APPENDIX 2
Memorandum submitted by GeneWatch UK
SUMMARY
The Medical Research Council is part-funding
a controversial new genetic research project called Biobank UK.
The project aims to identify the links between genetic and environmental
factors in common diseases by studying DNA samples taken from
500,000 volunteers between the ages of 45 and 69, linked with
lifestyle and medical data.
Biobank UK's
— aims are controversial;
— science is questionable; and
— there is a lack of legal safeguards
to protect the volunteers taking part from misuse of their genetic
information.
The main aim of the project is to be able to
identify individuals who are "genetically susceptible"
to common diseases, so that medication or advice on lifestyle
changes can be targeted at them. However, many scientists have
argued that this approach is unlikely to contribute significantly
to reducing the incidence of these diseases, and may even be damaging
to health. Underlying these concerns are serious question-marks
about the scientific validity of Biobank UK, including the danger
of identifying spurious links between genetic and environmental
factors and diseases. There are also serious concerns about confidentiality—particularly
potential access to the data by the police—and the need
to control commercial conflicts of interest and patent claims.
Biobank UK involves significant resources and
its full costs remain unknown. GeneWatch UK believes the project
should be shelved until the controversies surrounding it have
been democratically debated and resolved, and alternative approaches
have been properly assessed.
Some key questions for the MRC are:
— Why has there been no public, independent
assessment of the scientific validity or likely value-for-money
of Biobank UK? What are the full costs of the project expected
to be?
— The journal Nature has reported
that 36 projects given the top "alpha-A" rating following
scientific peer review have been rejected for MRC funding this
year because of Biobank UK. How was the decision made to fund
Biobank UK rather than these projects? Did Biobank UK receive
the top rating from its scientific peer reviewers?
— What assessment has the MRC made
of the relative costs and benefits of population-based approaches
to tackling the prevention of common diseases?
— What assessment has the MRC made
of the relative costs and benefits of smaller, more detailed (case-control)
studies as an alternative approach to identifying genetic factors
in disease?
1. INTRODUCTION
GeneWatch UK is a policy research group concerned
with the science, ethics, policy and regulation of genetic technologies.
GeneWatch works to promote environmental, ethical, social, human
health and animal welfare considerations in decision-making about
genetic engineering and other genetic technologies. We welcome
this opportunity to input to the Science and Technology Committee's
scrutiny of the Medical Research Council.
This evidence focuses on the issue of the planned
genetic research project known as Biobank UK. Biobank UK will
study the genetic information contained within DNA samples taken
from 500,000 volunteers between the ages of 45 and 69. The genetic
data will be linked with lifestyle information taken from an initial
questionnaire and information about subsequent sickness, medication
and causes of death taken from the volunteers' medical records.
The MRC and Wellcome Trust announced the allocation of £45
million start-up funding to the project in April 2002.
GeneWatch is not opposed to the use of biobanks
in medical research (we believe small, focussed biobanks can and
have been very productive sources of useful knowledge), but has
repeatedly called for Biobank UK to be shelved due to concerns
about both the scientific validity of the project and the lack
of safeguards to protect the public from the misuse of genetic
information. Our key concerns are that:
— prediction of future common illnesses
by testing people's genetic make-up is unlikely to be a successful
or cost-effective means of disease prevention;
— the serious scientific limitations
of Biobank UK mean that genetic factors in disease or drug response
will be hard to identify correctly, and spurious links between
genes and diseases may be made;
— Biobank UK will not contribute directly
to the development of new medicines but may allow commercial companies
to go "gene fishing", patent gene sequences and gain
excessive monopolies over future treatments;
— there is a lack of legal safeguards
to protect participants from future misuse of their genetic information.
Although the MRC and Wellcome Trust have conducted
a number of consultations on aspects of Biobank UK, widely raised
concerns have still not been addressed.
A copy of the GeneWatch UK report "Giving
Your Genes to Biobank UK: Questions to Ask", which covers
the issues relating to lack of safeguards in more detail, is enclosed,
together with three relevant GeneWatch briefings.
The limited health benefits of Biobank UK and
the lack of legal safeguards are considered in turn below.
2. BIOBANK UK—BENEFITS
TO HEALTH?
A major concern expressed by the public in recent
consultations about the Biobank UK proposals is the "unclear
benefits to individuals and society"[1]
If Biobank UK fails to deliver the promised benefits significant
resources will have been wasted that could have been better spent
on other projects. Public trust in genetic research is also likely
to be damaged if the downsides (such as potential negative impacts
on civil liberties, see Section 3) outweigh any benefits to public
health.
Key questions for the MRC include:
— What total resources will be needed
to fund Biobank UK and how will this impact on other medical research
projects?
— What is the likelihood of Biobank
UK meeting its aims and objectives?
— Is studying gene-environment interactions
in this way likely to contribute significantly to improving public
health?
— Why has the MRC refused to allow
an independent body to assess the scientific validity of the project?
— What assessment has been made of
the project's likely cost-effectiveness?
These issues are considered in more detail below.
2.1 Allocation of MRC resources
The MRC has reportedly already refused funding
for some other top-rated research projects because of the resources
required for Biobank UK[2],
yet the basis on which the project has been given such high priority
is unclear (see also Section 2.4). The total amount of public
funding likely to be spent on Biobank UK also remains uncertain—the
£45 million allocated to date is only a small part of the
likely budget. No public assessment has been made of the costs
of alternative approaches to studying the role of genes in common,
complex diseases (see Section 2.2). Nor has the MRC considered
alternative approaches to securing reductions in the incidence
of these diseases (see also Section 2.3).
Items which appear to be missing from the initial
budget are listed in Appendix A.
Key questions for the MRC:
— What are the full estimated costs
of meeting the project's objectives, including the costs of collecting
follow-up data, extending the time-frame and including nested
studies?
— Will the MRC provide a detailed public
breakdown of these costs?
— To what extent does the success of
the biobank depend on securing commercial funding of future (smaller
but more detailed) nested studies?
2.2 Will Biobank meet its aims and objectives?
The Biobank UK Protocol[3]
states that its main aim is "to investigate the separate
and combined effects of genetic and environmental factors (including
lifestyle, physiological and environmental exposures) on the risk
of the common multifactorial diseases of adult life". The
study will also examine whether or not the risk of adverse events
relating to the use of certain medications varies according to
an individual's genotype (genetic make-up).
However, the science behind Biobank UK has been
criticised by geneticists and others for three key reasons:
1. The study design (a "cohort"
rather than "case-control" study) may be inadequate
to identify the proposed gene-environment interactions;
2. The study is too large to allow sufficiently
detailed lifestyle and environmental data to be collected to test
hypotheses about the causes of complex diseases;
3. Medical records are inadequate to provide
the necessary follow-up data for meaningful research.
More detail on these concerns is given in Appendix
B.
"The scientific value of focusing on gene-environment
interactions has not been established, and in any case, the technical
advantages of cohort studies over case-control studies in detection
of statistical interactions between genetic and environmental
effects are less clear than has been assumed." David Clayton
(Department of Medical Genetics, Cambridge University) and Professor
Paul McKeigue (London School of Hygiene and Tropical Medicine)[4].
"Garbage in, garbage out", Chief Executive,
Gemini Genomics[5].
Key questions for the MRC:
— Will the MRC make a detailed, public,
scientific comparison of pros and cons of Biobank UK and the proposed
alternative of a smaller case-control study?
— How does the MRC plan to address
the likely poor quality of follow-up data obtained from medical
records?
— How has the MRC decided what data
to collect from the initial questionnaires, when it has not published
any hypotheses for the role of gene-environment interactions in
major diseases?
— Has the MRC made any assessment of
how complexity could limit the statistical power of the study
to detect gene-environment interactions?
— Why has the MRC not produced a detailed
response to the issues raised at its Protocol Workshop, particularly
the comments that the study would be "a poor vehicle for
study of cardiovascular and metabolic disease"?
2.3 Contribution to public health
The Medical Research Council (MRC) states that
Biobank UK will[6]:
— help in the design of a new generation
of drugs to counter most major diseases.
— help to explain why some patients
respond well to particular drugs and others do not.
— be used to predict the likelihood
that an individual would develop disease, so that medicines could
be used to prevent the onset of disease, rather than as a treatment
for symptoms once a disease develops.
However, the biobank will not lead directly
to the development of new medicines, nor provide detailed and
reliable information on individuals' responses to particular drugs
(see Section 2.2). The predicted benefits of Biobank UK therefore
rest largely on an assumption that "The more precise identification
of individuals at increased risk of disease through both exposure
and genotype will allow improved targeting of various interventions"[7].
These interventions are likely to include both advice on lifestyle
changes and preventative medication—used to prevent the
onset of a predicted disease.
However, this genetic approach to disease prevention
is a highly controversial one. Scientists involved in developing
the Biobank UK protocol have recognised that "the extent
to which the targeting of interventions in accordance with genotype
will ultimately prove useful is as yet unclear"[8].
Nevertheless, there is already sufficient knowledge about the
role of genetic and environmental factors in common, complex diseases
to make some estimates of the limited potential benefits of this
approach. The key issues are outlined in Appendix C.
If widespread genetic testing is not a useful
or cost-effective way of reducing the incidence of common, complex
diseases, then Biobank UK will be of little benefit to public
health. The MRC has not provided any public justification as to
why Biobank UK is a better use of resources than more research
on population-based measures—such as the most effective
ways to improve the diets of all school children.
"[The Director of the US National Human
Genome Research Institute] and many others have outlined scenarios
where all individuals have a battery of genetic tests early in
life so that the knowledge of `susceptibility' can be used to
avoid development of disease. Biomedical sectors would profit
from acceptance of the above approach, but it is doubtful whether
it is the approach most likely to increase the health of populations."
Professor Patricia Baird, Department of Medical Genetics, University
of British Columbia, Canada[9].
"Genetic tests for markers that may not
result in disease for half a century or more could be new examples
of a process of premature medicalisation of attaching the `disease'
label before it has been established that prevention or treatment
is clearly beneficial. . . Unless it is established that a genetic
variant is a pointer to beneficial action, there is a potential
for inappropriate medicalisation through the spread of poorly
understood tests." Dr David Melzer (Research Associate) and
Professor Ron Zimmern (Director, Public Health Genetics Unit),
University of Cambridge[10].
"Genetic research. . . might improve our
understanding of individual susceptibility to disease, but it
will not contribute to population cardiovascular disease control."
Dr Robert Beaglehole, University of Auckland[11].
"...it is vital that the more conventional
approaches of epidemiology and public health, particularly as
they relate to tobacco-induced diseases and other aspects of lifestyle,
continue to be pursued with vigour. This is particularly important
as there are still major uncertainties about the predictive role
and cost of genomics for controlling common diseases." World
Health Organisation, 2002[12].
Key questions for the MRC:
— before funding Biobank UK, what assessment
has the MRC made of the likely health benefits of widespread genetic
testing for susceptibility to common, complex diseases?
— what assessment has the MRC made
of the likely costs to the health service of a genetic approach
to disease prevention, including estimates of the number of healthy
people that would need to be tested and treated to prevent each
case of disease?
— what assessment has the MRC made
of the relative costs and benefits of population-based approaches
to tackling disease prevention, compared to approaches based on
targeting the "genetically susceptible"?
— given the major future implications
for the National Health Service of using a genetic approach to
"predict and prevent" future common, complex illnesses,
why has there not been more public and parliamentary debate on
the advantages and disadvantages of this approach?
2.4 Scientific peer review
In the light of the extensive concerns about
the scientific validity of Biobank UK noted above, GeneWatch UK
believes the process of scientific peer review has been inadequate.
Funding was allocated in principle to Biobank UK by the MRC and
Wellcome Trust in June 1999. Yet the same bodies that proposed
the project also organised its peer review and took the decision
to allocate the funding. This has led to the unprecedented rejection
of MRC funding for a large number of other project applications,
which had been given the top alpha-A scientific rating following
peer review[13].
Key questions for MRC:
— In view of the scientific controversy
surrounding the Biobank UK proposals, why did MRC/Wellcome Trust
not seek an independent body to organise its peer review, independently
of the MRC/Wellcome Trust?
— Did the project's peer reviewers
include some of those who had been openly critical of its design?
— Did the project receive a top alpha-A
rating by its scientific peer-reviewers?
— How was the decision made to fund
Biobank UK, rather than the 36 alpha-A rated projects which have
reportedly been rejected for MRC funding this year?
— In view of the significant public
implications of the project, has MRC sought ways in which it could
make the peer reviewers' comments public, without revealing their
identities?
— If it cannot make them public, would
the MRC be prepared to provide the peer reviewers' comments to
the Committee, on an anonymous basis?
2.5 Value-for-money
The value-for-money of Biobank UK will depend
on:
— its overall costs (see Section 2.1);
— whether or not it can meet its own
objectives in correctly identifying gene-environment interactions
(see Section 2.2);
— whether or not these objectives,
if achieved, would contribute significantly to public health (see
Section 2.3);
— the comparative costs of alternative
approaches to studying the role of genetic factors in common,
complex diseases (such as a smaller but more detailed case-control
studies, see Section 2.2);
— the comparative costs of alternative
(non-genetic) approaches to securing reductions in the incidence
of these diseases (see Section 2.3);
— the costs of any negative impacts
on research due to lack of safeguards and loss of public trust
(see Section 3).
All these issues are highly disputed, yet there
has been no public assessment or debate regarding the value-for-money
of Biobank UK.
Key questions for the MRC:
— Why have the MRC not sought an assessment
of the project's value-for-money from an independent body?
— Has the MRC made any internal assessment
of the project's likely cost-effectiveness?
3. BIOBANK UK—LACK
OF LEGAL
SAFEGUARDS.
In addition to the concerns about the scientific
validity and usefulness of Biobank UK noted above, the lack of
proper legal safeguards for this type of research means that there
are considerable potential downsides for participants and for
society as a whole.
Although many smaller biobanks already exist,
Biobank UK needs urgent consideration because of its scale, the
involvement of large sums of public money, and the potential that
it could be the first step towards a larger, national biobank,
involving the entire UK population[14],[15].
Unless the necessary safeguards are put in place before the recruitment
of volunteers begins, there is potential for widespread loss of
public trust in genetic research and perhaps medical research
in general.
Key issues are:
— the role of commercial companies
in Biobank UK;
— the need for public involvement in
deciding how the biobank is used;
— the potential for genetic discrimination;
— the potential for forensic use of
personal genetic information and/or other breaches of confidentiality.
These issues are considered in more detail below.
3.1 The role of commercial companies
The Wellcome/MRC consultation with primary health
care workers found that they felt that the patenting of genes
should be avoided[16].
The Wellcome/MRC GP's focus group also wanted assurances that
the involvement of commercial organisations would be strictly
controlled[17].
In the same study, religious and community leaders raised concerns
about profiteering by companies. A more recent public consultation
for MRC/Wellcome Trust[18]
reported that "The idea of access by commercial organisations
raised concerns and generally the first reaction was to reject
the idea" although the researchers then noted that "Further
debate brought the realisation that if medicines are going to
be developed, pharmaceutical companies must have access"
and that some participants therefore "became resigned to
their involvement".
GeneWatch is particularly concerned that the
MRC has not yet adequately addressed two key areas of concern
regarding commercial involvement in Biobank UK:
3.1.1. Gene patenting and profiteering
Current legislation does not require people
to be informed if their genes are patented. The patenting of gene
sequences is controversial and allows unprecedented monopolies
over future genetic tests and treatments, which can stifle future
research and innovation and increase costs[19].
The Nuffield Council on Bioethics recently concluded that the
granting of patents that assert rights over DNA sequences should
become the exception rather than the norm[20].
In GeneWatch's view the patenting of genes should not be allowed,
and new legislation is needed to regulate the commercial use of
biobanks.
"There should be no patenting of gene sequences,
period. They were invented by nature." Affymetrix (US biotech
company), 2002[21].
"The intellectual property arena is nothing
less than a minefield . . . if a gene sequence is patented, you
can't necessarily design around it. What type of discovery associated
with the gene sequence would entitle somebody to lock up a whole
area of research and prevent competition?" Dr Elliot Sigal,
Senior Vice President of Early Discovery and Applied Technology,
Bristol-Myers Squibb's Pharmaceutical Research Institute[22].
3.1.2. Commercial conflicts of interest
Conflicts of interest are particularly likely
if commercial companies are allowed to undertake research seeking
to identify those who are genetically susceptible to diseases
associated with their own products or pollution (eg tobacco companies
seeking to identify those smokers who are genetically susceptible
to lung cancer; employers seeking to identify employees who are
susceptible to diseases associated with hazardous chemicals or
radiation in the workplace). This could lead to attempts to avoid
liability by blaming individuals' genetic make-up for diseases,
the possible exclusion of some people from some hazardous environments,
and resulting negative impacts on public health (see Section 3.3).
People could quickly lose trust in this type of research if it
is conducted against the interests of those that they are trying
to help.
"[Biobank UK] is a big gamble . . . People
who opt into this study have to know exactly what is being done
with this DNA. They need to know its relationship to any industrial
exploitation." Professor Sir David Weatherall, Oxford University[23].
Key questions for the MRC:
— does the MRC intend to allow gene
sequences contained in Biobank UK to be patented?
— what steps does the MRC plan to take
to prevent the patenting of gene sequences identified using Biobank
UK leading to excessive monopolies over future genetic tests and
treatments?
— what steps does the MRC plan to take
to ensure any new medicines developed using genetic sequences
identified in Biobank UK are accessible and affordable to the
NHS?
— what steps does the MRC plan to take
to control potential commercial conflicts of interest and ensure
public involvement and support for the research projects undertaken?
3.2 Public involvement in research
Some types of genetic research are particularly
controversial—for example research on behaviour, race, criminality,
or intelligence. There is particular controversy about the results
of such research being used to influence future reproductive decisions
(including the use of abortion and the concept of "designer
babies"). People may also object to companies using their
data for research where there may be potential conflicts of interest
(see Section 3.1.2). The general type of consent (for "medical
research") proposed for Biobank UK, increases the likelihood
that people's genetic information will be used for research they
disagree with. GeneWatch would like to see new democratic mechanisms
established to ensure effective public involvement in deciding
how biobanks are used and operate. This would also allow people
more say in deciding the limits of any commercial involvement
(Section 3.1).
Key Questions for the MRC:
— Does the MRC expect research on genetics
and behaviour, and other controversial topics, to be part of Biobank
UK?
— Does the MRC expect companies with
potential conflicts of interests (for example, the tobacco, fast
food, insurance, chemical and nuclear industries) to undertake
research using Biobank UK?
— How does the MRC expect decisions
about such research proposals to be made?
— How does the MRC intend to keep the
public involved in these decisions, so public trust in "medical
research" is not compromised?
3.3 Genetic discrimination
People will be asked to donate their samples
to Biobank UK on the grounds that the research will benefit those
who are susceptible to future illness because of their genetic
make-up. Yet genetic tests developed using Biobank UK could be
used in future to discriminate against people identified as "genetically
susceptible"—for example by refusing them insurance
or a job.
A particular concern in relation to Biobank
UK is that genetic tests will be developed to identify those at
higher risk of cancer or other diseases when exposed to particular
hazardous chemicals or radiation. Such genetic predictions are
likely to be poor, so that many people with the high-risk gene
would never have developed the disease, and many without the high-risk
gene will still be at risk. However, employers may still perceive
genetic screening as a means of reducing liability, potentially
cutting the costs of compensation claims[24].
Job-seekers have little power to influence decisions
by employers and are likely to be particularly vulnerable to decisions
to exclude them on the basis of genetic test results[25].
Existing employees, although somewhat protected by employment
legislation, may also be the subjects of genetic discrimination
rather than gaining more control over their environment. There
is therefore widespread concern that this type of research could
lead to the exclusion of people with a particular genetic make-up
from hazardous workplaces, or from compensation, rather than reducing
or removing the hazard and cleaning up the environment for all[26].
Although genetic testing is not currently widespread,
dozens of cases of exclusion from insurance or employment on the
grounds of genetic test results have already been reported in
the US[27],[28].
A recent US case involved the Burlington Northern Santa Fe Railroad
Company testing employees who had reported wrist injuries for
a common genetic variation associated with carpal tunnel syndrome.
The company abandoned the testing after it was sued by the US
Equal Employment Opportunity Commission on the grounds that the
tests were discriminatory and carried out without consent[29],[30].
Research on genetic susceptibility to hazardous chemicals has
not yet led to widespread genetic testing, but the option has
been and is still being considered by some employers[31],[32]
,[33].
A recent survey by the Institute of Directors[34]
found that 50% of employers who responded to the survey thought
it would be appropriate to conduct genetic testing with an employee's
consent "to see if they are at risk of developing an occupation-related
disease due to exposure in the workplace". 16% thought it
would be appropriate to make genetic testing compulsory in such
circumstances.
Currently, the costs and logistical and technical
difficulties of screening are highly uncertain—however,
one study concluded that screening 1,429 workers to hire 1,000
would be a cost-effective way of preventing an employee contracting
chronic beryllium disease[35].
Excluded workers (potentially a high proportion of the workforce
in some industries) are likely to suffer negative impacts on their
health and economic prospects, and those who are employed may
also suffer if exclusion policies are used as an alternative to
improving health and safety standards. Such policies could also
exacerbate health inequalities, since they will have a disproportionate
effect on those manual workers and technicians who are dependent
on employment in such industries.
There are currently no laws in the UK to prevent
insurers or employers using genetic test results to set insurance
premiums or decide who gets a job, although there is a moratorium
on the use of most genetic test results by the insurance industry.
"Ethical considerations, and legal, are
fundamental to the whole issue of genetic testing . . . The consequences
for individuals with regard to insurance and employment are also
of the greatest importance, together with the implications for
stigma and discrimination." Expert Working Group to the NHS
Executive and the Human Genetics Commission[36].
— Has the MRC urged the Government
plan to sign and ratify the European Convention on Human Rights
and Biomedicine (1997), which prohibits discrimination against
a person on the grounds of his or her genetic heritage?
— Has the MRC urged the Government
to legislate to ensure that genetic tests developed using Biobank
UK are not used by insurers or employers in the future to set
insurance premiums or to decide who gets a job?
— In the absence of such legislation,
does the MRC feel confident that research seeking to identify
those defined as "genetically susceptible" to environmental
hazards, particularly in the workplace, will be of benefit to
health?
— Does the MRC plan to warn volunteers
donating samples that the results of the research might be used
in this way?
— Does the MRC expect employers to
undertake this type of research using Biobank UK, and if so, how
will it address potential conflicts of interest?
3.4 Forensic uses and confidentiality
There is also concern about potential failures
of strict privacy and confidentiality for genetic information,
including the need to clarify the grounds on which the police
or the courts could gain access to the information. Addressing
these issues is particularly important since the Government sees
Biobank UK as a pilot project for a national genetic database,
potentially involving all NHS medical records.
A person's genetic-make up is unique, unless
they have an identical twin. This means this information has powerful
forensic uses. Biobank UK will also contain personal information
about people's lifestyles, linked to their genetic information.
Although there has been some discussion of computer and other
systems that would need to be in place to keep the data secure,
there are some additional factors that require particular consideration.
These are:
3.4.1 Access to the data by the police, courts
or government
A recent case in Scotland highlighted the fact
that the courts are able and willing to over-ride guarantees of
confidentiality in research studies. In this case, a prisoner
who had volunteered to take part in a study of HIV was convicted
of knowingly infecting his girlfriend, who had taken part in a
separate research project[37].
Samples from both projects were used by the prosecution to show
that the prisoner's girlfriend had the same form of the virus
as that found in the prison.
Current legislation allows the police to ask
the courts for access to such databases when this is deemed to
be in the public interest or if it could prevent serious harm.
There is a real danger that Biobank UK could come to be seen as
a forensic resource to identify individuals associated with increasingly
trivial offences. Political pressure for access to the data could
also increase, particularly in times of terrorism or war, with
potential negative impacts on civil liberties.
GeneWatch believes legal clarification is need
regarding when genetic information and other data in Biobank UK
can be release without consent, particularly to the police, courts
or government.
3.4.2 The inherent limitations to the "anonymisation"
of genetic information
Current research into the use of DNA techniques
in crime detection includes ways of identifying commonplace characteristics
so that in future a "genetic photo-fit" could be built
up from a sample left at the scene of a crime[38].
Techniques are already available to attempt to predict whether
or not a suspect has red hair, on the basis of a DNA sample, and
to attempt to predict a man's surname from his DNA[39].
This type of research renders the idea of truly "anonymous"
genetic information rather meaningless, particularly when DNA
data is linked to other personal data, such a postcode or job.
Both postcodes and employment information are likely to be used
in research by third parties using Biobank UK, since they are
the only data relevant to many of the environmental exposures
that the biobank is claiming to investigate—but they may
also help reveal an individual's identity. A public assessment
is needed of the extent to which the data from Biobank UK, even
when supplied on an "anonymised" basis, could be used
to identify individuals. This issue is also of relevance to decisions
about commercial access, since the public may have more concerns
about some types of researchers discovering their identity than
others. For example, academic researchers are likely to be more
trusted than companies who might be using the database for marketing
purposes, or than employers or insurers, who may—perhaps
inadvertently—discover personal information about their
own customers or employees.
GeneWatch believes there is a need for specific
legislation for the protection of personal genetic information
before members of the public are asked to donate genetic samples
to Biobank UK.
"The linkage of genetic and health information
and the potential for using the [Biobank UK] database for a wide
variety of analyses aimed at determining susceptibility to disease
raises important issues about confidentiality, security of data
and informed consent. These concerns were raised as recently as
Monday by GeneWatch, and they were the subject of a debate in
the House of Lords yesterday. The Government takes these concerns
seriously and will not allow the work to proceed until they have
been satisfactorily addressed." The Rt Hon Alan Milburn,
Secretary of State for Health.[40]
Questions for the MRC:
— In what circumstances does the MRC
envisage the police or government seeking access to the data in
Biobank UK?
— Will the MRC be warning participants
that the police will be able to apply to the courts for access
"in the public interest"?
— What studies has the MRC undertaken
in relation to the possibility that an individual's identity could
be deduced from supposedly anonymous data released to third parties
by Biobank UK (eg genotype plus postcode, or genotype alone)?
— Does the MRC plan to warn participants
about the limitations to "anonymising" personal genetic
information?
4. CONCLUSIONS
Biobank UK should be shelved until the serious
question-marks over its scientific validity have been resolved,
and proper public safeguards have been put in place.
— Considerable resources will be needed
to fund Biobank UK, over and above those already allocated, and
this will impact negatively on other medical research projects.
— Biobank UK is unlikely to meet its
aims and objectives, given the complexity of gene-environment
interactions, and the likely poor quality of the data in it.
— Studying gene-environment interactions
in this way is unlikely to contribute significantly to improving
public health.
— There has been no truly independent
assessment of scientific validity of the project or of whether
it will be good value-for-money in terms of improving public health.
— There is considerable public unease
about commercial involvement in Biobank UK, yet the MRC has provided
no public information regarding limits or controls on such involvement.
— Gene patenting has not been ruled
out, although it is highly controversial and could hinder future
research and development and increase costs of future treatments.
— Serious conflicts of interest could
arise if companies responsible for various exposures (via their
products or pollution) are allowed to determine the research agenda—yet
no steps have yet been taken to prevent this.
— There is a need to ensure public
involvement in research decisions, so people's samples are not
used for research they disagree with—yet plans for a general
form of consent to "medical research" sidestep these
concerns.
— There are currently no laws in place
to prevent insurers or employers using genetic tests developed
by Biobank UK to set insurance premiums or decide who gets a job—harming
instead of benefiting those the biobank is claiming to help.
— There is an urgent need to clarify
on what basis the police, courts or government might be given
access to the data held in Biobank UK for forensic purposes.
— Insufficient consideration has been
given to the possibility that an individual's identity could be
deduced from supposedly "anonymous" information given
to researchers from the biobank.
Annex A
COSTS NOT
INCLUDED IN
THE BIOBANK
UK BUDGET
Missing core funding
The original set-up costs for Biobank UK were
estimated to be £60 million, and costs were confirmed to
the project's Protocol Development Workshop as being in excess
of this[41].
Although £45 million has been allocated, it is GeneWatch's
understanding that another £20 million is required to implement
the Scientific Protocol as planned. The potential source of this
funding remains unclear.
Costs of updating exposure records
The Biobank's Protocol Development Workshop
noted that there may be crucial changes in exposures (including
environment and lifestyle) over time and recommended that returning
to the cohort for updated exposure information should be budgeted
for from the outset. It is now planned that all study participants
will undergo some form of re-survey approximately five years after
recruitment, however funds for this have not been included in
the budget[42].
Nested project costs
Even if follow-up questionnaires are funded,
there is considerable scientific concern that the data collected
within Biobank UK will be insufficient to be of scientific use
(see Section 2.2). Participants in the Protocol Development Workshop
noted that the true costs of meeting the academic objectives of
the study might be in excess of £1,000 per participant, and
that the costs of sub-group studies needed to be included from
the outset. At the Workshop, a much more detailed nested study
of between 20,000 and 100,000 individuals was proposed, without
which many scientists believed that the usefulness of the biobank,
particularly for studies of cardiovascular disease, was "very
weak". Funds for this study, or other nested studies, have
not been allocated, although they could be expected to cost many
millions. It is clear that the MRC expects commercial bodies and
perhaps medical research charities to contribute funds to nested
studies. However, it is unclear to what extent such funds will
be forthcoming. Over-reliance on commercial funding raises additional
ethical issues (see Section 3.1).
Cost of extending follow-up period
Even scientists that support the project believe
that Biobank UK would have limited value as a ten-year study of
the causes of disease. They see its value as providing a 20-30
year resource, particularly for nested case-control studies[43].
In addition to the absence of a budget for such nested studies,
no budget has been given for the costs of extending the project
for 20-30 years.
Administration and recruitment costs
Concerns have been expressed by health professionals
about the considerable time, effort and training required to properly
administer the biobank, particularly the necessary input from
GPs and nurses[44].
Although nurse time per patient is included in the budget, it
is unclear whether the full costs to the NHS have been considered,
particularly if participants need to be re-contacted for follow-up
or nested studies.
Annex B
SCIENTIFIC LIMITATIONS
In April 2001, the MRC, Wellcome Trust and Department
of Health held a Protocol Development Workshop for Biobank UK[45].
The overall impression given by the Workshop was that the aim
of the project to identify interactions between genes and other
risk factors (environmental/lifestyle exposures) is highly unrealistic.
The workshops on both cardiovascular disease
and on diabetes and metabolic disorders both suggested that the
potential of the project was poor or very weak, and that smaller
cohorts and more focused and detailed research was necessary.
The other workshops all produced lists of detailed (different)
questionnaires that would be necessary to investigate the likely
environmental/lifestyle factors involved in different diseases,
and highlighted the probable need for repeated screening, history
data, and a wide range of physiological measurements. Medical
records were widely considered to be inadequate to provide the
necessary data for meaningful research, and the need for considerably
more money and time in order to obtain meaningful results was
also highlighted.
For example, "The opinion was very strongly
expressed, that the current proposal for the 500K study (focusing
on a relatively elderly cohort, and powered for incident cases
of common cancers) would be a poor vehicle for study of cardiovascular
and metabolic disease". Participants discussed the need to
measure intermediate traits such as blood pressure, lipids and
insulin and concluded that "The potential contribution of
the proposed cohort to studies of cardiovascular disease was thought
to be very weak without the more extensive phenotyping" and
"If such an intermediate trait analysis were not possible,
the possible uses of the study for diabetes/metabolic disease
would be limited". Participants proposed a studying a smaller
group of 20,000 to 100,000 people in more detail, however funds
for such a study have not been allocated. Some scientists have
argued that funding this type of smaller study on its own would
not require the larger Biobank UK and would provide much better
value-for money and a better test of the causes of disease[46].
However MRC has not provided an alternative budget or comparative
assessment of this type of approach.
In relation to mental illness and hearing and
visual impairment, the Protocol Development Workshop noted that:
— GP recording and diagnosis of common
psychiatric or neurological disease and hearing or visual impairment
is known to be unreliable;
— It is estimated that fewer than 25%
of all cases of dementia are known to their GP;
— Death certificates are of little
or no value in studying mental illness;
— Direct screening of much or all of
the cohort may be the only feasible approach for accurate follow-up.
The issue of complexity has also been sidelined.
The Biobank UK Protocol includes calculations of the statistical
power of the study to detect the expected risks associated with
genetic and environmental factors. However, the authors recognise
that the situation will be complicated by variations in the risk
associated with a particular gene (known as "penetrance"),
interactions between multiple genes, interactions between different
one environmental factor and another, and confounding factors
(hidden variables in the population being studied). The real ability
of the study to quantify the risks associated with particular
genetic and environmental factors therefore remains highly questionable.
The Protocol Workshop participants also considered
the data on drug prescriptions and side-effects, and noted that:
— follow-up data on prescriptions would
be difficult if people moved practice;
— data on drugs given only for a short
period are likely to be less reliable;
— sufficiently detailed information
on clinical response to drugs might be difficult to obtain from
medical records;
— information on drug levels will not
be possible, but is necessary for detailed study;
— it would not be feasible to obtain
data on compliance in a study of this scale (it is estimated that
only around one-half of people with chronic diseases take their
medication as recommended[47]).
People's response to medicines may certainly
vary according to their genetic make-up, although many other factors
are also likely to be involved. However, the ability to assess
genetic variations in drug response without being able to measure
drug levels in individual patients, and without accurate records
of either drug intake or side-effects will be very limited.
Because the links between genes and diseases
are complex, it is common for genetic research to identify spurious
links between genes and diseases, which cannot be replicated in
subsequent studies[48].
If research in Biobank UK leads to many such spurious claims,
it will not only be a waste of money but may also lead to poor
or dangerous health interventions.
Annex C
GENETIC TESTING
AND DISEASE
PREVENTION
Current evidence suggests that, even if Biobank
UK correctly identifies the links between genes, environmental
factors and diseases (and this is doubtful, see Section 2.2),
for most people genetic tests are likely to be poor predictors
of future health[49].
For example, one paper has found that the more
common a genetic variation is in the population, the higher proportion
of cases of a disease it may contribute to, but the lower predictive
value it will have[50].
This means that for common, complex diseases many people with
the "high risk" form of the gene will not get the disease
and many people without it will.
There are therefore concerns that for many people
genetic testing combined with preventative medication ("predictive
medicine") could do more harm than good—by worrying
them and giving them medicines that they do not need—perhaps
with unpleasant or even fatal side effects[51].
Giving preventative medicines to all those identified as "genetically
susceptible" means that many people would be treated—perhaps
for their whole adult life—who would never have suffered
the predicted illness.
The pharmaceutical company, GlaxoSmithKline
has identified predisposition profiling—"the ability
to assess an individual's risk for a disease or diseases so that
medicine can be given to prevent illness"—as one key
area of the new "predictive medicine" and has argued
that integrating genetic testing and pharmaceutical products will
increase market size for pharmaceutical products and services[52].
This marketing strategy—using genetic tests to target medication
at those identified as at "high genetic risk"—is
sometimes known as "pills for the healthy ill". As well
as being expensive and causing unnecessary side effects, it could
also take resources away from treating the sick and from preventing
the underlying causes of many diseases (such as poor diets, lack
of exercise, smoking and pollution). The costs to the NHS of using
this approach to try to prevent common, complex diseases have
not been evaluated.
On the other hand, targeting lifestyle interventions
at those identified as "genetically susceptible" could
miss most potential cases of a disease, which many people other
than those identified as "genetically susceptible" could
suffer from. This "targeted" approach to limiting exposures
will always have less effect than policies that seek to limit
everyone's exposure, particularly as one type of exposure (such
as smoking) is often associated with many different diseases.
Genetic testing will also not necessarily increase targeted individuals'
motivation or ability to change behaviour[53].
Finally, there is potential for genetic discrimination, for example
via the exclusion of people with a particular genetic make-up
from insurance or employment (See Section 3.3).
As an example, at the current rate of increase,
more than one in four adults will be obese by 2010 and at a higher
risk of a range of major health problems including heart disease
and adult-onset diabetes[54].
The cause is obviously not an increase in "genes for obesity",
but unhealthy diets and lack of exercise—problems which
need tackling across Government departments, using measures to
integrate healthy eating and exercise into daily life, for example
by expanding the Government's fruit in schools programme. Too
much emphasis on susceptibility genes for obesity could lead to
the underlying causes of the epidemic being ignored, and the targeting
of expensive diet or anti-cholesterol pills to those identified
as "genetically susceptible" instead. The approach of
targeting interventions at a minority of high-risk individuals
(even if such individuals have been correctly identified) is well
known to be limited in comparison to population-based approaches[55].
This does not mean that seeking to identify
genetic factors in disease cannot sometimes provide health benefits.
Finding a genetic factor that plays a role in a disease may in
some cases help the development of new medicines, via an improved
understanding of how that disease develops. It may also help identify
the small minority of people at particularly high risk of the
rare familial forms of heart disease and cancer. However, Biobank
UK's main aim is not simply to identify a gene that may play a
role in a disease (which may be done more efficiently in other
ways), but rather to try to quantify the combined effects of genotype
(genetic make-up) and exposure on the risk of disease. If widespread
genetic testing is not a useful or cost-effective way or reducing
the incidence of common, complex diseases, then Biobank UK will
be of little benefit to public health.
1 "BioBank UK: A Question of Trust: A Consultation
Exploring and Addressing Questions of Public Trust", Report
prepared for the Medical Research Council and Wellcome Trust by
People Science and Policy Ltd, March 2002. Available on http://www.wellcome.ac.uk/en/1/biovenpopethtwo.html Back
2
"Top projects suffer as medical funding falters", Nature,
418, 714, 15 August 2002. Back
3
"Protocol for BioBank UK: A Study of Genes, Environment
and Health", The Wellcome Trust/Medical Research Council
and Department of Health, February 2002. Available on http://www.mrc.ac.uk/index/public-interest/public-consultation/public-biobank.htm. Back
4
Clayton, D, McKeigue, PM (2001), Epidemiological Methods for Studying
Genes and Environmental Factors in Complex Diseases, The Lancet,
358, 1356-1360. Back
5
Coghlan, A (2000), A Wild Gene Chase, New Scientist, 9
December 2000. Back
6
http://www.mrc.ac.uk/index/public-interest/public-consultation/public-biobank/public-biobank-why.htm. Back
7
Protocol, p8. Back
8
Burton, P, McCarthy, M, Elliott, P (2002), Correspondence-Study
of Genes and Environmental Factors in Complex Diseases, The
Lancet, 359, 1155-1156. Back
9
Baird, P (2001), The Human Genome Project, Genetics and Health,
Community Genetics, 4, 77-80. Back
10
Melzer, D, Zimmern, R (2002), Genetics and Medicalisation, British
Medical Journal, 324, 863-864. Back
11
Beaglehole, R (2001), Global cardiovascular disease prevention:
time to get serious, The Lancet, 358, 661-663. Back
12
WHO(2002), Genomics and World Health. Back
13
Nature, 2002. Op Cit. Back
14
"Fury at plan to sell off DNA secrets", The Observer,
23 September, 2001. Back
15
House of Lords Select Committee on Science and Technology, Human
Genetic Databases: Challenges and Opportunities, Session 2000-01,
4th Report. Minutes of Evidence, para 39 (Examination of Sir John
Pattison, Department of Health). Back
16
Hapgood, R, Shickle, D & Kent, A (2001), Consultation with
Primary Care Health Professionals on the Proposed UK Population
Biomedical Collection, Wellcome Trust and MRC, April 2001. Back
17
Wellcome Trust and MRC (2000), Qualitative Research to Explore
Public Perceptions of Human Biological Samples, October 2000. Back
18
"BioBank UK: A Question of Trust: A Consultation Exploring
and Addressing Questions of Public Trust", Report Prepared
for the Medical Research Council and The Wellcome Trust by People
Science and Policy Ltd, March 2002. Back
19
Econexus and GeneWatch UK (2001), Patenting Genes-Stifling Research
and Jeopardising Healthcare, April 2001-available online at http://www.genewatch.org/publications/briefs/brief14.pdf Back
20
Nuffield Council on Bioethics (2002), The Ethics of Patenting
DNA: A Discussion Paper, July 2002. Back
21
San Francisco Chronicle, 25 March 2002. Available on http://sfgate.com/cgi-bin/article.cgi?file=/chronicle/archive/2002/03/25/BU9 Back
22
http://www.bms.com/news/features/data/pf-features-1196.html. Back
23
Highfield, R (2002), DNA Bank is a Big Gamble, Says Expert, The
Telegraph, 6 May, 2002. Back
24
Schulte, PA, Lomax, GP, Ward, EM, Colligan, MJ (1999), Ethical
Issues in the Use of Genetic Markers in Occupational Epidemiologic
Research, Journal of Occupational and Environmental Medicine,
41(8), 639-46. Back
25
OTA(1990), Genetic Monitoring and Screening in the Workplace,
US Office of Technology Assessment. Back
26
TUC(2001), TUC Response to the Human Genetics Commission: Ban
Unfair Screening, 20 March 2001. www.tuc.org.uk/h-and-s/tuc-2921-f0.cfm
. Back
27
US Department of Labor, Department of Health and Human Services,
Equal Opportunity Commission, Department of Justice (1998), Genetic
Information and the Workplace, 20 January 1998. Available on www.nhgri.nih.gov/HGP/Reports/genetics-workplace.html. Back
28
Kaufert, PA (2000), Health Policy and the New Genetics, Social
Science and Medicine, 51, 821-829. Back
29
"Protecting Individuals and Promoting Science", Nature
Genetics, 28 (3), 195-196. Back
30
Gottlieb, S (2001), US Employer Agrees to Stop Genetic Testing,
British Medical Journal, 322, 449. Back
31
Mohr, S, Gochfeld, M, Pransky, G (1999), Genetically and Medically
Susceptible Workers, Occupational Medicine: State of the Art
Reviews, 14 (3), 595-611. Back
32
OTA(1990), Genetic Monitoring and Screening in the Workplace,
US Office of Technology Assessment. Back
33
US Department of Labor, Department of Health and Human Services,
Equal Opportunity Commission, Department of Justice (1998), Genetic
Information and the Workplace, 20 January 1998. Available on www.nhgri.nih.gov/HGP/Reports/genetics-workplace.html. Back
34
Day, G (2000), Testing Times: Directors' Views on Health Testing
at Work, Institute of Directors Research Paper. Back
35
Nicas, M, Lomax, GP (1999), A cost-benefit analysis of genetic
screening to occupational toxicants. Journal of Occupational
and Environmental Medicine, 41, 535-44. Back
36
"Laboratory Services for Genetics", Report of an expert
working group to the NHS Executive and the Human Genetics Commission,
August 2000. Back
37
Dyer, C (2001), Scientists Fear Breach of Confidentiality Will
Threaten Research, The Guardian, 17 March 2001. Back
38
Human Genetic Commission (2002), Inside Information, May 2002.
Section 9 (Forensic Uses). Back
39
"Surnames Found in DNA", BBC Online, 4 April 2000.
http://news.bbc.co.uk/1/hi/sci/tech/701286.stm. Back
40
Speech by Rt Hon Alan Milburn MP, 16 January 2002. Back
41
"Report of the UK Population Biomedical Collection Protocol
Development Workshop", held at the Royal College of Physicians,
London on Tuesday 17 April 2001. The Wellcome Trust, Medical Research
Council and Department of Health. Back
42
Protocol, p21. Back
43
Burton, P, McCarthy, M, Elliott, P (2002), Study of Genes and
Environmental Factors in Complex Diseases (Letter), The Lancet,
359, 1155-1156. Back
44
Hapgood, R, Shickle, D & Kent, A (2001), Consultation with
Primary Care Health Professionals on the Proposed UK Population
Biomedical Collection, Wellcome Trust and MRC, April 2001. Back
45
"Report of the UK Population Biomedical Collection Protocol
Development Workshop", held at the Royal College of Physicians,
London on Tuesday 17 April 2001. The Wellcome Trust, Medical Research
Council and Department of Health. Back
46
Clayton, D, McKeigue, P (2002), Study of Genes and Environmental
Factors in Complex Diseases (Authors' reply), The Lancet,
359, 1157. Back
47
Audit Commission (2001), A Spoonful of Sugar: Medicines Management
in NHS Hospitals. Back
48
Ioannidis, JPA, Ntzani, EE, Trikalinos, TA, Contopoulos-Ioannidis,
DG (2001), Replication Validity of Genetic Association Studies,
Nature Genetics, 29, 306-309. Back
49
Holtzman, NA, Marteau, TM (2000), Will Genetics Revolutionize
Medicine?, New England Journal of Medicine, 343, 141-144. Back
50
Khoury, MJ, Wagener, DK (1995), Epidemiological Evaluation of
the Use of Genetics to Improve the Predictive Value of Disease
Risk Factors, American Journal of Human Genetics, 56, 835-844. Back
51
GeneWatch UK (2002), Genetics and Predictive Medicine: Selling
Pills, Ignoring Causes, Briefing No. 18, May 2002. Available on
www.genewatch.org. Back
52
Gilham I, Rowland, T (2001), Predictive Medicine: Potential Benefits
from the Integration of Diagnostics and Pharmaceuticals, International
Journal of Medical Marketing, 2(1), 18-22. Back
53
Marteau, TM, Lerman, C (2001), Genetic Risk and Behavioural Change,
British Medical Journal, 322, 1056-1059. Back
54
National Audit Office (2001), Tackling Obesity in England, 15
February 2001. Back
55
Rose, G (1985), Sick Individuals and Sick Populations, International
Journal of Epidemiology, 14 (1), 32-38. Back
|
