Memorandum by the Medical Research Council
1. As the UK's principal public-sector funder
of genetics research relevant to human health, MRC welcomes the
opportunity to contribute to the Committee's inquiry.
2. Human Genetic Databases play an important
role in medical research and will do so increasingly over the
coming years. The major benefits of knowledge of the human genome
sequence for public health will come from the discovery of genetic
factors predisposing people to develop the common, multi-factorial
diseases of later life, and those affecting an individual's response
to treatment of such disorders. Knowledge will allow the possibility
of more effective use of existing treatments and also the development
of new treatments and interventions to prevent the progression
of disease. Databases bringing together health information and
genetic information are one of the key tools we can use to make
What current projects involve collecting genetic
information on people in the UK?
3. We believe that policy on the operation
of human genetic databases needs to reflect a broad definition
of research in this area, that includes not only collection of
genetic sequence information, but also collections of human biological
samples and tissue that could be suitable for genetic analysis.
4. A survey of MRC and Cancer Research Campaign-funded
scientists in 1998-99 suggested there are well over 300 collections
of cell or tissue samples in use, within this research community
alone. Many were developed with support from more than one funder,
and are likely to have been counted in the estimates submitted
by the Wellcome Trust and other charities. The diseases addressed
included cancer, heart disease, asthma, diabetes, dementia, stroke,
Parkinson's disease, malaria, HIV, influenza and many others.
Most of the existing collections are small: two thirds involve
samples from 500 people or less, and only five exceeded 10,000
samples. DNA sequence information from these samples is invariably
accumulated in the course of research.
5. The collections and their uses are very
diverse. Some collections of tissue samples are used purely for
studies of the physical, molecular and cellular processes involved
in the disease. Those used in genetics research could be grouped
Research to identify genes involved in disease
6. Collections of tissue or DNA from people
with and without a particular disease can be used to identify
the gene or genes key to processes underlying the disease. Sometimes
the comparison of samples from ill and healthy peopleor
whole familiesis the first and most important method of
identifying the genes involved. More often, a number of approaches
are pursued, and the collections may be used to test ideas about
genes identified through animal studies, or more basic studies
of variations in the properties of molecules or cells. Most common
conditions, such as hypertension, cancer and mental illness, are
influenced by variations in many different genes, each modest
effect located in different parts of the genome: in these cases,
research studies require greater numbers of samples than when
the disease is caused by a single gene of large effect.
7. Example: the "BRIGHT" study
is a major UK collaborative study, co-ordinated from Glasgow University,
which aims to identify the main genes affecting the risk of hypertension.
Volunteers from 1,500 families nation-wide have been recruited
through MRC's General Practice Research Framework, and the genetic
analysis is being carried out at the Wellcome Trust Centre for
Human Genetics in Oxford.
Research into gene/environment interactions
8. To gain a full understanding of the causes
of common diseases, research needs to address genetic factors
and external factors, such as diet, lifestyle, and environment,
in parallel. The effects of diet on health may be strongly influenced
by an individual's genetic make-up; conversely, when it is discovered
that a variation in a particular gene may predispose to disease,
the best preventative measures may prove to be a dietary or lifestyle
change, rather than drug therapy or regular check-ups. Such data
collections usually need to be large, and to collect detailed
information about people's health, lifestyle and medical care
over a long period of time.
9. Example: the Avon Longitudinal Study
of Parents and Children (ALSPAC), now in its ninth year, was specifically
designed to analyse this interplay between genes and environment
in areas such as childhood, infection, allergy, asthma, growth
and development. The project is based on information about 14,000
children born in 1991 and 1992, and their parents. Information
is derived from the analysis of questionnaires completed by their
parents, measurement of the environment in the home, non-genetic
assays of biological samples, and tests at age seven and eight,
as well as analysis of DNA. Over 127,000,000 items of information
are already on the main database, and some 70 research papers
have been based on the survey.
Collections linked to evaluations of medical treatments
10. Patients participating in clinical trials
are being asked for samples for genetic studies in a few cases
at present, but we expect that such samples will increasingly
be used to analyse the genetic influence on disease progression
and the response to treatment. A current example is the collection
of samples from adult patients with leukaemia entered into MRC-sponsored
clinical trials. Analyses of the DNA and RNA, linked to knowledge
about clinical outcomes and response to treatment, have already
allowed identification of mutations associated with differences
in prognosis in different patients. This research will allow therapy
to be tailored to specific tumour types, and already allows identification
of patients with a poorer prognosis, in whom more aggressive therapy
might be appropriate.
11. As well as the scientific diversity,
it is also important to recognise the different contexts in which
the collections are assembled. The relationship between the individual
and the collection is as important as the nature of the science
in determining the ethical issues that need to be considered in
each case. Some depend on healthy people providing samples and
information about their life and medical history through altruism,
sometimes on many occasions over the years. Others are based on
samples given by patients and their families, sometimes in the
hope that the research results may directly benefit them in their
lifetime. Whereas others may be collections of anonymous samples
linked only to basic information about the diagnosis of the person
from whom they came.
Are there collections of material (eg tissue samples)
that could be used to generate databases of DNA profiles?
12. It is technically possible to extract
DNA from most recent collections of human tissues (including blood)
that have been stored for research, and from archives of samples
taken for routine diagnosis. Whether it is acceptable and worthwhile
doing so depends on the sample's size and storage, the quality
of the accompanying information about the donor's health, and
the nature of the agreement with the original donor about how
the sample should be used. In most existing collections, the donor
consented to the sample being used only for a specific project
or line of research.
13. Current MRC guidelines stipulate that
old samples taken for routine clinical purposes, where the individual
was not asked to consent to research, can normally only be used
for research if anonymised, and with the approval of an ethics
What is the genetic information that is being
collected? How is it being stored?
14. Most newly established collections of
material for genetic analysis are based on samples of blood, from
which white cells are extracted and frozen (mature red cells do
not contain chromosomes). Either DNA can be extracted from the
white cells, and used until the supply runs out, or the sample
of cells can be immortalised (ie changed so that they divide and
reproduce indefinitely) to provide a larger supply, though this
is a costly procedure. For some research, such as work on gene
mutations in cancer cells, samples of frozen tissue are needed.
15. The nature of the genetic analysis varies:
in some studies, the aim is to gather information about the likely
position within the genome of genes that appear to influence particular
diseases, as a prelude to more detailed work to identify the gene
and sequence it. When research explores the links between known
genes and health, the study might typically involve analysis of
samples of DNA or RNA to detect variations in the code of anywhere
from one to 20 genes, out of the 100,000 or so genes that a person
has. New technology is beginning to allow larger scale analyses,
however, that might allow analysis of the key variations in hundreds
of genes at a time from each person.
What other projects are about to start? Why are
these genetic databases being assembled?
16. Genetic databases and collections have
a central place in MRC's strategy for developing the research
skills and resources to apply advances in genetics to better understanding,
prevention and treatment of common, complex diseases. MRC's plans
reflect the following principles:
the UK's strengths in population
based research and clinical trials, the almost universal population
coverage by the NHS, and the UK's strengths in genetics, mean
we are well placed to play a leading international role in large
studies of common diseases;
as the UK's principal public sector
funder, MRC has a special responsibility to develop resources,
such as large collections, that individual research centres or
small charities could not;
collections of samples donated through
public goodwill should be developed in a co-ordinated way, and
access and sharing of collections must be managed to maximise
the scientific return and reflect the wider public interest;
the ethical codes governing the use
of collections should be transparent and as consistent as possible
across the UK.
17. To clarify the standards that MRC-funded
research must meet, and to help build a wider ethical consensus
across the various different stakeholders in the UK, MRC has prepared
draft guidelines on the use of human tissue and biological samples
in research (Annex A) (not printed), which will be finalised
at the end of this year. A separate guide on general aspects of
the use of personal data in medical research will be published
on 5 October 2000.
18. The guidelines underline researchers'
responsibilities to ensure optimum use is made of human tissue
collections. MRC expects all those holding collections to co-operate
with reasonable and ethical requests from other research teams
for access to the collection, and, for newly funded collections,
will only release funds once clear plans for managing access to
the collection have been agreed.
19. To help make best use of existing collections,
MRC, with the Cancer Research Campaign, will be making details
of over 300 collections available on the Internet based on information
derived from the survey mentioned in paragraph 4. However, the
consent given by the donors will in many cases limit the new uses
to which these collections can be put.
20. Next year, MRC expects to commit funds
to establish national DNA banking facilities, to provide central
and secure storage for DNA samples, to facilitate sharing of samples
and data for use in medical research within the original terms
of the participant's consent.
21. In many cases, identifying the many
genes involved in common diseases, and their relative importance,
requires more fully characterised collections of samples and medical
data than exist already. In the 1998 Comprehensive Spending Review,
MRC bid successfully for additional funds to develop new DNA collections.
As part of this initiative, we invited proposals for large-scale
collections on the Web earlier this year of samples from patient
cohorts or case-control studies.
22. Priority has been given to collections
relating to common diseases of public health importance, and where
sample collections are not already available. The call generated
considerable scientific interest; out of 155 outline applications,
27 research groups were invited to submit full proposals, and
funding decisions will be taken in October. This call complements
the Wellcome Trust's recently advertised Functional Genomics Development
The proposed UK population biomedical collection
23. In May 1999, following preliminary discussions
within MRC and the Wellcome Trust, the Council and the Wellcome
Trust jointly organised a workshop, with representatives from
other major UK research funders, to consider whether the UK would
benefit from one or more new collections, large enough not only
to help identify genes, but also to analyse the interplay between
different variants of important genes and individual lifestyle
and environment factors in the development of common complex diseases.
The workshop concluded that a large, long-term collection would
provide opportunities for advancing understanding of health and
disease that could not be achieved in other ways. In the UK and
world-wide, most existing collections are too small to allow statistically
meaningful research, do not have enough high quality health information,
have too little DNA left, or are not based on full consent for
this sort of research.
24. Following further analysis, a Working
Group involving MRC, the Wellcome Trust and the Department of
Health recommended establishing a long-term study of about 500,000
volunteers, aged 45 to 64, and representative of the UK population.
MRC Council supports this proposal, which has also been agreed
in principle by the Wellcome Trust. Firm plans for the study are
still being developed, but the expectation is that it would involve
each volunteer providing a blood sample via their GP from which
DNA would be extracted. In addition to a detailed interview to
establish the individual's agreement to participate, basic measurements
(height, weight, blood pressure, lung function, etc) would be
made and participants would complete a questionnaire about their
medical history and lifestyle. With the participants' consent
and involvement, follow up data on their health and lifestyle
would be collected over the succeeding years, and the results
of genetic analyses carried out on all, or subsets of, the volunteers
by separate research groups would be held centrally, so that the
collection would become a much more valuable resource with time.
25. The 45 to 64 age group was chosen for
the survey because this is the stage in life when the major diseases
of adult life become common: the sample size is expected to be
around 500,000 so that within a few years of the start it should
be possible to undertake meaningful research into significant
numbers of ill people, in many different disease areas.
How do the organisations involved see their responsibilities
regarding privacy, consent, future use, public accountability
and intellectual property rights?
Consent and ethical guidance
26. In 1998 MRC decided that additional guidance
was required on the ethical, legal and management issues relating
to the use of samples of human tissue (including DNA) in research,
recognising that such samples would be increasingly important
in understanding the function of human genes and translating that
knowledge into benefits for health. A Working Group including
researchers, ethicists, and lay members, and representatives of
the Department of Health and other funders developed new guidelines,
with widespread consultation with research and consumer groups.
These were published as an interim version for further public
consultation in November 1999 (Annex A) [Not printed].
The revised version will be available by the end of this year.
They have been broadly welcomed and endorsed by other organisations
such as the Wellcome Trust and the Cancer Research Campaign.
27. The law on ownership of samples of human
tissue is unclear in the UK. The MRC guidelines highlight the
need for secure and responsible custodianship of samples and the
data associated with them, and recommend that in most instances
formal responsibility for custodianship should rest with institutions
rather than individual researchers. They highlight the importance
of obtaining proper informed consent from donors, and detail issues
that must be addressed in the consent process to ensure that participants
have a full understanding how any planned and future research
using their sample might affect their interests. The guidelines
recommend a two-stage consent process, first consent for the experiments
immediately planned, and second for a broader (but limited) range
of possible future experiments. Consent must always be obtained
for genetic research, and any secondary use of samples must be
approved by an ethics committee, which must decide whether the
use proposed is within the terms of the consent given by the donors
or whether new consent is required. However, genetic tests of
known clinical predictive value should never be done on samples
that can be linked to individual donors without their express
28. The responsible use of old sample collections
presents difficult issues. Such collections can be very valuable
or impossible to replicate, but often consent for secondary use
was not obtained and it can be impractical or impossible to approach
the original donors for further consent. The MRC guidelines recommend
that old collections can be used for secondary research if the
samples are anonymous to the researchers, and if the rationale
of the research is approved by a suitable ethics committee.
29. Databases of genetic information can
be anonymised, but for much medical research, including that on
the genetic factors affecting disease risk or response to treatment,
it must be possible to link individual data to names, contact
addresses, or other information that can identify individuals
in order that the database can be updated with follow-up surveys
about people's health and lifestyle from time to time. However,
there is no need for any more than a few people to have access
to this identifiable information. For databases used by several
research teams, the majority of researchers should only have access
to anonymised data. The MRC guidelines stress that any researcher
with access to identifiable data must have a formal duty of confidence
to research participants.
Public accountability and supervision
30. There are a range of mechanisms in place
to ensure public accountability for medical research involving
human participants and materials, including genetic databases.
As well as consenting to participate at the outset, generally
(excepting anonymous databases) any individual is entitled to
ask to withdraw from any research project at any time and for
data on them to be removed. Individuals participating in research
should also be told who to contact if they have any complaints
about their treatment, and how to take the complaint further if
31. All research involving human participation
or materials is already reviewed by independent Research Ethics
Committees. The Department of Health's proposals to strengthen
research governance should help ensure that, even where a project
has no external funder, there are local mechanisms for supervising
the quality and conduct of the work.
32. Special governance arrangements are
felt to be necessary for the proposed UK population biomedical
collection, not least because of its scale and the longer lines
of communication between researchers and participants. Detailed
plans for a management structure are still being developed and
consulted upon, but the funders expect that it will be appropriate
to form an independent body, with lay and public representation,
to oversee custodianship of the DNA samples, management of the
collection, application of the ethical codes, the principles guiding
priority-setting, the quality of information available to participants,
and the handling of complaints. Members of this overseeing body
would be recruited through public advertisement.
33. The UK PBC can only be formed if the
initiative enjoys widespread public support: without volunteers,
the research cannot proceed. Arrangements for public consultation
and feedback are therefore central to the initiative, and the
consultation has to be a genuinely two-way process, with the aim
of developing the sort of collection that the public will support.
MRC and the Wellcome Trust have already commissioned a preliminary
study to explore perceptions of the use of biological samples
in research. The executive summary of the report, from Cragg Ross
Dawson Ltd, is at Annex B. [Not printed]
34. Further consultations are planned over
the coming months both with the public and with primary care practitioners.
Commercial use and intellectual property rights
35. While some research using genetic databases
will point towards possible new public health measures, other
research will identify genes important in disease and help clarify
their role, providing new targets for drug design. The UK PBC
may also contribute to pharmacogenomicsie establishing
how genetic variations make different individuals respond differently
to drugs or other treatmentsknowledge that is potentially
of great economic importance to the health service and to industry.
36. Attitudinal research carried out thus
far suggests that participants in the UK PBC will need to appreciate
that the knowledge gained will often be used by industry in developing
new treatments, and that this is one of the expected benefits.
At the same time, they will want assurances that decisions about
the management and use of the collection are being made in the
public interest, rather than in the interests of industry alone.
37. The exact terms of engagement whereby
industry, or academic groups funded largely by industry, will
be able to analyse samples or data in UK PBC, are still under
discussion. However, the principles are likely to be that:
industry will not provide core funds
for the creation of the collection;
industry users will be expected to
meet the full costs of survey work or analysis needed for their
since DNA samples are a finite resource,
and there is also a need to limit the volume of surveys which
participants are involved in, a basic prioritisation system will
need to be established to ensure all studiesfrom academia
and industryare well designed;
to ensure all the knowledge gained
is available for use in the public interest, all users will be
expected to add their results to a central database after a set
period of time.
These basic principles will be developed further
in discussion with industry at workshops later this year and in
38. Intellectual property arising from research
using samples of human tissue is expected to be protected by academic
or industry researchers in the normal way. Where sample collections
are established as shared resources, the Funders may act to ensure
that IPR arising from use of the UK PBC data is handled centrally
and exploited effectively, with the ultimate aim of ensuring that
potentially beneficial new treatments or diagnostic tests are
developed and made available for the public good as soon as possible.
What practical considerations will constrain developments?
(The comments below reflect the factors which could constrain
development of the proposed large UK biomedical and large patient
collections. All are being actively addressed by the funders and
further details can be provided as required.)
Development of databases of human genetic variation
39. Although the sequence of the human genome
is almost complete, there is currently insufficient information
available on the extent of sequence variation between different
individuals. The availability of a public database of single nucleotide
polymorphisms (that is, individual bases within the human genome
sequence at which variation is known to occur) will simplify the
search for genes in which variation is associated with the development
of disease. MRC strongly supports the principle that such information,
along with all basic DNA sequence information on humans and other
organisms, should be placed in the public domain, as free and
equal access by all researchers will speed up the translation
of such basic knowledge into improvements in human health.
40. To realise the benefits of this approach,
improvements in the technology to identify genetic mutations are
required, in order to increase the speed and decrease the cost
of genotyping, and to allow accurate results to be obtained with
smaller amounts of DNA. Using currently available technology it
is not feasible to screen thousands of individuals for hundreds
of thousands of possible mutations, so attempts to find genes
related to disease must rely on a targeted approach. The amount
of DNA available also constrains the amount of genotyping that
can be done on a single sample; work is ongoing to develop the
technology for whole genome amplification by the polymerase chain
reaction, but currently expansion of an individual DNA sample
requires the creation and maintenance of transformed cell lines,
which is both expensive and time consuming. Developments in statistical
methodology and the recruitment and training of bioinformaticians
are also required for analysis of the very large amounts of data
Public and professional support
As discussed above, the development of collections
of genetic and health information depends on widespread public
41. Since recruitment for UK PBC will be
through general practices, the initiative also depends on the
support and expertise of general practitioners and other practice
staff. In addition, a large number of research nurses will need
to be recruited and trained to carry out the initial interviews
and assessments of the 500,000 volunteers.
Availability of NHS records
42. A current constraint for studies following
up large numbers of individuals to study the effects of genetic
variables on disease is the cost and time involved in obtaining
the information on clinical outcomes. It would be significantly
more efficient if it were possible to obtain such information
from data collected by the NHS in the course of the research participants'
normal clinical care, rather than through questionnaires completed
by participants and their doctors. This would of course require
the participants' informed consent, and very careful measures
to protect confidentiality.
43. Currently such an approach is constrained
by the uneven quality of the data recorded and the lack of comprehensive
electronic health records. In planning the UK PBC, one option
that will be assessed will be the collection of data on health
outcomes through the electronic health records increasingly being
kept in General Practices.
How are activities to be funded?
44. It is currently assumed that MRC and
the Wellcome Trust will be partners in funding the setting-up
of the resource, and will provide the majority of the necessary
funds, although it is hoped that the Department of Health, the
Scottish Executive and other charities may also contribute.
45. The establishment of DNA banking facilities
and assembling large collections of samples for patients with
particular diseases will be funded by the MRC, in some cases the
funding is in partnership with medical research charities. However,
these resources are intended for use by all bona fide researchers
and it is envisaged that funding for research using the samples
will come from multiple sources. For example it is likely that
a number of disease related research charities would fund programmes
of work based on these resources.
How are activities in the area of genetic databases
developing in the future? What advances in sequencing, screening
and database technology are they anticipating?
46. The key technological and scientific
developments that will shape future use of DNA databases are:
advances in the speed and affordability
of genotype analysis (ie identification of genetic variations
in specific genes in individuals);
advances in the quality of electronic
advances in the ability to transfer
and link health records between centres securely;
development of new mathematical/statistical
tools, and software, for analysis of large databases;
expansion of the UK's pool of skilled
personnel in health informatics and bioinformatics.
What lessons should be learnt from genetic database
initiatives in other countries?
47. The key lessons are that:
large collections will only win widespread
support if they are publicly accountable, and managed for the
the public needs to be engaged in
discussion about the aims, safeguards, and ownership of a survey
at an early stage, and the dialogue needs to continue throughout
participation in genetic surveys
should be based on consent, opt-in rather than opt-out, which
can be withdrawn at any stage;
plans should take account of the
need for genetic counselling or confirmatory tests if and when
results are fed back;
there need to be clear, visible boundaries
between medical and forensic databases;
collections need to be linked to
high quality health and lifestyle information if the genetic factors
influencing health are to be fully understood.