Towards Genome Medicine: UK Perspective - Timothy Hubbard

Uploaded by GenomeTV on 16.12.2011

Timothy Hubbard: So after Eric's introduction I just want to
give a little U.K. perspective about how we're thinking about this whole area, across multiple
committees in the U.K. Obviously, as Eric said, it's all being driven by this cost here,
it's of course in pounds rather than dollars, but it's the same sort of picture: full genome
sequencing is approaching the cost of the kind of clinical tests which are already being
done inside our health system. Of course, we have an NHS system, comprehensive health
care in the U.K. You can see a point where those lines are going to cross and it's going
to be worth thinking about storing individual sequences and processing over those.
So there've been a number of committees looking at the kind of things you need over a number
of years. These are a few of the high points. Firstly, linking those kind of research activities,
we have the Medical Research Council, the Wellcome Trust, funders like that -- sorry?
Male Speaker: A little closer to the microphone.
Timothy Hubbard: A little closer to the microphone, right.
And then we have the NHS, which also has a research side. And so there was a creation
of an office to kind of link those two together. And one of the things that came out of that
was a board looking at health records. And so, you know, the need to have, you know,
a generic ability to research over health records, we have some legal provisions that
makes that feasible in terms of ethics. It's a question of actually putting in place the
infrastructure, and so we have some representatives from the Scottish program here and also from
the, one of the English programs, John Parkinson. So that's one side.
Then, of course, there's the genomic side. So there was a House of Lords report in 2009
saying that something was happening in genomics and the medical system would -- should start
taking that into account. The government's response to that has been the creation of
a committee which has been looking at this whole area, this human genome strategy group.
So what do you need coming out of some of those discussions? What do you really need
to make this feasible? You know, the top, you've got health care professionals; the
bottom, you've got the kind of research activities in yellow, sequencing-type things and then
the sort of patient record, researching over those. So, you know, down here we have the
kind of research activities, NHGRI, Wellcome Trust, Sanger, EBI, involved in the U.K. sort
of side of things is EBI and Sanger. We have the reference genome. So you can start thinking
about the health care professional producing, asking for the whole genome sequence and comparing
that with the reference sequence and ending up with just the variant file, and attaching
that to the patient record. It's small enough to attach to the patient record. Those patient
records, you know, they exist inside GP surgeries. You could attach files for images there, so
there's no reason why you can't attach, you know, a summary of a -- summary of variants.
So once you've got that, so what's actually missing in terms of being able to analyze
that, use it in a clinical setting, in a sort of, you know, large-scale way as opposed to
the sort of individual gene tests we have right now? Well, certainly this sort of annotation,
I think that's the crux of this meeting. In our discussions in the U.K., if you go and
look at the House of Lords report, it was proposing a whole institute, saying that this
needs some sort of institute to address that kind of delivery at the edge of regulation.
You also need this actual infrastructure to do the comparison. We think, well, that could
be this institute providing some of those services, but it could also be a whole load
of small companies, SMEs, providing different competing services to GPs, to medical professionals.
But in order to enable that, actually it's important that that underlying dataset, the
database, is open data, so that lots of different people can get access to that and they can
interoperate with other countries, you know, other countries' datasets.
Then there's the other side, which is research over this, because whatever you put in that
database is going to be driven by what you can extract from the information across, you
know, maybe the whole country's records, if you start sequencing everybody. And so we
have this research capability program which is being set up to aggregate patient records;
if those patient records have genetic data, that could be a point for doing that research
as well. Of course, there's other types of data, there's cancer data, pathogen data,
could all kind of be fit into the same sort of framework.
But the bottom line is, if you've got these systems, you've got the research databases,
lots of variants, you've got very few that have been identified as clinical ones, in
terms of the funders, in terms of the policy makers within the national health service,
they want to know: am I going to get benefit; if I start sequencing people, at what point
am I actually going to get improved health care? And you know, it's really a cost-benefit
analysis, where do these two lines cross. And I think that's an area where we don't
know. It relates directly to how we define clinical variants, you know, that are significant,
and how many we can define.
So this is the sort of, this is the overall picture. It's clearly, there's a feedback
loop in the middle. And I think one thing that's going to come out of this is that,
if you look at the funding that's been going on up to now, in the past, all the genome
data that's been feeding into this process has been coming through, you know, the kind
of research activities Wellcome Trust funded, NIH funded. But if this connection to a large
number of individuals within the health system is made, that's going to become the dominant
source of information in the future. That's going to drive this kind of analysis rather
than the research-led things coming out of doctor -- of hospitals.
I think that the other thing is that this is just the start of digital medicine, so
this is genomic medicine, you know, looking at variants, but there's talk within the EU
of much broader projects, building, you know, digital integration of lots of other types
of data around the patient, images, other types of high frequent data. And there's,
in fact, a project, there's a proposal for flagship projects, large-scale developments
funded in the next framework program, and there's a pilot going on at the moment called
"IT Future of Medicine," which is around, producing virtual patients, around digital
So that's all I'm going to say. Just acknowledge the different groups that have fed into this.
There have been lots and lots of discussions over the last five years in the U.K. around
what's necessary to make this happen. I think this is going to be an important meeting,
in terms of this, defining this, how do you capture the variants which are going to feed
into the clinical value. Thanks very much. Now, so I'd like to introduce Rex, who's going
to come up after me, who's one of the co-chairs --