Tip:
Highlight text to annotate it
X
Andres Metspalu: So, I want -- so, good, let's say, morning
or afternoon, whatever. But I'm really happy to talk and continue another small country.
So it's a small country [unintelligible].
So I have to, first of all, I want to make sure that you all know where Estonia is; it's
a small country here, because otherwise, you know, sometimes you are here and sometime
here.
[laughter]
But this is three Baltic -- small Baltic countries, and we have a northern one close to the Finland,
which over here.
So health care -- we have one medical university, one single payer for health care, and our
budget is about €1 million, 1 million, and pays for health care. No other [spelled phonetically]
location, no research. From last week, January 1st, exome sequencing is reimbursed by this
system. And in order to get your testing through the system, you have to apply in January,
now, and it takes a year for them to evaluate it, and then you'll get your code and price.
And price in our case is €1,500 per genome -- per exome -- and they accept three years
also. If you're one patient and parents, they pay all three.
So this is -- think about exome sequencing. So what I'm talking is that medical genetics,
this is a newborn screening, micro array, this is going on, and, you know, exome sequencing
most recently, so I'll call it. But I'm going to talk about some medical genomics. And this
is a common and complex disease.
So, first of all, we started about 14 years ago, year 2000, biobank was adopted in Estonia,
and they started a biobank. It took many years. You know, 2010, we had a biobank; 5 percent
of Estonian people were in the biobank, and we used this for research. We are publishing
this small place. We have a biocenter -- our genome center has 40 people. And they are
publishing like 30, 40 papers per year now because we have this resource. And, of course,
in addition to the Biobank, we build up sequencing and genotyping, and also have a good team
of bioinformaticians that teach us.
So, main thing is that we have Estonian Biobanking, Estonian Human Genes Research Act, which is
a law. And it tells that we have to use results in order to improve the public health. So
it's by law we have to do it.
So this is how people are actually reacting. When we started the biobank, you know, we
had lots of people who didn't know anything, and so still don't know too much about biobanking.
But, you know, there are lots of people who actually support genomic medicine, support
biobanking, and genetic research. All these people who basically are against it. And this
is also normal, you know. I know that some people you offer color TV and they refuse.
[laughter]
So this is -- we did like everybody else in Europe did, a PC [spelled phonetically] map,
and so, you know, Estonians are over here. So next to Latvians, Lithuanians, Poles, and
Russians from northwestern part. Fins are over here, and Pusuma [spelled phonetically]
which is famous [unintelligible] is over here. The rest of Europe is -- Swedes are here.
So this is -- tells us that we are pretty close to rest of Europe but still have to
look some specific alleles. Not so many like in Finland; we know that Fins have their own
diseases here which we don't have.
So then just to tell that our biobank is available for research. Everybody can get access. You
have to go through -- to get it you have to go through Ethical Committee and [unintelligible]
Committee, and we can send samples and papers, if approved.
And -- okay, so it tells that -- so, next thing. Now we have a biobank. We can do research.
We know a little bit already about the population and we know what is going on.
So next thing, what is kind of needed in order to do genomic medicine for complex diseases,
which is for the full population, we need e-health solutions. And everybody has, in
our country, so-called ID card. And as you see, we have active cards 1.2 million, and
we have about 70,000 people don't still have this card for different reasons. And this
is now -- in addition to this one, we have also mobile ID. So from my mobile phone, I
can do all what is needed. I can do bank transfer. I can go to my health records. But there you
can do. You can use your mobile ID also.
So it tells you that in Estonia we have all databases, like here. So, prescription database
-- all prescription [unintelligible] all are recorded. So we have national health information
system. We have digital images on everybody. And you can access this database from everywhere.
So everything is connected, so-called X-road, which is secure way to connect different databases,
and even Fins now actually took this over. And everything is [unintelligible]. It's really
shame to accept something from Estonia. But it's nothing to do because they have been
doing it for so long and still can do better.
So we're -- so national with the personal key infrastructure, which means that everyone
can access and give a digital signature, which is legally binding, using your phone or ID
card. And what is good is that over 10 years, nothing happened. No misuse. And people are
very trustful, and using this, banks are closing offices because nobody's going to bank anymore.
So, this is a national patient portal, which everybody can access. And, you know, this
is, of course, Estonia here. But this is my data. And here you can sign. If you have accidents,
you can donate your organs, so you can make obligation here, vaccination, so on.
Here I just would like to add one box, which is just genetic data. And I tell you what
-- how we are going to get this data. This is website telling that -- the OACD. Estonians
are really good putting health information into the Internet. But what this missing this
is decision support, which is very small here. And this almost everyone could see small because
what we need is we need the diagnostic support software -- decision support software.
So do we have enough information? So we -- I don't have to explain, you know, what's GWAS,
but combining several markers, I still believe that we can do some -- can use this information
in some useful way.
So let's take this -- all information here is coming from Biobank. We have a child that
-- let's say Type 2 diabetes patients here. And we have 20,000 arrays, so we have information.
And you see that these were just lots of patients according to the genetic risk. So you see
this second number of patients have a higher risk than others. And if you just take it's
more simple way, you'll see that actually it's -- of course, BMI is important. But genetic
information is also important. You'll see how much risk is increasing. You have a high
genetic risk, let's say, the top 20 percent.
So I guess this is the most telling figure. So if you are young, let's say 25 or 30, and
your BMI is 25, and you are not in service group because in our country, if you are 40
years old and your BMI is 25, above 25, then you are in risk group and GP should do something.
But say can't do because there's too many of them, and as they have tons of other things
to do also.
So this type of analogy actually can [unintelligible] number or they can focus. And if they focus
for 20 percent, it's already probably preventive measures are more kind of fruitful also.
So if you are young here and you are high risk, so your BMI goes up or your test keep
it low, you'll get it here. So it probably five times difference in this. And so if you're
young, I tell you, it's much easier to keep your BMI lower than to reduce it, than to
have it already 35 or close to 35 like I do. So I tell you, this is something useful what
we can do.
Okay, the [unintelligible] don't tell you much also because you -- they are almost the
same, but still better if you use the genetic information. But, again, genetic information
is known already day number 1. So BMI is going -- when you are already going on. Of course
relevant [spelled phonetically] test is already too late. You already have the disease.
So this is now slide I borrowed from Markus Perola, and you see if you just take some
myocardial infarction, and usually, say you have certain numbers of 20 percent risk, they
will say you will be treated like normal. But then the ones, which below screen here,
if you are [unintelligible] extra genetically, you'll still find out certain number of patients
which can be treated. And according to the Finnish data, which is coming from Fin risk
[spelled phonetically], so they can prevent 130 deaths in 10 years. So per year, it's
13 but this is only 400,000. In Estonia, have 1.2 million. So almost every third day, somebody's
dying from myocardial infarction because they are not tested. So these are not [unintelligible],
this is Finnish data.
So what I proposed to the government plan that how to use a resource, what we have,
and the information from the biobank [unintelligible] really use genetics in public health. It's
always telling. You have to use information in order to improve [unintelligible]. So I
thought that let's do the bio project first and sequence 5,000 people in order to get
those haplotypes and variants up to -- down to 0.1 percent, and then use this information
to build a new array, which Illumina, let's say up to 100,000 markers to the existing
700 snp/ChIP and to everybody on the Biobank.
We know this information, health information on these people in the Biobank. We know all
events, what happened after they were recruited because we got all the information back from
the different databases, including hospital databases. So we can actually analyze that
data, what we see from the arrays and what are the actual real clinical picture, and
put everything into this here system and physicians, GPs can use it over there, can access their
-- patients owns the data and now we're going to patients owns of it. And they can tell
who can actually get access. And they also can make access or research.
So if this would be -- of course we discussed it through many organizations, including the
parliamental social committee, but then they decided let's do this part first and put everything
into health center people are using. And then main project would go -- we offer this test
to everybody 35 to 65, and we did that survey, and about 75 to 80 percent would like to have
it. And so we have like 500,000 people in the database, so it's e-health database but
this only database.
And basically we have child types, we have a family structure, and we have prescription
database, we have images. And it's updated daily, basically, hourly. And so now can you
imagine? We have a database half a million people. You'll see somebody's getting drug.
And after one month, he finds the same drug. Or, actually, after one week, happened to
be [unintelligible] and so on. You can something -- you can see like from the airplane on town,
you can actually follow the system. This can be used, I guess, both for research and also
for the real prediction.
So, type 2 diabetes or glaucoma -- glaucoma's pretty easy. With few markers you can get
the risk and just send to the eye doctor the image of your eyeball pressure. We have 28,000
people blind because of glaucoma, so on.
So, basically, this is most important thing because we have only 10 medical geneticists
in [unintelligible]. There's no way that we can offer genetic counseling for everyone.
So physicians and GP, primary care provider, have to learn a little bit more. But basically,
we have to do everything, the ChIP analyst has to be in automatic [unintelligible] in
this software. And when GP is opening the computer, just the probably three lengths.
If glaucoma risk is high, go to eye doctor. Your warfarin risk is -- warfarin level is
very low. You just have to go into certain -- basically it has to go into the prescription
database and doctors are prescribing [unintelligible]. They have to look what are those, and so on.
Not so many things and not very complicated. And then computer can land the plane in the
-- with 400 people in the [unintelligible]. So computer can also help GP doing his decision.
Okay, so basically what happens, we have a kind of circle. And we just improve the database,
learn or just makes the data better, we get better results, and this is a circle, some
of you were showing it before. I guess Dr. Robin [spelled phonetically].
So survey of PC is actually a very interested in this and patients are -- actually it's
a main [unintelligible] because patients are even more interested in this. So they would
like to. Of course we have challenges. Hospital directors are against it because they have
to change their life, so not -- they have to follow probably thousand of these pre-diabetic
patients instead of doing some [unintelligible]. And of course we don't have enough knowledge
but we always learn and lots of things to do, but in small country like Estonia, this
can be a deficit because we need a one positive case, one positive example how things can
work. And with this 1 million people, I guess we can't do it.
So I guess also [unintelligible] in place. And the government decided basically to add
some more location part and of course some business part. But this is mostly approved,
and we have to come up with more detailed plan in -- by summer and fall when you finally
will be open. We are ready to move on.
So, thank you very much. [unintelligible]
[applause]
Reed Pyeritz:That was very nice. Reed Pyeritz. We were talking amongst ourselves and wanted
to suggest to you that you market your electronic health record system to the rest of the world.
My question has to do with the histogram of type 2 diabetes risk that you showed to the
tail of high risk. How many of those individuals would be detected by family history alone?
Do you have those data?
Andres Metspalu: No, I don't have it here. And from the biobank,
family history, I don't know the data now from my heart.
It's a problem here is. So I tell you what the problem is. Now the problem is that 50
percent of kids are born are single parent. So father's side is gone. This is one thing.
And basically I don't think that family history are so good that we can really -- I believe
that this is much better because half of family in many cases gone. Yes?
Marco Abramowicz: Yes, Marco Abramowicz, Brussels. I understand
the beautiful opportunity to have a nationwide database. But I have a question about the
Estonian gene ChIP that you devised with 1 million ChIPs. I'd like to know how many of
these are validated for clinical use? How do you intend to use them when you mentioned
applications for the patients?
Andres Metspalu: We are going to validate to [unintelligible]
pilot phase.
Marc Williams: Marc Williams, Geisinger. A comment on the
family history question that Reed brought up, and then a question. The comment is, is
that, you know, given the infrastructure that you have developed, I think you have the opportunity
to systematically collect family history and have that persist in your system, which would
allow, actually, linkage to develop over time so that you wouldn't necessarily have to rely
each time on the patient, you know, providing that information. That could be actually systematically
collected and updated as some have proposed, not the least of which Dr. Ginsberg and the
work that he's been doing.
The question I had related to the clinical decision support, which I think is elegant,
what I wanted to know is do you have, then, a centralized clinical decision support group
that when something is submitted that is approved, that -- is that then automatically pushed
out across the entire country, or how is that adjudicated?
Andres Metspalu: Yeah, there is a kind of a task force that's
being discussed. This will be just one place, and like sequencing or genotyping that going
to be in one place. And we'll have a genetic testing this one place in hospital [unintelligible]
something. You know, a small place, you know. Everything -- we don't have to divide everything
by two, you know. It's just one.
[laughter]
But coming back to this family thing, you know, doctors are using family information
-- all information about we can use to date. But number -- the percent of type 2 diabetes
is increasing. So why it's not reducing before, which [unintelligible] instruments what we
have. I believe that we need something additional in order to put the ceiling on this.
And I tell you one more thing. If you are just give this personal information to people
-- so people are more likely to change their lifestyle than just put a poster up and lose
faith around more, or eat what they're -- so if you tell personally, then it's different.
And we did a survey and not it published but Dr. Sing [spelled phonetically] is published,
but it was interesting. Of course, it's easy to put that box -- a tick in the box. But
after [unintelligible] go to run. But I still -- there are also data also showing that people
are willing to change something if they're told personally.
Male Speaker: Thanks.