Examination of Witnesses (Questions 440
- 459)
WEDNESDAY 16 JULY 2008
Professor Sir John Bell and Professor Sir Alex Markham
Q440 Lord Winston:
How important do you think are the epigenetic and environmental
factors in healthcare and healthcare-related research, compared
to knowledge of the genome? We tend to think in terms of genes
in the environment but we forget, for example, the development.
Professor Sir John Bell: The epigenetic story
is fantastic. It is about 10 years behind where we are with conventional
germ line variations, but again the new tools, particularly the
new genetic sequencing tools, are really providing a fantastic
window on epigenetic modifications in a way that you can systematically
screen them around genome, and you can do it pretty efficiently.
I think we are soon going to get maps about where genetic modifications
are and how that changes, potentially how that changes in development
and certainly how it changes in different types of common disease
and what you might do about that because there is an interesting
question whether by modifying some of those epigenetic changes
you cannot actually have therapeutic opportunities as well. My
view is that it is very important but we are not yet at the discovery
stage where we actually know how it will apply. We know, for example,
the Barker view about metabolic disease, that almost certainly
it comes from some intrauterine event that actually sets programmes.
Because the germ line stays the same we know it is not that so
it has to be epigenetic; the real question is exactly how does
that work. We know in a variety of other settings that epigenetic
modifications are important, there is great work in animal models
showing that rats who suckle infants, those progeny get really
profound epigenetic modifications and may go on to have quite
a different behavioural phenotype. It would be easy to under-estimate
the impact of the epigenetic piece but it is also fair to say
it is early.
Q441 Lord Winston:
It is not only behavioural of course it is things like diabetes
as well.
Professor Sir John Bell: That is right, and
hypertension. The whole intrauterine environment story rests around
epigenetic changes that dictate your likelihood of developing
those phenotypes that we find.
Q442 Lord Winston:
I know we keep on asking you to push out the predictive boat but
how do you think these could be and how soon could they be related
to healthcare? Alex, do you have a view on that?
Professor Sir Alex Markham: We may learn quite
a bit from looking carefully at what happens to the first generations
of agents that work epigenetically in cancer. Clearly, there are
the H-stack inhibitors and a whole raft of therapeutic approaches
that are targeted at the fact that in some malignant disease the
change is all epigenetic. My take on that work is that there are
some of them in mid-phase development, phase two, and I guess
there are one or two now in phase three, and what I think we might
learn from looking carefully at those patients, should they have
a profound effect on the cancers, is whether they go on to have
peculiar, unanticipated side effects potentially several years
down the track. So if one extrapolates from the baby in utero
to the development of type 2 diabetes, I guess epigenetic modifiers
used as treatments for cancer might start to turn up some unexpected
consequences decades later. People are looking at tools to change
the epigenetic signature of the cell and the first evaluation
is in cancer and I think it is watch this space. I do not think
there is a need for UK Plc to be catalysing more research in this
area because we have got some of the most distinguished basic
science in Britain in epigenetics and, as I say, we do not have
much of an option right now but to look very carefully at what
happens to the cancer patients.
Q443 Lord Winston:
Are you thinking about the supplies of micro RNAs or do you think
that is going to be so complicated—there is a plethora of
data presumably. Management of patients with diseases like haemochromatosis,
which again is not all that uncommon, now is reliant on genetic
testing.
Professor Sir Alex Markham: Management of haemoglobinopathies
is absolutely reliant on genetic testing. It has strayed back
into monogenic diseases. Where we will see the next big demand
might be unexpected. If you want a prediction, it would be in
one of the cancers that has rather escaped notice. There is an
enormous amount of investment in the genetics of breast cancer
and it is very interesting and controversial. It tends to draw
attention away from a big challenge we face in colorectal cancer,
a very common disease for which we have just implemented, I am
delighted to say, a national screening programme. That national
screening programme currently tests faeces for blood. What we
need for a patient aged 60 who has a polyp in their colon is some
sort of categorisation tool that says, "This 60 year old
patient now has absolutely nothing else to worry about. We have
removed the polyp. They have had an £800 colonoscopy at the
NHS's expense. We do not need to bring them back in two years'
time for another one, another one two years after that and so
on." We need a tool for colorectal cancer, for the screening
programme, that says that we cannot afford to colonoscope every
at risk patient who is 60 years old every two years indefinitely.
We did not have any of those tools and the genetic studies that
are just publishing now on colorectal cancer genetics are giving
us unique and unexpected insights that I think are probably an
order of magnitude more exciting and important than what we have
so far out of the breast cancer screening programme, unexpectedly.
The application may not be very high wire/high tech but I think
it will have a massive impact on the way we go about delivering
a national screening programme. My hope is that we can screen
a lot more people a lot more effectively for a lot less money.
I would be excited about that. I think the genetic studies that
are going on right now in prostate cancer are going to offer us
some of those kinds of opportunities too. Breast has been a little
bit disappointing after a huge euphoria about the BRCA1 single
disease mutation story and the breast cancer people are going
to have to get their thinking caps on about how all these new
genetics work. John's suggestions about sub-segregating breast
cancer in terms of what genes are expressed is probably the way
to go. I think we are going to get some really exciting things
to chew over from some of the other malignancies. I understand
there are some very interesting things coming out of melanoma
genetics right now. I do not know whether it is interesting or
quite amusing but the first genetic studies of predisposition
to lung cancer have just published and they all show that what
predisposes you to lung cancer is polymorphism in your nicotinic
acetylcholine receptor genes. If you are predisposed to rather
liking your nicotine, you are predisposed to lung cancer. You
might say, "Whoopee do, we could have told you that anyway,
Professor", but it does show you that genetics do get things
right, I think.
Professor Sir John Bell: To give you two other
examples, it is really important not to be narrow about the use
of genetic tests. HIV and viral sequencing to detect resistance
is very common in HIV. It is the way you manage many of the new
medications. People have been using it for five years. It is highly
effective. The herceptin test used to be based on the presence
of a particular receptor in certain breast cancer cells and is
now almost all done by FISH, so it is there in a variety of different
formats, not always in a chip.
Q444 Lord Krebs:
I wanted to ask briefly about the translational gap in genomic
medicine. Quoting from the Academy of Medical Science's submission
to us, they said, "It is far from evident that the UK environment
for translating advance in genomic medicine into health care practice
is optimal." We have heard a number of concerns about translation
and the perceived lack of funding for evaluating the utility of
new genetic tests—for example, the NIHR is not funding research
into the evaluation of laboratory tests. I wanted to ask whether
you think that is a fair assessment about the translational gap.
Given your position, what is OSCHR's vision in relation to this
issue?
Professor Sir John Bell: Everybody has this
problem. The delivery of a new set of genetic tools into the clinic
has proved really difficult in every jurisdiction. The Americans
have had a very hard time doing it. The Europeans have had a hard
time doing it. We were having a hard time doing it and there are
several reasons for that. One is that I do not think you can rely,
in this setting, on the diagnostics companies to do what is done
in therapeutics, which is to demonstrate clinical utility, because
the cost of a clinical utility programme is such that, at the
prices paid for diagnostics, they would never get the money back.
Diagnostics companies say, "If you want clinical utility,
you are going to have to do it yourself because we cannot do it."
That is a really important difference. We do have the advantage
that the regulatory framework in Europe is much more attractive
for moving diagnostics. The approach to CE marking rather than
various forms of FDA approval makes it much more likely that you
can move down that road efficiently with all the diagnostics.
The real obstacle comes in who is going to do the diagnostic evaluation
and who is going to pay for it. If you look at the host of different
activities and technologies that you could apply at the moment,
you can see why the Department of Health has not jumped immediately
in and said, "Yes, we will do it", because the bill
is going to be pretty substantial to deal with that problem. There
are things going on funded through the NIHR which are very positive.
They funded for many years genetics knowledge parks which did
a certain amount of this in the very early days of genetic testing.
A number of the biomedical research centres brought genetic teams
which are funded to try to implement some of these genetic activities.
Where we lack real oomph is setting up large scale prospective
studies to try and evaluate for example transcript profiling in
breast cancer and prostate cancer. What information does it give
you? Does it save lives? Does it save money? What is the real
data you get from that? Those are complicated, costly projects
and they are probably not the same as randomised phase three studies
for therapeutics because you do not have to do the randomisation.
You can do it in a cohort fashion, running the new tests alongside
the old tests to see how you differentiate people as they go forward.
Clinicians may be blinded to what results you get in the lab but
you do not have to do it in a test. My view is that this is a
really important thing to get right. I have wondered whether the
NHS should not consider itself the laboratory for the world for
developing clinical utility in new diagnostics. We have a fantastic
setup that would allow us to approach the commercial suppliers
and say, "You have the test. We will help you do the clinical
utility. If we do it together, we are expecting a pricing on the
diagnostic which reflects our partnership in the process."
In the end, the payback is that we, unlike the rest of the world,
would get our diagnostics at half the price of everybody else
and the clinical utility data which we generated to the benefit
of our patients in the NHS would be used worldwide for people
to get registration and sell their products internationally. There
potentially is a deal to be done there that probably requires
a bit of reflection to see whether we might do that. One could
do it to the long term financial benefit of the NHS as well as
providing clinical utility data.
Q445 Chairman:
Who is doing this thinking? What structure will be required?
Professor Sir John Bell: That was my idea. I
floated that by the DH and Andrew Dillon at NICE because I thought
NICE should be involved in that as well. You could conspire to
make sure whatever studies you do are the ones that deliver the
health and economic data for NICE to make decisions on. The discussion
went rather well, as far as I can see. I think they have gone
back to talk within the Department of Health about how they might
advance that. There are a variety of ways to play this but it
seemed to me that would position us right in the front of this
field and this whole issue about how you get innovation delivered
to the NHS quickly—nothing does that better than being in
a place where you do the utility testing in the first place. I
think it has many benefits.
Q446 Lord Winston:
Is there a model for that so far in the history of—?
Professor Sir John Bell: Not that I know of.
I do not think there is.
Q447 Lord Winston:
I could not think of one.
Professor Sir John Bell: Neither could I. If
I come up with one I will let you have it but I cannot think of
one.
Q448 Lord Winston:
Current genetic tests for single gene disorders are approved for
NHS commissioning by the UK Genetic Testing Network. Do you think
the role of that network could be expanded to include genetic
tests for common disorders?
Professor Sir John Bell: The UKGTN has done
a really good job. As you know, it is a voluntary organisation
where people participate in the evaluation of single gene testing
and have been pretty effective in providing a clear understanding
about whether it works or does not. The decision about commissioning
does not sit with the UKGTN. It sits with the Genetic Commissioning
Advisory Group and then it gets passed down to local commissioners.
That process needs another look because the time frame from getting
from the top to the bottom is at least three years, so the time
frame to get stuff to patients, in my view, is far too long. When
you enter the arena of some of these much more complex tests in
common disease, I am not sure that the skill base that is housed
within the UKGTN is the right skill base. You are talking about
often epidemiological data. You require health economic data to
work out whether this stuff works or not. I am not persuaded that
the structure which I applaud is necessarily transferable into
this rather more complicated, complex world where clinical utility
testing will have to be done on thousands of patients in large
prospective cohorts. The methodology for analysing that may have
to be new because some of this is relatively new methodology.
My feeling is that the structure that is needed might have to
dock with a NICE like agency which has some of those other capabilities
to think about what evidence you would need and how does it work.
My view is it might need to move.
Q449 Lord Winston:
Effectively you need a new agency?
Professor Sir John Bell: Yes.
Professor Sir Alex Markham: I am always reluctant
to start new agencies because they always seem to me to have a
tendency to slow everything down. I think the pathologists as
a community have to take some ownership of this. The Royal College
of Pathologists has to step up to the plate here because they
have responsibility first of all for managing the scientists in
the genetic testing networks anyway. That is a discipline that
the College of Pathologists sees itself as being top of the food
chain for. Once you get into the realm of polygenic disease, there
was a question in the preamble you sent out to us: does all this
genetic advance imply that the profession of medical genetics
needs to massively expand? I would argue no. If you have to manage
a patient with psoriasis, you have to be a dermatologist. You
as a dermatologist have to understand the implications of the
new genetic understanding. I would put pathologists on the spot
and say, "What are you going to do about this? What do you
think of the role of the UK Genetic Testing Network?" which
has done a terrific job. What they did successfully was they divvied
up the cake and let the whole community develop individual tests
that they were good at and then shared that together. The sociology
of the thing was good. When you get into looking at updated versions
of cytogenetics in whatever form they evolve, when you start looking
at profiling expression levels in particular tissue, diseased
tissues like tumours, I think you have to be given the pathologists
because they are going to have to advise front line clinicians
on issues of prognosis. They are going to have to give advice
about optimal clinical management routes. This is where multidisciplinary
clinical teams have such an important role to play. I would urge
against another body until you have exhausted all the possibilities
of making the one we already have do its job a little better.
Q450 Lord Warner:
That brings us neatly to an issue that has been raised with us
on a number of occasions about the competence and costs of the
current pathology labs doing genetic tests. I am interested that
you say they should step up to the plate. There has been a bit
of nervousness about stepping up to the plate in this area. Should
genetic tests be carried out in the specialised regional genetics
centres or will most pathology laboratories need to be carrying
out these tests in the relatively near future?
Professor Sir Alex Markham: That is a very timely
question. The regional laboratories have played an important role.
I get the sense that the regional genetics laboratory service
is a little uncertain of the future. It feels to me that people
in them are intrigued and a little bit worried about what the
future holds in terms of all the new advances that can be anticipated
and what their place in that world is. I get the impression that
it is becoming more difficult to recruit into regional genetics
laboratories. That may be incorrect. Lord Winston will probably
have his finger even more on the pulse than me on that. The time
might be right for moving to a system where they are more integrated
into "mainstream pathology". The time might be right
for that to be considered. A lot of the funding streams that have
driven the genetic testing laboratories over the period since
the White Paper will probably be coming to an end now. A lot of
what went on in the genetic testing world was underpinned by things
like the genetics knowledge parks and the funding for that cycle
is ending. I do think there is scope to look very carefully at
how the NHS structures those laboratories and goes about commissioning
that activity and paying for it in the context of pathology as
a whole. Lord Carter is reporting on pathology and I think he
would miss a big trick if he does not address that question of
genetic testing and its incorporation into mainstream pathology.
Professor Sir John Bell: Alex and I did not
conspire on this question before we came in here but I think we
are on the same spot. I would go a little further. We have got
ourselves silo-ed in a really unhelpful way in laboratory services
generally. One of the things that has emerged in my local domain
is that pathologists want to do a little bit of molecular pathology
and some genetic stuff but clinical genetics often use the same
technology. Haematologists want an array machine. I would just
say, "Forget it; you are not doing that." Let us have
one really good molecular pathology lab and the best facilities
in there. When you want DNA sequencing done in the clinical setting,
it gets done there. It can go out to the sub-specialties. Moving
everybody to a different space I appreciate will be difficult
but I think you have to do it because the waste in the system,
if you just let everybody fool around, is going to be terrific.
I would really urge you to take a serious go at this one. It is
not complicated. You just have to break down some of the old barriers.
Q451 Lord Warner:
What levers would you use in this area? The two levers you can
usually use are money and regulation, money in the sense that
you can incentivise people. That becomes the only show in town,
which is a bigger grouping and bigger competence. The other is
regulation which says to people, "You have to pass certain
thresholds of competence to be able to do these tests." Which
combination or use of those levers would you be saying to the
Committee we ought to be pushing?
Professor Sir John Bell: There is another piece
to this, which is another dimension that I think might be a lever.
My suspicion is you will not want to start, even if within a single
general hospital you put all the lab services together. I am not
sure you will want every little district general hospital to have
a DNA sequencing facility in the back room. It is not straightforward
technology. Having some significant regional laboratory capacity
is going to be part of this. You are going to have to say, "We,
the NHS, are going to fund these in a certain number of places",
ideally hooked up to the capacity to move novel developments into
them in an efficient way. There are some good examples out of
the White Paper where you spread technology all over the country
but they are still using it in the way they were five years ago,
which is unhelpful. Being hooked up to a stream of innovation
is really crucial and also to have the capacity to provide that
set of services at some kind of regional level. You may be able
to dictate how many of them there will be and what sort of area
they will cover. By doing that you will end up with the right
result because you will say, "We are not going to do it unless
they have dah-di-dah." It has to soak up all the activities
in those areas and provide services for the surrounding regional
hospital arrangements. The Americans have done this really effectively.
If you go to the molecular pathology labs in Germany, it would
knock your socks off. They are fantastic. They are really slick.
They have all the bits of kit. Every record is on an IT system.
There is no paper. The big American reference labs are like that
too. I would look at those.
Professor Sir Alex Markham: I would just remind
us of a model that was used for the genetic testing labs five
years ago. Each regional health authority has a genetic testing
lab but only two of them are resourced and recognised as the ones
that develop the new, special tests for relatively rare diseases.
That was done, as I recall, in Salisbury and Manchester. Now,
with modern IT and the way this technology is developing, one
of your levers is to say to a region, "If you cannot get
your act together to deliver an integrated molecular pathology
service that covers all these bases, the region down the road
can eat your lunch" and get some competition in like that.
The history here is that the clinical medical genetics community
rather used the regional genetic testing laboratories as part
of their empires. That was what came with the turf of being a
clinical geneticist. You had the regional clinical genetics lab
in your fiefdom. I do not think that is necessary any more. The
clinical laboratories need good interaction with the clinical
geneticists but they do not necessarily need to be down the corridor
any more.
Q452 Lord Winston:
Lord Warner mentioned two ways but is not the best way to be getting
research done?
Professor Sir John Bell: Exactly. You get certain
expertise and if it is hooked up particularly to people who are
using these skills in the research environment then they can implement
them in the clinical environment and the whole thing should be
seamless. It is really important that you do not recommend that
we just establish another set of regional genetics labs. What
both Alex and I are saying is that this is not a regional genetics
lab; this is a regional molecular pathology lab that brings the
geneticists in, the haematologists in, and if you give it to the
geneticists to own it will cause a lot of trouble. It has to be
everybody working together and that means they will probably end
up in different places as well.
Q453 Baroness Finlay of Llandaff:
With the philosophy of devolution of the NHS and non-central control,
picking up Lord Warner's point, where do you think an obligation
to participate in research and provide research led data should
sit in terms of supplying and having central labs? I am picking
up Alex's comment that it is unethical perhaps to treat patients
without considering the research information and the knowledge
that you gain from it.
Professor Sir John Bell: It is really helpful
that the NHS constitution is going to have within it a commitment
to making research a major pillar of what we do. We have said
this time and time again. Patients do not dislike research. In
fact, patients like research a lot. In a sense, one of the opportunities
we have within the NHS is to make it available to patients in
the same way you make other things available to patients. My view
is you can do the distributive model, although I am not a great
fan of it. There are a few things where you will have to say,
"You are going to have to do this" and it relates to
quality, what patients want, and most sensible people will implement
it anyway, but I think there needs to be central management to
make sure that it is a main pillar of the whole organisation.
Q454 Lord Broers:
You have talked about the potentially very competitive position
that the UK is in. Perhaps this question relates to that. UK Biobank
and Generation Scotland are studying very large numbers of individuals
to understand more about their propensity to develop common diseases.
How will genome technologies and genomic information be useful
in studying these populations? Are these study populations large
enough? Could there be a need for a larger, Europe-wide study?
Professor Sir John Bell: I was very involved
in the initial stages of conceiving the Biobank study. When we
set it up, we felt that the major argument for doing it was the
likely availability of genetic tools and technologies to allow
you to interrogate large numbers of people for their genetic susceptibility
factors, for genetic factors that relate to disease and ultimately
potentially their response to therapy. The numbers we chose to
recruit were to some extent defined by the budget, but they did
provide the sort of numbers that you see in a variety of other
genetic studies that have been emerging recently. They will probably
be just enough to start to interrogate some of these interactions
between genes and environment which is obviously the fundamental
thing that we want to try to deliver. We could have done a million
but 500,000 is not bad. We are now five years ahead of everybody
because everybody said they were going to do it and nobody did.
The Germans and the Canadians have just announced one. They are
300,000 each so each one of those is smaller than us. Obviously
we are trying to align their protocols with ours so we could add
the results together over time. I would put this to you: one of
the great powers of the structure that is being set up and one
of the reasons that Biobank will be successful is because of our
ability to manage and handle data in large numbers of people,
which really relates to the Connecting for Health programme, and
the Research Capability programme analysis alluded to. If it evolves
as we all hope it will, given the ready access to that kind of
data, I am not sure why we would not expand the Biobank concept
much more widely in the UK, where one gave all patients an opportunity
to deposit a bit of DNA that would be used in an anonymised fashion,
to link the data system. When you say to people, "Would you
like to be part of a system that will allow you (a) to inform
the next generation about their diseases much more effectively
and (b) would you not want to know about the opportunities that
you have for understanding disease in a much clearer way and understanding
your response to therapy in a much clearer way?" I think
you will find that, if it is no imposition in time terms and it
really just involves a sampling of blood, you might find that
you could expand Biobank using largely a DNA sample across a much
larger population in the UK. I see Biobank as phase one, a pilot
study. We may end up eventually with ten million people who are
all participating in the programme. The IT makes it possible.
Q455 Lord Broers:
Surely, if you included particularly southern European countries,
you might reveal some of the environmental factors more clearly
than a monolithic, northern European population?
Professor Sir John Bell: You are right. You
will also get diversity of genetic inputs. There are two other
studies which are under way, which are already contributing to
that. One is the Kaduro Study, being run by Richard Peto in China.
They have already ascertained 500,000 people and have biological
material, including DNA, on half a million people with health
records. There is also a Mexican study which again is led out
of the UK but which is being done by the Department of Public
Health in Mexico, which was greatly advocated by Julio Frank and
others to develop essentially the same bio-repository. My view
is: let us get as many of these as we can. I am sorry to say that
I do not think in Europe southern Europe is likely to be the place
where that will happen. It will happen in Scandinavia and Germany.
Q456 Chairman:
In the context of the UK and maybe the rest of Europe, who will
fund such large population studies? We are talking about £60
million or £70 million.
Professor Sir John Bell: UK Biobank kicked off
with £65 million. That came, as you know, from a number of
funders including the Department of Health, the MRC and the Wellcome
Trust, who were the founder contributors. My view is that this
gets easier over time because the availability of IT records which
come as part of the overall system makes the acquisition of data
on patients that you are following a great deal easier. Genetic
technologies are now a tenth the price of what we anticipated
they were going to be when we started the study. The price of
doing the analytical side is falling. The price of getting access
to the information of patients is falling. I am not pretending
we will get this for nothing, but I think it is well within the
reach of national governments now to fund this pretty effectively.
If we had five national governments each of whom put half a million
people into the system, it would be impressive. It would be a
good start.
Q457 Chairman:
Who from the UK would coordinate this to make it happen?
Professor Sir John Bell: Not me. That is for
sure.
Q458 Chairman:
OSCHR?
Professor Sir Alex Markham: No. I fear something
coming by me here. The answer is no. Rory Collins and Richard
Peto would be seen as the head of the food chain I think by most
of their international peers. The British leadership of this would
be welcomed in most other countries.
Q459 Chairman:
In the United States, as you well know, the National Human Genome
Research Institute coordinates all the activities related to genomic
science and genomic medicine, including ethical, legal and social
issues, not just the genomic technologies and research and its
implications. Do you think we should have such a body to coordinate
all of the activities related to this, because again we have all
kinds of different bodies involved.
Professor Sir Alex Markham: Your Lordships may
not know this but the director of the NIH Human Genome Research
Centre has recently resigned.
| 
