Ræðismannaráðstefna: Kári Stefánsson

Ráðstefna fyrir kjörræðismenn Íslands erlendis
Reykjavík, 2.-5. September 2001

Ræður og erindi

Dr. Kári Stefánsson, forstjóri DeCode
"DeCode Genetics and Bio-Technology in Iceland"

3. september 2001


It was many, many years ago that Masters and Johnson pointed out that size does'nt matter. Now Geir has pointed out to us that in economy it does, the smaller it is the better it is. That was not what Masters and Johnson said, however.

I would like to thank you for inviting me to address you today to tell you a little bit about DeCode Genetics, or genetics and bio-technology as an addition to the Icelandic scenery. It is interesting that bio-technology is basically a service industry to healthcare so the goals of bio-technology reflect in many ways the goals of healthcare. If you look for example at the Icelandic society and we look at one of the studies that we have recently done which is on the genetics of old age we have found out that in Iceland longevity is inherited. It goes in families, that is a genetic component to our ability to become old.

It is absolutely fascinating that you can see that in a population, we took the entire Icelandic population, we took everyone above the age of 90 in Iceland and we took 500 control groups of people born approximately at the same time, and we asked the question whether the people who managed to become 90 years of age were more related to each other than the ones who died before 90 years of age, and indeed, the people who managed to become 90 years of age are as a group much more related to each other than the people who died before the age of 90.

Then we actually asked the question of what does the lifespan of parents reflected in the lifespan of the children. And actually if you are an Icelander and above the age of 65 your probability of dying within the next year is significantly less if one of your parents became 90 years of age. So not only is longevity inherited in Iceland but the lifespan of your parents has a significant impact on your own lifespan.

And think about it, this happens in spite of anything you do in your life, this happens in spite of the trucks that hit some of us and terminate our life before we can take advantage of the longevity gene.

Then we asked the question of how is longevity inherited. There are basically two ways, fundamental ways in which you can inherit longevity. One of them is that you can avoid the disease genes, the genes that give you the diseases that kill you early, or you can inherit something specific that allows you to live long. If you avoid the disease genes it means that the inheritance pattern is very complex because the disease genes are so many. If you, however, inherit a positive asset, this could be very simple and indeed in Iceland it is very simple. It is simple to the extent that it looks like the ability of Icelanders to become very old is dictated or conferred upon them by a single gene.

So, why am I mentioning specifically the story of the genetics of longevity. I am mentioning it because of the difficulties that I have, of the difficulties that most people have in defining the objectives of healthcare, what is the goal with healthcare, what are we trying to achieve with healthcare. Are we trying to achieve longevity, or are we simply trying to make sure that this little time that we are allocated on this earth we will lead without pain or discomfort or without disease.

There's actually a very dramatic scene in one of Shakespeare's plays. Shakespeare has Caesar sitting in a room before he is killed and then one of his servants comes because he was beginning to feel that something was going to happen to him there, and the servant asks if he isn't afraid to go in there and he answers with these very famous lines: "The cowards die many times before they face their death, the valiant tastes of death but once. Of all the wonders that I yet have heard it surprises the most that man should fear, knowing that death, a necessary end will come when it will come."

But the goal of healthcare, it actually turns out that Shakespeare who put together these wonderful lines, and expresses this very heroic view of what life is and what the end of life should signify, is not exactly in keeping or in tact with the goal of modern healthcare in our times because we are expending enormous amounts of money, we are putting enormous amounts of effort towards understanding diseases that kill us, tackle these diseases and try to extend the life that we live here.

It's absolutely clear that the healthcare system is not only trying to make sure that the distinction between life and death is clear, which means that when we live we live, and then we die. It tries to extend life and has done it successfully, and there are some of my colleagues here in this room who have successfully extended life of a good number of people. And in that have served society well.

One of the things that have happened over the past couple of years, or particularly over the past year, is that there have been taken very large symbolic steps towards allowing us to understand health and disease and giving us thereby an opportunity to do something about the loss of health or do something to allow us to retain health.

Probably the most significant of these steps has been the sequencing of the human genome which contains the information that goes into the making of man. This information is extraordinarily important, it lies in about 3,5x109 bases. There are 4 bases, A's, C's, G's and T}s and the altering of these bases, the altering of the sequence of these bases basically contains the information that goes into the making of all life on this earth, not only human life but all life. All life can be traced to the sequence of A's, C}s, G}s and T}s. Terribly important. Life has evolved by random changes in this sequence and then selection and that is what evolution tells us although there are without any question some in this room who believe that there may be some place, somewhere, for a divine intervention in the changes of these bases for the selection, on the basis of which these species have risen.

One of the things that becomes obvious, when you begin to look at the genetics of human disease is that they basically fall into two categories. One category are the socalled Mendelian disorders where there are clear so-called mutations in genes that cause very rare diseases but cause diseases in a clear, distinct manner. You have a mutation, and you will develop a disease, you don't have a mutation and you won't.

But this is not the situation in the common diseases of man. In the cases of the common diseases of man the diseases are basically caused by several genes that come together or caused by an interaction between genes and environment. This is important to recognise because it means that these diseases can skip generations, they can affect grandparents and grandchildren but not the parents etc. and in that their study becomes very different from the study of the Mendelian or the simple disorders, because the diseases can skip over generations.

And this has considerable implications when it comes to the work that we have been doing at DeCode Genetics in Iceland, because once you begin to study diseases that skip generations you have to begin to study the diseases or the genetics in the context of a population rather than a nuclear family. The nuclear family is the appropriate unit in which to study Mendelian or simple disorders, it does not suffice when it comes to the common diseases.

It is also important, when you look at the study of the genetics of common diseases versus the Mendelian disorder is that if you take the rare Mendelian disorder you probably could have removed them from man, you could have taken the mutations that cause them, and you could have removed them from the genome and man would have evolved just like man has done without them.

However, if you take the common disorders of man I am absolutely convinced that man would not have evolved without them, because in many ways the common diseases of man, or the genetics of the common diseases of man, reflect the limitations in the design. So if you e.g. would try to take all of the mutations in all of the genes that participate in causing the common diseases of man you would most definitely limit dramatically human diversity, and you probably would get rid of man altogether by doing this.

But we have to limit the diseases, they are a part of our limitations, something we that we have to adjust to, and when you begin to think about the common diseases in this way, it completely alters the way in which you look upon your attempts to deal with them, to diagnose them, to treat them and even to prevent them.

Once you recognise the difference between the Mendelian or the rare disorders and the common disorders of man, once you recognise that you have to study them in the context of a population rahter than a family, there are certain things you have to begin to demand from the population that you are going to study the disease in, one of them is that you have to have extensive genealogical records in the population if you are going to be able to study genetics in the population.

Why is that? Because the study of human genetics is the study of the information that goes into the making of man and then the study of the flow of this information from one generation to the next. The geneology gives you the avenues by which the information is passed from one generation to the next. So having a complete geneology is extraordinarily important because once you go beyond the nuclear family the anecdotal reporting of relationships is pretty much useless.

In our company, at DeCode Genetics, we have the genealogy of the entire Icelandic nation going centuries back in time, on a computer database, and I will give you an example of what you can do with that geneology, in a minute.

You also need extensive medical information that you can cross match with the genealogical information and I will mention ever so briefly a law that was passed in Iceland in 1998 allowing the Ministry of Health to commission the building of a centralised database in healthcare in Iceland. You also need extensive molecular data that allow you to look at variations in the genome and correlate them with variations in the development of disease, variations in the ability to resist disease and cross reference them with geneology.

The Icelandic population has all kinds of other advantages or qualities that give us some advantages when it comes to the study of genetics. One of the qualities that the Icelandic population has that I haven't mentioned here, and it is terribly important, at least when it comes to our little company, is that it has turned out at least for us to be fairly easy to recruit absolutely outstanding scientists. We have in our company mathematicians who form our data mining group, we have an astonishing collection of great mathematicians, we have great statisticians, we have wonderful molecular biologists, we have historians who are working on the compiling of the genealogical records and I am absolutely convinced that in our company, which has about 600 employees today, we have the collection of talents that we are not going to find anywhere else in the world.

And to refer again back to the talk of the Minister of Finance, he was mentioning the necessity of developing industries that allow us to retain highly educated people, we have repatriated somewhere between 100 and 200 Icelandic scientists who were working all over the world and we brought back to our company, and what is more, and no less important, is that about 15% of our 600 employees are foreign scientists of very, very high caliber that we have been able to bring in to Iceland, and although we are partially making use of the genetic pool of Icelanders in our study of genetics of disease I believe that we have served our community extremely well by increasing the ethnic diversification in Iceland, because if you are going to thrive in a globalised world we have to have fluency in our interaction, cooperation with people of all kinds of colour, ethnic backgrounds etc., so I think it is extraordinarily important for Iceland not only to retain the best people, but also to steal the some of the best people that you guys have in your home countries.

Let me now show you what we can do with computerised genealogy and what I am showing you here is the use of an extraordinary instrument that is absolutely key, when it comes to doing genetics of common diseases. It is an instrument that allows us to systematically use data mining to figure out who is related to whom and in what manner so we can follow the flow of genetic information to see how it segregates with disease, or ability to resist disease, and I am going to trace my own noble lineage back to Egill Skallagrimsson, who was born the year 910.

The yellow colour here siginifies consanguinity, and I refuse to be held responsible for what my relatives did during the dark, cold and long winter nights, but if you continue to move down the line, and we are just continuing to walk down, here is one of my relatives who was a contemporary of Leonardo da Vinci. This guy here, Gisli, was one year younger than Shakespeare, not quite as good a poet, but Snorri here who was a contemporary of Dante was really his match, when it comes to poetry at least.

But we continue here and we end up with Egill Skallagrimsson, born in the year 910, and I can tell you that this is not only a guess. This is the way in which I came into this world, this is exact, accurate tracing of my family back to Egill, who was one of the greatest warriors, one of the greatest poets and also said to be one of the ugliest men ever to live in Iceland.

This is said for fun, but since this is the denominator in everything we do, since we use the genealogy to cross reference everything with, we start out with it. It is extraordinarily important to make sure that the data are accurate.

And what we have done is that we have taken modern Icelanders who are the descendants of a common foremother and a direct maternal lineage, and we have done it with a large number of Icelanders, and keep in mind, mytocondria, the energy stations of the cell are passed from mother to offspring. So you can trace maternal lineage going by trancing mytocondria.

So we have looked at if there is DNA in the mytocondria and we have looked at variable regions there and we can pick up patterns that are much more accurate than fingerprints, and we can ask the question, is the relationship that is reported, does it match, the relationship that is documented in the database?

For the first couple of hundred years it was written on calf skin and when we look at this the reported relationship match the biological relationship in more than 99,3% of cases. So this database, covering the genealogy of Icelanders for centuries, is almost completely accurate, which is astonishing. It is not only amazing, but it also shows what is obsessive compulsive nature of this nation that was starving for centuries but made absolutely sure that the genealogy was accurately documented.

Another problem with a database like this, if you look at it in a horizontal plan, if you look at it in modern time it is a question of non-paternity or false paternity, because there is, as you may remember, an English saying "a mothers baby, a fathers maybe", how often is that a problem. We get all data that comes into our building encrypted, for personal ID's encrypted and all of the biological samples encrypted. And we just isolate the samples and we do our studies and then at the end of them the first thing we do is so-called inheritance check.

And, acutally, if you add together lab error and non-paternity or false paternity that is a little bit less than 1,5%, which is astonishingly low. I mean, for most countries this is reported between 10 and 15% which means that either Icelandic women are not as loose as their reputation has or they are good bookkeepers.

I honestly did not think that this was funny because this is one of the things that we have to rely on in our work. But what comes out of this? When we run through this information on patients with diseases out come families like this that are extraordinarily important for us in our search for disese genes.

But what I am basically showing you, there is a list of about 50 diseases that we have been studying. We have been able to map genes in about 22 of these diseases, some of the most common diseases of man, including non insulin dependant diabetes, Parkinson's disease, Alzheimer's, schizophrenia, stroke, anxiety, chronic lung disease, and I could go on forever.

But what is the business in this? I have been going over time, but let me just flash by you the business model that we believe is adding a little bit to the Icelandic economy. Our business model is based on bringing into our data mining instrument a lot of genealogical data like I showed you before, what we call genotypic data which is data on variations in the human genome, phenotypic data, data on diseases, and then out of this we mine discovery services that we turn into development of drugs, database services and healthcare informatics and our company has very large alliances with the big pharmaceutical companies in all three of these major business areas.

One word on the Icelandic Healthcare Database because some of you may have read about it, some of you may even have been forced to talk about it a little bit. And the Icelandic Healthcare Database is a collection of selected information from the Icelandic healthcare system, encrypted and fairly well protected, where the information is used under presumed consent like the same kind of healthcare information is used all over the world. The idea is to use this healthcare information to help you to make correlations between variations in the genome, on the one hand, and diseases on the other hand. And the importance of this approach is that it allows you to study systematically the correlations between variations in genome and variations in disease and health without any a priori assumptions, without any hypotheses, and simply by using the pure, unadulterated power of modern computers.

The healthcare database was put together after community consent had been established as well as it can be. It was done in compliance with all international treaties that Iceland is a part of. It has the oversight of all kinds of ethics committees and data protection commissions and I am absolutely convinced that Icelanders are setting a ground breaking standard when it comes to the use of healthcare information for biomedical research. I challenge you to point to me a single place in the world where people are held to the same standard, not only in research of this ward in general, but in particular when it comes to the centralised database in healthcare.

And having said that, since I have run overtime, I am going to stop and give you an opportunity ask a question.