APPENDIX 4: SEMINAR HELD AT THE HOUSE
OF LORDS
19 March 2008
A seminar was organised at the House of Lords to
give the Committee an opportunity to discuss the Genomic Medicine
Inquiry with academic experts, representatives from the Department
of Health, the Department of Innovation University and Skills,
the Department for Business Enterprise and Regulatory Reform and
other organisations.
Members of the Sub-Committee present were: Lord Broers,
Lord Colwyn, Baroness Finlay of Llandaff, Earl of Northesk, Baroness
O'Neill of Bengarve, Lord Patel (Chairman), Baroness Perry of
Southwark, Lord Sutherland of Houndwood, Lord Taverne and Lord
Warner. In attendance were: Professor Tim Aitman (Specialist
Adviser), Elisa Rubio (Clerk), Christine Salmon (Clerk) and Dr Cathleen
Schulte (Committee Specialist).
The speakers were: Professor Tim Aitman (Specialist
Adviser to the Committee; Professor of Clinical and Molecular
Genetics, MRC Clinical Sciences Centre and Imperial College London);
Professor Sir John Bell (Regius Professor of Medicine,
Oxford University; President, Academy of Medical Sciences; and
Chair, Office for Strategic Coordination of Health Research (OSCHR));
Dr Ros Eeles (Reader in Clinical Cancer Genetics, Institute
of Cancer Research); Professor Wolf Reik (Associate Director
of the Babraham Institute in Cambridge; Professor of Epigenetics,
University of Cambridge); Professor Peter Donnelly (Professor of
Statistical Science and Director of te Wellcome Trust Centre for
Human Genetics, Oxford; and Chair, Wellcome Trust Case Control
Consortium); Dr Ewan Birney (European Bioinformatics Institute);
Professor Graeme Laurie (Professor of Jurisprudence,
University of Edinburgh; and Chairman, Ethics and Governance Council
UK Biobank).
Other participants were: Dr Adrian Pugh (Strategy
and Policy Support Officer, Biotechnology and Biological Sciences
Research Council); Dr Steve Sturdy (Deputy Director, Genomics
Policy and Research Forum, Economic and Social Research Council);
Nancy Lee (Policy Adviser, Strategic Planning and Policy Unit,
Wellcome Trust); Yvonne Gritschneder (Policy Officer, British
Heart Foundation); Dr Louise Jones (Experimental Cancer Medicine,
Cancer Research UK); Dr Peter Sneddon (Head of R&D Programmes,
National Institute for Health Research); Professor Peter
Furness (Vice-President, Royal College of Pathologists); Professor Peter
Farndon (Consultant Clinical Geneticist; and Director, UK Genetic
Testing Network); Dr Neil Ebanezer (Policy Manager, NHS Genetics
Team, Department of Health); Diana Paine (Team Leader, NHS Genetics
Team, Department of Health); Michael Davies (Research Councils
Unit, Department of Innovation, Universities and Skills); Dr David
Griffiths-Johnson (Bioscience Unit, Department for Business Enterprise
and Regulatory Reform); Dr Frances Flinter (Clinical Director
and Consultant Clinical Geneticist at Guy's and St Thomas' NHS
Foundation Trust; and Commissioner at the Human Genetic Commission);
Dr Rob Elles (Chairman of the British Society of Human Genetics;
and Director of Molecular Genetics, National Genetics Reference
Laboratory and Regional Molecular Genetics Service); Professor Richard
Trembath (Head, Division of Medical Genetics, Kings College London);
Professor Sandy Thomas (Head, Foresight Unit in GO-Science);
Dr Helen Munn (Director, Medical Science Policy, Academy
of Medical Sciences); Dr Sarah Bunn (Biology and Health Parliamentary
Adviser, Parliamentary Office of Science and Technology); Dr Hilary
Burton (Programme Director, PHG Foundation); Dr Ian Frayling
(Consultant in Genetic Pathology, University Hospital of Wales);
and Dr John Crolla (Chair, Joint Committee on Medical Genetics).
Introduction to genetics and genomic medicine
(Professor Tim Aitman)
Professor Aitman opened with a number of definitions.
Genetics was the science of heredity and variation in living organisms,
with the basic units of inheritance being called genes; and genomics
was the study of an organism's entire genome, its whole hereditary
information encoded in the DNA on the organism's chromosomes.
The word genome derived from the fusion of "gene" and
"chromosome".
There were two broad classes of genetic diseases:
- Mendelian diseases are rare diseases caused primarily
by defects in a single gene. Examples include cystic fibrosis,
haemophilia and Huntington's disease.
- Genetically complex diseases are more common,
with a prevalence of up to 30 percent, and are caused by an interaction
between genes and the environment. Examples include coronary heart
disease, diabetes, obesity, arthritis and common cancers such
as breast and prostate cancer.
Genome technologies had seen great advances in recent
years, driven in part by large-scale sequencing projects such
as the mapping and sequencing of the human genome. Automated DNA
sequencing and high throughput genotyping technologies detected
and measured sequence variations in the genome which were usually
inherited. DNA microarrays were a powerful method of genomic investigation
that allowed the expression of all the genes in the genome (20,000
- 30,000) to be measured in a single experiment. Use of DNA microarrays
had led to new and more precise molecular classifications of disease
states that were suggesting innovative treatment strategies for
a range of diseases.
New genome technologies had dramatically advanced
our ability to understand the inherited basis of common human
diseases. New generation DNA sequencers introduced at the end
of 2005 had led to spectacular increases in the quantity of data
output, as they were able to sequence 1,000 million base pairs
in a single run, often in a few hours. Similar increases in genotyping
capacity had, in just the last two years, led to a revolution
in identifying genes associated with common diseases. These technology
advances had enabled a new strategic approach, the genome-wide
association study, to be carried out. After the first publications
of this type of study towards the end of 2006, there had been
a flood of publications during 2007. Between 2005 and 2007 around
100 new genes for common diseases such as diabetes, arthritis
and cancer were identified. By the end of 2008 it was predicted
that another 400 will have been found.
Professor Aitman concluded that a new "genomic
information" era had arrived and was increasingly touching
healthcare professionals and the public. However, a number of
questions arose: how clinically useful and reliable was genomic
information in predicting and preventing common diseases? Were
we ready to put genomic information to good use? Were the costs
justifiable and affordable? And how should the UK take advantage
of these potential advances in healthcare?
Translation of genomics into healthcare (Professor Sir John
Bell)
Professor Sir John Bell described the impact
of genomics on healthcare in two main areas: diagnostics and therapeutics.
Advances in therapeutics were being driven by an increasing knowledge
of disease genes and mechanisms and through an enhanced ability
to predict drug efficacy and side effects. These advances were
mostly at an early stage of development. On the other hand, advances
in diagnostic testing using genomic tools were having a profound
impact on clinical decision making and many new tests had already
reached clinical practice.
Molecular diagnostic tests had led to an improved
ability to stratify common diseases, to predict risk of future
disease and to use drugs more effectively. For example, tests
in cancer patients that use DNA microarrays to measure gene expression
profiles and gene copy number could identify patient subgroups
with very different prognostic outcomes, and treatment could be
tailored to these different prognostic groups. This may not only
improve treatment outcomes, but may also lead to more efficient
use of existing therapies. The newly identified genes for common
diseases could also be used to test healthy individuals for their
risk of developing a range of common diseases, although these
genes mostly have a small effect on disease susceptibility and
the clinical utility of these tests in healthy subjects at present
remained to be defined. In most cases it was anticipated that
the range of new tests would be used in conjunction with existing
means of risk prediction and disease classification. However in
some specific diagnostic areas, such as the use of cytology screening
for cervical cancer, molecular diagnostics were rapidly gaining
ground as the method of first choice and may supersede conventional
screening tests.
The new discipline of pharmacogenetics aimed to personalise
drug treatment so as to optimise drug efficacy and to reduce the
frequency of adverse drug reactions. It was well known that most
drugs worked effectively in a minority of patients, and physicians
frequently relied on a trial and error approach to prescribing.
One way of improving drug efficacy was by using genetic tests
to distinguish responders and non-responders, and examples existed
where this approach had reached routine practice, for example
in the use of Gleevec in chronic myeloid leukaemia and herceptin
in breast cancer. Such genetic tests could be the most effective
way of establishing personalised treatment programmes, and by
increasing the proportion of patients who responded to a particular
therapy may also be effective in reducing overall drug costs.
Sir John identified five main obstacles to translation
of genetic testing more widely into the NHS:
- the hospital organisational structure, which
was currently not set up to use genetic testing across medical
specialties and different pathological disciplines;
- an increasing innovation gap in the NHS between
new tests becoming available and their delivery into clinical
practice;
- the commissioning system at the local level,
which was not oriented to the introduction of new diagnostic tests
and methods;
- the need to demonstrate clinical utility of new
tests—not only must tests be reliable and accurate, but
there should be evidence of clinical benefit and need; and
- costs.
These obstacles to translation would require innovative
solutions. For example, Sir John described the establishment
of a central Molecular Pathology laboratory in Oxford, where genetic
testing was carried out for all the conventional pathological
specialties. The requirement to demonstrate clinical utility of
new diagnostic tests posed significant regulatory challenges,
as present regulatory structures were not suited to adapting to
rapid changes in diagnostic technologies. New regulatory bodies
and procedures may therefore be required.
The Office for Strategic Coordination of Health Research
(OSCHR), chaired by Sir John, was a body set up by the Government
to oversee the translational agenda undertaken by the Medical
Research Council (MRC) and the National Institute for Health Research
(NIHR). The MRC undertook research leading to the discovery of
new diagnostics, but responsibility for proof of concept and clinical
utility trials rested with NIHR.
In conclusion, the likely impact of genomics on healthcare
was very large but a steady hand and clear vision would be required
to use genomics to deliver clinically useful and cost-effective
advances in healthcare across the NHS.
Genomics in cancer (Dr Ros Eeles)
There were about 300,000 cases of cancer per annum
in the UK, of which approximately 16 percent were breast cancer,
13 percent lung cancer, 13 percent bowel cancer and 12 percent
prostate cancer.
There were two types of common alterations in genome
sequence that were relevant to cancer susceptibility: somatic
changes, which took place in cancer cells only and were not heritable;
and germline alterations, which were found in sperm and egg DNA,
and were passed down from generation to generation. Major progress
was being made in cancer care by genomic profiling and sequencing.
The Cancer Genome Project which was being undertaken at the Wellcome
Trust Sanger Institute was looking at genomic changes in cancer
cells to determine patterns of DNA sequence changes that related
to cancer diagnosis and treatment outcome. Gene expression microarrays
were useful molecular tools to refine pathological diagnosis,
determine prognosis, guide treatment and predict response to treatment.
Dr Eeles gave two examples of ongoing genomic clinical trials
in cancer care: MINDACT (Microarray In Node-negative Disease may
Avoid Chemotherapy Trial) which was using microarray data in tumour
cells in breast cancer to ascertain whether chemotherapy could
be avoided; and a second study, being carried out in the USA,
that was investigating how the genetic make-up of patients determined
response to hormone therapy in prostate cancer patients.
Dr Eeles went on to talk about the genetic alterations
that were having an impact on public health and may lead to new
screening and treatment programmes.
There were several different types of DNA sequence
alterations that individuals could inherit, and examples were
cited for the breast cancer predisposition genes. Alterations
in the BRCA1 and BRCA2 genes were rare but they
convey a high cancer risk and a woman with alterations in one
of these genes was approximately ten times more likely to develop
breast cancer in her lifetime than women without such alterations.
BRCA1 and BRCA2 were therefore known as high risk
or "high penetrance" genes. By contrast, alterations
in the CASP8 and FGFR2 genes were much more common but the relative
risk of developing breast cancer for carriers of these genes was
very small. These genes were therefore of "low penetrance".
It was of interest that prostate cancer patients who had alterations
in the BRCA2 gene were twice as likely to die from the
disease as those who did not have BRCA2 gene alterations,
suggesting a common role for BRCA2 in breast and prostate
cancer. It cost £962 to screen for mutations in the BRCA1
and BRCA2 genes in NHS Genetics laboratories. Tests for
alterations in the CASP8 and FGFR2 genes were not
currently available on the NHS, but could be bought by the public
directly from genomic screening companies as part of a genomic
screen that currently costs around £500. As the cost of sequencing
and whole genome profiling had dramatically reduced, and continued
to do so, it was likely that DNA sequence alterations would be
detectable more quickly and cheaply in the future, permitting
wider use of targeted screening of high risk groups.
In the past two years, there had been an explosion
in genome-wide association studies in cancer that had identified
low penetrance genes for a wide range of cancers and other diseases.
Recently published studies covered breast, colon and prostate
cancer and there were on-going studies in lung cancer, lymphoma,
pancreatic, ovarian and testis cancer. However it was uncertain
how these discoveries could be applied to the clinic, and it was
also unknown how interaction of these low penetrance genes with
the environment may impact on disease susceptibility. Dr Eeles
and others had recently applied for a grant from the EU to investigate
these issues.
One potential application of genomic testing was
to guide the targeting of expensive screening tests to subsets
of the population who may have a higher than average risk of developing
a particular disease. For example, identification of individuals
from the general population who carried a significant alteration
in the BRCA1 gene, who therefore have a greatly increased
risk of developing breast cancer, could be used to target individuals
for screening with magnetic resonance imaging which was more expensive
and time-consuming, but also more sensitive than mammography.
Dr Eeles enumerated a number of issues to be
considered at the clinical interface in efforts to bring genomic
advances into health care. More research needed to be undertaken
in risk prediction and gene-environment interaction. On the other
hand, ongoing research should not stop clinical implementation
in cases of clear benefit. Access to genetic testing by specialists
and GPs needed to be clarified and the public and health professionals
needed to be educated on the potential value and implications
of genetic tests.
Dr Eeles finished with a word of caution, from
herself and colleagues at the Cancer Genetics Group, against over-regulation
of companies who sold genetic tests direct to the public. She
offered the view that, at present, the information that these
companies offered was of little value to consumers or healthcare
professionals but that, with further research, such information
would become useful in predicting disease in the future. Over-regulation
may impair or stop progress towards this objective.
Epigenetic factors and their importance in genome-wide
association studies (Professor Wolf Reik)
Professor Reik posed the question "who
do we think we are?" The answer should be in our DNA. All
the genes in the human genome were known but they were not used
all at the same time. Different sets of genes were switched on
or off during development of an organism to form different tissues
and organs. Epigenetics was defined as gene expression states
that were stable over rounds of cell division, but did not involve
changes in the underlying DNA sequence of the organism.
Epigenetic modifications generally turned genes on
or off, thus allowing or preventing the gene from being used to
make a protein. As a result cells would differ in their protein
content giving them different functions and forming diverse organs
such as the brain and heart. Epigenetic factors started working
as soon as the embryo was formed.
There were probably hundreds of epigenomes and what
was really important was that epigenomes were not only influenced
by genetic factors but also by the environment, nutrition, multigenerational
inheritance and by pregnancy. All these factors played a major
role in setting the shape that the multiple epigenomes had. There
were many associations beginning to emerge between epigenetic
marking and common diseases. Epigenetic factors had major influence
in cancer, it was suspected they had a major role in obesity and
psychiatric disorders and they were of key importance in the use
of stem cell therapies. There were multiple examples of environmental
influences resulting in altered epigenomes and possible disease
such as maternal grooming resulting in anxiety and altered methylation
of glucocorticoid receptors in children.
The challenge that we were facing was: if there were
hundreds of epigenomes, how could we determine what they were?
Given that the sequencing of a single genome took many years and
vast sums of money, this may seem like an impossible task. However,
next generation sequencing technology would make this a reality
very soon.
The UK was a World leader in epigenetics research
together with Japan and the US. However, we needed to build capability
in epigenomics and combine genetic mapping with epigenome sequencing,
as genetic variants interacted with epigenetic variants and the
nature of this relationship was largely unexplored.
Population genomics and insights into the genetics
of common diseases (Professor Peter Donnelly)
Professor Donnelly illustrated the pace of discovery
of genes associated with common diseases by explaining that until
October 2006 the number of common genetic variants that we knew
reliably to be associated with common diseases was very low. At
the end of 2006 the first results of a new generation of "genome-wide
association studies" started to be reported. While in 2005
only a handful of common genetic variants were known reliably
to be associated with common diseases, such as diabetes and macular
degeneration, in the year up to September 2007 more than 50 were
discovered that contributed susceptibility to a range of diseases
including coronary heart disease, prostate cancer and inflammatory
bowel disease. The pace of discovery was likely to continue to
increase, owing to better ways of analysing the data generated
in genome-wide association studies, to new and larger studies
being carried out across a range of common diseases, and to finding
ways of combining data from different studies.
Several factors had driven the current explosion
of genetic variants associated with common diseases. At the beginning
of the decade, the Human Genome Project provided a map where scientists
could start placing variants. The SNP Consortium was a private-public
partnership which aimed at finding single nucleotide polymorphisms
(SNPs)—letters in the genetic code that varied between different
human chromosomes. The International HapMap Project, a huge international
collaboration, then looked at the correlation of patterns of genetic
variation in different human populations. The most recent advance,
and a direct cause of the recent explosion of data, was the ability
to read as many as a million letters of the genetic code in different
positions in an individual's genome in a single chip experiment.
The Wellcome Trust Case Control Consortium (WTCCC)
undertook genome-wide association studies in seven different common
diseases, comparing the pattern of SNPs across the genome of 2000
people with each disease with the pattern in 3000 healthy people
in order to find sequence variants associated with predisposition
to each disease. This was the largest of the first generation
of genome-wide association studies and led to the discovery and
confirmation of more than 30 novel disease associations to date,
and around 20 more when combining data with other studies. After
decades of largely unsuccessful efforts, scientists had finally
found a method which was robust in terms of finding genes associated
with common diseases that could then be reliably reproduced in
other samples. Whilst it was agreed that these sequence variants
were robust markers of disease risk, at present it was not known
how most of these variants functioned to increase the risk of
disease development. Understanding the mechanism by which these
SNPs underlie disease risk was the subject of major ongoing global
research efforts.
Relative risk was a measure used to describe how
much a person's risk of disease was increased by a particular
genetic variant. For virtually all the loci found from association
studies, the estimated effect sizes were modest in some cases
and small in most cases. This supported the view that many genes
were likely to play a role in inherited susceptibility to common
diseases and that environmental risk factors, such as lifestyle
and environmental exposures, also had a major part to play. However,
since scientists estimated the effect sizes of the measured genetic
markers rather than the causative DNA change itself, it was likely
that the effect sizes had been underestimated. This underestimation
had consequences in the ability to use relative risk in disease
prediction, since relative risks at particular loci may increase
once the causative variants themselves have been identified.
It was also important to appreciate that although
the relative risk conferred by individual markers was not great,
combining information from many genetic markers and from conventional
measures of disease risk may identify segments of the population
who were at very significantly increased risk of individual diseases.
While most individuals would have an average risk for most diseases
almost everyone would be at very high risk for some diseases.
Professor Donnelly estimated that 95 percent of people would
be in the top five percent of genetic risk for at least one disease,
40 percent of people would be in the top one percent of genetic
risk for at least one disease and five percent of people would
be in the top 0.1 percent of genetic risk for at least one disease.
Professor Donnelly drew two main conclusions.
First, at a time when the new markers of common diseases had only
very recently been discovered, it was true that reliable disease
predictions were not possible, and therefore the clinical utility
of this new knowledge was uncertain. Nonetheless this may change
as we learn more about genetic variants, and when we were able
to predict disease for a range of, say, 50 diseases, each individual
was likely to be at a high relative risk for a few of those diseases.
It may therefore be useful to think of genomic tests, including
those sold "direct to consumers", as a tool for individuals
to identify the diseases for which they had the highest genetic
risk, based on current knowledge. However, in the context of tests
sold direct to consumers, if the information were to be of value
it was essential that suppliers of the tests were able to carry
out genomic tests accurately, and that they explained to their
prospective customers the pitfalls and limitations of such tests,
as well as the potential benefits.
The second conclusion was that while there was ongoing
uncertainty regarding the clinical utility of genomic tests for
disease prediction, the most important outcome of this new research
may be in advancing understanding of the molecular causes of disease
development, which was already providing new leads for ways in
which to prevent and treat common diseases.
Organisation and analysis of genomic data (Dr Ewan
Birney)
Dr Birney explained the key components that
would be required for an information infrastructure for Genomic
Medicine: a fundamental biology reference, patient-related information
and clinical knowledge. He further identified four basic principles
for a successful informatics infrastructure: (1) research infrastructures
were best constructed openly and should be coordinated on a national
and international level. The human genome was an example of open
infrastructure which was extensively used worldwide; (2) patient
information was not appropriate for public release; (3) informatics
hardware costs halved approximately every two years while the
sequencing capacity doubled every year; and (4) like most IT projects,
an information infrastructure for Genomic Medicine would require
complex management with the added difficulty that most informatics
developers were not trained in genetics and therefore the pool
of people with the appropriate expertise was very small.
With regards to the fundamental biology reference,
Dr Birney mentioned that the reference genome sequence would
be updated approximately every two years with minor updates (less
than one percent of the sequence), but the updated regions would
be disproportionately enriched for areas of interesting biology
and likely disease-associated regions. The reference gene and
biology resource were growing in stability and utility and were
also making advances in non-protein coding genes, but the dynamic
nature of these databases would necessitate a similarly dynamic
structure for clinical genomic databases, capable of adapting
to advancing knowledge. Dr Birney expressed the view that
information infrastructure in genomics was at present well funded,
though this needed constant investment from research councils
and charities and was coordinated worldwide.
By contrast, it was not yet clear how patient information
would be coordinated, assuming, for example, that we might have
the capacity to sequence the entire population's genomes in five
to ten years. This raised many questions: how should patient data
be coordinated with the reference genome? How would raw genomic
data be archived to allow for periodic recalling, for example
if technology advances yielded new information? Should genomic
information be part of SPINE (NHS care records system)? How would
genomic information be delivered in a useful way to practising
clinicians? Dr Birney suggested that useful answers might
emerge from comparison and dialogue with pilot projects such as
the informatics components of the 1,000 genomes project
Although resequencing projects have large storage
requirements, Dr Birney did not believe that disc space would
pose a problem for storage of genomic information compared to
other high density medically important datasets. Dr Birney
compared the disc space required for storing genomic information
to that needed to store a complex X-Ray digitally, and far less
space than that needed to store a CT scan. However, the challenge
in storing genomic information about patients was that the software
and the delivery would need to be custom made.
Finally, in resolving how to construct a usable "clinical
knowledge base" Dr Birney pointed to a number of existing
projects that could already provide partial or prototype solutions.
These include dbGAP/EGA and the EU-funded Gen2Phen projects which
linked genotype to phenotype; the Online Mendelian Inheritance
in Man database (OMIM) which provided clinician-friendly data
on genes and mutations underlying Mendelian (single-gene) disorders;
and a growing number of locus-specific and in-house databases.
Governance of genomic data (Professor Graeme
Laurie)
Professor Laurie identified six challenges for
optimal governance of genomic data: consent, confidentiality,
public confidence, commercialisation, collaboration and counselling.
Discussions on governance and genetics had taken
place over a long period. One of the first reports to be published
was by the Nuffield Council in 1993 titled "Genetic Screening:
Ethical Issues". The Human Genetics Commission (HGC) also
reported on a regular basis. We all share the same basic human
genome although each of us had individual variations that distinguish
us from other people. This highlighted our common interest in
the fruits of medically-based genetics research and a common public
good that could be achieved by optimum governance of genomic information.
However, an underlying assumption was that genetic information
was unique to an individual but that must not belie the fact that
there is a range of private and public interests at stake.
Consent had become the dominant paradigm in governance
of biomedical research, but there was a risk of over-dependency
on consent. Professor Laurie expressed the view that consent
was neither necessary nor sufficient to protect individual interests
and over-reliance on consent may serve as an obstacle to important
public interests.
There had been much recent discussion in the UK around
confidentiality and privacy. Protection of privacy under the law
in the UK was piecemeal. The Data Protection Act only protected
an individual's information when the individual could be identified:
anonymised information was not protected. Common law covered certain
aspects of privacy such as doctor-patient confidentiality, and
the Human Rights Act consolidated these protections. However,
an individual's privacy was not an absolute right: there were
exceptions when information could be processed to promote public
interests. Professor Laurie raised the question whether there
were sufficient flexibilities within existing law to promote such
public interests while adequately protecting the public interest
in individual privacy.
Security of genetic information was a serious public
concern. There may be informatics solutions to providing security
but these would not necessarily provide complete answers and did
not address all of the public expectations with respect to their
privacy. It was interesting to note how limited the current law
was in protecting the interests of others, such as family members,
especially when the genetic information of one individual may
have direct consequences for other family members. Another important
security issue was access to information: who had access, why,
and, particularly, where?
Public confidence in governance and government was
fundamental to progress in genomic medicine and genomic research.
The UK Biobank Ethics & Governance Council was set up to oversee
the legal and ethical implications of the UK Biobank project and
to monitor and advise the funders of the project on these issues.
The project had already recruited 100,000 participants of the
projected 500,000 and broad consent was given by the participants
based on robust confidentiality and a transparent ethics and governance
framework. There were ongoing questions about data access in the
future and commercialisation of the data which were incompletely
resolved at this stage. It is the role of the Ethics and Governance
Council to advise on these and other future developments.
With regards to commercialisation, Professor Laurie
highlighted the need for transparent access policies in today's
world of private investment in order to achieve fair, just and
equitable sharing. Open access may work for certain aspects of
research but not for others such as development of new drugs,
or other discoveries and inventions that may be commercially viable.
It may also carry unacceptable risks to individual privacy.
The UK Biobank was seen as setting the gold standard
in collaboration. Professor Laurie also referred to the Public
Population Project in Genomics (P 3G), an international collaboration
which aimed to promote scientific interoperability between biobanks
to ensure maximisation of scientific data generated, and to facilitate
scientific interoperability and data sharing. The P3G project
was also considering the governance regimes in different biobanks,
asking whether harmonisation of heterogeneous governance and data
protection regimes could facilitate progress and advance global
interests in genomics.
Professor Laurie ended by stating that optimal
governance was not yet with us. There were many examples of "self-help"
good practice such as UK Biobank and Generation Scotland. He suggested
that, in some cases, formal legal and ethical regimes were too
rigid and perhaps did not strike the best balance of interests.
There was an additional problem of a lack of regulatory "joined-up-ness"
across the various elements of innovation trajectories, from initial
conception through research, development, market and beyond.
Discussion
Discussion took place in which Committee members,
speakers and other attendees participated.
One participant gave the view that discussions on
clinical management run the risk of being hijacked by genetic
enthusiasts who overemphasised the importance of genetic factors
and genetic testing. Other conventional and possibly more important
factors in disease aetiology and management, such as lifestyle,
social factors and family history could therefore be disregarded.
In relation to drug therapy and pharmacogenetics, the current
trial and error approach to prescribing might be equally cost
effective in predicting efficacy and side effects as genetic testing,
which may only lead to relatively small advances in clinical practice.
Some participants agreed that the science was ahead of clinical
practice.
A commonly expressed view was that new genetic tests
were potentially of value but required evaluation before being
brought into mainstream clinical practice. The importance of assessing
the utility of genetic tests was highlighted by several participants,
though it was pointed out that it was not clear who would fund
research into clinical utility, as NIHR (National Institute of
Health Research) excluded funding of laboratory-based research
projects.
The issue of the practical end of implementation
was raised. Up until the present, almost all genetic testing had
been carried out in Regional Genetics Centres, and most genetic
tests have been for single-gene disorders. Now that genetic tests
were being introduced for more common diseases, the Regional Genetics
Centres would not have the capacity to carry out all new tests.
However, carrying out the test represented perhaps only half the
cost of the test, the remainder including genetic counselling.
Another participant concurred with the need for genetic testing
to expand beyond Regional Genetics laboratories, but pointed out
that there had been years of discussion about setting up molecular
pathology laboratories, but this was prevented by "silo budgeting"
to individual specialty laboratories. Data on new tests needed
to be evaluated in a timely manner, and to achieve this, a new
structure would be required to implement "health genomics".
The NHS was not set up to take on board the changes
that were associated with the move of genetic testing into the
mainstream medical specialties. At present, introduction of genetic
testing into mainstream specialties was piecemeal and reliant
on presentation of data on new tests to local funders who did
not have the necessary expertise or knowledge to make informed
decisions on these matters. A particular concern was the lack
of genetics expertise in Public Health, which was largely focussed
on environmental, rather than genetic issues.
It was pointed out that new, rolling funds would
be required to pay for updates to genetic testing equipment funded
under the White Paper "Our inheritance, our future"
of 2003, as such equipment only had a lifespan of four to five
years. The need to fund translational research for evaluating
new tests and bringing them into service was also highlighted.
The funding model that was in place in Wales for these activities
was commended.
The procedure for bringing new genetic tests for
single-gene disorders into NHS use was described. The UKGTN had
already approved 173 such tests. A similar mechanism needed to
be introduced for generating data on the utility of new genetic
tests such as single nucleotide polymorphisms associated with
common diseases and pharmacogenetic tests of drug utility and
responsiveness. No solution was currently in place to meet this
need. Attention was also drawn to the need for education of healthcare
professionals on genetic testing and commented that the National
Genetics Education and Development Centre had focussed to date
on surveying attitudes of health professionals and on learning
outcomes. However undergraduates needed to develop a concept map
of where genetics fitted into healthcare so that they were prepared
with appropriate knowledge when they started to practice.
With regards to IT, more money needed to be invested
in informatics and more IT experts would be required to set up
and manage genomics information systems if genomics was to be
useful in healthcare.
On the question on the desirability of "genomicising"
medicine arose, it was regarded as inevitable that much of the
medical profession would not like to move towards genomics in
healthcare, and the pressure for this change may need to come
from the science. Patients should also be at the centre of any
changes and should participate fully in these discussions. There
was resistance to change, and a significant fraction of the £2.5
billion spend on pathology services could be saved if sovereignty
of individual specialties were to be given up.
It was asked whether epigenetics could be applied
to modifying disease processes. Epigenetic contributions to disease
processes cannot at present be quantified as accurately as genetic
contributions but it was felt that dialogue between geneticists
and epigeneticists should be encouraged, though this might be
outside the remit of the Inquiry.
| 
