Event ID: 1040437
Event Started: 7/8/2008 7:47:05 AM ET
.Please stand by for realtime relay captions.
Please provide captioning service with telephone number to access audio for this event, thank you.
Good morning, everyone. And welcome back. We had an interesting session yesterday. Rick and I were just comparing notes. But before we get into that discussion today, I want to introduce, for those of you who may not know him our regular and close friend of the committee, rick camp Nelly. I say he's a close friend because he's the man who helps us with carrying the message to the secretary and helping us get our work done. We're appreciative of our support. Rick is here today on a bitter sweet mission, let me turn it over to you, Rick.
Good.
I'm going to go to the microphone now. [ Laughter ]
That's better. The bitter sweet mission is to say a few words on behalf of a person who is a friend to me and all of us. I wanted to come up here. It's great to be with you all again. It was interesting to you with -- to be with you yesterday. I know that you will continue that today. I've had a chance to visit with you a number of times. Most recently on the depart tour of [ Indiscernible ] and now on [ Indiscernible ]'s depart you're from the committee. Very very much appreciate and recognize the importance of your work and grateful for it. I do have the honor, today, on behalf of the secretary and my colleagues to say thank you to you Francis for your service to this community for many years. Francis in recent weeks with the passage of Gina we've had the time to recognize your work. If we had known that shortly after it passed you would announce your departure we may have left a few wrinkles in there. Actually, it was a great effort, a great effort for many years on behalf of many people. I know that Francis recognizes that. You and your team and you personally made a great contribution and really helped to push that towards the end. It took a great effort to get the ball over the goal line and to good affect.
All of us know, as well as though that our debt of gratitude goes far beyond Gina. You shaped the vision of the Human Genome Project. You advanced science achievemented that will continue to unlock so much potential. That phrase unlocking and continuing to unlock so much potential for human health that's an ewe day day shoes -- audacious statement. You walked me through where things were with genomic research. You got that smile when you talked about your work. I have thought that's the smile of the joy of the privilege of great discovery. You see it on the faces of children a lot when they discover. You don't often see it on the faces of adults. It's something that all of us can really look to and emulate. Francis, with you it's contagious. You said to me that in years to come people will look back on this era as the ushering in of a whole new era of science. You were not talking about your accomplishments but what it meant to all of us. Just yesterday I was talking to someone at the scripts medical school. They're the leading edge of the next generation of doctors. A new Army being formed with new weapons, weapons which you helped to give them. This is a wonderful gift that you have given all of us. There's something just as significant as the discoveries, all of these wonderful discoveries you have kept in sight and become a recognized voice. Gina is an example. In addition you have become, and we will trust you will continue to be a voice that ensures that ethical issues remain at the forefront. Recently I was at a dinner with a very significant, a person pretty high up in a government of the E.U. He had an important position there and was involved in some aspects of human molecular research that has been in the press. I asked is it that he was able to integrate the significant moral and ethical debate in the effort of pursuing science there. And in is what he did -- he turned his face sideways and waved his hand in a big arcane continued with the scientific discussion. So the scientific discussion went on, it was very interesting and valuable. We engaged in that scientific discussion. Then he said -- but I don't think you could do a lot of this in the U.S. I said, how is it then that you deal with the moral and ethical questions? He turned his head away and waved his hand. You know, Francis, you have contributed so much to science. No one can deny it. Just as importantly your life and voice have reminded us that we can and must always be able to entertain these questions. You've kept the big picture of health in mind. You've enriched our science and public debate. Not only are you open to true scientific inquiry and debate, but you are a good debater, I know. It's always been a joy. It's always been a great education for me. It has been one of the wonderful privileges. 15 years is a long time at anything. Your work and contributions to this committee has had an remarkable influence. Now after reports, meetings and calls at the end of this journey I think all of us can acknowledge that you have taken us to a new land. So, we here want to acknowledge your many valuable contributions. We thank you for bringing together the discipline and mind of a great scientist and bringing it together with the joy and exiewb ranls of a boy on the edge of a great adventure. Thank you very much. [ Applause ]
[ Speaker/Audio Faint or Unclear ]
I want to say a few words on behalf of the committee, as well. I have known you as long as Rick has. But nonetheless, from a distance it's been extra order. Yesterday we talked about the things that have happened over the last few months that this committee is proud of. We talked about those events, reports that have gone to Rick and the secretary, which we hope will have an impact. The one thing that we didn't talk is the amazing news of the next steps in your career, or what we hope to learn about them. And the accomplishments that you have going forward.
I hope to learn, too. [ Laughter ]
I wasn't here when these committees were created. I understand fully that you are the pivotal person who recognized the importance of advisory groups and helped to create the committee on genomic testing and now this committee. It's been important to us that this committee is able to tackle some of the complex issues that you have been facing. We appreciate your wisdom that you have brought to us and the country. It's the kind of thing we hope will continue to be Fostered here in the committee. We know you will go on to great new achievements. On behalf of the committee we want -- we do hope that you will stay with us so we can continue to benefit from your ideas and suggestions. As a token we wanted to present you with this gift. Um, which we hope you will find to be a fitting symbol of your extraordinary vision and leadership. This is not the first nor the last star that you will receive we wanted you to be assured this one comes with the committee's enduring respect, gratitude, best wishes and admiration. Many thanks to you. [ Applause ]
I'm still a government employee I assume this is entirely okay with all of our various rules.
Yes, we've had it fully vetted.
Wow. My goodness. How beautiful, thank you all. [ Applause ] I didn't expect this at all. I guess it's a good thing I didn't show up late this morning. Having been in a different time zone the last couple of days . I'm glad I made it here in time for the nice words. This decision of mine to move on to the white space of unemployment is not arrived at easily. But after 15 years of this incredible privilege I felt it was time to seek some other opportunity to serve in a way that is yet to be defined. For several reasons, one it's difficult to do much in the way of exploration without bumping into potential conflicts. It seemed reasonable to step out to do search. Also, it will give me a chance to do some reflection about where we've come from and where we may be going and my next opportunities. I could talk for a long time about the way in which this particular committee, the SACGHS, and it's predecessors occupy a critical place for the deliberation on this issue. I have say, Rick, it's been delightful and wonderful with your arrival on the scene to have such a comrade as a point of contact for those many issues where serious attention needed to be paid. You have just been wonderfully helpful in that regard. I will miss the chance to interact in those ways. I think we should all recognize that the secretary has really envatted the importance of these issues into a place that it needed to be and was finding its way into that kind of visibility. This has really come into this own because of his strong leadership and his interest in personalized medicine. That has been wonderful to see happening. I remember those years going back, there's a lot of them, where it was hard to get that track shun. I see Greg in the back, want to give a real shout out to him for the way in which he has played such a critical role in the department in terms of being a leader in trying to wrestle through these issues. And Steve, I thank you for your kind remarks and your willingness to step into the role and to do so-so effectiveness. The work is not done. Clearly, as you all wrestled with the agenda items yesterday. There's many things that need attention in a pressing way. The American public needs you to make sure that we take this remarkable moment in history where we have the opportunity to transform science. That moving these disciplines together in a way. It's been a great honor, a great privilege. I will miss that. I may take up your invitation and sit in one of those back chairs. I will not stop my interest in personalized medicine. One of the things I hope to do is write a little book about personalized medicine. I think perhaps there's a need. Maybe I could make some contribution in that regard. I don't know where I will land. That's part of the adventure. Thank you, all of you, especially Rick, Steve, all of you for the wonderful friendships. And flishments that -- accomplishments that we've managed to do. It's been breath taking. Much more to come, thank you all so much. [ Applause ]
Thank you so much, Francis. Thanks, Rick.
Let me turn it over to Silvia Mann Au who will continue the discussion.
Yesterday the committee had updated [ Speaker/Audio Faint or Unclear ] understanding of personal genome services offered to consumers. This morning we have the opportunity to continue the discussion of the personal genome services by taking a -- looking at the broader landscape and including the state of the science. And the public policy considerations. We'll also have the chance to look at another angle on the consumer interest in these services. Following the presentations by the invited speakers the ky will have a chance to discuss next steps in this area. Specifically what SACGHS might do for further study. So, if you look in your book under tab four you will find that our wonderful and capable staff have prepared summaries of the talks and background information in there. Which you have found very useful. Especially the summaries of each of the companies for later on.
So, this morning we're fortunate to have Dr. Teri Manolio come to speak to us. She is the director of the office of population genomic of the national human genome research institute. Wow, that is is a big title. She's had this position since 2005. She will be providing us an introductory overview of the current state of the genomic research with a particular focus on genomewide association studies. Dr. Manolio will have 20 minutes to present. Then we will have 10 minutes following the presentation for questions and answers. Thank you, Dr. Manolio.
Good morning to you. Thank you very much. You can see in your books the questions that I was asked to address. This could take a week or so to go through. I have but 20 minutes, I will do my best. This is an interesting time to be discussing this. I think we would all recognize as of many 4 or 5 years ago the genome was a complex place. Maybe a couple of associations that were known through linkage studies and that. It wasn't until 2005 when this first finding in age related macular degeneration was found. In 2006 three more associations were added. Then things really began to pick up.
In 2007 the journal of science dubbed that year the year of genome wide studies. This was based on the [ Indiscernible ] and the sequence of the genome and the development of a [ Indiscernible ] map. That showed the relationship of snips and the relationships among those so one doesn't need to measure all of them and use that to infer relationships. The second generation map was just published last year.
The goals for the happen map were to use the density of the snips needed to find associations needed. Not to miss chrome sow mal. And recognizing that one would need more snips for more complete coverage for populations. In parallel with the dpoa. of a happen map the cost of genotyping has fallen dramaticically. In 2001 we thought this was a good deal. The costs fell dramatically as the number per test increased. In 2005 it was about a penny per gene owe type. The costs are now are at about the $400 for about a million snips. To truly a remarkable reduction in cost.
This has led to a huge host of diseases and traits. As of last night there were 58. This sort of onslaught of information is referred to as drinking from the fire hose. I think we can all agree with this. This is an incredible time to be in genomic. We've been trying to keep up with this. We have a genomewide association catalog that is put together by my colleagues. You can find this on the genome website. Or you can Google it. It provides information on the study, the number of people examined, the region studied, as much information as we can pull from these papers. This is just about a full-time job for two people. We would refer you to this if you are interested in knowing the most up to date findings. Although we're a little bit behind.
We were looking particularly at the snips that were identified in terms of what they do in the genome. One would have expected that most of these would have been in coding regions.In the areas that one would have expected to be the most powerful, those that change the protein, the amino acid, or to cause the protein to be trunk Kateed, only about 13 were in these regions. Another three were synonymous, there's no change in the amino acid. Maybe about 40% of them in the introns and not in particular splice regions. And then a few in the untranslated regions. A few more in the five prime promoter region. And the three prime role. And nearly half of them in other or unknown regions. This is one of the big surprises of this.
Lessons learned, gened that were not on anybody's list are now popping up. Macular degeneration everyone was an as keepic disease. Coronary disease, childhood asthma, type 2 diabetes, prostate cancer has been shown to be in a gene desert. Also in Crohn's disease a similar kind of thing. Diseases you might have thought these two were similar. They all share this strong association. All these forms of breast cancer. Crohn's disease and type 1 are not associated, yet they share a strong association. Lots of surprises in here. And many more surprises to come.
In addressing the first question, the recent advances. Probably the low cost gene owe typing. That has generated these incredible findings. In just three years, really an unbelievable bounty of findings. Genotyping costs are now in the reach of consumers, that's a considerable change.
Things that I don't have time to talk about that are on the horizon are associations with copy number variance, next generation sequencing, the thousand genomes project -- it's an international project to identify even rarer sequence variants and alleles. And then DNA methylation. All of these are things that will probably provide even more information about genetic association.
For which diseases are strong genetic associations. I would turn back to you and say define strong. One could have a lot of debates about it. Is it a large odds ratio? Is it a very small P value? The association is very, very unlike to other by chance. Is that is good metric? Is it a risk allele in the population? Is a large proportion of the disease attributable to the risk allele? There are certain assumptions that go into that don't apply to genomewide case studies. Or do they explain a large proportion of the genetic variance?
Let's took a few of these. Again, from the catalog. Here are the odds ratios for a variety. You see the vast majority are below 1.3 to 1.4. The only reason this drops is the power to detect these is low. Probably this is a disability that goes up through the ceiling at these low ranges. Maybe those are ones that you want to focus. Similarly for P values here. Here's the number of associations. Here at 10 to minus 10th. There are some that are much less, less unlikely than that. Even out to 10 to the minus 80. Are those much more believable? Probably. How about the frequency of the risk allele? You can see there's a range of these. Most are in the 20 to 40% range. Some are much more common. Some are associated with fairly large odds ratio. These might be risk variants that might be important to focus upon.
I was asked for which diseases are there a lot of genetic associations shown? Crohn's disease is probably the winner. This paper shows 32 Crohn's disease loci. Until that paper they also show the proportion explained. You can see the vast majority explain .1 to .2% of the genetic variability. You will also notice this drops off a fair amount. This dotted line is the power to detect. It drops off quite steeply. As we're able to detect more of the variance we will probably find more. What those mean at the very low is another matter for debate. Maybe we important in terms of identifying pathways.
Next I was asked what criteria should be used [ Speaker/Audio Faint or Unclear ] is strong enough for the marker to be included in genetic testing. Here we have it strong enough. Because we're in the district, I will say I don't like this question. I would rather answer a question that I would pose, what criteria should be considered? And then also like a politician I won't answer it but say it depends on the testing. What would the purposes of genetic testing be? One could provide targeted risk proven strategies. We could identify persons at high risk for later rapid implementation. This is being done in eye disease. We don't yet have interventions. Wouldn't it be wonderful if we did? To improve the cost efficiency. Other ways of reducing risks may be targeted at people with higher risk. Possibly to provide information to individuals regardless of if we consider it actionable.
So I'm sure many people around this table could provide much better criteria than I could. Just to throw some things out for you, here we're are at strength again. Availability and acceptability. If you don't have a way of reducing this risk one can say how valuable is it to identify it? The availability of the test. The confusion it may give to their physician. Tradeoffs that cannot be paid for in a fixed budget. This was explored in great depth by Paul [ Indiscernible ] from the U.K. They have a fixed budget for healthcare. Showing what they estimated to be the distribution of genetic risk in the U.K. population for breast cancer using seven known LOCI. And what they estimated was taking those seven genes you could have 2200 different combinations. You could look at women who have none of the risk alleles, they would be at this lower end. This is a relative risk of .4. In the U.K. the risk of breast cancer here is about [ Speaker/Audio Faint or Unclear ] there are only about 60 such women. This is somewhat population specific because of the' legal frequency differences. If you had the risk alleles for all seven, you would have a risk of about 2.5. You would be at about 60 per thousand. Which is considerable. Only about 7 to 10 women would be in this group. What is shown here is a nice estimate of the heavy black line, baseed on the current loci, here's the proportion of the population. It would include about 60% of the cases. Down here it would include about 28% of the cases. You have enriched the population. If you were and to identify all of the genetic loci, you would have 88% of the cases. In the top 20% you would have three fold enrichment. You could do quite well. How could one improve the efficiency? The U.K. recommendations are to offer mammography to all women over 50 per year. Women in the 40 ort percent tile of current risk have a ten year risk of 2.1%. Maybe you don't want to test them at 50, you might want to wait until age 50 or 60. How about the women in the 5th percent tile in they have a ten year risk of 1.5%. They never risk a 2.3%, they die from other things. This is based on only seven loci. Would one not offer mammography to these women?
And then what are the limitations of genetic markers? Most markers are not deterministic. Many that don't have the markers will develop the disease. Much of the risk remains ux explained. Plus, there's little or no evidence that detection will change the outcome. Genetic markers may provide risk information. There's little evidence to support this. I have likened this risk information to the cholesterol information. One would never not measure a cholesterol level. It's something that would be perhapsly equally useful.This evidence took 30 to 40 years to put together. Do we want to wait that long to develop that kind of evidence for genetic variance? Probably not. Do we want too biopass this step entirely? Also, probably not. This is just one example. It's cited on page 8 of your brief. It's designed to test a number of risk variances, about 15 or so, it provides risk information back. And see what changes they make in their lifestyle and behaviors. Also to create infrastructure to facilitate this research. I realize you are probably used to researchers coming and saying they need nor research. We have to realize this we're early in this research. We have a fair amount to do before we learn how to refine the techniques. Thank you. [ Applause ]
Thank you, Dr. Manolio. Do we have questions from the committee? Julio?
I have a question, I mean the work is important. Coming up with new [ Indiscernible ] is fundmental. One thing that people often forget is that the issue of cholesterol lowers the drugs were found through working through families that were rare. From those rare kind of genetic findings you can generate the most commonly used drug. I think discovering the target is crucial. I think what people don't understand very much, I wonder how to improve that, the concept of risk. If you have 2% higher or lower risk of having diabetes or higher weight or whatever you could have that genetic component and not have the risk at all. 1% risk of dying of something, you could be 1 of the 100 or the 99. It's very hard to communicate that to people. There was another meeting yesterday, someone came from the audience. They said that someone was tested, they had a low risk of colon cancer. Her father had colon cancer. You can have 1% risk of developing something. You could have a 99% of developing something and be in that 1% that does not get it.
I don't have good answer for that. I'm a cardiovascular epidemiologist. People would say that's only 1 in 10, heck, what does that mean? Others around the table can probably comment on that better than I.
Wanted to ask a question. This is a very clear expose of the field right now. Thank you for your comments. I want to make one comment and ask a question. What I heard you say is the application of the low P values and the low odds ratios are not necessarily translated into clinical validation yet. We have predictive values. You increase your risk. More importantly, I think I heard you say something about clinical utility. If you take the example of cholesterol which took years. You know, I think you said do we want to wait that long, perhaps not. Where is that right balance? That's the first question. The second, where does the field of genetic association lie compared to more traditional ways? I can cite you one date. People like Roger Williams found that 15% of the families in Utah cluster about 50% of all cases of heart attacks in the whole state. It was based on family history score. What is the value added?
That's a critical question. Nobody really has looked at that well. There's a couple of estimates in diabetes that I'm aware of. It may be sort of a 58% area. Basically 50% is sort of dead even. You don't do much better than chance. 58% is not all that much better than that. If you use age, sex, BMI, you get it up to 88%. Family history for Crohn's disease do you need genistic testing? Perhaps not. On the other hand, cholesterol the LDL receptor variance are in less than 1% of the population. In certain people those are very high risk alleles. Athey lead to a pathway that works for everybody. Accept for the people that are homozygous. There are tradeoffs that one has to make. I agree the research needs to be done to prove that this information adds to what we currently know.
Paul.
Thank you. That was very interesting. I wanted to pick up one line of arguement that you raised. It's colored by experience during my training over the battles at establishing mammography standards for screening of women. As I understand your arguement it's conceiving there would be a recommendation not to provide mammograms for a subset of women who might not have the risk alleles that were identified. What kind of study would have to be done to undo what is a blanket standard that is now recommended for all women above the age of 50?
That's an interesting question and a very challenging one. A random used trial would be the only way to do. That would and very large trial. It would take a long time. Could one do this from observational data? Probably not in the U.S. There may be other places where mammography is more uniformly provided. In the military possibly or in the U.K. or in Canada. Those might be places to do it. A random used trial is what would probably convince people. Please understand that I'm not advocating that we not screen people. I'm just saying that's one possible conclusion that you could draw.
Other question ? Francis?
Teri has nicely explained this and the findings from the past few years. Opening up entirely new directions in terms of therapeutics, which is a wonderful aspect. In terms of inherentability. We're not in the position of being able to identify more than a small percentage. When you add it all up there's still a missing huge factor in there. Is it going to turn out those are rare alleles of large effect which you would not find by a genomewide association study, but you would find by sequencing? Are these copy number varrances? These are large segments of DNA that may include whole sets of genes. You can imagine that could have interesting facts. People are trying to look at that. The technology is still districty to be able to say that you have scanned the whole gene Nome. Have we missed on some gene/gene interaction? So individual affects don't look that impressive, but if you had enough power you would be able to see a combination. So far really not much evidence for the people that have looked at it. I think it's fair to predict that in the next couple of years this whole question of where the rest of inherentability will be pushed hard. You should expect that percent that is accountable is going to go up. It's going to up to a degree, whereas in the very nice example of the pharaoh [ Indiscernible ] study you will see that curve rising up more and more in the direction of being able to make more and more predictions. This is a bit of an unpredictable trajectory. That's the direction we're on. Whatever plans people are thinking about making in terms of the application of this to public health will have to be integrated with the sense this is a moving target. There's a lot for information around the corner. We're starting down a pat with interesting curves and turns.
I think our last question goes to Mike.
Teri, thank you very much for that. It's important to get beyond the press and get to the real science of the issues. Steve and I were at a conference last year at the Mayo clinic where you gave part of this presentation. You added a lot to it. I don't mean to put you on the spot. We talked -- somebody asked you is it even worth doing whole genome analysis studies. You said you were thinking about that. Where is your thinking now? I see this as a part of the complete disease signature. That history fiscal, all of these things come into play. Also many other biochemical programsters.
-- programMy personal prediction is that the genomewide association will probably not be the tool of choice for research and discovery in the next few years, maybe five or so. Maybe a little bit longer. But it may be of great value in terms of an individual's risk. Assessing an individual's risk for not one disease, but maybe hundred diseases. We may be at very high risk for some diseases, wouldn't you like to know that? Where that goes, I think as Francis pointed out, things are changing very, very rapidly.
Thank you, Dr. Manolio. Our next speaker is David Ewing Duncan. He's the director of an award winning, best selling author. Dr. Duncan is also the founder of the bioagenda institute. It discusses and analyzes critical issues in life sciences. Dr. Duncan will presenting for about 20 minutes. Then the committee will have about ten minutes for questions and answers again. Thank you, Dr. Duncan.
I want to thank you the committee for inviting me. I wanted to say briefly, add to the parade here thanking Francis. Francis is always accessible and clear in explaining things. We deeply appreciate that. Thank you for that. I have a lot to cover here in 20 minutes. I am here in morning as one consumer, a party of one, a subset of one, who is tested by all of the testing services talked about in these meetings. Not your average consumer. I am a journalist. And an author that is writing a book.For the book I'm having not only my genes tested, but also environmental impacts on the body. I will adhere that this is all noninvasive. I am also the director of a new program called the sfert for life science policy. We're launching this fall, there's a lot nor information about this coming out about this. In conjunction with my book the center is developing the experimental man project. It includes a website, and other activities. And the experiment is for one man to take a wide array of tests. It's a radical concept. I started this project with a visit to my physician. Who declared me to be a healthy why male, 50 years old. I also come from a mostly healthy family that lives a long time, some of us over 90, a couple over 100. We don't have a lot of heritable diseases. As we enter this brave new world I want to emphasize what we're talking about. In the past we've focused on the ill and the unhealthy. We face a future here that is a focus on the healthy individual, which is exciting. And we're trying to predict the future in a way that has not been done before. The committee asked me several questions. I will summarize them in groups. The first being being expectations.
I mentioned my primary purpose was as a communicator and as a journalist. I have have curiosity.Also a future insight to my future health. Coming from a healthy family. What sort of information did you anticipate receiving? I had fairly low expectations of getting extremely useful information. And the normal consumer I'm not sure would have the access to the information that I have. I knew going into this this is an early phase of the science. Also deep down I wanted confirmation that I'm as well as I think I am. Yesterday we had a presentation about different categories of patients. I'm probably the one that likes to keep myself healthy. Also, I went into this with some expectations about what my family results might be. That was a different set of expectations. I was asking others in my family, who were also tested by the way. It did my pause a minute bringing my family into my experiment. They very hartly agreed to go along with this. The next set of questions are tests and relates.
This is the bulk of the talk. What tests? The results. The differences. Overlapping results. A baseline of the tests that I have taken -- I have tested on most of the major snip array, some copy variance, I have been tested for several dozen individual genes. Coming in the next few months I'm hoping to get my full genome sequenced. By the way, that's plenty of information even without the full genome. I have been tested by many companies and academic labs, nonprofits. I was one of the few people around 2001, which is the Jurassic age for gene Mick testing. Very early on getting some results. I do want to emphasize some of these results and this information has been around for a while. We've been having this flurry of association studies coming out, this information and its attempted information has been around for a while.
The costs of my tests, if you add it up, for me it was around $16,000. If you add my family in it's about $20,000. I am gratified that many of the companies did this pro bono or they were covered by publications. If I get my full genome it will be perhaps $100,000.
My mother and father and brother and my daughter have been tested. I was a little reluctant to bring my daughter into this. She insisted, there we have. I will focus on three companies, what they call the big three. Navigenics, you will hear details of these companies. From a consumer point of view there's two or three differences in each of the companies. Navigenicss focuses on diseases. They do not do ancestorry. They do offer counseling. They're more expensive than the other sites. These images are my results. You will get to see them in a minute.
DeCodeMe has a few more diseases. They have incestry. DeCodeMe is also a company, it's a publicly traded company. They do drug discovery, their scientists have come up with some of the major studies that others use. They have scientists working in their back shop. At the moment they offer no counseling.
23andme hits the jackpot for the larger number of traits they feature. It is interesting they have a rating system they put into their large number of traits. They rate things, rate these traits whether they are preliminary or established. There's some criticism of that. But I think it's heading in the right direction for consumers to try to understand which of these traits has the best science.
Two others DNA Direct which is one of the older online companies. They offer only individual tests. I've had one test ordered from DNA Direct. It was a different experience than the others. They have a heavy counseling area to this. I talked to the counselor two or three times. Also, they have a rich site.
The [ Indiscernible ] institute, which is not represented here. I would recommend that the committee look into what they're doing. They're genome testing site will be up in the next few months. This is a nonprofit. I'm sure you all know of them for producing tissues and storing cell lines. They're testing for free 16, 17 diseases about a hundred thousand people over the next several years. They have NIH grant money. They're doing something interesting for those that consider doctors to be less educated in genetics, they're starting out by testing doctors in the Philadelphia area.
Now to the results. I will give you results for three diseases, for the three sites that the mentioned. The first two were randomly chosen. Macular degeneration, it's an A. And B, I if you look over to the right-hand side columns, if you have a behind you can study this chart more, the critical numbers and items for consumer would be the ones in color there. This is a threat level color code, which I have added here. You can see I have threat level colors that range from green which is low risk to risk which is high risk. The second to last column tells you which company. There's a number of different snips here for the same disease. On the far right are my lifetime, or overall risk factors that the site give. The far right is the average risk. This is interesting for a consumer. A bit head scratching. The average risk for the different sites are different. My risk factors are also a bit different. In this case it's not a huge factor. I am low on all of them. Which is really all that I want to know. It's worth noting there's a bit of difference there. Another way to present the data here for diabetes type 2.
19 different snips among the three sites for 15 different genes. My range were from yellow to red. My lifetime risks were within shouting distance of each other, about 4% apart. The average on all of the sites is around 25% for type 2. I'm below that, which is also really all I want to know.
It's worth noting there were four snips that overlapped. Out of them, two of them had different enough results they raised an eyebrow for me. Meaning the same snips had different risk factors associated with them on two of the sites. There were four additional snips on two out of the three sites, there were overlaps of the additional four. And 11 [ Indiscernible ] that were on one site only.
Is the data consistent? Basically to summarize this slide I would say within acceptable parameter. Most of it was fairly consistent. The one area that was strange was to have this variance within a disease from green to red. In talking to the companies they all correlate those and use them as modifiers. That is a little confusing to present such widely ranging results. There was one exception, that is my heart attack gene markers. I don't have heart disease in the family. I did get what I considered somewhat confusing results.
On one side I have low risk, one high and one in the middle. For something like heart attack this is probably not the most comforting results. Or for someone like me that wants to confirm I'm healthy it's not useful information to have. Again, you can see my color coding there. The range from yellow to orange to red. On the far right are these overall risks. And for DeCodeMe and Navigenics you have a 42% to 62%. The average is 49% according to the sites. It's worth noting my risk factors is almost twice the average on the 23andme site. I think I noticed last night that risk factor had slightly changed.
Why the different results? I do encourage you could talk to the companies about this. Some of the information that I gleaned as a reporter is that -- and can see it as a consumer. There's different snips used in studies. The different methods for determining risks for the different sites. You also have different methods for determining combined snips. There's also a reliance or [ Indiscernible ] snips. That has to do with linkage disequalibrium. Not everyone in a population shares these snips. Again, I'm not a scientist. These reflections are through a person who is trying to comprehend and understand this. I hope I get it right. This is how it's explained to me. In the end this heart attack result did leave me scratching my head.
I did with later tests find out that I did hit the mother load of a lot of high risk hoe mow zygote snips that indicate that I have a higher risk for heart disease.
The three generation study with my family I will go over one result, which is a bit surprising. My father and brother are hetero zygote for Alzheimer's. I was sitting in San Francisco one morning and turned on my computer and got these results back. To see anything to do with Alzheimer social security a Alzheimer's is a bit scary. I called my father who is 76, about to turn 77. He sort of shrugged and said I made it this far, I'm probably already. My brother equally shrugged. He seemed to be unconcerned. Hopefully that will continue.
I want to mention the difference between a rare and a common disease for a consumer for a patient. My brother does have a rare genetic it disorder. It would have been fantastic to have a test to know this early in his life. I'm not sure, however, I discussed this with my family, if we would want to find this out in a direct to consumer sort of process. In the future maybe one will not think of that. My brother would prefer to get this through a medical professional. The sites that we're talking about don't offer rare diseases.
Ancestry is fun. I encourage it for everybody. I don't have a bitter taste. I seem to have a lower IQ. I'm at a higher risk for heroin addition. I will live to be over 100 despite my risk for heroin addiction.
The crush of data is something also to think about here. We're talking about a very small number. It seems like a large number. From my project we're up in the upper hundreds now of markers. Many of these are for rare diseases and things I would have no business looking into normally. I'm trying to do everything for the project. If I printed out my chart right now it would be about 24 feet.
The final step here, the final list of questions are reactions and thoughts. Did you alter your behavior? I'm atypical here. As a journalist and having a much deeper knowledge of the topic. I did not really alter my behavior. I did have some tests which are not yet ready. Companies that do human modeling. I did find out I have a higher risk for heart attack. I did alter my diet. There was some talk of statins. WeThe pluses, one gets a great insight into personal and societal health. You get a personal empowerment. I believe that the appearance of these companies and websites is pushing along this process much faster than it was going. It's pushing the health industry to think more about applying this great research. Towards individuals and establishing guidelines and ethics, education and funding. Inwe need more of this. This is a fabulous discussion to be having here. Opening up new avenues for research and medical and drug development.
A few minuses at the moment. Some of these are temporary. This is early days. Association studies are not always applicable to individuals. Disease and nondisease results are sometimes mixed on these sites. At first it's jarring. You get used to it after a while. No standards that I can tell as a consumer that reassured me of the validity. The number at the end of the day is your risk factor. I think consumers are a bit smarter than is sometimes given credit for. If you explain to them even the fact that these risk factors may change over time people will begin to understand that and pick it up. There needs to be an educational process. Physicians really had no idea would to do with the division. There's a potential to frighten certain people. And high costs and who will pay.
A few thoughts and suggestions. Consumers should be free to access their information. And they will. Any attempt to harness in information or keep it away from consumers, especially with the web the way it is people who want this information will be able to get it. I love the discussion that is going on here. And would encourage that to help me and others to understand. I suggest looking into the core Ella approach. Especially the fact that they have doctors first. They're testing doctors first. We need to establish guidelines and standards for the tests and the information.
Kind of a good housekeeping seal of approval. I think most would like some assurance that this information is accurate. A few others here, I think it would be helpful to come up with standards. Disease markers are more important and should be handled differently than ancestry. There's a way, I believe physicians should be involved with some of the decisions in the companies, in assessing people's results. And also 23andme did start offering counselors, a locating system online for finding someone in your area. To conclusion, I want to go back to the beginning. Genetics is just the beginning of this process. In my book and in the project we're also doing environment, brain and body. I have a feeling that a committee like this will be in session for many years discussing each of these developments. There's a lot of interesting research with using brain waves as biomarkers. They will begin to play a role in preventing medicine.
I would love to have you check our our new experimental man website. We're just getting it going. This will download my results. We'll have a wiki-style site. Others can participate in the portal. Various groups, companies, others who offer information about applying genetics to individuals will be able to have a link from the site.
Finally, the book comes out in March. I hope you will pick up a copy. Thank you very much. [ Applause ]
Thank you Dr. Duncan. Now questions?
Thank you so much. This is fascinating. One person's journey into the genome. Just a couple of comments and then ask you to tell this committee what it should do. This is an advisory committee to the federal government. Some of your closing slides were about people should access this information, but there should be guidelines and information. The seal of approval so to speak. And who would pay. If the guidelines for evidence-based applications of this, if we get an independent group to evaluate these technologies and then they will return don't use them until more studies are done. Okay, that could be one recommendation from one group. And at the same time consumers are free to do whatever they want with them, you went in with your eyes open, as to the potential limitations of this technology, which is early on. If you are advocating for guidelines and information, how do you reconcile the need for guidelines with the sort of individual freedom to seek whatever you want.You come from a healthy family. I'm throwing too much stuff at you. I would love to hear what you have to say.
I did this as a journalist. We almost didn't do this. It seemed like a reporting gymic. I learned from that. Also a story I did for national geographic. I had several hundred levels inside of me tested. This is a fabulous way to communicate science, using a real person. I said right off I'm a bit different than most consumers. I suppose it's always driven by a certain curiosity. In terms of the contradiction. I'm talking about in a pragmatic way it's not that hard to go out and get this information on yourself. You can get it tested in any number of ways. You get several discs with a lot of lines of data.
[ Captioner Transition ]
My new hat is similar for -- policy at Berkeley, in terms of policy, I think one has to balance very carefully the fact that this is a new technology, I am in silicon valley, San Francisco, where I live, we see --
The government at some level needs to step in. Things for you all to decide, in discussion with the companies. At the end of the day, as a consumer, one needs to have some sense, borne out of one of the surveys yesterday, I believe it was said accuracy is more important than the issue of privacy, which I haven't discussed at all. Obviously I am releasing my results to the world, not as important to me, the privacy issue, I may be an idiot for doing that, but we'll see. Maybe there are independent groups, a lot of discussion, our center, academic institutions, for creating rating systems. I think this will all naturally occur very quickly, and this committee could collect information on that and be very useful to have input from all of you and some guidance in figuring out how that will happen. I hope that answers the question.
Scott ask next, keep things short so we can get back on time.
Can I ask you to share, from the beginning, what experts did you consult and how did you find them?
I started out, the whole process in 2001 with my personal physician, and my personal physician, originally, for the original project, retired. These were very knowledgeable people at UCSF, my current physician is head of ambulatory medicine, he had to go online, look up information before he was able to tell me much. When I bring in brain scans he starts rolling his eyes. The second question?
You found some consultants, experts to help guide you when you got specific results, and how did you find them and what qualifications did they have to give you feedback on what you had come up with?
As a journalist reporting on biotechnology for major media outlets I have access as a journalist, that's mainly how I was able to access people. It's my job, to go find people like Francis, other other people who can comment -- these things. I have so much material, it's piled high in my office. I also try to get a balance of opinions on this as well, which is really important. There's obviously a lot of different opinions, and people, stakeholders, others, that need to be brought into the equation. Much of that will be in my book.
Paul -- next.
This is very interesting, thank you. I was struck by what I perceived to be an underlying assumption I heard in your remarks, actually heard yesterday also. That is about the term health, the search for I am in good health, as if people know, agreed on what is good health everybody is seeking, these technologies offer insight into. I was think being this question of health against the backdrop of two things raised by genetics I learned from Francis. Human variation is a natural part of the human experience. We have anomalies that are a natural part of human experience, and yet we have these technology and companies that are offering the search for genetic anomalies, to identify difference, with the expectation of what? To eliminate difference? Create better health? I have [indiscernible] plasia, I consider myself in good health, yet I have a genetic anomaly. I am curious from your perspective how you see that, if you thought about some of those questions in your journey, and particularly struck by your brother who has OI, osteogenic imperfect a, had they known that 40 years ago would they have done a gene screen to potentially look for an embrio owe with OI and eliminate it, is your brother in good health or bad health?
He's not in fabulous health. This disease has made him disabled, unable to work. Would have been nice to know, probably more for the family than him. When I pursued this trying to figure out what was going on with him through contacts, [indiscernible] a prominent geneticists, he actually sequenced both my genes and my brother's. It was a COL 1 A and 2 A, Francis can set us straight. 1 A 2. That made coul a collagen, on the conference call up in Seattle with Peter Biers, all very interesting, but at the moment no real effective treatment for what he has, taking drugs that slow down bone loss, but he wants a treatment. Doctor biers immediately said that's what we are after, why we are after the tests. It's a case, would have helped the family dynamic to know, we didn't know what was going on with him. We were fighting -- didn't recognize it as a problem because we come from such a healthy family. It's interesting to say what is healthy? Throughout these conversations, including yesterday, as a writer, working with words, many of the words we need to back up and define. I know I feel healthy, I know when I see it. Applied to other things too. Validity, valid, what does that mean? I thought a couple of times raising my hands yesterday, what is clinical validation, the other is usefulness. I understand that better, but as a writer and word person, maybe we could talk about Francis, putting him on the spot, he's become a wordsmith, we will get more information on that. It's important that we define these words in a way that everyone can understand, including the public.
You keep defining your family as a healthy family, not withstanding your brother's OI, and other issues, so I think health is generally relative in what is a concern, what are these technologies doing to relative assumptions about health, and unhealth, how is that related to disability and non-disability.
There's a philosophical component here too. I keep saying that, that's my mantra my whole life. At 50 I consider myself -- somewhat invincible, I am beginning to realize that's not the case, but that's my own -- I put that out there because we all have different ways of viewing our health. I think -- my brother's situation, he was very healthy until a certain point. I have to stop, pinch myself, remember he's not now, that's a real anomaly for my family. You are catching me up on very deep-felt sensibilities about who we are, who I am.
We have Paul, Joseph, Mike, Jim, Mara and we are going to cut it off there.
David, thank you very much. I have a three-pronged question for you. Prong one:
Get my pencil?
No, it's straight forward. As an experimental consumer, as well as experimental man, what's your perception of the error rate of the laboratories you were enengaged with? Low, never, what's your perception?
There are various levels of potential accuracy or errors here. The actual raw data on the arrays, I think, anyone that knows these tests, run in the clear labs, tend to be very accurate. I have been run more than once, same ri results on one chip, with no error rate. That's very accurate. I did have that slide, and we will do a more quantitative analysis of the sites, not only the three here, any others within our purview. I don't have the data on that, but my swag feeling is that as I said, I think there's a parameter in most of the results that were fairly consistent, accurate, another term we may have to define. We need to remember from a consumer point of view, any other point of view, these are moving targets. As more information comes in, something really important, to emphasize to consumers, the public, that this information is only what we know now, it will continue to change, and even the sites as I have been monitoring over the last several months, they change. Different SNPs come, my risk factors move around a little, new features are added. Again, I don't think consumers will be overly bothered with this as long as it's explained properly.
Second part vert of the question, for those with increased risks, how will you monitor them, what evidence are you using for your style of monitoring?
I don't realm really have that many. I am monitoring the heart situation. I would love to share with you the fascinating tests I have had that did convince me with a high degree of accuracy I need to watch myself. Funny enough, with a huge algorithm I was tested on, a couple of the genes that are on the sites we are discussing had an enormous impact on steering this entire algorithm. I had ultra sound, CT scan of heart, lots of chemistry, other genetic tests, and association study SNPs, especially those on chromosome 9, had an enormous impact. We ran the algorithm as if I was hetero heterozygous instead of -- the test will probably be out in a couple years.
Are you intending to disclose the results of this test when you apply for life insurance?
Contacted them, and someone yesterday said they tried to share this with their insurance company, I haven't gotten much of a response. Insurance industry seems to be sitting back, trying to analyze, figure this out. I did speak to the major national actuarial group about a month ago, ask about this. They said at the moment that the association studies are not yet, don't yet have the predictive power an actuary would need to apply. You can see there will be an enormous change. My results, I am extremely lucky, I have low risk factors for most of these disorders. If I am looking around, saying who should pay what, I no longer have the same risk factors of everyone in the room, everyone has a different risk factor. How do you determine shared risk system for insurance, determine that? If we all have different risks. That's a challenge coming up, for how we go forward with insuring ourselves.
We have Joseph, Mike, Jim and Mara, can we keep them to single questions, please?
Just so you get to me, okay, my question, thank you for the presentation, appreciate that. My question is interrelated. Switch the questioning a bit, in terms of where we sit as a committee, our real concern is what it is we can recommend for the population as a whole, public health question. I will ask a public health question. In terms of use of the information, what is your expected out comes for use of what you are doing, beyond this? The other part of the question really is, whether you have one or not, what would you recommend -- prevention, I noticed argument yesterday for a number of the DTC panel, a lot of added value to knowing in terms of motivating behavior. We know that's not really the case, you yourself confirmed it wasn't. In terms of use of information, what recommendation would you make to the population in terms of prevention, that sort of --
We are in the interim phase here, things are in great flux. In the future, what I believe will happen here is that we will have batteries of tests, already beginning with some neonatal tests, beginning of life for some people, continue to have expanded batteries of tests earlier and earlier in life, giving some indication, I pull out my iPhone, and I believe sometime in the future we will have something called an I House, give us probabilities, possibly measure things like what's in the environment around us, other measurements that will be factored in, do the work, tell us what's coming up. I don't know about live nothing living in that world. Some people will be perhaps frightened by it. When you are talking about predicting the future, essentially what we are doing, we are a long way from really doing that for most people. The rarer the disorder, obviously, as you all know, the more predictive power there is. But in terms of guidelines for preventive care, I mentioned we need to accelerate dramatically the process to Val validate, whatever that words means, some of the resources we spend on pure science, shift rapidly to applying the information to individuals, to find out what's going on. If one comes out with a -- for prostate cancer, mentioned yesterday, a consumer wants to know to make it effective, do a 1000 of those, whatever the number is in a clinical test. Find out if people with that, risk factor, get a biopsy, you come up with results that are meaningful. I am hoping that the federal government and others who have resources to apply to this will focus more on the individual now, how the science applies to the individual. Does that answer the question?
You can have my question, I am going on talk about standards, standardization later, but you got my question.
I appreciate that, I am save integrity saving it.
Jim?
Fascinating presentation. You did something right, because you look far younger than 50. I am interested in this idea of personal empowerment. It came up yesterday as one of the justifications for doing this, and doing it now. My feeling is, I want to get your take on it, that personal empowerment only follows from the prior demonstration that this is useful information. I mean there are large percentage of people in this country who find their horoscope personally empowering, but I don't think we should be in the business of encouraging personal empowerment based on ile losery lusoryy ideas of what -- right now, getting your genotype at 500 -- do you think it needs to wait until we have evidence you can do something before it's truly empowering?
I think my story is that it's not been particularly personally empowering. Most of this information for me has proven to be, one of the possible outcomes of this experiment, most has not been particularly useful, not designed to be particularly useful for an individual, a primary point here. I am talking more about the future. What people would like this to be. Any information a patient or consumer is interested in their health can have, understand, use on their own, is important. Of course, extremely important role for caregivers, physicians, others. In this age when consumers have so much access to information there are those as we heard yesterday, that seek it out, almost desperate to have it. I think it is incumbent on this committee, others, in positions of responsibility to figure a way to create an atmosphere where people feel that these are accurate without squelching, throwing the baby out with the bath water. There's an ongoing experiment on how to apply the information, which is useful and exciting, three or four sites, all have different methods, the -- institute, non-profit model, I know of several others out there about to start. It's one of the delicate moments where in this transitional phase, responsible parties need to make sure there's some accuracy, guidelines, to explain to people this is a transition period, alongside allowing the experiments to go forward in a responsible way.
Mara?
I am going to keep to the future-reaching simple question. The role of genetic counseling, you spoke about that on a couple of sites, throughout the initial part of your, obviously, what was the role of counseling, how did that make you feel more or less comfortable, and how do you see that playing out in the future?
This phase where people are not used to getting results online, some are, silicon valley culture people are much more used to receiving information and it's always interesting to come back to Washington, where I lived and worked a loping time, East Coast, we are more involved with getting our information back in -- perhaps online in the San Francisco area, West Coast. I liked having the option -- first, as a journalist, being able to talk to people, as a consumer, with site that's offer genetic counseling. Of course that was reassuring. I was getting services pro bono, I don't know if I would be willing to pay thousands of dollars for that. I may have to think about the price point that's might effect my answer. I think having some access to somebody that you can talk to if you need to, why I mentioned in my thoughts, ideas, having given careful thought that maybe one way to deal with this is to have physicians that review these results, possibly hired by the companies, maybe independently, not sure which. Somebody watching to make sure there isn't something a consumer misses. There may be a result they don't fully understand that's important medically. There needs to be a system for that. Of course, be great if all the sites had genetic counselors to call. I don't know if that fits into some of the business plans, or perhaps it should.
I would like to thank Mr. Bun Duncan,n, can we try to make it back by 10:00? We will take a break now.
Thank you very much.
APPLAUSE.
[Recess until 10:00 a.m. eastern time.]
Can everyone please take their seats? Thank you.
Okay, we are going to continue the session on personal genome services. Next session is a roundtable with each of the five companies, each will present approximately 10 minutes, less if you can do it, then we will have a Question and Answer period after everyone presented. Just like the NFL, we will give a two-minute warning before the end of their talk so that we can, with no time-outs, though. Our first speaker is from Navigenics, Dietrich Stephan. He's the cofounder and chief science officer at Navigenics. Prior to his current role he was Deputy director of discovery research at translational genomics research institute and holds faculty A pointment there. He's identified genes, contributed to the understanding of genetic predisposition for multiple diseases. Thank you for being here.
Dietrich Stephan: First, thank you very much for the invitation, it's a pleasure and honor to be here. I would like to congratulate Dr. Kolenbrander ins on the if Dr. Dr. COLL ins on the first half of his career.
We have monogenic disease on one side, a broken gene causes the disease, there is no environmental component. On the other hand of the spectrum we have trauma or infectious disease, even those have genetic drivers with the healing -- and pathogen. It's the goal of Navigenics, if I can get these slides working, I guess I could do it without slides -- thank you. The goal of Navigenics, original vision of Navigenics was to, early in life, completely articulate your genetic risk for all human diseases, and across your life span unmask portions of that information that may be useful within that window of life, and use that's what information in conjunction with a physician to avoid environmental stim stimuli that might activate or a serum biomarker for cancer you could ascertain if you knew afternoon individual was genetically loaded, or put on prevention therapy, or treat the disease early so that you could reduce the burden of disease for that individual, and do that on a public health level. I should mention, I was trained in a public health department, with genetic counselors, my first -- I come from a monogenic testing background. I would like to perhaps convince you what we are talking about today is not so different from what we have been doing the last 20 years in the field of medical genetics.
So, for the first 15 years of my career I was involved in doing linkage analysis to identify broken gene that's definitively caused disease, identifying the genetic basis of about a genetic monogenic diseases. You could toss a mutation across the fence to a molecular testing facility, adopt it, medical geneticists knew how to interpret, lots of mutation with a -- associated this, is what will happen to you, your unborn child or your planned children. We have that entire infrastructure, but we didn't at one time have that. There were no genetic counselors at one time. It's recent history. Now that we are doing medical sequencing of genes in people uneffected we are starting to understand that penetrant itself is a concept that will be modified dramatically going for forward. We will turn up people who are compound hetero -- zygotes, the field we understand is evolving as well.
What I would like to posit, alleles of low-effect -- between one and 10 are not so different than monogenic mutations with penetrant variables associated with them. Before we get into that, I guess I would like over the next eight minutes communicate we are facing a healthcare crisis. I believe that should be the primary motivator for all of us. The key driver of mitigating that potential crisis, we are on the trajectory, everyone would agree, the key driver of mitigating that crisis is going to be prevention, I believe. If you believe that, then we should use all of our -- information to maximize prevention and improve outcomes across the population.
I would also like to convince you that genetic risk factors this, is a term I really like, because it embeds all of our understanding of environmental risk factors into what we will talk about today, they are not so different, genetic and environmental risk factors, and they can be used to refine risk in a clinical setting, in addition to other types of risk factors.
That we need a new delivery vehicle for these types of genetic risk factors. You can look at someone, say they are obese, smoke, but you can't look and say they harbor a 9 T -- you can't introduce, the environment is not geared to do that.
So we have done a lot of monogenic disease identification and here are a couple of examples. We have also, in my group identified the alleles that drive several dozen common complex genetic disorders, this in Science, over 500,000 SNPs to paint the genome, with disease, rich in people with diseases versus those without. Continued to flush out the Alzheimer's story, withstanding replication by the community, doing this on a national level. I chair an NHI consortium that provide these type of -- for the scientific community, done over 400 project, many are whole genome studies. We come from an understanding of the technical nuances from which this information is derived.
I couldn't stomach going out to raise another million bucks or two to do one of these studies without an implementation infrastructure waiting on the back end. It seemed like a frivolous exercise to me. Don't get me wrong, on the therapeutic side this is incredibly useful information side, but on the risk assessment side there was no infrastructure. I have taken a sabbatical from T Jen to found natch gens, Navigenics, to understand how we can use this hard wired information to alter the course of chronic diseases so we don't see the explosive rises, Alzheimer's disease, any of the common chronic, age-related diseases. Really, the only way to alter the course of this massive trajectory, looking at millions of people costing the healthcare system trillions of dollars is through early detection and strategy. I already articulated, we have a genomic sample early in love, sample for common and rare, and de novo, do holistic copy number analysis, sift through the modifications and sequence the mi to condrial genome, we are building the infrastructure to do there. The interpretation doesn't exist yet for the vast majority, put all that in a big computer, push a button and get a list ordered, preventions --
For a few the information exists and is robust. I would posit,age related macular degeneration, hemachrome owe to sis, Alzheimer's. We were building a new industry, understand there were ethical concerns, counseling learnings that hadn't been done, clinical paradigm shifts, from the very beginning of the company, we built from the board of directors to clinical advisory board, really understanding how does medicine need to change, evolve, should it, how do we interface with the medical community in the appropriate way, to the advisory board, people like -- trying to guide us through the complex science, to really understanding how to provide genetic counseling. This can be very important information to an individual. We have, for example, the past two presidents of the national site for genetic counselors on our task force, policy, ethics, and an important component of that, Paul Shroff ac, a risk communication expert, as 1%, 1 over 100, how do you use colors, words, to maximize the accuracy of the risk information you are providing to someone.
We have also taken great pains to build a team of genetic epidemiologists in-house. And epidemiologist to vet all the literature that comes down the pike with respect to -- I will call this validity. We are calling them our curation criteria, these are more practical than the [indiscernible], without a loss of accuracy of significance, they allow us to click through studies, identify what is real and not by fully reading those and implementing on the risk assessment panel. From day one we decided quality is of the utmost concern. I should mention the first year and a half of the company's existence all we did was try and understand the regulatory environment, probably spent, chime in, $10 million trying to understand how we would click into the established environment. We feel like we have really digested, understood how to do that. For example, this is an example of that, from day one, we decided we needed to create a certified laboratory with extremely stringent QC and QA standards, measuring quality on a per-SNP basis with across the population, Harvey Weinberg -- using retrospective control data, given it's a germ-line constittative -- modeling this out, can do it fairly accurately, I can point you to a website where all of our information is fully transparent. You have seen this morning, given the current common risk variance we have, how much predictive hower we have. What you see are a bunch of ROC curves, downloading the data from Framingham and -- applying our algorithms on to those data sets to understand the AUCs and ROCs, and these are also going to be available.
I would like to segue now into talking about briefly the fact that the common argument is these genetic effect sizes are solo so low you shouldn't bother. Between one and five, on the same scale, orders, magnitude that public health -- are commonly messaged. And the comment is they haven't been studied enough. Some of the papers published, the 9 P vearnlt -- myocardial infarction, studied in close to 17,000 people, a fairly large data set, replicated twice in sets of 1000 each. The numbers here are approaching, our often on par with environmental risk factor studies. But what you see here in black are home zygote -- you see hetero zygote hetero swrieg effect hetero het -- all between the ranges of one and 10. Practical example and I will end. This is the state-of-the-art of clinical assessment, the primary physician's office, the physician will test your blood pressure, do a cholesterol test, do you smoke, have diabetes, and based on this information, if you have three of these things you go on a stat in, be careful, you will get a heart attack. Very rarely are they plugged into the framingham risk calculator, two of these, one of these, go home, watch yourself, be good. These are the effect sizes we are talking about that drive clinical decision making. If you drive in the two genetic associations, the effect sizes are exactly on par with this. So to say well, we are not going to use that information because it has the word genetics in it, or not use ing it because there's nothing you can do, I don't buy that argument. There are a lot of things you could avoid to balance out your genetic load. We decided we need professional access to counselors, physicians. The whole reason we went DTC in the first place, conscious decision to avoid, minimize the possibility of health insurance, life insurance discrimination. Now we are relaxing that a little, building out physician channel, you will see that moving forward. Basically we are trying to use genetic information to motivate behavior change. There are evidences of this being published. The Reveal study shows minimal distress to people with E 4 results, a study just published that shows people who are at high risk for melanoma got genetic testing, those went and got screened more often than those who just had a positive family history, showing that the word genetic can motivate behavior change.
I will end with this slide, saying that Navigenics, we don't make tools, don't decide on what to do in the clinic after a person gets their screen done, but we believe we can digest all of the information in the literature accurately, provide that in a transparent and accurate way to an individual so they can use that risk information moving forward, and we will turn on monogenic testing shortly for hundreds of genes, and the capability to capture rare variance and common variance for common disease, moving forward, which will hopefully explain more of the heritability Dr. Coul ins talked about, and how to understand how the information changes an individual's outlook and how best to indicate it, as well as Boston University. So thank you very much.
APPLAUSE.
Next speaker is cofounder of 23andme, Linda Avey, developed translational research collaborations with academic and pharmaceutical partners, spent time at SpotFire, helping scientists understand the power of data, visualization and applied biosystems during the early days of the human genome project. Welcome.
I would like to remind committee members there's information about the companies in tab 4 of your briefing book.
Linda Avey: Thank you, everyone. It's really great to be here in person. Last time I was on the phone, so it's always better to be in person. I think we will have problems here with Adobe acrobat. Yikes, I might need to buy a new -- there's not a program for this computer, we have a credit card let's hope this works.
As you all heard, especially from David's wonderful presentation, that was fantastic. We are really on the very forefront of a journey a lot of us are going to take together. I am so encouraged to see you all here today willing to have a conversation with us as we embark on this experiment really, the way we look at this, 23 and me is a new way to do what we hope will be an effective tool in research, the ability to really engage with consumers for the first time in a very large-scale and web-based way to conduct genetics research.
So I may skip the slides, as they say in that one commercial, lose the slide and talk. Last November, to enable individuals to get access to genomes for the first time in a very broad way, we see ourselves as a company providing an interface to that genome, and what we would like to do is tell people this is really the early stages. We have been doing genome studies for a very brief time, looking at history of research, discovery of the DNA model, and Francis and others work, to give people access to it, give them dynamic tools and ways to interact with that, information. Move and change the field of how we conduct research.
What we have done is, we have a system where you can order our service through sending in a saliva sample, we extract DNA, work with a facility, generate a set of 600,000 data points for you. There's no way to get this up on the screen? Everybody has copies of the slides, I am on number three, talking about enabling consumer access to research. As pointed out, I would love the slide with all the chromosomes and papers. It's been a year in -- coming from -- seeing the wonderful tools companies like them and Illumina created, we are seeing the fruits of the labors of the researchers, doing the study and publishing results. We crated an interface, through the gene journal, scientists go through comb the literature, go through crit ya, how do we present the information to consumers with very little understanding. We have established research versus preliminary research. We did this because we first started out with 14 of what then was the ER category, left it at that, wanted to stay with these studies we felt would pass the test of time. Right after we launched, everybody, including scientific board members, really wanted more information. The minute you get access, the next thing you want to know is what more can I learn about myself. We took that into being account, caveats around that. More studies coming out. We don't pick the people. The whole thing about learning from your mistakes, I don't know who did the study, it did get publish, maybe in Science, what's going on in the research community, we give that to consumers, don't judge based on the phenotype or category, we want to be really honest, open about the studies.
The star system we employ now with our gene journal gives people the idea that if it only has one star, take it with multiple grains of salt, at one point we were going to have salt shakers, but people she see more of something they think better, so -- we are taking this information to consumers, putting it in the right light, this is new, a lot more to learn.
When you look at one of our gene journals, we have three genes associated with something, we leave the line open, there could be a lot more to fill in, be a reflection of that, continue to fill in the gaps, get a broader picture, better understanding and then be able to incorporate that in eventually, hopefully to our overall care.
We think in discussions, especially like these, we need other definitions of things. Would this be considered a genetic test, per say, when you are getting your whole genome, do you call that a test or just information? If we center 580,000 points and only know about 40,000, we think of them as information about you, now you have access through your computer, a different way of looking at it.
Then the other idea is what is prognostic, what are the definitions, how can we work together as a community to come together with a better regulatory means of reviewing it.
So going back to the slide Terry showed, now that we have our gene journal filled with great studies, we try to keep it current, people go through the literature, we now have 78 of the gene journals, 25 are of the established research category, the bulk of them are considered the preliminary research. As new studies come out, in this preliminary category, look like they moved into the ER, or established research category, we make these changes, could explain why David saw a bit of a change in the risk profile. He brought up a good point. I point at us, and at the other people that are in this industry together, we are willing to do our job to work with other organization to create standards. There are things we can do, we think, assumptions we make, David noted, we make different assumptions when we come up with risk profiles, but we should come together as industry, settle on the same assumptions, it's a straightforward thing for us to do. We will work with the program medicine coalition, Abrams and his browm, take the charge as neutral body to organize the thoughts, desires of the company and come up with a set of standards. The plan, I hope, by the end of the year. You can stay tuned, and we are really excited about where that will go.
I had planned to go through one of our Gene Journals, but you can, don't have to pay any money if you come to 23andme, set up an account, the tools, data at your disposal, we have done this through the family, eight people, three generations, grandparents, three children. From the beginning we felt it was important to allow people to come together as families, friends, compare things about your genome. We have seen a remarkable amount of uptake in sharing, things like maternal -- assignment, Y chromosome markers, how much you compare to someone across the genome. It's been phenomenal to see that going on, how many people are really doing that.
You can look through the site we have on data security, something we take very seriously. We wouldn't have a successful company if we didn't stress the privacy of the information we are generating for customers, that's a very important aspect of what we are doing. So what I want to touch on is something we just introduced. Every month we have a new release of software, new tools, features coming out. Within the last month or two we released the ability to do surveys with our customers. Started with fairly straightforward, simple, left or right-handed. We took a survey used, validated by epidemiologist, it's been really fascinating to see when you take this, do you write with right or left hand, how do you sweep with a broom, open a jar? I found out I am right-handed, but with a moderate preference for left handedness, I would have had no idea. If you sweep with your right hand on the top of the broom or left hand on the top of the jar -- different answers on surveys. Trying to take these validated surveys, move to the web, looking out to research community, working with epidemiologist to develop these, see what results we get. Starting with eye-color, handedness, will help prove the model of can we do research in a web-based format like this. We don't know. It's a new experiment we want to embark on. So far we are happy with the number of people responding, partly because the surveys appeal to pretty much anyone. Most of us can write, sweep, although that one shocked me how many people don't know how to hold a broom. So it's been the next step we were excited about. We had a study funded by the Michael j. fox foundation, the program has been studying Parkinson's many years, developed their own tools for diagnosis. They are very interested in looking at how can we move these validated instruments online. Beyond that what can we do to develop tools using web 2.0, technologies that will enable people with movement disorders we can measure over the web. This is the kind of thing Michael J. Fox is really interested funding. We have had meetings with Stanford research, measuring WII game, understanding, measuring, monitoring disease. We are at the early stages, but quite excited about the study. Had since then a lot of people coming forward, submitting grants, and once we get a number of people in the database and they show they are interested in volunteering information about themselves we believe it could be an interesting mechanism for conducting that type of research.
Another important thing, you look at people online, they want immediate feedback. You fill out the survey at 23andme, you find out how you stack up to other people in the database. It's instant feedback people get a lot of satisfaction out of. It's a mechanism we will always have, are you more all right handed or --
To finish, really, what this goes back to is the interest in translation. How do we translate all of these wonderful studies going on with all of this great information, get that into the clinic? It does seem to be a bit of a gap we have in our current system of making this mechanism happen. Very interesting to come up with C TSA, funding mechanism, major universities to come together with a clinical side, say you have to work together or you won't get this funding at this level. How we can match that, meet that, work together, let consumers be more active participants, get data, access to information because they want it. Studies, maybe Kathy Hudson is here, some people are, asked, part of the research study, they think they should get the data. It's been pretty obvious that is going to be the case going forward. If that is the interest of consumers we want to be here to give them some sort of access, interface to that information. The overall goal is to improve healthcare and work together, work with you, and we are really excited about continuing this conversation, thanks, again.
APPLAUSE.
Thank you. Our next presenter was Kari Stefansson, but he's stuck in Iceland, not Switzerland. So Jeff -- will present for Dr. Stephan son.
Hello, I am shorer than Cary, have less hair. I found a decode, the goal to -- [indiscernible] also targeting novel drug pathways, which might be most important.
We set up in iso lands so we could focus on one population, collected about 140,000 Icelanders with informed consent, specifically for diseases we ask about. We all vs also have collected about 200,000 non-icelandic samples from Europe, Asia. When we make discoveries with Iceland with a large cohort we want to be able to rapidly replicate, determine in there's -- outside Iceland, so we have cohorts to do that.
Over the years we have been, started doing linkage studies using the DNA database, made discovery for TCF 702, and the 8 Q for prostate sans cancer. We have geno typed about 45,000 Icelanders with high-density systems to allow us to do a combination of what linkage-based studies or genetic association studies.
The whole goal of DeCodeMe is to study -- common diseases, where the relative risk to the general population might be high enough to have an impact on prevention, early does he detection, at least in some niches. Why we launched the disease risk test for individual diseases like myocardial infarction, Type II diabetes, glaucoma, prostate cancer and -- the risk ratios were high enough to have an impact in certain circumstances. Using only markers validated, replicated across, in sick the 60 different populations around the world. In some cases only calkation -- Caucasians and in others crosses ethnic lines. Some use this information in practice to stratify in -- the Type II diabetes, we include that in a complement of four Type II diabetes chains, about ready to upgrade to eight. Shown in a collected sample cohort, the DPP study of pre-diabetes, to double one's risk. The absolute riff being of overload, pre-diabetes -- converting is 50 to 70% within three to four years, based on the DPP study and the -- study done in Europe as well.
Here's a special niche for a diabetes variant widely replicated in 60 cohorts, but has a certain potential clinical utility of identifying patients with pre-diabetes, at high risk for converting. The risk is 30 to 35% baseline, and -- [indiscernible] obesity, and overweight, certain ADAs recently identified pre-diabetics, and encouraging those patients to lose weight, those who have additional risk factors, they might benefit from pharmacologic management, with Actowes.
Personal genomics, a direct access to consumers with or without their physicians, we saw a way in which, like the other companies, to make that accessible, not to say patients work with this information, or individuals work in a vacuum. We encourage them to talk with genetic counselors, and offer them, have for the last six months of our service, for free. We encourage them more specifically to work with their physician. They have a bigger role to play when it comes to prevention or detection of cancer or other diseases. We have been able to pick and choose, add some of these diagnostic risk tools, putting together large systems of information to do that. For DeCodeMe we offer about 30 diseases now. Since there was a large discussion yesterday at the HHS meeting any analytical validation, I want to convince you for genetics it's simpler than for CRP or LDL cholesterol elements, you can document the accuracy of your phenotyping, whether SNP or -- you have 15 -- sampling, beads accessing the genotypes, evidence you can make use of in quality control. We think we are complients with CLIA, and the accuracy can be measured, documented, through repeated testing, marching the gold standard of -- sequencing, which we do for all the vearnlt variants we annotate, and do -- testing, which is also a requirement of CLIA, we do clinical validation, document those validations we define the way FDA defines, that is replicating the markers, and dem sthraiting those markers are consistent across populations. That's not a formal requirement by CLIA, surprisingly, but something we voluntarily add. We send clinical validation reports to CMS as part of our demonstration that these results are reliable. When it comes to the genetic markers, we [indiscernible] the individual disease test, they are all well-validated. We don't have preliminary data or diseases where there might be one or two studies, they need to be replicated widely before we put them in to risk classification. Then, I should emphasize, the relative risk is such these markers replicate, the thing also, clinical validation, how do you access the appropriate relative risk to -- phenotype, based on a few hundred patients, thousands? Most of these are based upon data sets we use for, that include up to 10,000 patient and controls.
So in some cases it may be 5,000 and 30 controls, others 12,000 patient and 30,000 controls. The assessment of relative risk is based on data sets larger, used for FDA approval, diagnostic tests used for approval by the FDA.
Then the question becomes what can you do with the information, use that reliably in a consistent manner? We convert each variation, odds ratios into a relative risk, compared to general population to have a consistent reference population to attach to the risk. We and others demonstrated for the vast majority of these diseases from the various studies that Terry mentioned, you see no interaction whatever. Even with the large large data sets, combined with others, we fail to find significant interaction terms, suggesting they are redundant or synergistic interactions and fits a multiplicative model, we are justify in applying these risks together in a multiplicative model way, useful to assess the risk for a particular individual. Can comes to clinical validation, Dr. Cory questioned that yesterday. Take prostate cancer, these markers have been replicated in large cohorts, even in initial discovery papers, 500 patients versus 1400 controls, cohorts in United States and Europe.
17 Q, another 45,000 controls, so on. Chromosome X replication, discovery, the patient population of 10,000 patients, 28,000 controls. I put this up as an example. We are in a different era now than we talked about yesterday. You can get published in Nature, the journal, you have to have a wide -- encouraged by Dr. Coulins and others. This shows how with the eight markers discovered for -- should mention other groups have replicated as well, yndz beyond our studies -- you can multiply these individual geno type specifics together, like done in the fusion study in genome-wide study on diabetes with colleagues, you can multiply, come up with a composite genetic risk for the individual that maps across the general population. The eight markers across the Caucasian population. About 10% has an average risk of two-folds compared to the generation population. 16%, life-time of 32%. Other risk factors? Nothing but family history. Markers shown by us and others for the common variants are independent of family history. Less than 5% of the population has a risk for prostate cancer, but if you have this, another 10%, doubling of risk, independent of their family history.
We try to comiewrn communicate results, patients and physician can understand better than table odds ratios, they don't need to know what a SNP is, much about genetics, these are risk fact offers, not determinant risk factors, much more analogous to LDL cholesterol. If you can give a reliable risk score they can use that along with other risk factors.
When it comes to consist encey among different factors, David's talk, the risks were consistent across the companies. The geno typic accuracy seems to be high, but when it comes to accessing the combining of markers, we do it a little different way. And we are getting together with PMC to get -- with academia, to do it in a consistent way. It's actually fairly consistent, but the variation is in where we use [indiscernible], different for marker to marker initially reported. Although those are still well-validated as surrogate markers. There's a way for additional consistency, but I think already the results are quite consistent across the different platforms, depends on whether they updated the latest prostate cancer Juniors genes in their profile, as fast as some of the others.
David Duncan did a great job demonstrating the comparison. Are the tests useful today? That's the debate. 10% or less, few percentage points, that's debatable. But we agree the vast majority of genetic information is not captured by this. Is it still useful to identify those at highest risk or not with these tests? We maintain it appears to have an effect that's independent of your Framingham score. You talk about LDL cholesterol by itself, very little impact, 50% or less with just LDL cholesterol, you comimine combine with other risk factors, the opportunity of moving a low-risk patient based on ATB 3 intermediate or to high risk. A recent perspective study in the U.K., 15% of patients, Steven Humphries, shows in the 15-year cohort, adding these genetic markers, 9 P, reclassified 15% of the patients that were originally classified with just Framingham, 5% of those were in the intermediate category, to high-risk category. It can have utility. What would you do differently? You would target the cholesterol to a different level based on that risk. I gave you my example of prostate cancer yesterday. For A trial fibrillation, very strong, 25% of us in this room are homo swriegeous for -- it turns out these are the strongest-acting genes, why? Only --% of stroke is due to -- fibrillation, more contributes to scroak, the -- stroke of unknown cause or large vessel -- not realized until we did these studies. Now doing a very large observational study to -- estimate 150,000 stroke and -- patients are misdiagnosed as having a stroke of different cause when it's really ATRIal deFibrillation -- if you did cardiac monitoring for a few extra weeks, you could, based on prevention, correct drug, moving from anti-platelet to -- can save the healthcare system a bill yorch yon billion dollars a year.
The case study yesterday, me, 48 years old, I have no business testing -- my father had prostate cancer -- I was compulsive enough to get it at age 42, low/normal. Got the test results back, using eight markers I showed you, my risk was 1.88, 30% based on this. Extra markers suggesting [indiscernible] the non-aggressive form of prostate conferencier by 1.3 fold, not dramatic. That prompted my primary care physician to get another PSA test. This time still in the normal range, zero to, but my -- was high. That high genetic risk referred to you -- can we improve the specificity of the biomarkers we all agree are not perfect, improve that specificity, sensitivity by adding extra information, used to guide these types of managements.
Thank you very much.
APPLAUSE.
Thank you. Our next speaker is George Church, Dr. Church is founder of Knome, and -- is a professor of genetics and director of center for mutation aal genetics at Harvard medical school. At the fore fronts of DNA sequencing technologies, including the first sequencing method in collaboration with Walter Gilbert in 1984. His research continues to foster high throughput technologies -- welcome, Dr. church.
Thank you. I want to thank all of these government agencies, as well as companies that we work with very closely, and also to disclose possible influences we have had. I guess from the discussion so far I wanted to say when we say personal empowerment requires prior validation was a conversation that came up. One of my take-homes is a lot of what we are doing in the personal genome project, at Knome, in some extent my advisory role at 23andme is empowering people to do research, rather than their medicine. Important in the sense there's a very strong attitude among many people, not everybody, where we want to learn about the world at some risk to ourselves. We will explore the planet at risk I areas, look at investments, will look at the Internet, all risky environments, that effect your quality of life. And the decisions that are not necessarily any more, or are less complicated than genetics.
This is the other end of the spectrum from the big four, five we are talking about here today, where we have various ways people are doing their own genetics. Bhean Many people know about the -- project, Mape mainly an testily. But -- was trying to understand his daughter's illness, in his home basically. But there are people hiding from their personal genomics, there is no cure, becoming activists, saying we can do something for people --
[indiscernible] my family has diabetes, about lipids, biochemistry, nick note became a representative of -- and his -- we heard about Michael Fox, so forth. :
So in this context, some of the people we heard about today are saying privacy is not the top concern. Etch when Even when it is, there are many ways privacy is compromised when you put things anywhere other than a vault. You can have laptop where 26 veterans information got out. A case where a 15 year old person wanted to know his anonymous sperm donor father, took a cheek swab, narrowed to an individual, found, confirmed was his father. There are many ways data gets out, it's unrealistic to overpromise. We certainly want to try to make it as private as possible.
When we talk about the research landscape here, we have standard research on the far right here where we have open access, as long as there's no trait data, such as the genome project -- various types of brooch approaches that increasingly -- the Reveal study, my close colleague, and so forth. Typically the data are kept de-identified or safe from the individuals that donated it. At the other end you have it only available to the individuals with marginal -- public domain, exploring in the middle where we can make it both publicly available, connecting done DNA and traits, and yet not over-promise on privacy, particularly recruiting people who have passed an exam with 100% on the questions at item four. One of the goals of the project, had -- approval since 2005, to try to get ahead of the curve. Of course the curve caught up with us at this point, but the idea is to bring technology to bring down the cost, not just because of the regulatory -- talking about R DNA regulations, closer to 90% in the rarer diseases that populate inheritance. We want full participation, not unusual in this context, and we have multiple samples, to make sure we have the identity, open access, trait questionnaire, stem cell -- I will talk about, and 100,000 individuals.
So we are focusing our sequencing, and I think Brian Bryan feel an will talk about -- the chip-based analysis, when you really want to go into detail on a test -- you talk about sequencing, not chips, that's not ins fore ever, but what is practiced now, you actually will see causative alleles, which change the reading -- for tumor suppress or -- have to be very carefully interpreted at the DNA sequence level because the consequences are very serious in a preventive sense where people will do a bilateral mastectomy if they trust the data.
Not touch odd yeted on yet, this sequencing, possible, factor of 1000 drop in price, from the -- slide. There are at least two classes of chemistry, and we are trying to produce a platform to support both of these -- [indiscernible] work on the -- version in 1999, and Illumina and -- biosystems uses this, same thing could be done for -- where you use flor easessently colored -- now things in -- [indiscernible] complete genomics.
Now, in addition to those commercial instruments, we have an instrument which is at the -- in between academic and commercial, we call a pollinator, intended to be an unusual model, completely open-source, opening to community, empowered to change any part, intended to be easily modular. Maybe only 5% of the research community will want to, but that one 5% will greatly aid the other 95%.
This is $15 5,000, less than our previous -- which was $600,000, similar achievement, lowering in price in reagent used and --
What does that technology result in? Plummeting costs, faster than the already rapid More's law of computing, a 24-month improvement in service for given price point for computers. This is more like six to 12 month, going from fairly low estimate cost of $100,000 per million based pairs to a logarithmic plot, getting down to $1000 very very conge. soon. Slightly different points, pathway. You can see how it plays out in the consumer market, the geno graphic -- a very hi-priced tag per bit of information, people are still very curious about ancestry, willing to pay a lot, $99 for 12 bits of information. DNA Direct has quality, actionable information, mostly with DNA technology, sequencing technology historically excessive, but plummeting. Personal genome project, not just genomics, but coding, replicating -- roughly similar price. And Knome is the only company that really offers full genome sequencing, currently $350,000, likely to go down on the same curve very soon. As David mentioned earlier, there's environmental components, very important, we want to be able to offer this to the same -- when you say personal genomics you should think about the regulatory elements which might be less expensive and more interpretable if analyzed at the RNA level, and some of the environmental components can be measured, either by measuring the microbiological component, microbes, vier uses, or their impact on immune system, rather than being a spike of microorganism, will be a longer term persistence.
Don't just go -- go through this regulatory, environmental filter. So in order to get at some of the regulatory, RNA regulatory interactions with the environment, in the personnel genome project we have included multiple [indiscernible] adults with -- healthy or with disease. We -- different tissues from the PGT volunteers, very gun-ho, draw the line at 1000 biopsies. You take one from the skin from which we have established stem cell lines, making these available to the community from which you can reprogram to almost any tissue you want. This is, of course, a very fast-moving target as well.
So we want to be able do do biology on these individuals, as well as inherited germ line genomics, the extreme of that is looking at the resistance or microbiological components of vier us and bacteria, but selected parts, different assays for inherited genome that go beyond SNPs, might want to go beyond SNPs for microorganisms. Here we study resistance level to 18 major classes of antibiotics over 140-some days, and personal genome project volunteers, and found you -- big solid blue means each along the X axis is -- surprising, ought outlier -- the background information you can do by highly targeted --
Could you wrap it up in the --
This is the last slide. So the questions I wanted to -- the questions given, how do we fund, is there a role for direct to consumer companies? Not just how do we celebrate the -- scare the -- reinforce the oldest, what about do-it-yourself genetics, completely outside what we are talking about, the riff risk of gene information, not educating, I don't think anybody would seriously consider that. What kind of model do we have, is insurance an interesting model. Healthy people like David still have a finite risk. Thank you.
Applause.
Thank you, Dr. Church. And I will turn it back to Steve, and Rick for a little -- Dr. Feel an Phelan will talk after the secretary does his thing.
As all of you can see, we are extremely fortunate today to be joined by secretary Leavitt. As many of you know, the Secretary has shown enormous -- personalized healthcare, working very diligently for a long time. We are really grateful that the reports, efforts have reached your office and you are giving the scrutiny we hoped you would and really appreciate your leadership. For those of you not aware the secretary was formerly the governor of Utah, born and bred in Utah, it's a privilege to welcome you to the group and look forward to having your thoughts.
Thank you very much for your leadership of this group. I had a chance to meet with Reed V. Tuckson a couple of months ago at the interchange, and I didn't have a chance to publicly thank him, but I want you to know I appreciated his service, expressed it to him, and I thank you for indulging my unscheduled visit. I thank you for your service, this is an area of interest to me. We are struggling with so many different issues where there is a need for balance, finding that place between fostering innovation and at the same time giving people a sense of confidence they need. It is so seldom that we are ever close to the curve or ahead of the curve, and I think many of the issues you are dealing with are on the leading edge of the curve, and there's a good possibility we will be able to stay somewhat ahead in terms of policy decisions. I was very pleased and appreciative of all the work that within the into the passage of Gina, and I want to acknowledge my great admiration for Francis Collins, as I am sure you mostly know, will be leaving the panel, but also HHS, and well, he's going to go on, do great things, and we will all have a chance to see him, be involved, but Francis, I want to tell you how much I value our relationship, I have said privately and want to say publicly, thanks for all your contributions.
I mentioned in the same spirit of staying ahead of the curve on these policy issues, today I happened on my desk landed an FDA release. U.S. food and drug administration approved a novel genetic test for determining whether patients with breast cancer are good candidates for treatment with the drug Herseptember in. We will see a lot of this. If we don't have the capacity to stay ahead of the policy issues related to it, there's so many still to be dealt with. I have spent a lot of my time as secretary dealing with what I think is going to be a convergence of information, available electronically, that we can use to provide patients with unparalleled information about the quality and cost of their healthcare. I think that will lead us into an era where we will be able to make judgments on the effectiveness of drugs, and the effectiveness of different procedures, and the ability then to be able -- I was in Singapore a couple weeks ago, and saw a collaboration going on there with NIH. We were looking at the various generations of being able to sequence, lay out an individual genomic history or profile. It called to my mind how rapidly this is changing. We talked about the speed with which that could be accomplished over the course of just a few years. It's evident to me that at some point this will be very common, and if we don't have the ability to manage this information in a standardized way, the capacity to protect the privacy of those, that there will be a great opportunity lost. I think we are ahead of this curve. So I want to thank all of you being willing to struggle with these significant dilemmas. It's a great value to me as Secretary and will be to future Secretaries. Steve and I didn't want to disrupt the meeting too much. If there are subjects your colleagues would like to inform me on directly, this would be a very good time to do that, or if you have a suggestion or comment, I would love to interact with your group.
Please do.
We have a standing-room crowd, there's interest, please take advantage of this --
There's a fight --
So far, so good.
I have a comment -- from University of Miami. As we are, particularly today's meeting, there has been a lot of advancing genetic and genetic information made directly by the private sector, to consumers, and special academic health centers, traditional medicine has been hesitant. The two sides look at each other with some degree of suspicion. Do you have an idea of how to move ahead with one approach --
I won't resolve that today. [Laughter] but I do think that's the type of question I was talking about. How do you give people the sense of security they have at the same time allowing innovation to go forward. If people are given information, I believe, if it's high-quality, consistent, they will use it in a way to drive their own interest and in doing so will drive the quality up and the cost ut ultimately down. The capacity for consumers to have access will be of importance. I don't want to weigh on one side, to say it's a positive struggle, in the tension between those both sides of the debate will ultimately be improved. So I don't have anything beyond that to offer, but except to say the struggle is good. It's a very positive, will create a positive outcome because both sides have legitimate interests and points of view. It's often the case that consumers are underestimated in terms of their capacity to sort through these things, and I think it's often, there's often a sense of well-intended protection that we want to provide that sometimes can constrain progress. So somebody needs to be pushing the envelope a little, at the same time, someone needs to have the brake on enough to keep us on the road.
Secretary, can I ask a broad overarching question, first thank you for your kind comments a moment ago. It's been a privilege to be a part of this department, to serve under your leadership, to have someone as Secretary with such an interest and knowledge about personalized medicine. Truly gratifying, and I speak for everyone on the committee --
I had great teachers.
One of the reports the committee put out was about genetic tests, and we are excited about prevention, wondering how it will get implemented. At the present time the medical care system seems to be more devoted to reimbursement than the actual prevention, the economic model, we have to change if we are going to reign in what is otherwise a scary curve of what proportion of GDP will go to healthcare. Can you project the opportunities here for taking the science, which is putting us in a position to do a better job of prevention, and actually putting that into an economic framework that will enable implementation across the board, not just for people with a lot of personal resources to invest in this with their own pocketbooks, but as a more general public health strategy?
I will say it has to happen. And it's a function, really, of I think when the science, people are confident enough in the science, and we have an economic model that will demonstrate the capacity for this to provide long-term savings. I am aware of this issue, get lots of mail about it. And I suspect some from people in this room. I know CMS is still wrestling with this. I had a briefing, conversation not long ago where they laid out what they saw as the competing considerations. But it must happen, and ultimately will. It's a function of when are we confident enough in the science and when is the economic model warntd. One thing that gets into this discussion is the way the federal government scores its budget We don't score prevention very favorably. We don't give credit in the development of budgets in the scoring model for good thoughtful preventive fiduciaries. It's not likely to happen in the next 19 seven-days.
I think IT will happen.
Mr. Secretary, thank you for your work as well. I would is ask how the committee can be useful to you in your last months of Tennessee your, tenure, how you think we might be helpful to the incoming --
The oversight of genetic testing, to have a place to -- give me the best solution. You can expect the work of this panel will just increase. There's endless number of thorny difficult policy issues I believe I and future secretaries will be addressing to this group for solutions. So I would just say what you can do is show up, give your best advice because there's going to be a lot of it required.
Bap
William Cory from the CDC, public health genomics, thank you for your leadership the last few years, following up on the oversight report, the committee really worked hard over the last few months, give you that report, lots of good information, and it's befitting you show up in a session we are talking about the personal genomic services, because the field is moving so quickly, and we need that balance between innovation and at the same time high-quality information to the consumers, the need for oversight, the federal government to do just the right thing, not too much of one thing versus the other. I was wondering whether or not there's going to be movement on that report during the next few months, or would that's what be --
The place you will see money decisions being made -- little question in my mind, as I suggested, we have to ultimately deploy a model where we can use these tools in prevention. The issues really will boil down to when does the science reach the point that we can reliably depend on it, and when can we get government policies aligned with it. I don't want to bias or improperly inform about it, won't say more, except I do want you to know the report is both useful and being used and moving us toward the point where those cross in making that determination.
Just thank you, Mr. Secretary for being with us today. Quick question, somewhat logistical, practical, more and more important to this committee as you do what I say, try to address the multiplicity of issues. Logistically in the past our responses have tended to be focused on the larger reports, which I am sure are very helpful to you, if you have insomnia. It might be also useful if we focus perhaps more on brief responses, directed responses, things that might have to be done in between our semiannual meetings. The question we have been wrestling with a little today and yesterday, there other ways we could get information to you in a timely manner, perhaps a more focused way, would that be helpful, and to say Rick and Greg have been wonderful in giving direction from your office, helpful advice on how to best serve the Department of DHHS S. Is that something we should be looking at more or --
Secretary: Let me be truthful about this, the executive summary of the reports are the most important thing to me. [Laughter].
I don't weigh through the depth of things I don't understand, but I find the summaries very useful. Where I turn is to Rick and Greg, who are very much involved in helping me have the information I need. In some cases defining the questions I need to ask. Let's pose that to them.
[Laughter.]
Well, I -- the question was do we, the longer versions of these things, more targeted things? Right now I would say that for our tenure here we have a lot to chew on in the report you have given us, and are chewing on that. But I think it is a good question, looking forward, and we certainly have the relationship developed we have where we could come to you directly with particular questions, than we have in the past, targeted more specifically. I think, from my perspective, Greg I don't know if you want to say more about that, but that we really would avail ourselves of both methods if it's convenient for the committee.
I would like to hear Greg's response as he comes to a microphone, I have found the larger reports often anticipate questions that haven't yet been posed. Or at least they are not mature or ripe yet in terms of policy. Generally, I will end up with a policy question that frames up an issue, and it's very helpful to have one of my colleagues to say we had a very thoughtful group look at that. The SACGHS has examined it and here's what they said. They may then extract some thoughtful piece from that report. I get it in a two-pager. They write it at a level I can understand it. So I don't think you should assume that because it's big, and thick that at some point in time the essence of matters related to direct policy don't find a way to the Secretary. They do.
Now that you revealed all of the secrets of how it works, the one part that's not told is the community that develops within the organization across the agency, this forum, leadership, and your vision, I think has created, at least in my 15 years of public service, a unique tune opportunity to share science, technology, but also human elements, and the important challenges. It's still a little fuzzy sometimes as to how the decision processes ultimately work, but what tried under this initiative is new ways in which the confluences, challenges of agency, many aspects, I didn't even know of, new where ways of collectively getting to -- two or three years ago, we are all about that. The visit to -- how to bring the community to a new way of understanding. Won't be the vision of everyone, but the journey is an important part of the story. I think we have valued, Sheila is here too, the forum is an important way to ask critical questions, help us keep in front of everybody everything that's coming. In your role of Secretary it's important to convey that to those who follow it will be an important asset for the change of science, technology, and aspects of human health.
Your referencing the mountain, when I was governor of Utah, Steven, there was, out behind the state capitol there's a peak called InZen peak. You know well, then, that it's a favorite place for people to hike. At lunch, often for my exercise I would leave the state capitol, hike to the top of this peak. It was about a 45 minute-forced march, and your heart was pumping pretty good by the time you got to the top. It wasn't just the exercise I enjoyed. It was being able to be up high enough you could see the entire expanse of the valley in Salt Lake City area. You could see freeways, schools, universities, hospitals, houses, playgrounds, you could see the way a society, all the forces of a society have to come together to make a community. It was a great place to think, because you could see so much of the way things interact. I would say that has been the significant privilege of being governor of a state, or being Secretary of health. It has allowed me to see enough of the confluence of different things, that a picture of how this is going to unfold begins to form in one's mind. I can see this, I know you can too. It's a much different future than we now live in, but it's happening one dot at a time. The key is figuring out how to connect them all. We spent a lot of time with electronic medical records. Why? Because it's the electronic medical record that allow us to have the quality of information and ability to provide the personalization that will ultimately use this tool.
So moving all of these along in parallel, it isn't a neat, clean, process. It doesn't happen in an order orderly way. Having the capacity as you come to a point to connect two or three of these dots, having a group to be able to have a, not just a place to toss questions, but a body of thinking that can inform various parts as it comes together is a very important thing. My message is don't think you're just responding to a question from the Secretary. What you are doing is creating a body of information and thinking that will connect a lot of different parts of this system as we move into the future. I am disrupting your meeting longer than I intended to.
Thank you very much for coming, Mr. Secretary, it's a privilege, important, and you should know the staff supports you and -- [indiscernible] APPLAUSE.
Captioner transition.
Thank you for allowing me to be here today. I started this company, DNA Direct, in 2004. I spoke with many of you in the year before I actually developed it. Trying to think through everything from the ethical issues involved to the clinical and scientific issues to the challenges of doing internet commercial. It's a pleasure to see over this period of time how much is changed in our industry. And really, I believe, how progressive society is. The mission of DNA Direct is clearly to bring the power of perch used medicine. As mentioned in my introduction I have been involved in healthcare information for 30 years. It is an exsentence of a belief and commitment that I have.
We launched in 2005, which working with a distinguished and important group of scientific advisors that looked at how we could innovate without innovating around the science, but innovating around the delivery medium. And using standard clinical protocols that was enabling them. That meant medical oversight. It meant clear guidelines for our genetic counselors. And follow that same procedure from beginning to end. DNA Direct now works with healthcare providers as an exsentence -- extension. DNA Direct is never the lab in any of this. We work with all CLEA labs depending on the tests.
Our medical genetic testing is prenatal to carrier to drug response testing. We look for, in everything we do, whether or not this is a test that will be clinically actionable. We spend yesterday defining what that really means and how these terms can be used and how they can be confusing. I think we all know whether or not something has medical and healthcare significance. I will walk you through a few slides about how our testing works. When you go through a testing area there's significant pretest information to help consumers understand the pros and cons of testing. DNA Direct is clearly committed to making sure when people know when a test will be helpful to them and not. They determine where testing might be promote. It helps us, the information we glean from that regarding personal or family medical history is used to build a personalized web enabled report with their test result.
All of our customers, their medical charts are reviewed along with the test result by a medical geneticist. A unique identififier is sent to CLEA certified lab. DNA Direct is not marking up a lab fee, we pass that cost on to the customers. This is just an example of a report for, in this case BRCA result that is positive. The test result is custom used based on family history. All of our BRCA customers go through an indepth phone counsel. We want to identify that the right person is getting the right test. And phone counseling will or will not work for this patient. One of the things we talk is whether or not they would prefer to see an in-person genetic counselor. We have a network of over 60 that are affiliated with us. We can refer somebody if we think that a phone consult is not as appropriate. All of our reports include a physician letter that is two or three pages. What they want is really specific, easy to read, fast information about the clinical guidelines and the significance of the tests for that patient. All of our data is secured from the get go of the company. We look at the protocols to make sure we set up a system where our genetic data regarding a patient is kept separate from any e-commerce transactions.
Just a few minutes about why consumers use us. First of all, they often come because they've had a problem obtaining a genetic test for some reason or another. Sometimes it's just geography. There's a derth of expertise out there. As consumer awareness increases there will continue to be this bottle neck. Some come to us because the physician doesn't want to take significance time to do the consulting or counseling around family history and everything else. Some purely want pure anonymity. I suspect some of that will change with Gina.
I thought I would take a second to explain our experience with customers. First of all, 46% of our customers have a personal family history for the condition they're testing. 18% have a personal diagnosis with the condition. 21% have a known family mutation. And a combined 53% have both. 34% of our customers across the board since we started test positive for mutation. You recognize in a traditional genetics clinic positive rates are much closer to 5 to 10% in the general public. That means that DNA Direct is testing appropriate people for clinically valid tests.
I would like to end on this one slide, which is something that I showed yesterday. Unfortunately, this acrobat is not reading this correctly. Those little squares were meant to be check boxes. I'm trying to help the committee think through this whole field of testing. Not all genetic tests are equal. I started the bottom with serious diagnostic testing like huntington's. Most of us would agree that's best tested in a bricks and mortar facility. As you go up this ladder you have to think broader. Up the ladder I say predictive testing for serious health disorders. They have clear clinical indications. At DNA Direct we believe that this can be done with a genetic consult in phone. We're under correct with [ Indiscernible ] where we do for part of our research with black African women all of their consults on phone as part of their research trial. The academic centers are already acknowledges that phone consult can work well. Web enabled education will be a wave that is moving and integrating into healthcare.
As we go up this ladder we look at genetic screening for common things. That probably doesn't need to be done in a physician's office. We might want to say that a healthcare insurance plan could look at how to target particularly select populations in making sure they could get a genetic test easily facilitied as appropriate. As we continue to go up this ladder to full genome sequencing that's where we have to think out of the box. We don't provide those. We do provide support services for anyone who had information gleaned from one of these tests. What I think is important here as we look at this area is we have to request whether or not there should be a different guideline for people who are engaging in genome wide arrays for different purposes than clinical genetics. This is something [ Speaker/Audio Faint or Unclear ] I just came back last week from the United Kingdom's genetic's commission. They, too, are wrestling with. They're calling for a voluntary code of practice regarding the delivery of genetic information services. One of them was clear to say that -- this is a re