data from this resource to be conducted simultaneously. In addition, the stored tissue may in the near future be analyzed using nano-sensor devices, capable of generating hundreds of measurements per research participant. So while the results may be clinically insignificant at first, their clinical significance for the individual concerned may become rapidly clear over time. As the amount of data and the meaning to be attached to them will rapidly increase, the accumulation of genotype and phenotype data, in combination with the research results will become strong indicators of an individual participant’s health. This may warrant faster and more routine reporting of the resultant “personalizable” research results to individual participants. 4 Research on Biobanks is longitudinal, requiring long-term research participant involvement. It will involve continuous updating of participant’s (sensitive) health information and the provision of fresh biological samples. The participants may expect to benefit somehow from their long term contribution and one such benefit could be individual feedback of results.

1.4 Methods

To answer our research question we took a two-pronged approach. First, we wanted to ﬁnd out the attitudes and preferences of biobank researchers and participants in respect of reporting and receiving research results. Do participants want to be informed and are researchers willing to report their results to individual participants (social aspects of reporting biobank research results to individual participants). To that end we conducted empirical research to identify the information and communication attitudes and preferences of participants and researchers. For this empirical study, questionnaires were drawn up and submitted to three groups of stakeholders. Second, we wanted to ﬁnd out the legal rights and obligations of biobank researchers and participants in respect of reporting and receiving research results. Do participants have a right to be informed and are researchers under a legal obligation to report their results to the individual participants (legal aspects of reporting biobank research results to individual participants). The legal analysis involved a survey of international legal instruments, guidelines and policies, policies in use with existing biobanks, analysis of case law and statutes governing analogous situations. The legal analysis is presented

here in the form of a case study, followed by an assessment of whether participants ought to have a right to receive individual results. Both, the legal and empirical study, have been informed by our review of literature and our examination of deﬁnitional issues. As a preliminary issue, we tested our hypothesis that the rationale behind the traditional non-disclosure policies may no longer hold in the context of biobanks, by surveying developments in both research and the perspectives of participants.

1.5 Deﬁnitional issues 1.5.1 Biobanks

Quest for a definition Biobanks are not entirely new. The Framingham Heart Study for instance, which is considered the ‘mother of all biobanks’, dates back to 1948 (Framingham Heart Study). But what, exactly, is a biobank? Obviously what is included in the definition of a biobank has legal consequences as well as consequences for the preferences and attitudes of the stakeholders concerned (Cambon-Thomsen et al., 2007). Hence, a portion of our preparatory work was devoted to discussing the initial definition set forth in the proposal, i.e. biobanks are large-scale collections of human biological material of a representative part of a population, and associated genetic, proteomic, metabolic, clinical, health, genealogical, lifestyle and environmental data of, a large number of research participants, to be stored long term (Sallée and Knoppers, 2004; OECD, 2006). Examples in the Netherlands are the UMCG Groningen initiative LifeLines, (LifeLines) and the initiative of the academic-medical centres String of Pearls (String of Pearls). Variations in biobanks The various other terms that are used to describe these collections, such as ‘gene banks’, ‘tissue banks’, ‘DNA banks’, ‘biorepositories’ reflect not only their diversity but also demonstrate a lack of consensus on what exactly is a biobank (ALRC, 2005; IBM, 2004). Biobanks differ, inter alia, in terms of the population included (patients or unaffected individuals, specific or general community), the nature and size of the biological specimens included (blood, tissue, urine), the data included (genetic data, and/or clinical data, and/or health data, and/or genealogical data, and/or lifestyle data, and/or environmental data), the context of the collection (clinical or research setting), the form of storage of the sample, the underlying scientific purpose (screeningprograms, association-studies, genetic epidemiology, pharmaco-genetics) and funding (public, private or public-private) (Bovenberg, 2006). Variation in creation Biobanks also differ in the way they are being created. There are basically two ways

of creating a biobank; either de novo, or by converting one or more of the many pre-existing collections of human biological material and associated medical data into a biobank. In practice, most biobanks are ‘hybrids’ in that they combine novel collections of bodily material with existing data or the other way around (Bovenberg, 2006). Distinguishing features While no single deﬁnition can be given, biobanks can be distinguished by the following characteristics. Unlike most existing cohorts and patient-registries: • the information is both structured and extensive (OECD Draft Biobank Guidelines); • as a resource, biobanks are accessible not only to the initial researcher who built the collection, but to a whole range of researchers, both domestic and from abroad, both academic and commercial; • biobanks contain a combination of both genotype and phenotype information for long term multidisciplinary research; • the biobank often has the ability to link research ﬁndings to individual research participants and, in many cases, to participants’ blood relatives (Boggio, 2008). 1.5.2 Results

Different results What exactly, are biobank research results? Biobanks will generate results at various stages: enrolment measurements with evident clinical significance (blood pressure, cholesterol, BMI), biochemical measurements of the samples taken, and research findings. Wide variations exist between the different research results that may be derived from biobank research. A distinction is often made between the so called ‘aggregate research results’ (results that have a meaning for a group as a whole) and ‘individual research results’ (results that have consequences for the health and treatment of an individual). This distinction can however be challenged for genetic research, as aggregate results can be interpreted as individually relevant by participants (Miller et al., 2008). The so-called ‘incidental’ or ‘unexpected findings are also relevant; these are the findings that are discovered in the course of conducting research but are beyond the aims of the study in question (Wolf et al., 2008). Much of the current population-based research is not expected to reveal clinically relevant information yet, as it will focus on lower-penetrance gene variants and generate results at the population level that are rather exploratory than confirmatory (Beskow, 2006). Progressive continuum of results However, biobank research is expected to identify risk factors related to more than one disease (pleiotropy or common denominators) and the occurrence of various diseases within one individual (KNAW, 2006). Also, biobanks are set up to facilitate large-scale genomic epidemiology pursued as ‘discovery research’. In such discovery research any genomic pattern correlating with pathology may be captured and studied,

blurring the line between incidental findings and intended findings or research results (Wolf et al., 2008). Consequently, the ‘status’ or ‘meaning’ of results from biobank research is not fixed. At some point, the weight of existing evidence for a gene-disease association or a gene-environment interaction will mean that the next generation of studies will be confirmatory rather than exploratory (Beskow et al., 2008). We submit that at least some results, while lacking direct clinical utility, may serve to generate information that may be relevant to individuals’ health or reproductive choices. Even more importantly, while the results may not meet the threshold of analytical and clinical utility, the participants may perceive them as relevant health risk information and expect to be informed about them. Therefore, the concept ‘research results’ as used in this study includes not only unexpected, incidental findings, but also results that were aimed at in the research project concerned. Furthermore, it includes not only individual level results (i.e. observations resulting from examination of the participant or of his samples/related data) but also aggregate results (resulting from the analysis of (sub) populations) since also the latter may be of importance, at least to certain (sub)groups of research participants.

1.6 Hypothesis: Always Expect the Unexpected

As a preliminary issue, we tested our hypothesis that the rationale behind the traditional non-disclosure policies no longer holds. We submit that our hypothesis has been veriﬁed by the following subsequent developments. 1.6.1 Research perspective: oceans of translatable data

Linking genotype and phenotype Biobank research is expected to identify both risk factors related to more than one disease and the occurrence of various diseases within one individual, the so-called ‘common denominators’ (KNAW, 2006). The inclusion in biobanks of DNA material will allow genotyping to identify new genetic risk factors in the population at study. The number, the breadth, the diversity and the continuity of the samples and information in a biobank will allow the use of high-through-put approaches and technologies. These technologies will enable analyses of large numbers of samples simultaneously, and in record time and so generate large numbers of data to accurately describe patients (KNAW 2006). As evidence of the predictive value of these genetic variants accrues, investigators may face growing pressure to report findings that have an influence on risk (Bookman et al., 2006). Translational research and international standardization Biobanking is believed to enable two new methods of conducting biomedical research: translational research and molecular profiling (IBM, 2004; Gezondheidsraad, 2007). Research results of biobanks may rapidly become clinically significant as the

proliferation of data and the meaning to be attached to them rapidly increases with the use of nano-technology (Galas, 2006). In response to the need for international coordination and standardization, a number of standard setting or coordinating initiatives have emerged (P3G; ISBER; OECD Draft Guidelines, 2008). The combined effect of these efforts is to facilitate and speed up biobank research and related translational research, leading to faster applications in clinical practice. Hypothesis driven research > data-driven research The traditional policy of non-disclosure is based on the concept of hypothesis driven research. As noted in the Explanatory Notes to the Code Goed Gebruik, the chances of making incidental findings are considered to be minimal and the Code even prescribes that the research must be designed so as to minimize the chances of an incidental finding (Code Goed Gebruik). However, one of the consequences of the recent data-explosion in biomedical research is that (part of) the research has become data-driven rather than hypothesis-driven. Data driven research involves the correlation of millions of data-points by correlating molecular measurements of thousands of genes, proteins and metabolites with all sorts of clinical data (such as blood pressure, cholesterol, diabetes and MRI-scans). Subsequently or sometimes initially, these data are aggregated with data from databases resulting from comparable research data stored in other databases (LUMC, 2001). To the extent research is no longer hypothesis-driven, one of the rationales behind the classical policy seems no longer valid. Genome-Wide Association Studies A related aspect is the recent arrival and, partial, validation of Genome-Wide Association Studies (‘GWAS’). A GWAS study is defined by the US National Institutes of Health as a study of common genetic variation across the entire human genome designed to identify genetic associations with observable traits (NIH, 2007). For purposes of our hypothesis, the point here is that the GWAS approach permits interrogation of the entire human genome at levels of resolution previously unattainable, in thousands of unrelated individuals, unconstrained by prior hypotheses regarding genetic associations with disease (Hirschhorn and Daly, 2005; Pierson and Manolio, 2008). Even if the purpose of some whole-genome research is not to provide results about individual participants, the work may generate such results, ranging on a continuum from clinically significant information to information relevant to ancestry and genealogy to information that is merely of recreational interest (Caulfield et al., 2008). 1.6.2 Participant perspective: Great Expectations

Beneﬁt-sharing In addition to the above scientiﬁc considerations, biobank operators and researchers will have to address the information rights and information preferences and attitudes of those contributing samples and data to their resource: the research participants

and their families (HUGO, 2000, 2002; UNESCO, 2003, 2005). While their motives may be entirely altruistic, research, participants typically expect to gain health benefits from their participation in research. As they become more aware of their position as stakeholders in the collaborative research enterprise, donors of cells and tissues are increasingly abandoning altruistic participation and claiming a benefit-share in exchange for their contributions (Bovenberg, 2005). In the case of biobanks then, the participants’ long term, indeed life-time, involvement, coupled with the breadth and the sensitivity of the information they might provide, may raise these expectations of receiving benefits in the form of individual feedback of results, regardless of whether these were ‘expected’ or ‘unexpected’. Personal Biomarker and Genotyping Services The past couple of years have witnessed the introduction of a whole new series of ‘self tests’, ‘collect at home tests’ and ‘supermarket tests’, available directly to consumers (Raad voor de Volksgezondheid, 2008; www.diagned.nl; www.prescan.nl). In addition, 2007 saw the market introduction of personal genomics services, which claim to bridge the gap between peer reviewed and published findings on the one hand, and the individual, or at least, his genotype, on the other hand (Navigenics, deCodeMe, 23andMe). Using current scientific, published knowledge, these commercial suppliers analyze samples collected at home for individual predispositions for a variety of common conditions. They also use the scientific knowledge base to help individuals understand what their genes have to say about their future health and what steps they can take to prevent, detect or diagnose them early (Navigenics, deCodeMe, 23andMe). Participants may argue that if these companies can make sense out of the scientific literature for individual customers, then why can’t the scientists, who are the primary producers of the findings, do the same for them? Great expectations Sense and sensitivity of some of the above tests and services remain to be seen. Whether these commercial offerings can and will be applied in scientific biobank research also remains to be seen. The point is, however, that the widely advertized availability of both personal genotype and biomarker information, in unprecedented detail and breadth and direct to consumer, is likely to raise the level of expectations of those who participate in a biobank. These expectations will put additional pressure on biobank operators and investigators to translate and report “personalized” scientific findings to individual participants fast and routinely, rather than reporting incidental findings only.

1.7 Literature overview: bioethical discourse

Duty to disclose Questions have been raised regarding a generalized duty to disclose results obtained from research with biobanks to individual participants. In their seminal paper, Ravitsky and

Wilfond (2006) argue that the ethical principles of beneficence, respect and reciprocity justify routinely offering certain results to research participants. They subsequently delineate a framework to assist in deciding whether and how to offer individual genetic results to research participants. Their proposal presents an alternative for two polarized approaches regarding the disclosure of research results. The first approach is a general non-disclosure policy based on the argument that the primary goal of research is to produce generalized knowledge and not information about individuals. The alternative approach suggests that all results should be offered because participants should have a right to access information about themselves. In reactions to the Ravitsky and Wilfond paper, the extent to which ethical principles lead to an obligation, rather than a consideration, to disclose research results is challenged (Miller et al., 2008). Such obligation would lead to and/or maintain the therapeutic misconception, i.e. the blurring of the relevant distinction between research practice and clinical care (Meltzer, 2006; Dressler and Juengst, 2006; Ossorio, 2006). Additionally, the research results are considered too complex (e.g. risk information, variety of results in terms of severity and treatability of condition, pleiotropy) (Parker, 2006) and the costs too high to ensure proper disclosure, i.e. correct understanding by research participants (Ossorio, 2006; Klitzman, 2006). A researcher would not be the appropriate person to disclose but rather the treating physician (e.g. GP) or a genetic counsellor (Knoppers et al., 2006). Moreover, a fundamental lack of clarity exists about ‘what’ may constitute ‘research results’ (Miller et al., 2008). The most common and undisputed distinction is between aggregate findings (representing the sample of individuals) and individual results (findings regarding the person herself). Yet results on an aggregate level may be highly salient to an individual. Another hard to define distinction is between definitive (validated) and preliminary (invalidated) results (Miller et al., 2008; Dressler, 2006). Genetic associations with disease are often not validated in more extensive studies and could mislead participants to overestimate the significance of results (Knoppers et al., 2006). Finally, clinical validity of results is frequently called for but is again hard to define in the research context since it may diverge, relying on different appraisal processes or achieved at different stages of the research program (Miller et al., 2008; Facio, 2006). There is general consensus that empirical data are needed about what research results participants might prefer to receive, taking changes over time, motivation for participation and level of understanding into account. To some, it is an illusion that we will ever fully appreciate such preferences (Parker, 2006) to others it is important to query participants’ opinions to inform the informed consent procedure regarding the issue of disclosure. (Sharp and Foster, 2006; Manolio, 2006; Klitzman, 2006).

2 Preferences of (potential) participants and researchers: an empirical study 23

2.1 Introduction 2.2 Methods 2.3 Information preferences of citizens and patients 2.4 Attitudes of citizens, patients and researchers towards the communication of research results 2.5 Discussion 2.6 Interim con clusion empirical research regarding Guiding Principles

2. Preferences of (potential) participants and researchers: an empirical study

2.1. Introduction

This part of the investigation entailed a cross-sectional questionnaire study addressing (potential) participants and researchers’ information preferences and attitudes associated with the communication of results from DNA-banking. We asked citizens (as potential participants) and patients (as participants, as they had already donated blood or saliva to biobanks) about their preferences with regard to receiving various research results. Additionally, we asked citizens, patients and researchers about their views on the duty of researchers to inform DNA-bank participants about research results. Here we provide a brief summary of these questionnaire studies.

2.2 Methods

Sample Citizens (N=1163, response rate 78%) were consulted by means of the ‘Dutch Health Care Consumer Panel’ of the Netherlands Institute for Health Services Research (NIVEL). Patients (N=515, response rate 53%) comprised asthma, rhinitis and thrombosis patients from the University Medical Center Groningen (UMCG)1 and Leiden University Medical Center (LUMC)2, respectively. All patients had previously given informed consent for participation in research involving banking of tissue. Researchers were mainly recruited via a database composed by the Royal Netherlands Academy of Science and Art (KNAW) foresight committee ‘Multifactorial Disorders in the Genomics Era’, on behalf of their survey among principal investigators involved in the banking of tissue as related to common complex disorders. In total 164 researchers responded to our questionnaire (response rate 59%), nearly half of this group (N=80) had been involved in biobanking during the last year and was therefore eligible for analysis. Instrument To develop the questionnaire we carried out two preparatory focus group interviews: one with citizens and one with patients. We also interviewed researchers who were project-leader of a biobank, including small collections in academic hospitals and largescale repositories. Citizens’ and patients’ information preferences for receiving research results were assessed using six ﬁctitious research results (see table 1). They were instructed to answer the questions assuming they would contribute to a biobank, described as ‘long-lasting health research on genes and environment’. The results were formulated as ‘chance-results’, e.g. ‘research shows that people who have a speciﬁc mutation in their genes have a somewhat

1 2

UMCG: LUMC:

Prof.dr D Postma; Dr G. Koppelman Prof. F Rosendaal, Dr A van Hylckama-Vlieg

higher chance of getting stomach cancer than people who do not have this mutation’. It was explained that ‘a somewhat higher chance’ meant a 1% to 5% higher chance. The six fictitious results addressed the relation between a gene mutation and 1 a heightened chance of getting a disease, varied for the possibility of treatment to lower this chance, 2 a heightened chance that a specific medication would work well, varied for the situation that one had the disease or not, and 3 the heightened chance of a worse course of a disease, also varied for the situation that one had the disease or not. To also vary the seriousness of the implications, all six results related to stomach cancer (serious consequences) and to high blood pressure (less serious consequences), respectively. For the patients, results also addressed their own disease (asthma, rhinitis or thrombosis). Following the presentation of each fictitious result, respondents were first asked to indicate if they would like to be informed about the result on an aggregate level, described as ‘I would like to receive this general information’. Secondly, they were asked to indicate whether they wanted results on an individual level, described as ‘I would like to be informed if I have the mutation in the gene’. Both questions had a five point scale (definitely not (1), definitely yes (5)). Researchers received no questions about information preferences. Scores on the six items regarding the respondents’ preference for results on an aggregate level were summed, and a mean scale-score was calculated. The same procedure was followed for the items concerning respondents’ preference to know their mutation carrier status. Subscales were created (see table 1) to measure information preferences for fictitious results varying in availability of treatment, whether one had the disease or not and seriousness of the consequences. Citizens’, patients’ and researchers’ attitudes towards the duty of researchers to communicate research results were assessed using items as described in figure 1. The higher the score, the more respondents believed that researchers ought to communicate various kinds of biobank research results. Reliability of these instruments, as indicated by their internal consistencies, was satisfactory (Cronbach’s alpha >.89).

2.3 Information preferences of citizens and patients

For the individual items, a majority of 70%–78% of all respondents (citizens and patients) would like to receive information (‘probably’ or ‘deﬁnitely’) on aggregate research results (table 1). A minority of 13%-17% did not (‘probably not’ or ‘deﬁnitely not’) want to receive such general information. Most respondents preferred to receive the aggregate results via information letters (41%) or at meetings where results are orally explained (19%). Few wanted such information via a website on the internet (12%).

A majority of 66% - 88% of the respondents would ‘probably’ or ‘definitely’ like to be informed if they had the gene mutation themselves, whereas a minority of 5% - 20% would ‘probably not’ or ‘definitely not’ like to be informed. Respondents preferred to be informed about individual results by a physician connected to the research (34%), the family doctor (25%), the researchers themselves (8%) or a genetic counsellor (6%). Two percent wanted to have the ability to look at the results themselves. Respondents significantly preferred information on a) individual gene-mutations over aggregate results; b) the chance of getting a disease over information about efficacy of medication and course of a disease; c) the situation where treatment is known to lower the chance of getting a disease over the situation where the effect of treatment is unknown; d) results for medication-efficacy and course of the disease in the case they have the disease over the situation that they don’t have the disease; and e) results on high blood pressure over stomach cancer. It is of note, however, that all these differences are small, albeit significant (p<0,001). Influence of respondent characteristics (see appendix A) Citizens had a higher preference for receiving individual gene-mutation information than patients. Respondents (citizens and patients) who (would) contribute to a biobank to learn about their own health had a higher information preference compared to respondents who (would) contribute to a biobank for altruistic reasons, i.e. contribute to science or to be meaningful for others. Respondents who (would) participate because they were asked by the physician/researcher, had a lower information preference. Individuals with a monitoring coping style, that is the tendency to actively search for medical information, and who indicated a general desire for information about illnesses and treatment, more often preferred information. The stronger respondents believed that information about gene-mutations is meaningful to children or family, the more information they wished. Respondents, who considered gene-mutation information as potentially frightening, had a lower information preference. Within the group of citizens, those who did not want to contribute to a biobank had a lower information preference than those who did not know (yet) or did want to contribute.

Subscales ‘chance of getting a disease’: 1.1b, 1.2a, 2.1b, 2.2a; ‘treatment unknown’: 1.1b, 2.1b; ‘treatment known’: 1.2a, 2.2a; ‘medication-efﬁcacy’: 3b, 3c, 4b, 4c; ‘course of the disease’: 5b, 5c, 6b, 6c; ‘with the disease’: 3b, 4b, 5b, 6b; ‘without the disease’: 3c, 4c, 5c, 6c; ‘stomach cancer’: 1.1b, 1.2a, 3b, 3c, 5b, 5c; ‘high blood pressure’: 2.1b, 2.2a, 4b, 4c, 6b, 6c

Chance of getting a disease

Medication efﬁcacy

Course of the disease

Deﬁnitely yes

Probably yes

Maybe yes, maybe no

Statement

Probably not

Fictitious research result

Deﬁnitely not

Type of result

1.1. Suppose research shows that people with a mutation in their genes have a somewhat higher chance of getting stomach cancer than people who don’t have this mutation

a. I would like to receive this general information

11%

12%

33%

37%

b. I would like to be informed if I have the mutation in the gene

11%

28%

48%

1.2. Suppose supplemental research shows that, for people who have the mutation in their gene, a diet would lower the chance of getting stomach cancer

a. I would like to be informed if I have the mutation in the gene

28%

57%

2.1. Suppose research shows that people with a mutation in their genes have a somewhat higher chance of getting high blood pressure than people who don’t have this mutation

a. I would like to receive this general information

10%

32%

46%

b. I would like to be informed if I have the mutation in the gene

28%

55%

2.2.Suppose supplemental research shows that, for people who have the mutation in their gene, a diet would lower the chance of getting high blood pressure

a. I would like to be informed if I have the mutation in the gene

27%

59%

3. Suppose research shows that people who have a mutation in a certain gene have a slightly higher chance to react well to certain chemo therapy in the case of stomach cancer than people who don’t have this mutation

a. I would like to receive this general information

11%

32%

43%

b. I would like to be informed if I have the mutation, should I have stomach cancer

29%

57%

c. I would like to be informed if I have the mutation should, I not have stomach cancer

11%

13%

27%

44%

4. Suppose research shows that people who have a mutation in a certain gene have a slightly higher chance to react well to certain medication in the case of high blood pressure than people who don’t have this mutation

a. I would like to receive this general information

10%

32%

45%

b. I would like to be informed if I have the mutation, should I have high blood pressure

29%

58%

c. I would like to be informed if I have the mutation, should I not have high blood pressure

12%

11%

27%

44%

5. Suppose research shows that people who have a mutation in a certain gene have a slightly higher chance of a serious course of stomach cancer than people who don’t have this mutation

a. I would like to receive this general information

10%

14%

29%

40%

b. I would like to be informed if I have the mutation, should I have stomach cancer

11%

30%

50%

c. I would like to be informed if I have the mutation, should I not have stomach cancer

14%

13%

25%

41%

6. Suppose research shows that people who have a mutation in a certain gene have a slightly higher chance of a serious course of high blood pressure than people who don’t have this mutation

a. I would like to receive this general information

11%

31%

43%

b. I would like to be informed if I have the mutation, should I have high blood pressure

30%

55%

c. I would like to be informed if I have the mutation, should I not have high blood pressure

13%

26%

43%

Table 1: Responses related to the six ﬁctitious research results

2.4 Attitudes of citizens, patients and researchers towards the communication of research results

A majority of both citizens/patients and researchers were of the opinion that aggregate results should be communicated to participants. Items and percentages are depicted in ﬁgure 1. However, the attitudes of citizens, patients and researchers towards the communication of an individual result differed. Only 9% of the patients and citizens considered it acceptable when participants do not receive any information about mutations in their genes. Many of them consider information on results as a return for their contribution to a Biobank. In contrast, almost half of the researchers (40%) considered it acceptable when participants would not be informed about individual gene-related information. They deﬁnitely did not view such information as a natural return for participants’ contribution. A small majority of citizens and patients took the view that researchers must inform participants about individual gene-mutations, even when the health implications are unclear yet. In contrast, only few researchers agreed with this statement. Moreover, two-thirds of the researchers approved that participants only have to be informed about individual gene mutations when there are consequences in terms of treatment or prevention. Only about a third of the citizens and patients agreed with this position. Finally, almost half of the citizens and patients had the opinion that the more often participants donate data to a biobank, the more right they have to receive results whereas only few researchers had this opinion. 69

Researchers have to inform participants about general research results

I agree when participants do not get any information about mutations in their genes

40 9

Researchers have to inform participants about gene mutations in return for their contribution

4 50

Researchers have to inform participants about (all) mutations in their genes, even if the consequences for their health are not clear yet

1 59 74

Researchers have to inform participants about mutations in their genes only if treatment is available

The more often participants donate data during health research on genes and environment, the more they have a right to information about mutations in their genes

5 42 0

100

Researchers

Citizens and patients

Figure 1: Attitudes towards researchers’ duties to communicate research results (percentages agreeing with respective statements)

Inﬂuence of respondent characteristics on information preference (see appendix A) Citizens believed stronger than patients that researchers have a duty to communicate research results. Women more often than men endorsed this statement. Respondents who (would want to) participate in a biobank to learn about their own health ascribed more duties to researchers than respondents who (would want to) participate in a biobank for altruistic reasons. Attitudes on researchers’ duties were also positively associated with a monitoring coping style, a general desire for information about illnesses and treatment, and perceived meaningfulness of gene information for children or family. The more respondents anticipated to be anxious upon receiving information and the more they agreed with the statement that privacy will be harmed if gene-mutation information becomes available, the less they were of the opinion that researchers have a duty to inform. Finally, the more respondents believed that biobank participants retain ownership over their blood, the more they felt that researchers are obliged to provide information about research results.

2.5 Discussion

We found that most citizens and patients would like to be informed about aggregate findings and about a possible mutation in their genes. This high information preference is consistent with findings from other studies on the communication of research results of (genetic) research and clinical trials (Fong et al., 2006; Hoeyer et al., 2004; Partridge et al., 2005; Partridge et al., 2003; Wendler et al., 2002). Our study shows that, when asked, people on forehand want to know about any mutation in their genes. The question is why our respondents showed such a high information preference as it concerned relatively small increased chances. Respondents in a study by Wendler and Pentz (2007) indicated that the collection of test results on Alzheimer’s disease and cancer increased their desire to know the results because they did not want investigators to have information they themselves had not. We encountered this argument also in our preparatory focus groups; the mere existence and perceived availability of data supported respondents’ desire to have them. In a study on (genetic) health information, the general population expressed a right to choose whether to know health risk information and to control who else knows (Diergaarde et al., 2006). Another argument raised by participants in our focus groups was that a result might be meaningful to them ‘at some unforeseen point in time’ and participants did not want to have the feeling they ‘missed’ it. Such anticipated feelings of regret are known to motivate individuals to pursue genetic testing. Anticipated regret might explain why we found many respondents wanting to know research results on medication efficacy and course of the disease, even if they themselves did not have the disease of concern. It is of note however that a small but sizable group of respondents preferred no information about individual research results. We also found variation in information preference; respondents were less willing to know their mutation status when a) it is unknown

whether treatment is available, b) they do not have the disease the result relates to and c) when it concerns a more serious disease (stomach cancer). Likewise, in the study by Hoeyer et al. (2004), a small majority indicated that they would like to know genetic predispositions only if some kind of treatment or preventive intervention was available. We found that researchers are not so eager to communicate information about gene variations to individual participants. They are more inclined to provide such information when treatment or prevention is available. This is in line with the bioethical literature about the communication of research results and with existing guidelines. Though many citizens and patients indicated that they would like to receive various kinds of research results, fewer were of the opinion that researchers should make this possible. Nevertheless, a sizable group of 59% had the opinion that researchers have to inform participants about results with unclear health consequences. More than one third of citizens/patients took the view that researchers only have to communicate results when treatment is available. Clinical significance is an important argument in the debate on the communication of research results of biobanks. Our findings indicate that for (potential) participants not the health consequence per se, but the possibility of treatment is the most important argument in this regard. The opinion of (potential) participants diverges from the opinion of researchers; almost none of the latter agreed that results should be communicated with unclear health consequences and two thirds agreed that results should only be communicated when treatment is available. Our results should be interpreted within the limitations of the study. First, we spent much effort in developing the questionnaire to do justice to the complexity of the issues involved. Our questionnaire seemed relevant for citizens and patients as suggested by the satisfactory response rate. Our finding that answers to the information preference questions are associated with information preference in general and a monitoring coping style supports the validity of our approach. Nevertheless, we are aware that respondents may not fully understand the fictitious results, especially the implications of a gene-mutation result. In our view, we encountered the same difficulties as any researcher will when developing an informed consent procedure for recruiting participants for a biobank. Our recommendations therefore, also reflect our own experience in this regard. Secondly, our findings regarding differences between the fictitious cases should be interpreted prudently, as differences in mean-scores were very small and may be due to our large sample size.

2.6 Interim conclusion of empirical research regarding guiding principles

What are the implications of these results for the policy of biobanks regarding the communication of research results, specifically the communication of individual gene mutations? The fact that many (potential) biobank participants want to know about mutations in their genes (even when there are no immediate health implications) does not necessarily imply that they must receive such information, as other legal, ethical and practical arguments might overrule this wish. We conclude that on the basis of our results a biobank should give attention to the managing of 1. expectations, 2. diversity and 3. the gap in opinions between researchers and participants. 1 Managing expectations There is a large group of (potential) participants who would like to receive information about research results. More specifically, they are of the opinion that researchers have a duty to inform them. People who want to be informed about research results are more inclined to participate in biobanks than people who are less eager to be informed. Though we do not know what participants actually want in the situation that a specific result is being offered at a specific moment, we may conclude that expectations in this regard may be high. The communication policy of a biobank must therefore be very clear, open and transparent to manage these expectations. Together with the fact that information about risks is difficult to understand, the communication should preferably be an ongoing process. 2 Managing diversity We found some respondents not wanting to receive research results and others wanting to be informed about some results, but not all. From this, we conclude that there is a variety of information preferences which must be taken into account for various ethical arguments (right to know and not to know). This implies that participants of biobanks should have the opportunity to indicate their information preferences in the informed consent procedure. As information preferences may change over time, the biobank must make it easy for a participant to indicate such a change in preferences. For instance by consulting their participants at regular intervals and/or make a website where participants can indicate (a change in) their preferences. 3 Managing the gap between researchers and participants There is a gap between researchers on the one hand and a large group of (potential) participants on the other hand with regard to the communication of results relating to individual gene mutations. Both parties must be aware of each others expectations for the sake of the communication process. Researchers in particular must take this

discrepancy into account, when they communicate to (potential) participants as participants might need more and/or more frequent clariďŹ cation than expected. To verify that (potential) participants understand (the consequences of) the policy of a biobank, oral communication is recommendable.

3 Rights and obligations of researchers and participants (legal research) 3.1 Case study 3.2 Whom to sue? 3.3 Xâ&#x20AC;&#x2122;s brief 3.4 Biobankersâ&#x20AC;&#x2122; defence 3.5 Expected outcome of case study

3. Rights and obligations of researchers and participants (legal research) General principles are easier to state than to apply. In contrast to the debate to date, which is typically cast in abstract, conceptual and ethical terms, the legal research will subject the application to biobanks of the recent declaration of an ethical imperative to return research results to individuals (Fernandez et al., 2003), to a legal reality check. It will do so by exploring whether such an imperative could be legal, binding and enforceable in a court of law. A hypothetical court case (X vs. Biobank) was framed as a means to identify all relevant stakeholders and their interests and to unravel both the legal and the policy arguments pro and con a reporting obligation a judge is likely to consider when confronted with a claim brought by a research participant for individual feedback of research results. The case study presents the major elements and issues in a real life context, as it draws on the design of a number of existing biobanks.

3.1 Case study

Consider the following scenario. A 45 year old, healthy woman (X) receives an invitation letter from her Family Doctor (GP) to participate in a major study called the “Biobank”. The letter begins: L.S., Not everyone who is obese will develop diabetes, not every person with high blood pressure will suffer from renal failure and not every member of a family of asthma-patients actually develops the disease. How come?

To find the answer to this question, her GP explains to X, the study aims to unravel the ‘nature & nurture’ of common complex disorders, such as diabetes, cancer and Alzheimer’s. These complex disorders are thought to be caused by a large number of small, often additive effects, representing the outcome of the interplay, at various levels, of genes, lifestyle and the environment (Kaiser, 2002; Bouchie, 2004; Hagen, 2004; Manolio 2006). To reveal these complex interactions, the Biobank will collect and study genetic, clinical, biological, epidemiological and molecular information and corresponding blood and urine samples from a large number (some 150,000) of participants (patients and healthy persons) and their family members for a period of 30 years. The Biobank’s collection of samples and data is to serve as a resource by multiple researchers for multiple studies into a host of common complex disorders. Eventually, the researchers hope to find out what determines the effect of a universal risk factor for a given disorder in a particular individual, such as X. Having read the Participant Information Brochure and the consent form, X decides to join the study. At an appointment at the assessment centre, a nurse practitioner measures her height, weight, BMI, pulmonary function, bone density and blood pressure. A specially trained staff member collects 3 tablespoons of her blood and she provides a urine sample. She fills in a questionnaire, answering detailed questions about her

education, employment, physical activity, nutrition habits, general health condition, smoking and alcohol consumption, hospitalisations, diseases suffered, medicaments used, hormonal contraceptive preparations and menopause medicaments, and pregnancies. There are also questions about her nationality and native language, as well as detailed questions relating to her parents, grandparents’ and great grandparents. X signs a consent allowing the Biobank to re-contact her (e.g. to answer some more questions and/or to attend another assessment) and to monitor her health for the term of her participation, directly through her medical record and through other records that may be related to her health (e.g. occupational or residential information). She goes home with a print-out of her measurements, alongside ‘standard ranges’ to give meaning to these measurements. Halfway into the study, X has developed diabetes and suffers a heart attack. As cardiovascular diseases run in the family she is fearful of an imminent stroke. Her daughter has just been recruited into the study and was informed that both her cholesterol level and her blood pressure were abnormally high. Fearing that this may all be related and ‘genetic’, X contacts the Biobank. For some years now, the Biobank’s newsletters have alerted her to a whole series of scientific findings by a host of research teams pertaining to these disorders, all based on research on the Biobank and all published in reputable, peer reviewed journals. She reasons that the Biobank is in a perfect position to translate those findings to her individual situation. The Biobank has comprehensive and superior knowledge of her family’s medical history, lifestyle and genetic profile, for a long period of time. That knowledge, she believes, combined with the Biobank’s permanent access to her complete medical and health records, should enable the bank to report to her “any risk factor that you may know about me, and whatever prevention strategies you have to offer”. In addition, she reasons that, as a loyal, life-time participant, she is entitled to some sort of care, or at least information, pertinent to her individual present and future health. In response to her request, the Biobank refers to the Participant Information Brochure which states that the Biobank will not provide participants with individual information (genetic or otherwise) about results or findings derived from examination of the database or samples by research undertaken after enrolment. X does not recall ever having read this brochure, let alone having consented to this non-disclosure provision. After some fruitless correspondence between her lawyer and the Biobank, she decides to initiate legal proceedings, challenging the Biobank’s non-disclosure policy. She demands that the Biobank report to her all findings, both intended and incidental, pertaining to her present and future health and to use all information available to the Biobank to translate these findings to her individual situation. Specifically, she demands to receive, on an ongoing basis, (i) a report on any findings of potential health or reproductive importance which have been discovered in the course of any of the studies that were within the aims of these studies (Intended Findings),

(ii) a report on any findings concerning her that have potential health importance and which have been discovered in the course of any of the studies but that were beyond the aims of these studies (Incidental Findings), (iii) a translation of said findings (both Intended Findings and Incidental Findings) to X’s situation, i.e. an assessment of how her and her family’s medical and lifestyle data and history impacts on her life-time risk for these conditions and on her genetic predispositions for a variety of conditions, including diabetes and stroke (Odds Ratio’s). X demands to receive ‘the Full Picture’, i.e. all of the above information, regardless of whether the findings are exploratory rather than confirmatory, regardless of whether the risks indicated by the findings are imminently, life-threatening, high, moderate or low, regardless of whether the findings concern a condition that is treatable, actionable, have reproductive importance or are merely recreational, regardless of whether the findings relate to conditions that are late or early onset and regardless of whether the finding has analytical validity or clinical utility. She posts her complaint on the Biobank web forum, which in no time is filled with expressions of support from thousands of other participants, who are happy to join the lawsuit.

3.2 Whom to sue?

A complex, if crucial part of any lawsuit is the identification of the proper defendants. Unlike the traditional research setting pitting one or more trial subjects against a clinician-researcher, the Biobank presents a whole range of potential defendants. A stakeholder analysis of the Biobank reveals a multitude of individuals, entities and institutions, who can all be claimed to be engaged, to a larger or lesser degree and at some various points in time, in the collection, generation, assessment and processing of the health risk information X is seeking. Selecting a particular defendant has major legal implications as X’s relations with each of them may be classified differently, as each one of them may owe her different duties of care and as each one of them may be subject to different standards of care. Identification of the proper defendant then, requires a defendant analysis, which, in turn, requires a stakeholder analysis. In our scenario, X could sue (i) her GP, who recruited her into the study and who shares his medical files on her with the Biobank researchers, (ii) the Biobank and/or its employees, academic institutions doing research on the Biobank, (iii) commercial companies doing research on the Biobank, be they companies developing DNA or biomarker tests or pharmaceutical companies developing drugs or checking postmarketing surveillance data against the Biobank’s database, (iv) public-private partnerships doing research on the Biobank, (v) the academic laboratories and the commercial laboratories carrying out tests or performing measurements. As regards testing, even the choice for a particular micro-array could give rise to liability, as the chips used on those micro-arrays may or may not target specific area’s (Wolf et al., 2008). The particular governance structure of the Biobank may suggest additional

candidate defendants, such as (the individual members of) the Management Board, the Supervisory Board, the Scientiﬁc Commission, and the Governance Council, as they may all, at some point in time, have had actual or putative knowledge, or at least could have been alerted to pertinent ﬁndings and related health information for X or the class or subgroup to which she belongs. In the US, injured plaintiffs are even reported to try to impose liability on Medical Ethical Review Boards and their members (Wolf et al., 2008). X might also consider suing the Ministry of Health, in its capacity of the co-organizer and chief funder of the Biobank. For the purpose of this case study and for reasons of space, we will assume that X selects the Biobank (hereinafter the “Biobank”).

3.3 X’s Brief 3.3.1 Theory of the case

Legal theories To support her claims, X advances a variety of legal theories. First, she argues that her demands are grounded in her terms of participation in the Biobank as set forth in the consent form (Contract). Second, she maintains that the Biobank’s nondisclosure policy has become obsolete, in view of scientiﬁc, technological and societal developments (Invalidity of Non-Disclosure). Third, she invokes a number of statutory, constitutional and international human rights and professional norms the infringement of which could give rise to an action for negligence (Negligence). 3.3.2 Contract

1 Consent and Participant Information Brochure X’s relationship with the biobank is, primarily, governed by contract. She participates in the Biobank on the basis of her informed consent. Her consent is based upon the information she has received from the Biobank: the Participant Information Brochure (PIB). Collectively, the consent document signed by X and the PIB set forth the contractual terms and conditions of her participation. 2 Right to withdraw A core provision of the contract, is X’s right to discontinue her participation and to withdraw her consent. X admits that she has the right to withdraw any time, at will. But that is not the point. Her point is that her right to withdraw entails the right to be informed on any relevant personal findings, so she can make an informed decision whether she has reason to withdraw. Her willingness to continue to participate might be affected by significant new findings, expected or unexpected, developed during the course of the research, which may relate to her. The individual implications of these findings, no matter how qualified and limited in terms of analytical validity and clinical utility, may lead her to reconsider her initial consent and to drop out of the study. In support of her theory, X refers to US legislation. In the US, most academic

and institutional research in human subjects is governed by the Federal Common Rule for Protection of Human Research Subjects, commonly known as the Common Rule (Common Rule). The basic requirements of the Common Rule for informed consent do not stipulate that any study data, including genetic information, must be returned to study participants (Renegar et al., 2006). However, one of the ‘additional’ elements in informed consent regulations is: ‘A statement that signiﬁcant new ﬁndings developed during the course of the research which may relate to the subject’s willingness to continue participation will be provided to the subject’ (Common Rule, 45 CFR46.116(b)(5)).

According to X, her willingness to continue to participate does require that she is informed of significant new findings which may relate to her personal health (Renegar et al., 2006; Fernandez et al., 2003). 3 Construction of contract: raised expectations Another core piece of the contract is the Patient Information Brochure (PIB). X refers to a number of paragraphs of the PIB which have raised her expectation that she would be informed of personal information. According to the PIB, the ultimate objective of the study was to bring about the era of personalized medicine. More specifically, the PIB explicitly declares as its purpose: •

• •

•

To link genotype and phenotype: Biobank research aims to link abstract genomic data with concrete patient medical records. Biobank research is expected to identify both risk factors related to more than one disease and the occurrence of various diseases within one individual, the so-called “common denominators” (KNAW, 2006). The inclusion in biobanks of DNA material will allow genotyping to identify new genetic risk factors in the population at study. The number, the breadth, the diversity and the continuity of the samples and information in a biobank will allow the use of high-through-put approaches and technologies. These technologies will enable analyses of large numbers of samples simultaneously, and in record time and so generate large numbers of data to accurately describe patients (KNAW, 2006); To bring about era of personalized medicine; To promote translational medicine: Biobanking is believed to enable two new methods of conducting biomedical research: translational research and molecular profiling (IBM 2004, Gezondheidsraad, 2007). Research results of biobanks may rapidly become clinically significant as the proliferation of data and the meaning to be attached to them rapidly increases with the use of nano-technology (Galas, 2006); To find out what determines the effect of a universal risk factor in a specific individual. (emphasis added).

X argues that the above language provided for her a major incentive to participate in the study. For her, these objectives were part of the “deal”. The Biobank could have known, from their own consultation surveys among prospective participants, that a primary reason for participants to participate was their wish to find out about their own health, rather than contributing to the health of future generations. In addition, the widely advertized availability of both personal genotype and biomarker information (Navigenics, 23andMe, Personal Genome Project), in unprecedented detail and breadth and direct to consumer, has raised her expectation that the Biobank would translate and report “personalized” scientific findings fast and routinely. X’s expectations to be so informed must be deemed part of the contractual relationship and not informing her constitutes a breach of contract. 3.3.3 Invalidity of Non-Disclosure

X has to admit that, by signing the consent form, she has agreed to its “non-disclosure” policy, which reads as follows: Participants will NOT be provided with information (genetic or otherwise) about their own individual results or incidental ﬁndings derived from or made in the course of examination of the database or samples by research undertaken after enrolment.

According to X this policy is no longer valid, as its rationale no longer holds. The traditional policy of non-disclosure rests on a number of considerations, which include, but are not limited to: 1. the fact that most research findings in epidemiology or cohort studies are aggregate findings of an exploratory nature, with little or no analytical validity or clinical utility for the individual concerned, 2. the costs, competence and complexity involved in proper reporting to individuals and, 3. the fact that these findings ordinarily cannot be linked to identifiable participants (Beskow et al., 2001; Code Goed Gebruik, 2002; Knoppers et al., 2006; Boggio, 2008). Also, the policy is based on the traditional concept of hypothesis driven research. For this type of research, the chances of making incidental findings, i.e. findings discovered in the course of conducting the research but beyond the aims of the study, were considered to be minimal. Traditional disclosure policies even prescribed that the research must be designed so as to minimize the chances of an incidental finding (e.g. the Dutch Code Goed Gebruik). X argues that the above rationale has become obsolete, in view of the nature of characteristics of the research on the Biobank, and in view of developments in technology, science, medicine and society, described below.

1 Characteristics of Biobank research • The Biobank is designed so as to form a resource that will enable the conduct of not just one, but hundreds of different research projects into all kinds of diseases, some of which yet to be formulated, by a host of multi-disciplinary research teams rather than focusing on a specific disease (KNAW, 2006; Knoppers and Sallée, 2004); • The Biobank research is expected to identify risk factors related to more than one disease (pleiotropy or common denominators) and the occurrence of various diseases within one individual (KNAW, 2006); • The Biobank is set up to facilitate large-scale genomic epidemiology pursued as “discovery research”. In such discovery research, any genomic pattern correlating with pathology may be captured and studied (Wolf 2008); • The Biobank is data-driven rather than hypothesis-driven as a consequence of the recent data-explosion in biomedical research (Bovenberg, 2005); • The Biobank researchers have the ability to link research findings to individual research participants and, in many cases, to participants’ blood relatives (Boggio, 2008). 2 Technological developments X points to a number of new, high throughput technologies used by research on the Biobank, which are capable of generating massive amounts of information, such as low cost and ultra fast sequencing techniques and tandem mass spectrometry which can detect dozens if not hundreds of metabolic abnormalities (Wolf, 2008). These technologies have enabled the conduct of Genome-Wide Association Studies (GWAS). Genome-Wide Association Studies (GWAS) A GWAS study is defined by the US National Institutes of Health as a study of common genetic variation across the entire human genome designed to identify genetic associations with observable traits (NIH, 2007). The arrival and, partial, validation of GWAS permits interrogation of the entire human genome at levels of resolution previously unattainable, in thousands of unrelated individuals, unconstrained by prior hypotheses regarding genetic associations with disease (Hirschhorn and Daly, 2005; Pierson and Manolio, 2008). Reportedly, GWAS have important limitations, including their potential for false-positive and false-negative results and for biases related to selection of study participants and genotyping errors. But while they are clearly many steps removed from actual clinical use, specific applications of GWAS findings in prevention and treatment are actively being pursued (Pierson and Manolio, 2008). Even if the purpose of some genome-wide research is not to provide results about individual participants, the work may generate such results, ranging on a continuum from clinically significant information to information relevant to ancestry and genealogy to information that is merely of recreational interest (Caulfield et al., 2008).

3 ‘Translational’ developments According to the Biobank, most findings lack either analytical validity or clinical utility or both. X points out, however, that the past couple of years have witnessed the introduction of personal genomics services, which claim to bridge the gap between peer reviewed and published findings on the one hand, and the individual, or at least, his genotype, on the other hand (Navigenics, deCodeMe, 23andMe). Using current scientific, published knowledge, these commercial suppliers have developed genotyping and software tools to analyse samples collected at home, for individual predispositions for a variety of common conditions. They then use the scientific knowledge base to help individuals understand what their genes have to say about their future health and what steps they can take to prevent, detect or diagnose them early (Navigenics, deCodeMe, 23andMe). X argues that if these companies can translate the scientific literature into personal information, in a matter of weeks, then the Biobank, as a primary producer of such findings and with superior long term access to all relevant phenotype data, could do the same for her. 4 Scientific developments: P4 medicine X points to the emerging paradigm shift that biology is an informational science and that most biological information is mediated by dynamical biological networks (Hood, 2008): Systems Biology. The traditional approach to studying biology and human health could only yield relatively limited insights about the human body, as it focused on identifying individual genes, proteins and cells, and studying their specific functions. In contrast, a systems approach to biology and medicine has emerged (Systems Biology) which studies an organism, viewed as an integrated and interacting network of genes, proteins and biochemical reactions which give rise to life. The view of biology as an information science, the systems approach to disease, the new measurement and visualisation technologies have catalyzed this paradigm in medicine. P4 medicine Unlike traditional medicine, which is reactive and myopic, this new medicine is predictive, preventive, personalized and participatory (P4 medicine). P4 medicine examines the unique biology of an individual, to assess their probability of developing various diseases and then designs appropriate treatments, even before the onset of a disease. It uses more sophisticated measurements, as well as more measurements overall, thereby yielding accurate health assessments for truly personalized treatments. As a result of this personalisation, medicine will become participatory. Patients will actively participate in personal choices about illness and well-being .

X argues that for her to reap the beneﬁts of P4 medicine and especially to be empowered to participate therein, the Biobank ought to provide her with all information and measurements pertinent to her present and future health.

5 Societal developments Know Your Number and Scan Thy Self Every American knows his or her ‘number’, i.e. his or her (good or bad) cholesterol level. This habit to measure and monitor personal health indicators seems to have spilled over to the Netherlands. The past couple of years have witnessed the introduction of a whole new series of ‘self tests’, ‘collect at home tests’ and ‘supermarket tests’, available over (or sometimes, under) the counter (Raad voor de Volksgezondheid 2008, www.diagned.nl, www.prescan.nl). Consumers can measure their blood sugar and cholesterol levels, have their kidneys tested through an online kidney test (www. nierstichting.nl) and order biomarker tests to screen their blood and faeces for colon and prostate cancer (www.mirates.com). In addition to those ‘single disorder’ tests, the private sector has developed miniaturized “test batteries”, which enable the examination of hundreds of biochemical markers simultaneously and thus provide information about a broad range of (latent) disorders or abnormalities. deCodeMe As noted under the translational developments, personal genomics services have arrived to bridge the gap between peer reviewed and published ﬁndings on the one hand, and the individual, or at least, his genotype, on the other hand (Navigenics, deCodeMe, 23andMe). Naturally, the (clinical) utility of this type of “direct-toconsumer” susceptibility testing has been questioned (Hunter et al., 2008; Van Ommen, 2008). However, the notion of giving individuals their own genotype has not been entirely discarded by the scientiﬁc community (Nature Genetics, 2008). It would give them a personal stake in the ongoing research effort and could persuade them to participate in longitudinal research (Nature Genetics, 2008). The rapid move toward reporting all the abnormalities detectable by ms/ms, not just those of known clinical utility, was driven in large part by the view that everything that can be revealed by a technology must be sought and disclosed (Clayton, 2008). 3.3.4 Negligence

The third legal theory X could advance, is an action for negligence, in casu for failure to exercize due care. For a negligence lawsuit to succeed under Dutch civil law, X must establish that the non-disclosure policy of the Biobank amounts to a wrongful act or omission (DCC 6:162). To be wrongful, the Biobank’s non-disclosure policy must ﬁt into one or more of the categories of ‘wrongful acts’: 1. infringement upon a personal right, 2. an act or omission in breach of a statutory obligation or, 3. a failure to act conform an unwritten, generally accepted duty of care (DCC 6:162). Below, X cites a number of statutory, constitutional and international human rights and professional norms the infringement of which could give rise to an action for negligence (Negligence).

1 Right to Know Health Information X enjoys a number of personal rights. A personal right could be either a personal property right or a personality right (persoonlijkheidsrecht, Allgemeine Persönlichkeitsrechten). X asserts that she has a personality right to know health information. She refers to the Convention of the Council of Europe for the Protection of Human Rights and Dignity of the Human Being with regard to the application of Biology and Medicine (Oviedo Convention on Human Rights and Biomedicine). This convention covers all medical and biological applications concerning human beings, including preventive, diagnostic, therapeutic and research applications (Oviedo Explanatory Report). The Convention’s starting point is that the interests of human beings must come before the interests of science or society. The Convention stipulates that all patients have a right to be informed about their health, including the results of predictive genetic tests. Pursuant to article 10, “everyone has the right to respect for private life in relation to information about his or her health”. In addition, everyone is entitled to know any information collected about his or her health. According to the Explanatory Report, a person’s “right to know” encompasses all information collected about his or her health, whether it be a diagnosis, prognosis or any other relevant fact (Oviedo Explanatory Report). X also refers to the Additional Protocol to the Convention on Human Rights and Biomedicine. Article 13 of the protocol provides that, before being asked to consent to participate in a research project, the persons concerned shall be specifically informed, according to the nature and purpose of the research, of available preventive, diagnostic and therapeutic procedures; of arrangements for access to information relevant to the participant arising from the research and to its overall results. In addition, the protocol provides that research participants shall be entitled to know any information collected on their health in conformity with the provisions of Article 10 of the Convention. If research gives rise to information of relevance to the current or future health or quality of life of research participants, this information must be offered to them. That shall be done within a framework of health care or counselling. In communication of such information, due care must be taken in order to protect confidentiality and to respect any wish of a participant not to receive such information. As to the availability of results, article 28 of the Protocol provides that the conclusions of the research shall be made available to participants in reasonable time, on request. X then asserts that, to the extent either the Biobank findings contain information about her health, not disclosing them amounts to a breach of her rights under the Biomedicine Convention to know ‘any relevant fact’ collected about her health. And she points out that, unlike her Article 2 and 8 rights under the European Convention of Human Rights discussed below, this right is limited neither to information on “real and immediate risks” to her life, nor to information on risks which are treatable or actionable.

2 Standard of care for biobank research X argues that researchers owe participants a certain duty of care (Miller and Weijer, 2006 and Richardson and Belsky, 2004). X claims that the appropriate standard of care in the context of large-scale observational Biobank research are offered in the International Guidelines for Ethical Review of Epidemiological Studies issued by the Council for International Organizations of Medical Sciences (CIOMS 1991 and CIOMS 2007). CIOMS In the preamble to the CIOMS 1991 Guidelines, the drafters acknowledge that the general conduct of biomedical studies is guided by statements of internationally recognized principles of human rights, including the Nuremberg Code and the World Medical Association’s Declaration of Helsinki. They reason, however, that these and similar national codes are based on the model of clinical medicine, and often address interests of “patients” or individual “subjects”. Epidemiological research, however, concerns groups of people, and the above codes do not adequately cover its special features (CIOMS 1991). The aim of the Guidelines is to apply the four basic ethical principles governing all research involving human subjects: respect for persons, beneficence, non-maleficence, and justice (CIOMS 1991). Respect for persons incorporates autonomy, which requires that those who are capable of deliberation about their personal goals should be treated with respect for their capacity for self-determination. Beneficence is the ethical obligation to maximize possible benefits and to minimize possible harms and wrongs. This principle gives rise, inter alia, to norms requiring that the investigators be competent both to conduct the research and to assure the well-being of the research subjects (CIOMS 1991). Non-maleficence (“Do no harm”) holds a central position in the tradition of medical ethics, and guards against avoidable harm to research subjects. Justice requires that cases considered to be alike be treated alike, and that cases considered to be different be treated in ways that acknowledge the difference. When the principle of justice is applied to dependent or vulnerable subjects, its main concern is with the rules of distributive justice: the class of persons bearing the burden should receive an appropriate benefit.

The preamble to the CIOMS Guidelines provides that part of the benefit that communities, groups and individuals may reasonably expect from participating in studies is that they will be told of findings that pertain to their health (CIOMS 1991). In informing individuals of the findings and their pertinence to health, their level of literacy and comprehension must be considered. Research protocols should include provision for communicating such information to communities and individuals (CIOMS 1991). In addition, research findings and advice to communities should be publicized by whatever suitable means available. This may entail that, where feasible, specific testing and individual counselling ought to be available, for example when HIV-prevalence studies are conducted by unlinked anonymous screening (CIOMS 1991). And when findings indicate a need of health care, those concerned should be advised

of means of obtaining personal diagnosis and advice. 2008 Draft “International Ethical Guidelines for Epidemiological Studies”. The 1991 Guidelines are currently being updated. In February 2008 the CIOMS, in collaboration with the World Health Organization (WHO), published a pre-final draft called the International Guidelines for Epidemiological Studies (CIOMS 2008). Pursuant to Guideline 5 the investigator must inform, prior to requesting an individual’s consent to participate in research, the individual that, after the completion of the study, subjects will be informed of the findings of the research in general, and individual subjects will be informed of any finding that relates to their particular health status. In addition, prospective participants must be informed that they have the right of access to their data on demand, even if these data lack immediate clinical utility. As noted in the Commentary to Guideline 4, when a study employs anonymization, which makes it impossible to notify individuals (and, in some cases, even identifiable groups of subjects) of research findings or personal test results, the ethical review committee should take this into account in deciding whether to approve the study in the first place (CIOMS Draft 2008, Commentary). Even when they have not anonymized the data, epidemiologists have often not notified individual subjects of test results. In light of contemporary standards for informed consent, however, epidemiologists should make subjects aware of findings that are clinically relevant to their individual health. In all cases, the extent to which findings will be disclosed to subjects as a group or as individuals should be clearly conveyed in the informed consent material (CIOMS Draft 2008, Commentary). CIOMS CIOMS

3 Right to Know Research Results Under Dutch law, there is no enumerated statutory right to receive personal feedback of results from observational research. According to X, however, her situation can be analogized with the situation of human subjects in clinical trials, which are protected by the Dutch Act on Medical Scientific Research involving Human Subjects. Notably, neither the Act nor the EC Directive on which the Act is based, grant participants an affirmative right to receive the results of the trial, whether at an individual level or aggregate findings. However, the Act does provide that participants in clinical research enjoy certain information rights during the trial. Specifically, in the event the trial has unsuspected serious events that turn out to be more adverse than foreseen in the research protocol, the investigator must immediately notify the trial subject. In addition, article 10 of the Act provides that the investigator is responsible for informing the subject about the course of the trial. 4 Right to Know Medical Record As a patient, X has a statutory right to know, inter alia, all the health information kept by her treating physician in her medical record. That would include information about diagnosis, test results, prognosis, risks and treatment (WGBO). Obviously, this statutory right applies to the traditional patient-physician relationship in the traditional clinical

setting. However, X now asserts that this right should be extended to the Biobank setting at hand, for the following reasons. First, as shown in the defendants’ analysis, the Biobank forms an integral part of a series of academic hospitals. In a sense, the Biobank can be seen as an extension of the healthcare infrastructure available to X’ clinician and therefore is under an obligation to provide her with all relevant health risk information. Second, X has been paying repeat visits to the Biobank for repeat donations of blood and urine and repeat examinations. Third, a number of the examinations, analyses and diagnoses performed by the researchers during and after those visits do qualify as medical examinations. Fourth, research on her samples and data in the context of the Biobank has not been an isolated, incidental and one-time affair, performed by a post-doc in a remote university lab. Rather, the line between research and care has been blurred by the fact that the Biobank has had permanent, 24/7 access to her health records, family history and life-style data for ongoing research. In this setting, X asserts, she has a right, similar to her rights as a patient, to receive pertinent personal health information from the Biobank. 5 Right to Benefit X refers to a series of international instruments in the area of biomedical research that all call for benefit sharing (HUGO 1998, 2003, UNESCO 2003, 2005, CIOMS 1991). For example, pursuant to Article 12 of the UNESCO Declaration 1997, benefits from advances in biology, genetics and medicine, concerning the human genome, shall be made available to all, with due regard for the dignity and human rights of each individual. The applications of research, including applications in biology, genetics and medicine, concerning the human genome, shall seek to offer relief from suffering and improve the health of individuals and humankind as a whole. In addition, the International Declaration on Human Genetic Data (UNESCO 2003), which specifically applies to the collection, processing, use and storage of human genetic data, human proteomic data and biological samples, requires that the benefits resulting from the research on the data be shared with the society as a whole. And article 3 of the Universal Declaration on Bioethics and Human Rights (UNESCO 2005) provides that, in advancing scientific knowledge, direct and indirect benefits to patients, research participants and other affected individuals should be maximized. Pursuant to Article 15, benefits resulting from any scientific research and its applications should be shared with society as a whole and within the international community, in particular with developing countries. In giving effect to the principle of benefit-sharing, benefits may take, inter alia, the following forms: (a) special and sustainable assistance to, and acknowledgement of, the persons and groups that have taken part in the research; (b) provision of new diagnostic and therapeutic modalities or products stemming from research; (c) access to scientific and technological knowledge; (d) capacity-building facilities for research purposes.

X argues that her participation in the Biobank amounts to a significant contribution. In addition to providing, bi-annually, blood and urine samples, she frequently fills in questionnaires, answering detailed questions about her education, employment, physical activity, nutrition habits, general health condition etc. She has also allowed the biobank access to her health records and other records that may be related to her health. This contribution to science deserves a proper quid pro quo (CIOMS 1991, Preamble). X maintains that, in consideration of her loyal, life-time participation and her granting access to a whole range of sensitive data for a prolonged period of time, the Biobank owes her a benefit in the form of personal feedback. 6 X’ Right to Life X also invokes her personality right to life. Notably, the right to life is not enumerated in the Dutch Constitution. It is, however, laid down in article 2 of the European Convention on Human Rights (ECHR), which provides that ‘everyone’s right to life shall be protected at law. No one shall be deprived of his life intentionally’ (ECHR article 2).’ As well as a negative obligation not to take anyone’s right, this imposes a positive obligation on the state and public authorities to protect the right to life (Johnston and Kaye, 2004). Applying the Convention to the non-disclosure policy of UK Biobank, it has been argued that article 2 ECHR could impose a positive obligation on UK Biobank to put in place measures to avoid a risk to the lives of the participants (Johnston and Kaye, 2004). More specifically, they argue that this positive obligation entails the provision of feedback of risk of a serious genetic disease, if this is revealed during the course of the research project (Johnston and Kaye, 2004). In support of their theory, Johnston and Kaye refer to the case of Osman vs. the uk (Osman, 1997). In this case, Mrs Osman and her son complained that the police had failed to act on the warning signs that her son’s former schoolteacher was a serious threat to the physical safety of Mr Osman and his family. Having extensively stalked and harassed the son and the family, he eventually killed the father and wounded the son. The ECHR stated the following principles. It noted that it was not disputed that article 2 of the Convention might in well-defined circumstances imply a positive obligation on the authorities to take preventive operational measures to protect an individual whose life is at risk from the criminal acts of another individual. It considered that, bearing in mind the difficulties involved in policing modern societies, the unpredictability of human conduct and the operational choices which must be made in terms of priorities and resources, any such obligation had to be interpreted in a way which did not impose an impossible or disproportionate burden on the authorities. Accordingly, for the Court not every claimed risk to life could entail for the authorities a Convention requirement to take operational measures to prevent that risk from materializing. The court also considered the need to ensure that the police exercised their powers to control and prevent crime in a manner which fully respected the due process and other guarantees which legitimately placed restraints on the scope of their action to investigate crime and bring offenders to justice.

Where there was an allegation that the authorities had violated their positive obligation to protect the right to life in the context of their duty to prevent and suppress offences against the person, it must be established that they knew or ought to have known of the existence of a real and immediate risk to the life of an identified individual, from the criminal acts of a third party. It must also be established that they failed to take measures within the scope of their powers which, judged reasonably, might have been expected to avoid that risk. It is sufficient in this regard for an applicant to show that the authorities did not do all that could be reasonably expected of them to avoid a real and immediate risk to life of which they had or ought to have knowledge, a question which could only be answered in the light of all the circumstances of any particular case. X asserts that, to the extent the Biobank knows or ought to know of the existence of real and immediate risk to her life or those of her daughter and her grandchildren, she can reasonably expect the Biobank to take measures to avoid that risk, as the Biobank has or ought to have superior knowledge and access to all relevant records, and thus is uniquely equipped to (offer to) report and translate their scientific findings into an assessment of her individual risks. 7 X’ Right to Health Next, X invokes her personality right to health. Such a right is not laid down in the Dutch Constitution. Failure to protect a person’s health, however, may amount to a breach of the right to respect for one’s private life, set forth in Article 8 of the European Convention of Human Rights (Johnston and Kaye, 2004). ‘Private life’ includes not merely a right to control personal information, but also protection of privacy interests in physical and moral integrity (X and Y v. the Netherlands, Laurie 2002). The right imposes on the state both negative and positive obligations, including a right to have assistance in the fulfilment and enjoyment of one’s private life (López Ostra v. Spain - 16798/90 [1994] ECHR 46 (9 December 1994, Guerra and Others v. Italy, Laurie 2002). In the case of Guerra and others vs. Italy, the applicants lived one kilometre away from a chemical factory, which was classified as “high risk” according to the criteria set out in an EC Directive 82/501/EEC on the major-accident hazards of certain industrial activities dangerous to the environment and the well-being of the local population. Ascertaining whether the national authorities had taken the necessary steps to ensure effective protection of the applicants’ right to respect for their private and family life as guaranteed by Article 8, the European Court of Human Rights reiterated that severe environmental pollution may affect individuals’ well-being and prevent them from enjoying their homes in such a way as to affect their private and family life adversely (Lopéz Ostra v. Spain). The court considered that Guerra and others waited, right up until the production of fertilisers ceased in 1994, for essential information that would have enabled them to assess the risks they and their families might run if they continued to

live at Manfredonia, a town particularly exposed to danger in the event of an accident at the factory. The Court held, therefore, that the respondent State had not fulfilled its obligation to secure the applicants’ right to respect for their private and family life and had breached Article 8 of the Convention. X argues that, to the extent that the Biobank possesses essential information that would enable her to assess the risks she and her family might run, failure by the Biobank to disclose such information would constitute a violation of her ‘right to health’. 8 Disclosure of findings is supported by international precedent X refers to the Estonian GeneBank, a project similar to the Biobank. The Estonian GeneBank is a national GeneBank consisting of tissue samples, descriptions of DNA, descriptions of state of health, genealogies of the Estonian population. The objective of the Estonian GeneBank is twofold: ( i) to enable gene and health research to find genes that influence the development of illnesses and (ii) to provide a gene donor with an opportunity to assess his or her health risks and diagnose illnesses more precisely, prevent falling ill and receive more effective treatment in the future (Estonian GeneBank). Gene donors participating in Estonian GeneBank have the right to access personally their data stored in the Gene Bank, except for their genealogies, for free (Human Genes Research Act). In addition, the Estonian gene donors have the right to genetic counselling upon accessing their data stored in the GeneBank (Human Genes Research Act). 9 Disclosure of findings is advocated in the international literature X refers to a number of ethical and legal commentaries advocating broad disclosure of research results to research participants (Ravitsky and Wilfond, 2006). Commentators have noted that, as evidence of the predictive value of genetic variants accrues, investigators may face growing pressure to report findings that have an influence on risk (Bookman et al., 2006). Also, a collaborative clinical research regime, the Informed Cohort, has been proposed, wherein subjects are given a Web-based, interoperable personally controlled health record. Thus each patient owns an integrated copy of his or her traditional record data, plus high-throughput genome scale measurements made on his or her biomaterials (Kohane et al., 2006; Wolf, 2008). The claim that recognizing a researcher’s duty to handle incidental findings responsibly and to disclose them to research participants, is consistent with recent trends in US law. Such an obligation is not only consistent with the legal recognition of a researchers’ special obligation towards participants, but also with legal doctrines imposing a duty to warn of foreseeable harm. Based on their ethical, empirical and legal analysis, Wolfe et al. propose the following recommendations for managing incidental findings in human subjects research: 1. researchers must address incidental findings in the consent process; 2. they must plan for the discovery of incidental findings; 3. they must take steps to verify and evaluate a suspected incidental findings, with an

expert consultant if needed; 4. they must plan to determine whether to report an incidental finding, based on likely health or reproductive importance; 5. incidental findings with a strong net benefit ought to be disclosed, incidental findings with a possible net benefit may be offered and incidental findings with an unlikely net benefit should not be offered to research participants (Wolf, 2008). In conclusion, the (contractual) terms of participation, the characteristics of biobank research, its stated purpose and raised expectations of bringing about the era of personalized medicine, the biobank’s academic-clinical setting, developments in technology, science, society, statutory, constitutional and international human rights, the international standard of care for observational research, coupled with X’s significant contribution to the Biobank, all provide powerful arguments for her claim to receive findings pertinent to her health. In addition, these are powerful arguments not to limit any disclosures to those findings which relate to real, imminent and treatable threats to life, but all findings pertaining to her present and future health.

3.4 Biobankers’ defence 3.4.1 Preliminary defence: all data have been anonimized

Anonimization By way of preliminary defence, the Biobankers assert that the claim of X should be dismissed as a matter of fact. Even if a court were to ﬁnd some sort of obligation to provide feedback to individual participants, the feedback of information (genetic or otherwise) about their own individual results derived from examination of the database or samples by research is impossible. It is true that a wide range of tests has been done on X’s blood and urine samples. However, in accordance with the Biobank Research protocol, prior to those tests and studies, all identifying information has been removed. Most, if not all of the information requested cannot be traced back to X. 3.4.2 Defence against contractual liability

Clear policy The contractual relationship between X and at least one of the Biobankers (the Biobank) is formed by the consent form and the Participant’s Information Brochure (PIB). The PIB contains a clear explanation of the Biobank’s disclosure policy. By signing the consent form X has agreed to this policy, the terms of which are crystal clear and not in need of further interpretation. The contractual relationship between X and the Biobankers is governed by these express terms. The relationship does not ﬁt into any category of statutorily enumerated contracts. If anything, the contractual relationship between research participants and institutions and researchers has been deﬁned in US case law as an “unconditional gift(s) without expectation of a contemporaneous

return” (Greenberg vs. Miami Children’s Hospital Research Institute, Inc., Washington University vs. William Catalona et al.). Rationale for non-disclosure policy still valid Assuming that the Biobank researchers could legitimately adopt the dual role of clinicians, then their conduct, i.e. the offering of research results to an individual such as X, would be subject to the standard of care (WGBO, BIG). In order to meet this standard of care, the provision of feedback of (genetic) results to X would have to be evidence-based. Specifically, results should be analytically valid, clinically valid and have clinical utility (Ravitsky and Wilfond, 2006; Beskow et al. 2001). Or, as a Dutch researcher put it: It would be a disaster if we, as researchers would give genetic data back at an individual level without a proper context. Say, a single nucleotide polymorphism (SNP) contributes to asthma development for about 4% in a population based study (a risk that most SNPs confer in a complex diseases like asthma). If someone has this SNP, it simply does not tell the whole story as whether this person would develop asthma. Environmental exposures interact with this gene most likely to cause that the disease becomes clinically manifest, i.e. with respiratory symptoms. When not having this SNP does not tell in any way that someone will not develop asthma as well. Indeed SNPs in other genes may contribute (even greater) risks to a specific environmental factor (e.g. smoke exposure in utero) and this will determine the outcome. Thus the SNPs alone are not sufficiently specific to determine on an individual level whether someone develops disease. Providing an individual with such information on a SNP basis will either give an unrealistic, und unwarranted lifelong fear to develop disease, or the other way around an unjustified feeling of security that one will not develop disease.

The assessment of analytical validity would require the performance of independent conﬁrmatory testing. Clinical validity refers to the quality and quantity of empirical evidence regarding the association between a genotype and a particular clinical outcome (Ravitsky and Wilfond, 2006). The interpretation of reported associations requires a chain of evidence substantiating the validity of the association found in a single initial study (Beskow et al. 2001; Code Goed Gebruik). Results that do not meet this basic prerequisite do not constitute “information”. To oblige the Biobankers to offer these results to participants would force them to lower the threshold for clinical validity and thus to violate the standard of care and expose themselves to civil and criminal liability (compare Rechtbank Zwolle, holding radiologists performing population screening to the (WGBO) standard of clinical care). 3.4.3 Defences against negligence

No violation of professional standard for researchers The legal duty of care for the Biobankers must be determined by the research context. They do not have the same duty of care that they would have in a clinical setting. The appropriate standard of care governing the conduct of the Biobankers is set forth in

the Code Goed Gebruik (Code Goed Gebruik, 2002). It is considered undesirable to report new findings or results to individual participants (KNAW, 2006). International standards non binding The international instruments invoked by X are either non-binding (UNESCO, WMA, CIOMS) or non enforceable, as they have not been ratified (Council of Europe). They are directed not to individuals, whether private citizens or legal entities or corporations, but to the member states of the UN or the Council of Europe, which are ‘encouraged’ to take appropriate steps to implement the principles laid down in these declarations. International standards apply to intervention research, not to observational research Most of the international instruments invoked by X do not apply to the observational, epidemiological research that is being pursued by the Biobankers. Most instruments apply to interventional, clinical research. The Council of Europe Additional Protocol (2005) covers the full range of research activities in the health field involving interventions on human beings. For the purposes of this Protocol, the term “intervention” includes: (i) a physical intervention, and (ii) any other intervention in so far as it involves a risk to the psychological health of the person concerned. Council of Europe recommendation (Counsil Rec, 2006) on research on biological materials of human origin does indeed apply to both, research using biological materials kept at hospitals, and research using biological materials kept in population biobanks. Notably, however, it does not contain a provision on disclosure of the results from this type of research. The Recommendation even puts severe limitations on the type of screening to be done in such biobanks, by providing that screening for serious late onset diseases for which there is no treatment should remain exceptional, even when screening is related to scientific research: it would put too much strain on the free participation and on the privacy of individuals (Council of Europe Rec 2006). International standards conflate clinical care with research X’s invocation of the above international instruments implies that the duty of care applies to physicians equally in clinical practice and in clinical research. This implication reflects a pervasive failure to appreciate a basic and morally salient distinction between clinical practice and clinical research (Miller, 2003; Brody, 2003). The goal of clinical practice is improvement of the individual health of patients. In contrast, the goal of clinical research is development of generalizable knowledge for the benefit of society and future patients (Miller, 2003; Brody, 2003). The principles underlying the provision of feedback in the clinical trial setting are still in their infancy and they should not automatically extend to populations based research. On the contrary, some of those principles, such as respect for the person, seem to caution against reporting results which need replication and which typically lack clinical utility (Bookman, 2006). Imposing an obligation on (biobank) researchers to

feedback individual results would reinforce the ‘therapeutic misconception’ held by participants and sometimes also by researchers (De Vries and van Leeuwen, 2008) International standards limit reporting obligations Even if the declarations were enforceable and their scope to be expanded to include observational research on biobanks, they still clearly do not provide for an unconditional obligation on the part of scientists to report results to individual participants. • Article 10 of the International Declaration on Human Genetic Data (UNESCO 2003) explicitly provides that it does not apply to research on data that are irretrievably unlinked to identiﬁable persons or to data that do not lead to individual ﬁndings concerning the persons who have participated in such a research.

• The CIOMS 2002 Guidelines allow that an ethics committee may approve the nondisclosure of the data for a stated reason that will, itself, be given to the participant (CIOMS 2002). Reasons not to inform the patient are not specified but could include: lack of relevance of data, limitations of predictive capability of research data, concerns of misinterpretation by the participant, absence of ‘good clinical practices’ standards in exploratory research, or lack of feasibility (e.g., data are anonymized) (Renegar, 2006). Also, the duty to inform research subjects of any finding that relates to their particular ‘health status’ at the end of the study is open to wide interpretation – it is not clear that it would include most polygenic determinants of disease susceptibility, even if they had been validated (Renegar, 2006). • The CIOMS 1991 Guidelines also recognize that it is sometimes impossible to communicate study results. Often it may not be possible to extract from pooled findings information pertaining to individuals and their families. Therefore, it provides that subjects of epidemiological studies should be advised that it may not be possible to inform them about findings that pertain to their health, but that they should not take this to mean that they are free of the disease or condition under study. • The CIOMS draft 2008 seems to impose a higher standard: prospective participants must be informed that they have the right of access to their data on demand, even if these data lack immediate clinical utility. However, they too allow for an ethical review committee to approve temporary or permanent non-disclosure of data, provided the subject is informed of, and given, the reasons for such non-disclosure. One of these reasons may be the scale of a particular study. Investigators planning not to disclose obtain approval from the ethical review committee (CIOMS draft 2008).

International standard of benefit sharing may amount to improper inducements X bases her claim in part on her right to share in the benefits of the study in which she participates. Her claim is supported by some of the above international instruments (UNESCO 1997, 2003, 2006; HUGO 2000, 2002) and by a host of legal commentaries (Greely, 2001; Terry, 2003; Simm, 2005). However, the same instruments calling for benefit-sharing never fail to include language that prospective participants may not be lured into participation with improper inducements. The net worth of individual

feedback from the Biobankers can be estimated by comparison with the price of similar types of information offered by commercial enterprises; they currently range from $985 (deCodeMe) to $ 2,500,- (Navigenics). As a primary motive of participants in biobanks is the desire to ﬁnd out about personal health (Lifelines, 2007), then the right to receive individual feedback, in an academic setting could qualify as an inducement. If the Biobankers were obliged to provide this type of feedback to their participants, they might be accused of providing undue inducements and violating international and national standards. Foreign examples The Biobank’s policy of non-disclosure is further supported by a number of similar provisions in similar biobanks overseas. In the Singapore Tissue Network, neither donor nor doctor will receive the results of research done with donated samples (Singapore Tissue Network). The UK Biobank will not provide individual feedback to participants of results obtained through the research process for any reason (UK Biobank EGF). Participants in the Canadian biobank CartaGene will not receive any individual research results, unless they have opted to receive a document containing the measurements taken during the enrolment visit (CartaGene). Generation Scotland will give participants health information on some important clinical measurements such as blood pressure, cholesterol and kidney function, but no personal genetic information (Generation Scotland). 3.4.4 Defence against negligence: justiﬁcations for not reporting

No violation of the right to life The European Convention on Human Rights is directed at States and does not impose positive or negative obligations on individuals. It is true that the European Court of Human rights in Oslam found a positive obligation for the State to safeguard lifes, but it also held that this obligation must be interpreted in a manner that does not pose an impossible or disproportionate burden. Complexity, costs and consent to know Likewise, the Dutch Civil Code provides that no act or omission is wrongful in the event of ‘force majeure’ (overmacht), i.e. circumstances which justify an otherwise actionable omission. The Biobankers could advance a number of justiﬁcations which can be summarized as follows: complexity, costs, competence and consent to know. Complexity: the Incidentalome To support their claim that granting X’s request would impose a disproportionate burden and set a precedent for other participants to follow suit, the Biobankers submit expert witness testimony on the occurrence of the ‘incidentalome’. The broad array of new genome-scale tests has indeed led to the discovery of multiple abnormal or ‘unexpected ﬁndings’, analogous to the ‘incidentalomas’ that are often discovered in radiological studies (Kohane et al., 2006). The application of comprehensive genotype

and functional genomic measurements across the general population is likely to yield incidental findings for nearly everyone. Any large-scale genomic panel is therefore likely to routinely report false-positive results. In a series of hypothetical scenario’s, it is demonstrated that, as the number of genomic tests increases to 10,000, the fraction of the population that has a false-positive test result increases to more than 60% (Kohane et al., 2006). And even if genomic tests were to achieve 100% sensitivity and a falsepositive rate of 0, the risks of the incidentalome remain and will lead to iatrophic pathology. That is, these real findings without realistic clinical importance can lead to aggressive diagnostic and therapeutic investigations in an otherwise healthy individual (Kohane et al., 2006). The consequences of an incidentalome for biobank researchers are obvious. Biobank researchers will be overwhelmed by the complexity of pursuing unexpected genomic measures. To oblige them to look beyond the variables under study to potential unexpected findings of potential clinical significance for individual participants would place on them a disproportionate burden. And if the Biobankers were held to be under an obligation to report, X would be subjected to unnecessary follow-up tests and the cost of genomic medicine will increase substantially with little benefit to patients or physicians, but, with great financial benefit to the genomic testing industry (Kohane et al., 2006). Costs Assuming the Biobankers are competent (BIG) and qualified (WMBZ) to report results to individual participants, there still is the question of who will bear the costs associated with the provision of this feedback. Even the most ardent proponent of a reporting obligation has acknowledged that the disclosure of results has economic implications for researchers in planning their budgets and for funding agencies in determining an appropriate level and duration of funding (Fernandez et al., 2003). Or, as another ‘full disclosure’ proponent put it: “The problem, of course, is money” (Wolf et al., 2008). The consequence of requiring researchers to budget for managing incidental findings is “in the present financial climate ...that half as much research gets done, and that has, in my mind, a much greater impact on society than the very, very low incidence of incidental findings which are actually correct and an even lower incidence where there is something you could have done” (Ross, 2005). The setting in which Biobank results are generated, typically lacks the resources for additional research, replication of results and clinical counselling and follow-up of individual participants. Again, the price charged for commercial offerings of similar types of information could serve as a reality check. Even if the fee for the deCodeME services would drop by 50%, the total costs of providing feedback in the form of Odds Ratio’s at regular intervals, to the biobank individuals would still amount to $ 500,- times 150,000 or $ 75 mio. That is not including the costs of clinical geneticists and genetic counsellors. By comparison, the initial budget for the String of Pearls is € 40 mio. Competence: researchers are not licensed to practice To meet X’ demand for individual feedback would require the Biobankers to assume

the role of a clinician, if not the role of a clinical geneticist. The individual feedback to be provided could qualify as the provision of medical care for purposes of the Act on the Medical Treatment Agreement (WGBO). To the extent the provision of individual feedback involves complex genetic counselling, it would require a special license from the Ministry and it is reserved by law to a limited number of designated clinics (WBMV, Planningsbesluit). The Biobankers are not, at least not necessarily, licensed physicians and the biobank is not a designated centre for clinical genetics. The results would be communicated outside of a clinical setting and the significance of the observations might not be clear. The Biobankers are not in a position to interpret their implications fully. Consent to know results The international legal instruments relied upon by X all make it clear that the participants’ right not to know should always be respected (UNESCO, Council of Europe, CIOMS). Meeting this requirement of prior consent to know will place another undue burden on the Biobankers. While the science of biobanking is riddled with variations, the greatest variable may be the preferences and attitudes of the participants with respect to their wishes to know or not to know. Age, sex, children, education, employment, ethnicity, religion, health status, under treatment, self-esteem; all these factors impact, in most unexpected ways, on the attitudes and preferences of participants with respect to receiving individual feedback. Also, the preferences are likely to vary per individual, according to disease, probability, level of risk, and intervention possibility. Just as the participants mature over the lifespan of the bank, so may their attitudes and preferences in respect of feedback. Also, as data and results accumulate, their clinical utility may change. Therapeutic or life style modifications may have become possible, affecting upon participants’ desire to know. The combination of the wealth of possible results and the near infinite variety of participants’ desires to know or not to know these results would make it practically unfeasible for the Biobankers to meet the requirement of prior informed consent to know, each time a result has been found. Statutory restrictions The Biobankers refer to a number of statutes and regulations that govern the offering of individual health information and the offering of results of tests to (healthy) individuals. The provision of this type of information may be governed by the Act on Population Screening or the EU Ban on Drug Advertisements, (WBO, respectively EU Directive 2001/83, Reclamebesluit, CGR Code). No obligation to (offer to) report individual results can be imposed without a prior assessment of the implications and limitations under those regulations. In conclusion, a series of scientific, statutory, medical-legal arguments and practical barriers (complexity, competence, costs and consent) militates against the imposition of an (unconditional) obligation to report biobank research results and findings to

individual participants, with the exception of abnormal measurements which pose a present and imminent risk of a serious, treatable condition.

3.5 Expected outcome of case study

The outcome of the case is hard to predict. Much will depend on the facts and circumstances of the case and the outcome of other crucial variables such as procedural issues, issues of fact, statute of limitations, burden of proof, discovery, damages (loss of chance), and causation. The above discourse was limited to the threshold legal issue of whether there is a legal obligation on the part of the Biobankers to provide individual feedback. Anyway, in view primarily of the case law and current statutes, a court of law might very well reach the following opinion: 1. In essence, this case pitches an intuitively appealing claim of lifelong participants in a population biobank for disclosure of pertinent health information against the logically appealing notion that this biobank is neither their doctor nor their personal genomics company. The arguments to be made on either side can be summarized as follows. 2. According to X the following arguments support her claim for full disclosure of pertinent health information: the (contractual) terms of her participation; the characteristics of the biobank research; its stated purpose of linking abstract genomic data and concrete patients records; the expectations it raised of delivering personalized medicine; the embedding of the biobank research in an academicclinical setting which blurs the line between research and care; developments in technology, science and society; her (international human) rights to know health information, the right to know clinical research results, her right to life and to health; and, last but not least, her significant, lifetime contribution to the Biobank of a wealth of detailed and sensitive samples and data (right to benefit). 3. According to the Biobank the following arguments speak against (full) disclosure: the unambiguous language of its policy of non-disclosure and the rationale behind it; the lack of analytical validity and/or clinical utility of most findings; the need to avoid conflation of research and care; the inappropriateness of applying standards developed for clinical research to observational research; the Biobank’s lack of (statutory) competence to provide health information to individuals; the complexity and costs associated with reporting a potentially endless number of sometimes conflicting findings; and the disproportionate burden resulting from the need to respect the right not to know before any disclosure can be made. 4. It goes without saying that the legal duty to warn of imminent dangers implies that the more the requested data pose a real, imminent and actionable risk to X, the more likely it is that a reporting obligation on the part of the Biobank will be found. Also, the more the Biobank has raised the level of feedback expectations, the more likely it is that it will be required to meet those expectations. However, the terms of the non-disclosure policy at hand are indeed unambiguous. The

Biobank may have raised some expectations, but the references to personalized medicine are too global to uphold a claim for disclosure of findings at the individual participant level. Case law in the analogous area of population screening suggests that a proper ‘non-disclosure policy’ will be upheld. The case law also suggests that the Biobank is unlikely to be held to a clinical standard of care or even a standard of ‘semi-clinical’ care. Participants in the Biobank should and could have realized that the Biobank is not and cannot be their (collective) doctor. Rather, the standard of care of the Biobank must be determined in the proper context, i.e. the research context, taking into account all relevant circumstances. 5. As evidence of the predictive value of genetic variants accrues, investigators may face growing pressure to report findings that have an influence on risk. Some commentaries maintain that participants are even entitled to receive provisional results, with an explanation of the limitations of the data. Concurrently, a trend in international legal instruments towards the recognition of a right to feedback of research results has been reported in ethical and legal commentaries. These soft law instruments and opinions could be taken into account when assessing the proper standard of care in a tort action for negligence. Closer examination of these instruments, however, reveals that, assuming they are binding and enforceable in a private cause of action, most of them relate to clinical research, not to non-interventional research. Contrary to the situation in a clinical trial, X is neither vulnerable nor dependent on the Biobank. She can go to see her doctor at will. Unlike the Biobank, a doctor would be qualified and competent to make a diagnosis and to offer or refer to evidence-based interventions. This is also true for information bearing on reproductive choices, for which X could turn to regular counsellors. In addition, X could enrol in validated population screening programmes, which typically will be initiated or informed by validated findings of studies like the Biobank. 6. The features of the observational biobank findings further caution against the extension of these standards to population biobank research without appropriate adjustments: the sheer number of participants, the number of findings and the nature of these findings, which are likely to include both ‘good news’ and ‘bad news’. Notably, the CIOMS Guidelines, one of the few instruments which does apply to observational research, allows for an ethical review committee to approve temporary or permanent non-disclosure of data in view of the scale of a particular study. In addition, the 2009 OECD Draft Guidelines on Human Biobanks and Genetic Research Databases (which recommend that biobanks should elaborate an adequate feedback policy) provide that, as a general rule, non-validated results from scientific research using a biobank’s materials and data should not be reported back to the participants and that this should be explained to participants during the consent process. 7. Under the present circumstances and based on the above considerations, we are reluctant to honour X’s demand to be offered full disclosure of the Biobank’s findings pertaining to her individual health. However, that is not to hold that the

Biobank is not under an obligation to use best efforts to inform its participants, at the appropriate community or sub-community level, of its findings and their implications for individual participants. Indeed, providing as much validated personal health information to participants as is reasonably possible might also benefit the Biobank. It has even been suggested that disclosing to participants their own genotype would give them a personal stake in the ongoing research effort and could persuade them to (continue to) participate in longitudinal research (Nature Genetics, 2008). 8. As the Biobank acquires, over time, an increasingly enriched ‘picture’ of its participants, it would breach its duty of care owed to its participants if it failed to develop appropriate mechanisms to disclose pertinent health information to the appropriate community of participants. There is no reason to limit this feedback to genetic information regarding rare disorders that are life-threatening and for which clinical treatments exist. Any such mechanism should fit the type, the urgency, the context, the limitations and the validity and utility of the information to be so provided. Such mechanisms could include a website, a participants’ forum, regular meetings and newsletters updating and summarizing recent findings, accessible bibliographies of completed studies, a register of ongoing studies, webcasts, regular updates of FAQs, and web chat sessions with designated investigators. In due course, the Biobank should consider the provision to participants of a web-based, personally controlled health record empowering them to access and use any data generated by the Biobank (Kohane, 2007). 9. In brief, as the Biobank seeks “to track to their sources the causes of disease”, it should take note of its own stated ambition that research findings should be “quickly available for the prevention and cure of disease”. The Biobank is neither X’ doctor nor her personal genomics company. However, as a collaborative research enterprise, the bank must take care of her participants’ legitimate interest in receiving validated personal health information in a timely and appropriate manner. 10. By way of obiter dictum, the fact that participants have the right to request access/ change/delete any personal data contained in the biobank under data protection laws, should not be taken to mean an obligation on the part of the investigator or biobank to report research results to individual participants.

4 General discussion

4. General discussion

Over the last decade, feedback of research results to participants in medical research has received increasing attention. Whether, when and how research participants need to be informed about findings that may be relevant to their health (and to what extent researchers have a duty to communicate that information) has become an important issue, not only in the academic literature but also in the domain of research practice. Although initially the focus was to a large extent on clinical research, the question how one should deal with it in larger epidemiological studies and in population based genetic research has increasingly come to the fore. In the past, there has been some discussion about how to handle incidental findings that are discovered in the course of conducting research, but beyond the aims of the study in question (Wolf et al., 2008). More or less separate from this, there was the question as to whether and how the (intended) outcome of research projects should be communicated or at least made available to research participants. The emergence of large biobanks has added a new dimension to the issue, both because of their scale and because of the expectation that in due time biobank research will result in more findings with individual health relevance. Now, both questions have developed into the more comprehensive issue of communicating research results in a broad sense to individual participants. This is also related to the fact that the distinction between incidental findings and intended research results is becoming blurred. Biobank research is likely to produce widely varying kinds of findings. Potentially, many of these different sorts of research results may be relevant to the health of participants. Therefore, the concept ‘research results’ as used in this study includes not only unexpected, incidental findings, but also results that were aimed at in the research project concerned. Furthermore, it includes not only individual level results (i.e. observations resulting from examination of the participant or of his samples/related data) but also aggregate results (resulting from the analysis of (sub) populations) since also the latter may be of importance, at least to certain (sub)groups of research participants. Although the question to what extent researchers/biobanks have a duty to provide individual feedback is far from settled, there is a growing consensus that a simple, strict non-disclosure policy (in which there is no reporting whatsoever) does no longer suffice. It is generally acknowledged that the issue has at least to be addressed before research is undertaken, and that decisions have to be made on how to deal with it and, in particular, in what circumstances individual participants need to be re-contacted to advise them about salient findings. Nevertheless, the existing large scale, population-based biobanks still seem reluctant to provide feedback. As mentioned in the legal case study (see above), the UK Biobank for instance will not provide individual feedback to participants of results obtained during the research project for any reason. Like others (CartaGene; Generation Scotland) it provides only the results of clinical measurements (e.g. taken during the enrolment visit) but no personal genetic information. However, the strict non-disclosure policy of the UK

biobank has come under attack (see e.g. Johnston and Kay, 2004). It remains to be seen to what extent the restrictive policies of the others will resist the rising pressure to communicate relevant results to individuals. Looking at the conclusions of our empirical study, this pressure is likely to come in particular from the (potential) participants in biobank projects. Whereas almost half of the researchers felt that participants should not receive personal genetic information whatsoever, and almost all of them agreed that no such information should be provided if the health implications would still be uncertain, members of the general public and patients have different views. Many of them showed a keen interest in feedback, not only in individual findings, but also in aggregate study results that could be relevant to their health, including information on small, incremental risks, and irrespective of the nature of the disease and the possibilities for prevention or cure. This more or less confirms the outcome of other studies reported in literature, although it should be noted that most of these studies related to clinical trials rather than to biobank research. In the interim conclusion of our empirical research, we have already set out that it does not follow from this that, once recruited by biobanks, participants will always want the information they prospectively desire to receive. But at least at the outset, their expectations are rather high. The interim conclusions of the legal research underscore the need for biobanks to at least consider the feedback issue. They also show that the longer and more intensive the participation of biobank subjects, and the more biobanks themselves have raised the expectations, the more likely a reporting obligation on the part of biobanks will be found. Most important for the existence of a legal obligation is the nature of the available information, in particular whether it relates to a real and actionable risk to the health of an individual participant. For the rest, legal duties of care would seem very much determined by the research context and have to be established on a case by case basis, taking into account domestic law. At first sight, it seems easier to deal with study results in biobank research than in clinical/translational research, since the genetic epidemiologic population studies that are usually undertaken in the biobank context will less often result in individual findings of high clinical relevance. However, dealing with the issue in the latter context is complicated by the large scale and long term storage of data/samples and by the great variety in kinds of research findings that may become available, ranging from exceptional findings of high clinical utility to routine outcomes on gene variants and polymorphisms with low penetrance and very limited predictive value. This means that developing feedback policies at a more general level (i.e. at the level of biobanks in general or even at the level of individual biobanks) is only possible to a limited extent. Many different aspects have to be taken into account, not only relating to the nature of the information (such as amount of risk, validity of the results, the

seriousness of the disease, options available to participants) but also the feasibility, administrative burden and cost of informing participants. In the last resort, the details of when and how participants have to be informed need to be decided on the level of specific research projects. In spite of these limitations, biobanks (and certainly the general, national and international biobanks that are the focus of this study) will have to elaborate a policy in this field. Not only because of the need for transparency, public acceptance, the expectations of participants etc., but it has also to do justice to basic ethical and legal principles like protection of persons from preventable harm, respect for personal autonomy and privacy, and reciprocity (Ravitsky et al., 2007; Wolf et al., 2008). An interesting question concerns the normative value of expectations (in this case of prospective research participants) in addition to the normative principles just mentioned. In ethics and law, expectations are not just neutral facts. They imply a prima facie obligation on others to consider them, at least when these ‘others’ have some kind of relationship with the person having the expectation, and also where the ‘others’ have acted so as to raise such expectations. Apart from respect for the autonomy and preferences of their counterpart, reciprocity plays a role here. In the law, for instance, it is generally acknowledged that in a contractual relationship (and even to some extent in the so called pre-contractual phase) parties have to take into account their mutual interests, unless what is expected is unreasonable or when doing justice to those interests requires a disproportional effort. This kind of obligations become only more pertinent when the interests and expectations in question relate to goods or values that are generally considered to deserve protection (like the health of participants in clinical trials or in biobank research). Therefore, apart from factual consequences (e.g. ensuring the public’s willingness to cooperate), the expectations of prospective participants do have a certain additional normative value. At the same time, biobanks hold their own responsibility in providing feedback, or rather withholding it (if it would cause more harm than good and/or be an infringement of private life). Furthermore, they may have legitimate interests of their own (e.g. practical feasibility and cost) to provide less feedback than expected. In striking the balance between disclosure and non-disclosure, the basic elements to be taken into account in weighing the interests of participants have already been touched upon: basically, and in line with the bioethical literature on the subject, they can be identified as the nature and size of the health risk at stake, the validity of the research findings, and their clinical utility for the participant. While the potential health implications of a certain finding are a necessary condition to provide individual feedback, they are by no means a sufficient condition to do so. If a biobank would have to contact participants in connection with any research result that might have some relevance to their health, the efforts required from it would become disproportional. At the same time, participants would suffer from information-overload and confusion and the health care system might be burdened by requests for diagnostic procedures

lacking a sound basis. Individual feedback should therefore be limited to situations where the information available relates to a serious condition (including information of reproductive importance) and where there is a real possibility (also taking into account the validity of the information) that the participant can benefit from it. If one tries to envisage in what circumstances individual feedback is warranted (as distinguished from making general research results available to all participants collectively), two situations can be distinguished.1 The first situation is the classic case of a (usually – but not necessarily - more exceptional) finding of high clinical relevance for an individual participant. For instance, it reveals a condition that is grave but can be avoided or ameliorated, or it relates to a serious genetic disease and is a crucial piece of information in reproductive decision-making. In this case many commentaries in the legal and ethical literature tend to assume a duty to warn on behalf of biobanks/researchers. If they are able to avert imminent and serious harm, even if they have by no means the same responsibilities as a clinician, it is part of a researcher’s ‘ancillary’ obligations towards research subjects to do so (Richardson and Belsky, 2004). Depending on national law, this duty to warn may even become a legal one, actionable in a court of law when not complied with, for instance when the information available relates to a serious, treatable disease (Johnston and Kaye, 2004; Brownsword, 2007). The only possible exception to this (apart from the fact that it is practically impossible to retrace the person in question) would be that the participant has expressed a prior wish not to be informed. The second situation may more frequently occur in biobank research than the first one. It relates to the availability of less pertinent research findings (for instance because although they relate to a serious condition, the risks are more limited and/or the potential clinical utility is uncertain). In such a situation, according to Wolf et al., researchers ‘may disclose’ the finding to the research participant, unless he/she elected not to know. We agree that in such a case there is no overriding need to directly inform the participant; this may even encroach upon his private life as his actual preferences to receive or not to receive certain information may not be known by the biobank. But in stead of leaving the decision completely to the discretion of researchers, taking into account of the (potential) participants’ high expectations with regard to feedback as documented in our empirical study, respect for his person and autonomy would rather suggest that he be notified about the availability of research results with some (but limited) relevance to his health so as to enable him to access that information if he wants so.

1 This distinction is more or less similar to that made by Wolf et al. between findings with ‘strong net benefit’ and ‘possible net benefit’ in their consensus paper on incidental findings (2008).

In addition to providing a framework for which research findings should be considered for individual feedback and in what circumstances, a biobank policy on reporting research results should comprise other elements. A crucial one is that the policy leaves room – as far as reasonable possible - for the preferences of participants; a wish not to know should always be respected. Furthermore the policy should be transparent and be communicated adequately to prospective participants. Managing expectations should be an important aspect of communication with the public, not only in the sense that it is made clear what participants may expect but also that is explained that not all expectations can and will be fulfilled. Finally, since genetic information will also have consequences for family members, the policy should explain what position relatives of a participant have in terms of receiving information relevant to their health, also after the participant has died. Other elements relate to institutional arrangements to be adopted with a view to an adequate implementation of the policy. An example is who is responsible to decide on feedback and re-contacting individual participants; another one is how and by whom relevant research findings are to be communicated if such a decision has been taken. When research with data/samples stored in a biobank is conducted by external researchers, it should be ensured (in particular through access and/or material transfer agreements) that they act in conformity with the biobank’s policies. It should also be clear if they may contact participants themselves. Another important element relates to the role of ethics committees in reviewing the policy (and the arrangements implementing it) and in advising on hard cases and controversial questions arising in the course of biobank research.

5 Proposed guiding principles 67

5.1 Structural arrangements 5.2 Disclosure decisions

5. Proposed guiding principles

The national and international developments in this area do underscore a need for appropriate guidance, providing a proper framework for elaborating policies on feedback decisions and actions as well as clear terms of participation. Speciﬁcally, in what is likely to become one of the most authoritative international points of reference (the OECD Draft Guidelines on Human Biobanks and Genetic Research Databases, 2008), the OECD recommends that Biobanks (HBGRDs): 4.

HBGRDs should have a clearly articulated policy on the nature of the feedback that will be provided to participants, taking into account any domestic legal requirements. This policy should cover the feedback of individual-level results, if any, as well as aggregate results arising from research carried out using human biological materials and/or data from the HBGRD. 4.I. Participants should be provided with the opportunity to decide on whether or not to receive feedback of individual-level results arising from research. 4.J. As a general rule, non-validated results from scientiﬁc research using a HBGRDs’ human biological materials and data should not be reported back to the participants and this should be explained to participants during the consent process. 4.K. Where a HBGRD has offered and the participant has elected to receive feedback of individual-level results, and depending on the nature of the feedback, it may be appropriate for a trained professional to provide this feedback to participants or for counselling to be available to participants.

To meet the OECD call for a clearly articulated policy, we have developed the following Guiding Principles for a feedback or reporting policy. Taking account of the bioethical debate on the issue and considering the general discussion of the interim conclusions of our empirical and legal research in the preceding section, we suggest a number of Guiding Principles to be taken into account in the establishment and operation of biobanks, focusing on how they (and the researchers making use of such a resource) should deal with the individual feedback of research results. The Guiding Principles are addressed to those involved in the funding, establishment, operation, use, governance and oversight of such banks. While these recommendations are directed, ﬁrst of all, to the large scale general, national and international biobanks that were the focus of this study, they apply in an analogous way to other repositories of human biological materials and related data. Even if the application of some of the Guiding Principles is not fully feasible with regard to smaller biobanks, to the extent relevant and possible they should be taken into account in particular when biobanks are intended for sharing and samples/data may be used by external researchers for widely diverging research projects. The Guiding Principles relate to structural arrangements (1-13) and to disclosure decisions (14-23). Each of them will be followed by a brief explanatory paragraph.

5.1 Structural arrangements

1. Those planning to establish or planning a new biobank or operating an existing one must address the issue of communicating research results to individual participants. Explanatory note: Pre-established biobanks must consider the issue as well, although the application of some of the Guiding Principles may not be fully feasible.

When a biobank is set up, the responsible party should seek to identify reporting preferences of the most important stakeholders and take them into account in developing a disclosure policy. Explanatory note: In biobanking, many stakeholders are involved; for the purpose of this guideline, at least organizations representing the interests of future participants and/or patient organizations and of researchers should be involved (in general, also with regard to other issues, their involvement should be a continuous part of the governance structure of the biobank).

The parties involved in the creation, the operation and the governance of a biobank should undertake reasonable efforts to properly manage the expectations of the research participants in terms of feedback and benefits. Explanatory note: Research suggests that a primary incentive for prospective participants to take part in a biobank is to find out about their individual health risks. Biobank projects may, inadvertently, raise or reinforce such expectations. In addition, other scientific, societal and technological developments may help raise the expectations in terms of individual feedback to participants to unwarranted or unsustainable levels. The biobank should try to manage such expectations by providing realistic information. One concrete way to secure this would be by the involvement of an independent ethics committee.

4. Biobanks should have a clearly articulated policy on the nature of the feedback that will be provided to participants, taking into account any domestic legal requirements; this policy should cover the feedback of individual level results, if any, as well as aggregate results arising from research carried out using human biological material and/or data from the biobank. Explanatory note: This guideline echoes the central recommendation relating to the disclosure of research results in the recent Draft OECD Guidelines (2008). The policy should be submitted to independent ethical review.

5. The reporting policy must be flexible and include emergency procedures to inform both participants and non participants, as appropriate. The policy must have a built-in reassessment/evaluation mechanism. Explanatory note: While most biobank studies currently are exploratory and initial results are still to be confirmed, the next generation of epidemiologic studies may be more confirmatory. Also, special weight should be given that the very nature of the biobank could allow for rapid correlation and application of a certain study finding to a particular

individual, as the biobank may have ready access to pertinent phenotypic data of the individual concerned.

6. A disclosure policy should be compatible with the purpose and architecture of the biobank, and anticipate on its potential use; it should be achievable and resourced in the sense that any obligation can be actually met. Explanatory note: A prior assessment is to be made of implementation obstacles and costs, so as to ensure that it can meet the disclosure obligations resulting from the policy it has adopted.

7. The policy should address the implications of the termination of individual research projects on the biobank for the feedback policy. The policy should address the implications of the termination of the biobank for the feedback policy. Explanatory note: Biobanks should have a procedure dealing with the termination or foreclosure of the bank (OECD Guideline 10). These procedures should also deal with the implications of such termination or disclosure. Researchers cannot be held to owe reporting obligations arising after their termination of their research. The reporting policy should make it clear to participants that the biobank is not in a position to order or demand follow-on studies or monitoring studies after researchers have completed their research on the biobank.

8. Any policy should identify the roles and responsibilities of the parties involved. For the feedback to participants it should in principle make no difference whether the research is carried out by the biobank or by external researchers. Explanatory note: If research is to be carried out by external researchers, access and/or transfer agreements should seek to ensure that they comply with the feedback policy and take upon them the responsibilities that follow from it.

9. A reporting policy should provide a framework for disclosure/non-disclosure, to be elaborated at the level of individual research projects. Explanatory note: While the policy should provide the guidance possible at biobank level, it should leave room for (and require) further elaboration of details in line with the speciďŹ c features of any research project.

10. The feedback policy should take into account the various types of results and information to be generated eventually by research using material and data stored in the biobank. Explanatory note: As the objective of biobanks is to link genotype and phenotype, the policy should not limit itself to either type of information. As a clear deďŹ nition of what, exactly, constitutes genetic data is hard to give, a reporting policy should not distinguish between genetic and non-genetic data; this is not to say that no attention should be paid to the information interests of relatives where at stake (see Guideline 18 and 19).

11. Upon enrolment, as part of the informed consent process and within the framework of the biobank’s feedback policy, participants should be asked about their individual preferences concerning the feedback of research results which are of interest to them on an individual level after having been informed on the consequences of receiving such feedback. Any policy should also allow potential participants to express and have respected their right not to know any results. This would mean that in some cases their material and data may not be included in speciﬁc research. Explanatory note: Depending on the nature of the biobank, the options for participants may go beyond a simple yes or no. Respect for autonomy and privacy requires that participants can modify their preferences if they want so; the policy may provide that participants are being asked periodically whether they want to readjust their initial choice. Speciﬁcally, minors reaching adulthood should be recontacted to verify their wish to be informed or not.

12. Whatever the policy, participants must receive full information (including additional explanation upon request) about both the policy and its underlying rationale. Explanatory note: During the informed consent process, biobanks should provide potential participants with information on the nature, implications, foreseeable risks and benefits of their participation, so that they can properly assess the implications of participating. This includes full information on the possibility that research done with their samples/data may result in findings with individual health relevance and on when and how such findings will be reported back to the individual participant.

13. The policy (including the limits to disclosure) should be communicated to the public so that it is known what future participants may expect and not expect in terms of individual feedback. Explanatory note: In order to deal with unrealistic expectations, biobanks should – after careful consideration - not only make clear what kind of research results will be communicated, but also explain what they cannot and/or will not report (e.g. information on eventual non-paternity).

5.2 Disclosure decisions

14. Decisions to disclose or not to disclose research results should be based on carefully weighing the potential beneﬁts for the participants to receive the information concerned against the possible disadvantages, taking into account the feasibility to implement the decision. Basically, individual feedback should be given where the information relates to a serious health problem and where the possibility of an individual health beneﬁt is realistic. Explanatory note: Biobanks will have to strike a balance between disregarding potential health interests of participants and putting a lot of energy in efforts with no or marginal

benefit for the participants. The basic factors guiding disclosure and non-disclosure decisions should be: nature and size of the health risk, validity of the research finding, clinical utility for the participant, feasibility of feedback and integrity of the research and the meaning and ‘non’-meaning of findings for different groups. Clinical utility means that the information is about actionable risks, i.e. meaningful options should be available for the participant to deal with it in terms of cure, prevention or otherwise. Special consideration should be given to the fact that the research may regard samples taken a long time ago and that the samples may have aged during storage. Feasibility may stand in the way of providing (comprehensive) feedback; see in connection with this the draft CIOMS International Guidelines for Epidemiological Studies (2008): “… epidemiologists should make subjects aware of findings that are clinically relevant to their individual health. When (e.g. because of the scale of a particular study) an investigator does not plan to do so, he or she must obtain approval from the ethical review committee.” Integrity of the research means that researchers cannot be expected to explore the potential clinical relevance of every test outcome, in particular when that relevance is only theoretical or remote; basically, what they investigate should be guided by the goals of the research project and the related research questions.

15. In case of a ﬁnding with high clinical (or reproductive) signiﬁcance for an individual participant, any reasonable effort should be made to inform him or her about the risk and the possibilities to prevent or ameliorate the consequences. Explanatory note: If biobanks are able to avert serious damage, they are obliged to do so, unless the participant has expressed a prior wish not to be informed in any circumstances.

16. In case of findings where the health implications may still be serious but the risks are smaller and/or the potential benefit of being informed is limited or uncertain, participants should be notified of the availability of the relevant information only. Explanatory note: When health benefits of receiving feedback are still possible but less evident or substantial, participants should be advised that they can access the relevant information if they want so.

17. In other cases, there should be no individual feedback; validated research results should be made available on a collective basis to interested participants. Explanatory note: Aggregate results arising from biobank research should be communicated to all interested participants in easily accessible forms; this may include a newsletter, participant conferences or a website.

18. While acknowledging the interests of next of kin in being advised about some genetic conditions, it is the responsibility of the participant to contact them

when needed. The biobank’s responsibility is limited to inform the participant about a serious familial risk. Explanatory note: Biobanks and researchers owe more to participants than to their family members; furthermore, the efforts needed to fulﬁl obligations towards family members are likely to be disproportional. Therefore, although participants should be informed about serious family risks, it is their responsibility to pass that information to the relatives concerned. This should be clearly conveyed in the informed consent material.

19. If a ﬁnding of high clinical signiﬁcance for some relatives becomes available after the participant has died, reasonable efforts should be made to contact and inform them. Explanatory note: Although the obligations of biobanks vis-à-vis next of kin can be only very limited, they should undertake what is reasonably possible to avert serious and imminent danger if the participant himself can no longer do this.

20. Research results should be disclosed in ways and by persons designated by the biobank; external researchers should not directly contact research participants. Explanatory note: Communication with research participants should remain concentrated in the biobank so as to ensure that it takes place with sufﬁcient expertise, ﬁts into agreed procedures and does not exceed the consent given.

21. Disclosure of research results should be offered with appropriate supplemental information and with counselling when needed. Explanatory note: In addition to the health implications, supplemental information may pertain to limitations of the ﬁndings concerned, possible interventions, family implications, need for additional counselling, legal implications as well as to eligibility for insurance or employment etc. The relevant information may be provided by an independent physician designated by the biobank for this purpose, a general practitioner, a clinical geneticist, or other professionals with sufﬁcient expertise.

22. Biobanks should try to anticipate on disclosure issues that may arise in the future as a result from technological developments, especially in case new information could be derived that is not covered by the participants’ informed consent. Explanatory note: Due to technological and scientific developments, future research with the biological materials/data stored in the biobank may yield more or other findings with important health consequences than anticipated before, while exceeding the participant’s initial consent. Also, new analyses could be performed on previously collected material and data and result in a new meaning of previous findings. Biobanks should consider in advance how to handle this.

23. In case of doubt, decisions to give or to withhold individual feedback on research results should be submitted for prior advice to an independent ethics committee. Explanatory note: Biobanks should have a mechanism for independent ethical review, not only of their general policies, but also of speciﬁc decisions in case of conﬂicting interests or ethical dilemma’s. Unless the matter is obvious, decisions to disclose or not to disclose should be submitted for ethical review.

Samenvatting

De oorzaken van veel complexe aandoeningen zoals diabetes, kanker of de ziekte van Alzheimer, zijn, ook na het in kaart brengen van het menselijk genoom, nog altijd onduidelijk. Een wijdverbreide hypothese is dat complexe aandoeningen worden veroorzaakt door een groot aantal kleine, vaak bijkomende effecten, die het resultaat zijn van de wisselwerking tussen genen, levensstijl en de omgeving, op verscheidene niveaus. Een gerelateerde hypothese is dat het blootleggen van deze complexe relaties is gebaat bij het kunnen onderzoeken van grote hoeveelheden goed gedocumenteerde epidemiologische, klinische, biologische en moleculaire informatie en bijbehorend lichaamsmateriaal van grote aantallen patiënten en gezonde mensen. Daartoe worden deze data en dit materiaal ondergebracht in zogenaamde DNA-banken of biobanken die voor meerdere onderzoeksdoeleinden, door verschillende disciplines over een lange periode kunnen worden gebruikt. De afgelopen tien jaar zijn in veel landen dergelijke DNA-onderzoekdatabases ontwikkeld, soms in de vorm van op de nationale bevolking gebaseerde biobanken. Voorbeelden van grootschalige biobankinitiatieven in Nederland zijn het LifeLines initiatief van het UMCG Groningen en het Parelsnoer initiatief van de academisch-medische centra. Onze onderzoeksvraag was of en zo ja hoe, resultaten gebaseerd op onderzoeken die gebruik maken van deze biobanken aan de individuele onderzoeksdeelnemers moeten worden meegedeeld. Het uitgevoerde onderzoek was er meer speciﬁek op gericht om (i) na te gaan of onderzoekers juridisch gezien verplicht zijn om resultaten uit biobankonderzoek aan individuele deelnemers van het onderzoek mee te delen en (ii) om de standpunten en informatievoorkeuren van belangrijke belanghebbenden in kaart te brengen. Deze onderzoeksvraag houdt niet alleen verband met de opkomst van grootschalige biobanken, maar ook met de recente discussie in de internationale literatuur over de vraag of deelnemers moeten worden geïnformeerd over de resultaten van het onderzoek waarbij ze betrokken zijn. Volgens de ethische en juridische literatuur is er een trend waarneembaar in internationale aanbevelingen en vergelijkbare documenten die wijst in de richting van erkenning van het recht op informatie over onderzoeksresultaten . Van oudsher worden onderzoeksresultaten in principe niet aan individuele deelnemers meegedeeld, behalve als het onverwachte bevindingen betreft. Onverwachte bevindingen zijn incidentele (diagnostische) bevindingen die uitsluitend op één individu betrekking hebben en die niet zijn voorzien. Nieuwe bevindingen zijn generaliseerbare resultaten van wetenschappelijk onderzoek die een voorspellende waarde hebben voor de huidige of toekomstige gezondheid van een groep mensen. In ons onderzoeksvoorstel hadden we als hypothese dat het onderscheid tussen ‘verwacht’ en ‘onverwacht’, dat ten grondslag ligt aan het traditionele beleid van niet-meedelen, ondergraven wordt door fundamentele verschuivingen; verschuivingen in zowel het wetenschappelijk onderzoek (de ontwikkeling van grootschalige biobanken) als in de

perspectieven en verwachtingen van deelnemers van dergelijk onderzoek. We veronderstelden dat deze verwachtingen gebruikers en onderzoekers van biobanken onder extra druk zullen zetten om resultaten sneller en vanzelfsprekender te vertalen en aan deelnemers mee te delen. We menen dat onze hypothese door de volgende ontwikkelingen wordt ondersteund: • Het ruimer beschikbaar komen van biologische analyses in gestandaardiseerde opmaak, in combinatie met de aanwezigheid in een biobank van een brede reeks persoonlijke fenotype- en omgevingsgegevens over een langere tijdsperiode, kan de vertaling van de algemene bevindingen in klinisch relevante informatie op het niveau van de individuele deelnemer vergemakkelijken en versnellen. • Het uitvoeren van uitgebreide genotype- en functionele genomische metingen in de algemene bevolking zal waarschijnlijk voor bijna iedereen onverwachte, incidentele bevindingen opleveren. • Een ruim bekendgemaakte directe beschikbaarheid voor consumenten van zowel persoonlijke genotype- als biomarkergegevens in ongekend detail en breedte zal waarschijnlijk de verwachtingen van deelnemers aan een biobank doen toenemen. • De betrokkenheid van de deelnemers op de lange termijn, eigenlijk levenslang, gekoppeld aan de breedte en gevoeligheid van de gegevens die zij verschaffen, roept de verwachting op dat hun deelname voordeel zal opleveren in de vorm van individuele informatie over resultaten, ongeacht of deze “verwacht” of “onverwacht” waren. Als gevolg van deze ontwikkelingen is het probleem van individuele feedback in het kader van biobankonderzoek verschoven naar de meer omvattende kwestie van het informeren van individuele deelnemers aan onderzoeken in brede zin, over zowel onverwachte als beoogde bevindingen, en zowel bevindingen op individueel als op collectief niveau. In de jaren 2006-2008 hebben we deze vragen uitvoerig onderzocht, zowel middels een empirisch onderzoek om de standpunten en informatievoorkeuren van belangrijke stakeholders (d.w.z. (potentiële) deelnemers en onderzoekers) te inventariseren, als middels een juridisch onderzoek om erachter te komen of onderzoekers juridisch verplicht zijn om feedback aan individuele deelnemers te verschaffen. Ons doel was – rekening houdend met de bio-ethische literatuur en relevante internationale beleidsdocumenten – om een algemeen normatief kader te ontwikkelen dat op dit gebied als (nadere) leidraad zou kunnen dienen bij de oprichting en de exploitatie van biobanken. De empirische studie omvatte een vragenlijst voor een dwarsdoorsnede van burgers (als potentiële deelnemers), patiënten (als deelnemers, omdat ze al bloed of speeksel aan biobanken hadden gedoneerd) en onderzoekers. De conclusie van het empirisch onderzoek is dat het grootste deel van (potentiële) deelnemers veel belangstelling had

voor informatie zowel over individuele bevindingen, als over algemene onderzoeksresultaten die relevant zouden kunnen zijn voor hun gezondheid, waaronder informatie over kleine, toenemende risico’s. Deelnemers wilden informatie ongeacht de aard van de ziekte en de mogelijkheden tot voorkoming of genezing. Een aanzienlijke minderheid gaf er echter de voorkeur aan om geen informatie te ontvangen. In tegenstelling tot de mening van onderzoekers, dachten (potentiële) deelnemers dat onderzoekers de plicht hebben om deelnemers te informeren over allerlei resultaten. Op grond van deze bevindingen verdient het aanbeveling dat biobanken aandacht besteden aan de kwestie van feedback. Daarbij is van belang rekening te houden met het beheersen van verwachtingen van deelnemers (die hoog kunnen zijn), de verschillende informatievoorkeuren (d.w.z. een kleine groep heeft geen belangstelling voor het ontvangen van (speciﬁeke) resultaten) en de kloof tussen onderzoekers en deelnemers met betrekking tot verwachtingen. De juridische analyse betrof een onderzoek naar internationale beleidsdocumenten en richtlijnen, naar regelingen van bestaande biobanken en een analyse van de jurisprudentie en wetgeving die van toepassing zijn op vergelijkbare situaties. In het rapport presenteren we de resultaten van deze analyse in de vorm van een casestudy over een deelnemer aan een nationale biobank die, nadat zij verscheidene aandoeningen ontwikkelde, de biobank verzocht om haar te informeren over risicofactoren die over haar bekend waren en haar te adviseren over voorkoming/behandeling. Na een korte bespreking van de casus die aan de gerechtelijke procedure ten grondslag lag, behandelen we uitvoerig de juridische argumenten die de deelneemster zou kunnen aanvoeren om haar rechtsvordering te onderbouwen, evenals de mogelijke tegenargumenten van de zich verwerende biobank. De conclusie van het juridisch onderzoek is dat onder de huidige wetgeving, een reeks wetenschappelijke, wettelijke en gezondheidsrechtelijke argumenten, alsmede een aantal praktische obstakels (complexiteit, bevoegdheden, kosten en het vragen van toestemming) pleiten tegen het opleggen van een (onvoorwaardelijke) verplichting aan biobanken/onderzoekers om onderzoeksresultaten aan individuele deelnemers mee te delen of aan te bieden, met uitzondering van abnormale metingen die duiden op een actueel en dreigend risico van een ernstige, maar behandelbare aandoening. Echter, bovengenoemde wetenschappelijke, technologische, maatschappelijke en juridische ontwikkelingen zullen de balans op termijn waarschijnlijk meer doen doorslaan in de richting van een actievere houding ten opzichte van het meedelen van resultaten en gegevens. Deze ontwikkelingen zouden ook aanleiding kunnen zijn voor (het verplicht aanbieden van) mededeling van resultaten en gegevens over gezondheidsrisico’s die niet of minder dreigend zijn, niet levensbedreigend en niet meteen behandelbaar. Terwijl ze niet juridisch verplicht hoeven te zijn om elke bevinding te beoordelen en mee te delen, zouden biobanken in de toekomst nalatig kunnen worden bevonden als ze niet adequaat aandacht besteden aan de kwestie, geen duidelijk beleid hebben en/of deelnemers en andere belanghebbenden niet over dit beleid informeren.

Rekening houdend met de bio-ethische discussie en onze tussentijdse conclusies met betrekking tot de juridische en empirische aspecten en onder verwijzing naar de 2008 OECD Draft Guidelines on Human Biobanks and Genetic Research Databases (waarin wordt aanbevolen dat biobanken een adequaat beleid zouden moeten ontwikkelen), stellen we een aantal Guiding Principles voor die uitgangspunt zouden kunnen zijn voor een informatiebeleid bij de oprichting en exploitatie van grootschalige algemene, nationale en internationale biobanken. De betreffende richtlijnen zijn bedoeld voor partijen die betrokken zijn bij het ﬁnancieren, oprichten, exploiteren, gebruiken, besturen en toezicht houden op deze banken. Aan de ene kant hebben ze betrekking op ‘organisatorische regelingen’, aan de andere kant op ‘beslissingen over het informeren van deelnemers over onderzoeksbevindingen’. De eerste groep (13 organisatorische) principes omvat onder meer: • dat bij het oprichten van een biobank aandacht aan feedback moet worden besteed en er een omvattend beleid moet worden ontwikkeld; • dat de belangrijkste stakeholders hierbij betrokken moeten zijn; • dat het beleid een kader voor wel respectievelijk niet informeren van deelnemers moet bieden dat vervolgens op het niveau van individuele onderzoeksprojecten nader kan worden uitgewerkt; • dat deelnemers bij aanmelding volledig moeten worden geïnformeerd en dat ze de gelegenheid moeten krijgen om hun wens te uiten om wel of niet geïnformeerd te worden over bevindingen die relevant zijn voor hun gezondheid; • dat het publiek over het beleid moet worden geïnformeerd, zodat toekomstige deelnemers weten wat ze kunnen verwachten. De tweede groep van 10 principes met betrekking tot beslissingen over wel/niet informeren omvat onder meer: • dat individuele informatie moet worden gegeven als de beschikbare informatie van onderzoek betrekking heeft op een ernstig gezondheidsprobleem en als de kans op een individueel gezondheidsvoordeel realistisch is; • dat in andere gevallen gevalideerde onderzoeksresultaten collectief aan geïnteresseerde deelnemers beschikbaar moeten worden gesteld; • dat onderzoeksresultaten op een door de biobank te bepalen wijze moeten worden bekendgemaakt; externe onderzoekers moeten niet rechtstreeks contact opnemen met deelnemers van het onderzoek; • dat onderzoeksresultaten moeten worden meegedeeld met passende aanvullende informatie en indien nodig met advies; • dat in geval van twijfel, beslissingen om wel of geen individuele informatie over onderzoeksresultaten te verstrekken voor voorafgaand advies aan een onafhankelijke ethische commissie moeten worden voorgelegd. Het resultaat van het empirisch en juridisch onderzoek evenals de concept-richtlijnen zijn gepresenteerd en besproken tijdens een bijeenkomst van deskundigen georgani-

seerd door het Academisch Medisch Centrum van de Universiteit van Amsterdam op 12 september 2008. Het doel van deze bijeenkomst was om een aantal experts uit relevante disciplines de gelegenheid te geven om onze veronderstellingen,empirische bevindingen en argumenten te bespreken en om de door ons voorgestelde Richtlijnen te bekijken en van commentaar te voorzien. Daarnaast zijn de resultaten van het empirisch onderzoek en het juridisch onderzoek gepresenteerd aan en becommentarieerd door collega ELSI wetenschappers met verschillende achtergrond in een interdisciplinaire sessie van collegaâ&#x20AC;&#x2122;s en niet-collegaâ&#x20AC;&#x2122;s, georganiseerd door het Centre for Society and Genomics. De resultaten van deze discussies zijn in dit deďŹ nitieve onderzoeksrapport verwerkt. Tenslotte is commentaar op het juridische deel van het onderzoeksrapport gegeven door een buitenlandse deskundige, prof dr B.M. Knoppers, University of Montreal. De inhoud van dit onderzoeksrapport blijft uitsluitend de verantwoordelijkheid van de auteurs.

Sources

Sources Appendix A Factors associated with information preference for individual results and attitudes on researchersâ&#x20AC;&#x2122; duties to communicate research results Appendix B Supervisory Board Appendix C List of experts

Sources

ALRC, Australian Law Reform Commission,

Cambon-Thomsen A, Rial-Sebbag E, Knoppers BM.

Essentially Yours: the Protection of Human Genetic

2007. Trends in Ethical and Legal Frameworks for

information in Australia. Accessible at www.austlii.

the use of human biobanks. Eur. Resp. J 30: 373-

edu.au/au/other/alrcpublications/reports/96.

382.

Beskow LM, Burke W, Merz JF, Barr PA, Terry S,

CaulďŹ eld T, McGuire AL, Cho M, Buchanan

Penchaszadeh VB, Gostin LO, Gwinn M, Khoury

JA, Burgess MM et al. 2008. Research Ethics

MJ. 2001. Informed consent for population-based

Recommendations for Whole-Genome Research:

research involving genetics. JAMA 286: 2315-2321.

Consensus Statement. PLoS Biol 6(3): e73 doi:10.1371/journal.pbio.0060073.

BIG, Wet op de beroepen in de individuele g,ezondheidszorg, Wet van 11 november 1993,

CIOMS International Ethical Guidelines for

Stb. 655, laatstelijk gewijzigd bij wet van 10 april

Epidemiological Studies Prepared by the Council

2008, Stb. 164.,

for International Organizations of Medical Sciences (CIOMS) in collaboration with the World Health

Boggio A, Informing Participants about Research

Organization (WHO).

Results, Ethical and regulatory aspects of human genetic databases, B. Elger, N. Biller-Andorno, A. Mauron, A. Capron, eds., Ashgate 2008.

Code Goed Gebruik, especially 3.10, 3.11 and 4.2, in principle no communication, unless express wish participant to be informed and except in case of

Bookman EB, Langehorne AA, Eckfeldt JH, Glass

emergency and 5.4. and 5.5.

KC, Jarvik GP, Klag M, Koski G, Motulsky A, Wilfond B, Manolio TA, Fabsitz RR, Luepker RV.

Collins FS, Green ED, Gutmacher AE, Guyer MS.

2006. Reporting genetic results in research studies:

2003. A vision for the future of genomics research.

summary and recommendations of an NHLBI

Nature 422:835-847.

working roup. Am J Med Genet Part A 140:10331040.

Common Rule, Federal Common Rule for Protection of Human Research Subjects 10

Bouchie A, 2004. Coming soon: a global grid for

(45CFR46).

cancer research. Nat Biotechnol 22:1071-1073. Council of Europe, Convention for the Protection Bovenberg JA, 2005. Whose tissue is it anyway,

of Human Rights and Dignity of the Human

Nature Biotechnology Vol.37:929-933 .

Being with regard to the Application of Biology and Medicine: Convention on Human Rights and

Bovenberg JA, 2006. Property Rights in Blood,

Biomedicine, Oviedo, 4.IV.1997.

Genes & Data, Naturally Yours? Brill-Martinus

Additional Protocol to the Convention on Human

Nijhoff Publishers 117-118.

Rights and Biomedicine, concerning Biomedical Research, Strasbourg, 25.I.2005.

Bronsword R, The ancillary-care responsibilities of researchers: reasonable but not great expectations, J

Council of Europe Rec 2006, Committee of

Med Law Ethics 35 (2007) 679-691.

Ministers. Recommendation Rec (2006) 4 of the Committee of Ministers to member states on research on biological materials of human origin.

http://www.coe.int/t/e/legal_affairs/legal_co-

Framingham Heart Study, www.

operation/bioethics/texts_and_documents/Rec_

framinghamheartstudy.org

2006_4.pdf. Galas, Nanotechnology in Predictive, Preventive, DeCode, deCode Inc. (www.decode.com)

and Personalized Medicine: The Problem and the Promise, in: Perspectives on The Future of Science

Diergaarde B, Bowen DJ, Ludman EJ, Culver JO,

and Technology, Health and Nanotechnology: The

Press N, Burke W. 2007. Genetic information:

Promise and Challenge of Nanomedicine, A report

Special or not? Responses from focus groups with

from a Conference Convened with the Cooperation

members of a health maintenance organization. Am

of the US State Depertment and The European

J Med Genet Part A 143(6):564-569.

Commission, January 22-24, 2006

Dressler LG, Juengst ET. 2006. Thresholds and boundaries in the disclosure of individual genetic

Gevers 2006, AMC Public Health-Law (NWO-

research results. Am J Bioeth 6(6): 18-20.

Centre for Society and Genomics Grant 20062008), Project leader: prof J.K.M. Gevers.

ECHR, Convention for the Protection of Human

Researchers: drs T.M. Meulenkamp, dr E.M.A.

Rights and Fundamental Freedoms, ETS 5.

Smets, prof J.C.J.M. de Haes, mr dr J.A. Bovenberg.

Estonian Gene Bank, www.geenivaramu.ee.

Gezondheidsraad, Wet bevolkingsonderzoek: thuistest voor niet-deelneemsters aan de screening

Facio FM, 2006. One size does not ﬁt all. Am J

op baarmoederhalskanker, Den Haag 2006.

Bioeth 6 (6): 40-42. Gezondheidsraad, Translational research in the Fernandez CV, Santor D, Weijer C, Strahlendorf

Netherlands. Between laboratory and clinical

C, Moghrabi A, Pentz R et al. 2007. The return of

practice (the Hague 2007, nr 55).

research results tor participants: pilot questionnaire of adolescents and parents of children with cancer.

Greely HT, 2000, Iceland’s plan for Genomics

Pediatr Blood Cancer 48 (4): 441-446.

Research: Facts and implications, 40 Jurimetrics J 153-191.

Fernandez CV, Shurin S, Kodisch E. 2006. Providing research participants with ﬁndings from completed

Greenberg vs Miami Children’s Hospital Research

cancer-related clinical trials: not quite as simple as it

Institute, Inc., 264 F. Supp 2nd1064 (S.D. Fla.

sounds. Cancer 15;107(6): 1419-1420.

2003).

Fernandez CV, Kodish E, Weijer C. 2003. Informing

Hagen HE, Carlstedt-Duke J. 2004. Building

study participants of research results: an ethical

global networks for human diseases: genes and

imperative. IRB 25(3): 12-19.

populations. Nat Med 10:665-667.

Fong M, Braun KL, Chang RM. 2006. Native

Hirschhorn JN, Daly MJ. 2005. Genome-wide

Hawaiian preferences for informed consent and

association studies for common diseases and

disclosure of results from genetic research. J Cancer

complex traits. Nat Rev Genet 6(2):95-108.

Educ 21 (1 Suppl): S47-S52.

Hoeyer K, Olofsson BO, Mjorndal T, Lynoe N, 2004.

Klitzman R, 2006. Questions, complexities, and

Informed consent and biobanks: a population-

limitations in disclosing individual genetic results.

based study of attitudes towards tissue donation for

Am J Bioeth 6(6): 34-36.

genetic research Scand J Public Health 32(2): 224-229.

KNAW, Verkenningscommissie Multi-factoriële Aandoeningen. 2006. Multifactoriële aandoeningen

HUGO 2002, Hugo Ethics Committee: statement on

in het genomics tijdperk. Amsterdam.

human genomic databases, December 2002 www. hugo-international.org/img/genomic_2002.pdf.

Knoppers BM, Joly Y, Simard J, Durocher F, 2006. The emergence of an ethical duty to disclose

HUGO Ethics Committee, Statement on beneﬁt-

genetic research results: international perspectives.

sharing, April 9 2000, www.hugo-International.

Eur J Hum Genet 14: 1170-1178.

org/img/beneﬁt_sharing_2000.pdf Kohane IS, 2007. Reestablishing the ResearcherHuman Genes Research Act, available through

Patient Compact. Science 216: 836-837.

www.geenivaramu.ee. Kohane IS, Masys DR, Altman RB. 2006. The Hunter DJ, Khoury MJ, Drazen JM. 2008. Letting

Incidentalome: a threat to genomic medicine. JAMA

the genome out of the bottle—will we get our

296; 2: 212-215.

wish? N Engl J Med 358: 105-107. Lifelines, www.lifelines.nl/page/videos IBM 2004, biobanks: accelerating molecular medicine challenges facing the global biobanking

LUMC. 2001. Genomics and Healthcare, Value

community.

Added by research, education and entrepreneurship, position paper for the incorporation of a centre of

ISBER, International Society for Biological and

excellence for Genomics and Healthcare. Leiden

Environmental Repositories, a Division of the

University Medical Centre and the Faculty of

American Society for Investigative Pathology (ASIP)

Mathematics and Physics: 84.

www.isber.org. Manolio TA, Bailey-Wilson JE, Collins FS. 2006. Johnston C, Kaye J, 2004. Does the UK Biobank

Genes, environment and the value of prospective

have a legal obligation to provide feedback about

cohort studies. Nat Rev Genet 7:812-820.

individual ﬁndings to participants? Medical Law Review 12:239-67.

Manolio TA, 2006. Taking our obligations to research participants seriously: disclosing individual

Kaiser J, 2002. Biobanks. Population databases

results of genetic research. Am J Bioeth 6(6): 32-34.

boom, from Iceland to the U.S. Science 298:11581161.

Meltzer LA, 2006. Undesirable implications of disclosing individual genetic results to research

Khoury MJ, 2004. The Case for a global human genome epidemiology initiative. Nature Genetics 36: 1027.

participants. Am J Bioeth 6(6): 28-30.

Miller FA, Christensen R, Giacomini M, Robert JS,

Parker LS, Best laid plans for offering results go

2008. Duty to disclose what? Querying the putative

awry. 2006. Am J Bioeth 6 (6): 22-23.

obligation to return research results to participants. J Med Ethics 34; 3: 210-213.

Partridge AH, Burstein HJ, Gelman RS, Marcom PK,

Miller FG, Rosenstein D. 2003. The therapeutic

Winer EP, 2003. Do patients participating in clinical

Orientation to Clinical Trials, New England journal

trials want to know study results? J Natl Cancer Inst

of Medicine 348: 1383-1386.

95(6): 491-492.

Miller FG and Brody H, A Critique of Clinical

Partridge AH, Wong JS, Knudsen K, Gelman

Equipoise: Therapeutic Misconception in the Ethics

R, Sampson E, Gadd M et al. 2005. Offering

of Clinical Trials, Hastings Centre Report, no 33,

participants results of a clinical trial: sharing results

(2003):19-28.

of a negative study. Lancet 365 (9463): 963-964.

Miller PB and Weijer C, 2006. Fiduciary obligation

Pierson ThA, Manolio TA, 2008. How to Interpret a

in clinical research. J Law Med Ethics 34 (2): 424-

Genome-wide Association Study. JAMA 299;11.

440. P3G, Public Population Project in Genomics, www. National Institutes of Health, Policy for sharing

p3gconsortium.org.

of data obtained in NIH supported or conducted genome-wide association studies (GWAS). Federal

Ravitsky V, Wilfond BS, 2006. Disclosing individual

Regist. 2007;72(166):49290-49297. www.grants.

genetic results to research participants. Am J Bioeth

nih.gov/grants/guide/notice-ﬁles/NOT-OD-07-

6(6): 8-17.

088.html.

Raad voor Volksgezondheid en Zorg 2008, Screening en de rol van de overheid

Nature Genetics, 40, 119 (2008), Positively

www.rvz.net/cgi-bin/adv.pl

Disruptive, Editorial. Renegar G, Webster ChrJ, Stuerzebecher S, Harty OECD Draft Biobank Guidelines, Draft Guidelines

L, Ide SE, Balkite, B, Rogalski-Salter TA,Cohen N,

for Human Biobanks and Genetic Research

Spear BB, Barnes DM, Brazell C, 2006. Returning

Databases: Text for Comment, available at www.

genetic research results to individuals: points-to-

oecd.org/document/63/0,3343,en_2649_201185_

consider. Bioethics 20;1: 24–36.

40305535_1_1_1_1,00.html#11. Richardson HS, Belsky L, The ancillary-care van Ommen GJ, Popper revisited: GWAS here, last

responsibilities of medical researchers, Hastings

year, European Journal of Human Genetics (2008)

Center Report Jan-Feb 2004, 25-33

16, 1–2; doi:10.1038/sj.ejhg.5201970. Ross K, 2005. When volunteers are not Healthy. Osman 1997, Osman v. the UK, European Court of

EMBO reports 6, no.12:1116-1119 at 1118

Justice, (87/1997/871/1083)

available at www.pubmedcentral.nih.gov (quoting Alan Evena, neuroimaging researcher from McGill

Ossorio PN, 2006. Letting the gene out of the

University, Montreal, Canada, quaoted by Alan C.

bottle: a comment on returning individual research

Mistein in Wolf et al. 2008, at 359).

results to participants. Am J Bioeth 6(6): 24-25.

Sallée C and Knoppers BM, 2004 Existing Human

UNESCO, 2003. International Bioethics Committee

Genetic Research Databases, Organisation for

(IBC): International declaration on human genetic

Economic Co-operation and Development (OECD),

data. Paris: Unesco, International Bioethics

Directorate for Science.

Committee (IBC).

Sharp RR, Foster MW, Clinical utility and full disclosure of genetic results to research participants.

UNESCO, 2005. The Universal Declaration

2006. Am J Bioeth 6(6): 42-44.

on Bioethics and Human Rights, adopted by acclamation the General Conference, at its 33rd

Simm K, 2005. Beneﬁt-sharing: an inquiry regarding

session, on 19 October 2005.

the meaning and limits of the concept in human genetic research. Genomics, Society and Policy 1;

Wendler D, Emanuel E, 2002. The debate over

2: 29-40.

research on stored biological samples. What do sources think? Arch Intern Med 162:1457-1462.

Snowdon C, Garcia J, Elbourne D, 1998. Reactions of participants to the results of a randomised

Wendler D and Pentz R, 2007. How does the

controlled trial: exploratory study. BMJ 317(7150):

collection of genetic test results affect research

21-26.

participants? Am J Med Genet A 143A(15):17331738

String of Pearls, www.string-of-pearls.org/index. html

WGBO, Wet op de Geneeskundige Behandelingsov ereenkomst, Wet van 17 november 1994, Stb. 837,

Technology and Industry, Report on Human Genetic

laatstelijk gewijzigd bij wet van 13 maart 2002, Stb.

Research Databases, DSTI/STP/BIO (2005) 14;

121.

OECD,

creation and Governance of Human Genetic

Research Databases, (2006).

Wilson JMG, Jungner G, 1968. Principles and Practice of Screening for Disease. Geneva: World

Terry 2003, P F Terry, “PXE International:

Health Organization.

Harnessing intellectual property law for beneﬁtsharing”, in Populations and Genetics, eds., B M

Wolf SM, Paradise J, Caga-anan Ch, 2008. The

Knoppers, Martinus Nijhoff Publishers, Leiden/

Law of Incidental Findings, in Human Subjects

Boston 2003.

Research: Establsihing Researchers’ Duties. Journal of Law, Medicine & Ethics, SYMPOSIUM: 361-383

UK Biobank, www.ukbiobank.ac.uk

at 366.

UK Biobank ethics and governance framework,

X and Y v. the Netherlands, European Court of

Version 2.0, July 2006, www.ukbiobank.ac.uk/

Justice, (1985) Series A.,vol. 91.

docs/EGF_Version2_July%2006%20most%20upt odate.pdf UNESCO, 1997. United Nations Resolution on The Human Genome and Human Rights, Resolution 53/152, adopted by the United Nations General Assembly at its 53rd session (December 1998).

Appendix A

Factors associated with information preference for individual results and attitudes on researchers’ duties to communicate research results

Population (citizens or patients)

Gene-mutation information preference

Attitudes on researchers’ duties

Adjusted R square=0,280

Adjusted R square=0,249

� coeff (SE)

0,08 (0,03) (p=0,001)

0,12 (0,02) (p<0,001)

Sociodemographic factors age

0,09 (0,03) (p=0,001)

-0,05 (0,03) (p=0,047)

sex (0=male; 1=female) Reason participation biobank ‘because the physician or researcher asks me’*

-0,08 (0,03) (p=0,001)

0,07 (0,03) (p=0,07)

0,1 (0,02) (p<0,001)

monitoring coping style

0,13 (0,03) (p<0,001)

0,06 (0,03) (p=0,026)

general desire for information about illnesses/treatment

0,18 (0,03) (p<0,001)

0,16 (0,03) (p<0,001)

‘advantage is that information can be meaningful for health children/ family’

0,33 (0,03) (p<0,001)

0,34 (0,03) (p<0,001)

‘disadvantage is that information may cause anxiety’

-0,15 (0,02) (p<0,001)

-0,15 (0,03) (p<0,001)

‘privacy will be harmed when information becomes available’

-0,09 (0,03) (p=0,001)

‘remain in control over blood after donation to biobank’

0,09 (0,03) (p=0,001)

‘health and treatment information’** Psychological characteristics

Attitudes related to biobanking

0=’because the physician or researcher asks me’, 1=’altruistic’

** 0=’health and treatment information’, 1=’altruistic’

Appendix B

Supervisory Board Prof Dr DL Willems

Medical Ethics, Academic Medical Centre / University of Amsterdam

Dr C Oosterwijk

Director, VSOP (Dutch Genetic Alliance)

Prof Dr NJ Leschot

Clinical Genetics, Academic Medical Centre / University of Amsterdam

Prof Dr PH Reitsma

Experimental Molecular Medicine, Leiden University Medical Center

Prof T Caulﬁeld

Research director, Health Law Institute, University of Alberta, Canada

Prof J Bensing

Health Psychology, University Utrecht / NIVEL

Appendix C

List of Experts Participants expert meeting September 12th 2008 Mr M Boeckhout, MA

University of Amsterdam

Dr M Boenink

Center for the philosophy of Technology and science, University of Twente

Dr G Fobelets

Centre for Biomedical Ethics and Law, Katholieke Universteit Leuven

Prof Dr JCJM de Haes

Medical Psychology, Academic Medical Centre / University of Amsterdam

Dr ACJW Janssens

Department of Epidemiology and Biostatistics, Erasmus MC

Prof Dr LP ten Kate

Department of Clinical Genetics and Antropogenetics, VUMC

Prof Dr DA Legemate

Scientiﬁc Director ‘Parelsnoer’, Academic Medical Centre / University of Amsterdam

Prof Dr NJ Leschot

Department of Clinical Genetics, Academic Medical Centre / University of Amsterdam

Dr JG Loeber

National Institute for Public Health and the Environment (RIVM)

Prof Dr GJB van Ommen

Department of Human Genetics, Center for Human and Clinical Genetics, Leiden University Medical Center

Dr GJ Olthof

Ministry of Health, Welfare and Sport

Dr C Oosterwijk

Dutch Genetic Alliance / Verenigde Samenwerkende Ouder- en Patiëntenorganisaties

Mr EB van Veen

MedLawconsult

Ms M Verduyn

Astma Fonds

(VSOP)

Consulted experts Mr PW Francissen, MSc

Ministry of Health, Welfare and Sport

Prof Dr BHR Wolffenbuttel LifeLines Prof Dr DS Postma

University Medical Center Groningen

Prof Dr BM Knoppers

University of Montreal, Canada

CSG-reports 1. Genetica, genomics en gezondheidszorg – een toekomstverkenning, April 2008 (in Dutch) 2. Always expect the unexpected – Legal and social aspects of reporting biobank research results to individual research participants, November 2009

Always expect the unexpected Legal and social aspects of reporting biobank research results to individual research participants © Jasper Bovenberg, Tineke Meulenkamp, Ellen Smets and Sjef Gevers Centre for Society and Genomics PO-box 9010 6500GL Nijmegen The Netherlands www.society-genomics.nl Final editing: Frans van Dam Lay-out: Hannie van den Bergh, studio HB Photo cover: DigiDaan Printed by: Thieme Media Services ISBN: 978-90-79533-03-9 Correspondence Prof Dr JKM Gevers Academic Medical Centre/University of Amsterdam Department of Social Medicine PO-box 22660 NL-1100DD Amsterdam The Netherlands t + 31 20 566 46 36 j.k.gevers@amc.uva.nl

The emergence of large-scale genetic research - also called ‘genomics’ - has greatly boosted the study of tissues and patient data in biobanks. A side effect is that researchers can uncover results that may be highly relevant to the health of individual patients. This in turn gives rise to a range of questions of a factual, ethical and legal nature. How much do people want to know about ﬁndings that could indicate future health risks or beneﬁts? Is a biobank obliged to report research results to individual participants? In which cases should it - whether or not of its own accord - do so? And what form should this communication with participants of biobanks take? In this report, researchers of the Amsterdam Academic Medical Centre (AMC) conclude, backed by empirical and legal study, that biobanks have no alternative but to ask themselves these questions and develop clear policy in this area. The study formulates a number of ‘guiding principles’ for such a policy.

This report is issued by the Centre for Society and Genomics (CSG). ‘Genomics’ is a novel and fast developing ﬁeld of scientiﬁc research that will have an impact on health care, nutrition, industry and the environment. CSG analyses, assesses and improves the relationship between society and genomics research. By doing so, CSG improves the way in which genomics meets the expectations and needs of society. CSG, founded in 2004, is a national centre, based at the Radboud University Nijmegen in The Netherlands. CSG is one the 16 a genomics centres of the Netherlands Genomics Initiative (NGI). www.society-genomics.nl