Examination of Witness (Questions 1 -
19)
WEDNESDAY 24 JANUARY 2001
PROFESSOR MARTIN
BOBROW CBE
Chairman
1. Professor Bobrow, welcome to the Science
and Technology Select Committee. Thank you for giving up your
time to be with us this afternoon to help us in an inquiry we
are just starting into genetics and the insurance industry. I
wonder if I could ask you, to start our proceedings, we have your
CV in front of us and we know a lot about you, but I would like
your CV to be partially written into the record, so I wonder if
you would be so kind as to tell us about yourself: the expertise
you have; the position you currently hold; and the work you are
doing on genetics? I would be most grateful.
(Professor Bobrow) I am a medical geneticist,
by trade. I have been a medical geneticist, that is, I have worked
at the interface of genetics and clinical application, since the
mid 1960s, before the profession was invented, virtually. Initially,
I worked at a Medical Research Council unit, devoted to things
of this sort, in Oxford, subsequently became Professor of Human
Genetics in Amsterdam, and then Professor of Paediatric Research
at Guys, and although that name Paediatric Research is correct,
actually it was a Department of Genetics. I have been for about
five years Professor of Medical Genetics in Cambridge. I have
had a general interest in matters of medical genetics and how
genetics affect patients, therefore, for a long time. I have also
had a reasonably long-standing interest in issues of genetics,
ethics and public policy, have been a member of the Human Genetics
Advisory Committee, of the Clothier Committee, the Gene Therapy
Advisory Committee, the Nuffield Council on Bioethics, and probably
something else.
2. That is very impressive, and thank you very
much. It gets us off to a good start. In your present position
at Cambridge, do you have a team that undertakes any genetic research,
and, if so, what is that team doing?
(Professor Bobrow) I am head of a small department,
which has within it other, independent colleagues of mine, who
conduct research into the genetics of a variety of inherited diseases,
which include Huntington's chorea, polycystic kidney disease,
diabetes and various other neurological diseases.
3. This Committee, in the last Parliament, did
look at this subject of human genetics, as you may well be aware;
I was not Chairman at that time, nor, in fact, was I on the Committee,
but Dr Williams and Dr Jones were, and they reported in 1995.
For all our benefits but also for the benefit of Dr Williams and
Dr Jones, who probably need a bit of updating since 1995, can
you tell us what the important developments and important work
has been, in the last five or six years, in genetics?
(Professor Bobrow) The headline of the last five years
is that the sequence of the human genome is now essentially available
on the Internet; there is still tidying work to be done, but the
great bulk of that is there. That means that the groundwork for
identifying those parts of the human genetic sequence which are
responsible for variation between people will be unravelled, I
am not sure how soon but it will be unravelled. Part of that variation
is the genetic variation that will underlie many of the important
differences in health and health outlook between people.
4. Finally, before we go to Dr Gibson, would
you say the developments in the last five years have been, in
reality, greater and faster than you might have thought, if you
were in 1990 looking forward? If, in 1990, I had asked you to
predict what might happen between 1995 and 2000, has more happened
than you might have thought?
(Professor Bobrow) May I answer that in two different
directions. Scientifically, I think that is undoubtedly the case.
The speed with which technical developments, particularly in relation
to sequencing but also other developments, have occurred has been,
I think, very rapid, and that is easy to document. I think that
the speed with which our knowledge of health effects has accrued
has not been particularly rapid. If anything, I would say that
it is roughly what I would have expected, which is very little,
and I was very much at the pessimistic end of the market ten years
ago.
Chairman: Thank you very much indeed.
Dr Gibson
5. I believe there are something like ten tests
for seven sentinel conditions; could you say, just briefly, what
these tests are for and then perhaps follow it up with a statement
to explain to the Committee, where it says: "Genetic tests
are very good at distinguishing those who carry a particular gene
from those who do not. ... They are not at all good at predicting
when the disease will start, how rapidly it will progress, and
thus when death will occur." I would like you, if you could,
just to take time to elaborate on that: the tests first, please?
(Professor Bobrow) Clearly I am not prepared precisely
to describe the tests that are in the ABI's policy document, but
there are seven conditions, which include two that I have mentioned:
Huntington's disease, polycystic kidney disease, a number of nerve
and muscle degenerative disorders, one of the breast cancer genes,
from memory. These are all conditions which are well recognised
in the genetic clinic, that is, they are undoubtedly inherited
conditions. The genes which, when disrupted, lead to these conditions
are well known, and there are tests which are used in the clinic
in order to distinguish people within a family who are, and are
not, at risk of getting the disease because they have, or have
not, inherited the gene. Some of the tests are technically more
straightforward than others, but I do not actually think that
the technical capabilities of the tests are particularly contentious,
they are matters of record. The point at which I depart from the
interpretations that have been made by others is that, in studying
a genetic effect within a family, the question in general is,
you know that, to take an example, Huntington's disease is caused
by a specific kind of genetic change, and if you look for that
change in individuals who are at risk you either do see it or
you do not see it, and with a little bit of grey area in the middle,
and there is a bit of grey area, but with a little bit of grey
area in the middle, you are not left in doubt as to whether they
do or do not have the gene. So, for the purposes of having a medical
counselling interview, that one takes one a lot further forward.
However, within the group of people who have the gene, some will
get the disease, to take extremes, in childhood, and others, it
is becoming increasingly recognised, will not get the disease
at all until they are very old, and some, inevitably, will die
of other things before they get the disease, and, therefore, although
they have the genetic change, may reach a healthy age without
it ever manifesting. Some of that variation, in the particular
case of Huntington's, can be accounted for by differences in the
precise nature of the genetic change, but even within individual
classes of genetic change, which are otherwise indistinguishable
to our techniques, there are variations in age of onset, of ten,
15, 20 years, between people who genetically look identical, and
we do not understand the basis of that variation. But they still
have a medical problem, they are still at high risk of getting
the disease, but, in terms of a time-limited prognostic indicator,
there is a considerable amount of ambiguity.
6. Thank you very much. And if I can follow
that up, the other tests that are coming on line, as it were,
are there any predictions you can make about when they will come
on line, when they might be used? And what do you think about
the multifactorial conditions, like heart disease, for example,
where there are, obviously, many genes involved; are the insurance
companies ever going to be able to use that information, utilise
it, in any predictive way?
(Professor Bobrow) For most of the relatively simple
inherited conditions, the things we call Mendelian conditions,
of which there are several hundred, these seven are just exemplars,
there are already tests of greater or lesser precision. If one
takes the specific problem that we were discussing a few moments
ago, of whether tests will be developed that will allow us to
pick out speed of progression and time of onset of disease, I
would have to say that it is not obvious to me, now, how and when
we will learn about those things, nor even that that variation
is genetic. I suspect much of it is not. On your second question,
of the more complex disorders, which are more common, which most
of us, in the end, meet during our journey and which are not inherited
in a simple way, I think the jury is out. There is a very, very
active research field attempting to identify genetic factors that
contribute to diabetes and hypertension and rheumatoid arthritis
and asthma, and pretty well anything else you can think of, and
some genetic associations are being identified. But it is quite
unclear at the moment whether these will lead to individual tests
or groups of tests that will, between them, have a high predictive
power, that they will really pick out people whose risk is very
seriously elevated above the population risk, as opposed to being,
say, two or three times above population risk; which is not, in
my view, good enough, because if things are rare then two or three
times a small number is still a small number. And when you are
dealing with things like breast cancer, where the actual risk
is not that uncommon, nevertheless, differences of two- or three-fold,
we know they exist within the normal population, for all sorts
of reasons, it does not lead to this hundreds of fold difference
in risk that you get with the simply inherited conditions; and
I am pessimistic that it will get there soon.
Dr Williams
7. Can I come in with just a brief one? You
are speaking with great restraint and accuracy in what you tell
us, but are there other clinical geneticists out there that believe
that the blueprint will be more predictive than that, or in certain
instances, like in Huntington's or some of the genetic defects,
where maybe in five or ten years' time, from just the blueprint
itself, we will be able to say a good deal more, or are there
some enthusiasts in the field that think there is more information
here than you are telling us?
(Professor Bobrow) Of course, there is always a spectrum
of opinion. I think, if we take the specific question you asked
me, which is whether there are people who believe that we will
have tests that will predict time of onset and rate of progression
of Huntington's disease, I do not know any such people, nor do
I know of any scientific work that leads us to believe that we
are seriously going to crack that problem. Every scientist believes
that every problem will be solved, so I am not going to say never,
but there is nothing on the horizon, at this stage, in my view.
Dr Gibson
8. And you would not really think that the insurance
companies can predict epistatic effects between genes, we are
a long, long way from understanding that, we have never really
understood the phenomenon, although it has been around for years,
the interaction of genes and different chromosomes, for example?
So they are nowhere near that, they are millions of years away
from it, would you say?
(Professor Bobrow) Millions of years is not the kind
of thing that falls naturally from my lips; it is not easy to
see it happening. And, even if leads emerge, I think one of the
things that one needs to keep in mind is that finding genes that
have influence is a first step; demonstrating quantitatively how
large that influence is really requires studying people with those
genes over a reasonable period of time in order to draw sound
conclusions. So I do not think there can be any doubt that the
horizon between now and the time one could have a validated, verified
test that predicted the sort of the thing we are discussing must
be certainly decades away.
Chairman
9. Just before we go to Dr Jones, Professor
Bobrow, you did say, very early on, that, even if there is a gene
that is likely to lead to some disease, you cannot be sure whether
the person will get that disease in their childhood, in their
middle life, or indeed not at all. To what extent then is it valid,
and this is a qualitative question I am asking you, for insurance
companies to use genetic information at all, in setting premiums,
if scientists have no idea at all when this gene is going to kick
in with its trouble, how can insurance companies say that someone
should have a loaded premium, just because they have this particular
gene?
(Professor Bobrow) Can I go back and clarify something,
because I have clearly left somewhat of an overstatement in the
air. Some genetic conditions clearly manifest in childhood, pretty
well always, and if I gave the impression that there was no idea
of when things would start to have effects, that was not quite
accurate. For those that start in adulthood, there is, in general,
a relatively wide range of age of onset. The particular example
that I was discussing of Huntington's, which is obviously a prime
example, can have onset very early and very late, although those
are the extremes; but, in general, it is a disease with onset
about 40, plus or minus 15 years, and it is within that average
range, of 40 plus/minus 15 years, that we do not know where to
start or where to begin.
10. So that was when you were talking about
Huntington's?
(Professor Bobrow) Yes.
11. And, unfortunately, I generalised it across
a lot of things; forgive me for doing that. If we go back to other
genetic diseases, not Huntington's, for any one genetic disease,
have you got a spread of age onset for that disease, plus or minus
ten years, or so, which may be different, some may start at 25,
plus or minus ten, others at 45, plus or minus ten; is that roughly
the situation?
(Professor Bobrow) There is general information on
ages of onset; in general, the younger the age of onset the tighter
the distribution, so that conditions which are obviously not germane
to an insurance discussion, such as muscular dystrophy, the classical
type, always has a childhood onset, give or take a few years.
The older the average age of onset the more the variation. Most
of the conditions, as I recall it, from that list, are young adult
to middle-age type of onset, because those are the things that
are the most germane to insurance discussions, and they all do
have a spread of ages of onset, to a greater or lesser extent.
Dr Gibson
12. Would it be even more complex if different
mutations ended up with the same disease, for example, in sickle-cell,
you have different mutations, different parts of the gene; is
that true in these genetic situations you are talking about? You
say a single gene, but are they indeed single genes or are they
multiple mutations within the gene?
(Professor Bobrow) They vary. For some of the gene
loci, the changes, the mutations that cause disease, are quite
variable, and they are one of the components that probably make
it more severe in some people than others. For other conditions,
the gene change seems to be fairly uniform, and there are other,
as yet unknown, factors that influence progression.
13. Does it relate to geography as well? You
know how there is always this argument that certain groups of
people have one mutation, others do not, so it would vary between
different parts of this country, for example, there might be geographical
variations in the mutations, whereas the insurance companies treat
them all the same?
(Professor Bobrow) To the best of my knowledge, within
the limited group of Mendelian disorders, the really genetic disorders,
that are in the ABI's current policy document, that is probably
not so; they are probably not very sensitive to that kind of geographic
variation. Getting into the more complex, multifactorial disorders,
I would bet on it that you are right.
Dr Jones
14. Just one quick point on that. Wellcome Trust,
in their evidence, said that there were occasions when somebody
who had got the genotype which was predictive for a particular
condition did not actually develop that condition, and would you
concur with that as well, even in Huntington's, for example?
(Professor Bobrow) Yes. It is very well documented
that, at one end of the range of Huntington's mutations, there
are about 3, 4, 5 per cent of people who carry genes that clearly
are abnormal, by current definitions, but they have not manifested
the disease by age 70, or 75, something of that nature.
15. As you know, because you were a member of
the Human Genetics Advisory Commission, they recommended that
there should be at least a two years' moratorium on the use of
genetic tests. Could you explain the basis for that recommendation;
and, also, why do you think the Government did not accept it?
(Professor Bobrow) Clearly, I cannot speak for the
disbanded HGAC. My personal understanding of the—
16. Why can you not, you were a member of it?
(Professor Bobrow) I was a member, but it is a body
that is long gone and other people may have had different reasons
for reaching a common conclusion.
17. You must have discussed it amongst yourselves?
(Professor Bobrow) We recommended a moratorium, and
the arguments that persuaded me, and I suspect most others, that
that was a good idea were that we felt that there were two things
that weighed; one of which was a dearth of evidence, of actuarial
genetic evidence, if I can use that, that would allow insurance
companies to take well-founded decisions that were based on fact,
that could be made transparent and that people would, therefore,
presumably, feel more comfortable with. And the gathering of that
evidence was going to take some time, two years was probably an
underestimate, and it seemed to us reasonable that the tests should
not be used until such time as the evidence had been accrued.
Secondly, we were aware of the fact that there was already extant
practice in the field, and it seemed to me that—
18. So there was what practice?
(Professor Bobrow) There was practice in the field,
people were already asking for these tests. And it seemed a good
idea, if one was moving from people doing what they thought they
would like to do, to people having an agreed policy that would
be implemented under some form of open regulation, to draw a line,
stop the past and start the future again. And I think those were
the two arguments that weighed in favour of a moratorium. As to
why that did not appeal to Government, I fear that I have no insight,
that I am prepared to speak about.
19. You said in your memorandum that you remain
to be convinced that insurance companies will use genetic information
reasonably. Is that still your view, and, if so, why do you say
that statutory regulation is undesirable in principle, even though
you say it is probably unavoidable in practice?
(Professor Bobrow) I have been having this discussion
in various places for really a good number of years now. It seemed
to me, possibly ten or 15 years ago, ten years ago, that the simple
inherited diseases that we are looking at today are an area in
which people who are pretty seriously disadvantaged do not need
more trouble in their lives; and trying to load insurance issues
on top of their medical issues could only be justified if the
evidence was really clear and if the financial implications were
significant. And I would have to say that I have been grievously
disappointed at my complete inability to have any dialogue on
these issues with most people from the insurance industry. That
is one thing. The second thing is, I think that having agreed
to the establishment of the Genetics and Advisory Committee, which
was to look at, validate and agree a basis for using genetic tests,
for the ABI to continue recommending that these tests should remain
in use until they were declared invalid, rather than the reverse,
I thought was an irrational and unfriendly way to come to the
table over a difficult issue. Thirdly, there is a particularly
difficult matter over insurance companies' very long-established
practice of asking about family history, and it is one that has
not come up a lot in the conversation so far.
|
