 Good morning, everyone. Welcome to today's lecture, the 12th in the current topic series. It's my great pleasure to introduce to you Dr. Bruce Korf from the University of Alabama at Birmingham. He's the Wayne H. and Sarah Cruz Finley Chair of Medical Genetics, as well as the Professor and Chair in the Department of Genetics, and the Director of the Heflin Center for Genomic Sciences. Bruce is an internationally recognized leader in human genetics, and particularly in the study of the neurodevelopmental disorder neurofibromatosis. He's served on the Board of Directors of the American Society of Human Genetics, and is past president of both the Association of Professors of Human and Medical Genetics, and the American College of Medical Genetics and Genomics. He's currently president of the ACMG Foundation for Genetic and Genomic Medicine. Bruce is very committed to education and mentoring. The fourth edition of his textbook for medical students, Human Genetics and Genomics, was just published last year. He's also the co-author of Medical Genetics at a Glance, and is on the editorial board of current protocols in human genetics. Finally, Bruce has been a long-standing friend of the Genome Institute, having served on our Board of Scientific Counselors for many years. I'm very pleased that Bruce could join us today in sharing his perspectives on genomic medicine. Please join me in welcoming him to the NIH campus. Well, thank you. It's a great pleasure to be back here. And what I'll try to do in this next hour or so is to sort of paint a picture of, at least from my perspective, where things may be going in the world of genomic medicine. I guess I will show you my financial relationship disclosures. But now let me ask you for this next few minutes to engage in a little suspension of disbelief. I'm going to use a kind of approach to painting this picture that is, in some ways, kind of real world and in some ways, very not real world. We're going to focus on this individual, Laura, who as you see has two siblings, and follow her through her life, actually from birth through advanced age and speculate, in some cases, with a fair amount of experience because things are happening right now and maybe to some extent where things are going, how genomics might be useful in providing guidance to her in terms of health issues. Now, the suspension of disbelief is going to take the form that as she gets older and older, we're still going to be looking at the way things are done more or less now or now plus perhaps a few years, but not now plus 60 years. I would not even venture to guess what medicine will look like 60 years from now. So you have to realize that as she ages, the snapshot is more or less a contemporary one, maybe today plus a year or two, but probably not much beyond that. You're going to see a lot of pictures in this. These are all pictures that are actually purchased from a company that you can buy pictures from. They're not real patients. It's a real pleasure to completely ignore HIPAA in this context because these are paid models. And excuse the fact that these are all of a particular ethnic background. I didn't think I could ask you to suspend disbelief, to have them change their ethnicity as they grew older. So this is not meant to be any kind of statement about access to genomics or genomic medicine. All right, so with that lengthy preface, we're going to consider these six sort of epics of life. Start in the newborn period. We'll take a look at, in this case, pediatric genomic diagnosis, preconceptional screening as a couple is preparing to start a family, prenatal diagnosis, pre-symptomatic testing for risk of disease, and then predispositional testing. All right, so we'll begin the story. Wara is a newborn and blood is taken from her heel, sent to the state newborn screening laboratory, and her parents are told when the blood is drawn, they see the Band-Aid maybe, and they're told it's a routine test not to worry about it. So they never hear again about it, and in this case, no news indeed is good news. So as you imagine are aware, everybody born in the U.S. throughout more or less the developed world will have a heel stick done in the early days of life, and these days, but still put on a little card, these days it tends to go most times in the U.S. to the state laboratory, and tandem-esque spectrometry is the most typical technology used for most of the tests without going into any technical details. The sample goes in one end, and a spectrum comes out the other, and it can be used to identify analytes that are present in the blood that signal any of a few dozen at this point inborn areas of metabolism. And the theory behind this, in the 1960s, and the paradigm was in phenylketonuria, PKU, that if you wait for the onset of signs or symptoms, because of, in that case, high phenylalanine in the blood, which damages the nervous system progressively, if you wait for that to begin to occur, that damage is irreversible. On the other hand, if you can make the diagnosis at birth, and for most of these metabolic disorders, the placenta clears the toxic substance that you can't handle because of an enzyme deficiency. So until you're born, there's no accumulation. But the clock starts when you have your first feed. These toxic substances, toxic to you, will build up and progressively damage nervous system and other tissues for that matter as well. Well, if you can figure that out before the process has gone to any extent, these are treatable, usually by manipulation of the diet, in the case of PKU, a special formula deficient in phenylalanine. And there are now a couple of dozen, three or so dozen metabolic disorders that can be diagnosed this way because they can be treated, usually with diet, occasionally with vitamin supplementation, increasingly now with targeted medications even. But the point is this is a huge public health triumph that dates back into the early 1960s and is now routine. And there's a more or less standardized list of conditions that are screened for. American College of Medical Genetics and Genomics actually took the lead in defining the kind of panel that would be used around the U.S. Well, the question now is on the table whether this would be better done through a genomic approach rather than the tandem-esque spectrometry kind of approach that I just described. And in fact, this was actually a paper published a couple of years ago. It really wasn't focused specifically on newborn screening. In fact, it really was asking whether it would be efficient to use a genomic sequencing approach to diagnose 450 or so metabolic and other neurodegenerative disorders. But the point that was made in this was a significant proportion of annotated so-called disease-causing variants that were not, in fact, disease-causing at all. And it raises the point that as genomics are increasingly used in diagnostics and if they were to be used in newborn screening, we need to cope with the fact that you're going to find a lot of things you're not quite sure what they mean. And there are some unfortunate, even egregious examples in literature where somebody published a case report 30 years ago attributing a particular mutation to a particular phenotype. It got forgotten. Now, years past, you find that mutation. You read about it and say, well, this is the phenotype, but it doesn't really fit with the patient. And then you go back and you find out that the author had figured out eventually this was just a benign variant that they hadn't realized was that and they reported it and it never got re-reported or corrected in the literature. And literature is littered with these misannotations, which is the subject of a lot of effort now to try to have a much more well-curated database, which is going to be increasingly critical as genomic approaches are used in diagnostics. But it is a caveat in terms of newborn screening. You're probably aware that NHGRI is a sponsor of a set of grants that were issued over this past year or so. I think there are four, if I recall, that are actually looking at the question, I think, together with NICHD of the utility of a genomic approach to newborn screening to define the sort of paradigm by which that might occur and among the issues are things such as how do you know that a particular variant is pathogenic and also how do you deal with a point that we'll come back to in a few minutes, which is the inevitable finding of so-called incidental findings that are not necessarily the target of what you're trying to diagnose in the newborn, but nevertheless may be important to the health of the individual as they get older. And that's a point, as I say, we'll return to. All right, so from newborn screening, let's now move into the world of diagnosis. So Laura is now three and she has a brother, Seth, a couple of years older and he's been experiencing developmental problems and ultimately is diagnosed as having autism or autism spectrum disorder. A lot of people have asked in the past few years when is genomics actually going to hit sort of real-time medicine? There's been, I think, this perception that there was a lot of hope but how much of it's really making a difference. Yes, I would argue that one of the first and most significant advances that has certainly changed the world of medical genetics is the area of cytogenomics, which is depicted in this diagram here to the left. So until a few years ago, the approach to looking at chromosomes, which geneticists will tell you, by the way, they've been genomicists for 50 years because when you look at the human karyotype, you are taking, I guess it's the ultimate bird's eye view of the human genome. I often tell patients, it's like taking a picture of the Earth from a satellite 25,000 miles up and you can see volcanoes and you can see other very significant disruptions maybe and that's about the level of resolution that we had from about 1959 until fairly recently that you could see trisomies like trisomy 21 associated with Down syndrome. You could see translocations, big deletions that were large enough to be visible with the light microscope and then in the 1980s into the 90s, fluorescence and Cy2 hybridization came along permitting light up a particular region of a chromosome but you had to know in advance where to look. You had to have a phenotype that correlated with a particular genetic change and if you knew that you could often verify it at the molecular level. However, you often couldn't make a diagnosis because you didn't know in advance where to look and the level of change was too subtle to be visible with the light microscope and you might have been able to use fish if you had known which probe to use but you often didn't. Well, the cytogenomic technology I won't go into detail about it from a technical perspective. It's done by various approaches comparative genomic hybridization taking a patient sample, a reference sample having different color fluorescence for each and allowing them to compete for hybridization on a chip that contains oligonucleotides from across the genome would be a commonly used approach and these are the kind of readouts you get and way down here maybe visible is a shift of this curve everything on this zero line which you can't really read probably is essentially normal. There's a little bit of noise to either side but way down here is a region which is shifted to the left and reflects a deletion that would simply never have been seen with the light microscope and might be verifiable by fish but there would have been no a priori reason to look there and therefore it would have been very dangerous and I can tell you in our lab and I think this is pretty typical in most if you look at individuals with undefined so to speak intellectual disability so intellectual problems without other clues to help you make a diagnosis the cytogenetic pickup rate that or autism spectrum disorder was maybe 5% and now the pickup rate with this cytogenomic as it's been described is pushing 15% to 20% it's not 100% and there's significant gaps in our ability but it has made a big difference and actually there's a debate now that unfortunately has mostly occurred in the world of insurance is to reimburse ability of tests like this I'll probably return to this point a little later in another context which is how much difference does it make how are you making somebody better by establishing that they have a tiny deletion at this part of the chromosome you can't treat them differently because maybe someday we will if we knew what particular genes were relevant to this region and maybe there would be some kind of intervention that would be effective but honestly we're a ways off from that I think in most instances the benefit is well partly to provide a kind of peace of mind to the family it avoids the need to do lots of other testing that you might have otherwise done to try to establish a diagnosis and provides a basis for genetic counseling and recurrence risk assessment so these are I think to a geneticist significant benefits the insurance industry I think is moving in this direction at least in our state I think the willingness to reimburse this is actually improving and I see that as a positive sign but it has really been a kind of uphill battle over these past several years so this is the kind of experience you hear about pretty commonly the so called diagnostic odyssey that is if you start here you have a clinical problem you establish a differential diagnosis as well as you can you might do genetic testing it gets interpreted no that's not the correct gene is often the outcome and you try again and you keep going in circles and each turn of the cycle could cost depending on the test even a couple of thousand dollars perhaps sometimes less sometimes more takes time sometimes measured in many months even a year or more and as time passes families get increasingly frustrated because they often are motivated to establish a diagnosis and find that's quite elusive and so this kind of idea first of all of cytogenomics and now more recently of exome sequencing has been sort of posited as an approach this is a screen shot of a paper published in the New England Journal just in the last few months I guess October of last year that reports a 25 percent rate of diagnosis of single gene basically changes that are believed to be clinically significant which presumably is on top of whatever might have been detected by cytogenomics so it really begins to sort of push up the rate of diagnosis it's not an inexpensive test on its own on the other hand it doesn't take too many turns of the cycle in the context of ordinary testing before you've spent more than the cost of exome sequence and the truth is actually doesn't take many MRI scans especially in a child who needs anesthesia to similarly rack up a bill that will exceed the cost of exome sequencing well one of the things that has occurred that you probably have heard some about in this recent year or two is the debate surrounding incidental findings the exome sequence does not target obviously a particular gene of interest it is looking pretty well everywhere, close to everywhere not every exon as I'm sure you know but nevertheless a pretty broad search of the genome and therefore has the potential of picking things up that are not the target in terms of the diagnosis you were seeking but nevertheless may be important for the health of an individual I pulled this picture because it depicts what I sort of conceptualized to be the scenario which is guy in this case the passerby was too late hit on the head with a flower pot as he's walking as a passerby would you feel an obligation if you saw the flower pot teetering on the ledge to warn the individual of this impending potential disaster and I think most of us would say that we would I would hope especially if I were the passerby but nevertheless the question is you know what is the obligation of the genomics laboratory or the ordering clinician to report significant incidental findings that occur in the context of genomic sequencing well the ACMG took this on about a year ago actually Les B. Secker from this institute was one of the lead authors and co-chair of the committee with Robert Green this is the publication that appeared in genetics and medicine it sort of you know be careful what you wish for I think the genetics community has always wanted to have some stories that would at least the medical genetics community that would show up on the front page of major newspapers this one did but it generated a lot of controversy these are the kind of distilled recommendations mutations on a minimum list should be reported by the laboratory regardless of patient age even for children in spite of perhaps a condition that might be adult onset that they should report variants on a list I'll show you in a minute the ordering clinician would be responsible for the pre and post test counseling that was actually these three bullets were all I showed until about two months ago or something like that at the ACMG meeting that occurred in March this additional point was added the board of directors voted that patients should be able to opt out of having analysis of incidental findings which was recognizing a pretty significant controversy that had evolved so here's the list of genes and by the way there's a committee of ACMG working now and I think probably on an ongoing basis realizing that this list is a moving target but anyway here's I won't read the list but I will show you the kind of concept here so there are some tumor predisposition syndromes which is the largest list the poster case are the two BRCA genes that are associated with hereditary breast and ovarian cancer some connective tissue disorders cardiomyopathies disorders associated with arrhythmia hypercholesterolemia and malignant hyperthermia so what it took to get on the list was first of all you had to have a condition that might not have been diagnosed clinically so if it was an obvious clinical phenotype it wasn't on the list because the assumption was you should have already been diagnosed so if you carry a BRCA mutation you don't look any different from anybody else you hope that you'll realize that you carry this mutation before you develop cancer which would be the kind of sign that this is a condition that you're at risk for it also had to be a condition where there was a well established natural where we had a pretty good idea of what the particular pathogenicity of variance was so this can't be reporting variance of unknown significance and there had to be something you could do if you were found to carry this mutation you had to have an action which again this is the easiest one to describe because it's pretty familiar there's a lot of opportunity to intervene to reduce the risk of cancer in a person who is a BRCA one or two carrier whether it's surgery surveillance chemo prevention there's very well established paradigms now so the thinking was that if you come from a small family maybe or you're adopted perhaps you don't have any knowledge of your family history you would only figure out that you were at risk of breast and ovarian cancer by virtue of carrying this gene the day you develop one of those cancers wouldn't it be better if you knew this long before the fact and here there is real intervention that could be offered and the thinking was if you happened on this result even if it was a child and ACMG does not recommend testing children for adult onset disorders but the thinking was well even if you're a child you might ever otherwise never know that this risk occurred until maybe your mother developed breast cancer at some point and because of that the ACMG did recommend that this gene or these genes be scrutinized even in a child because of the benefit it would have to the family and indirectly to the child to have a parent who survives long enough to provide care as the child gets older so that was the rationale ultimately there was a lot of hue and cry about undermining patient autonomy and a lot of debate about the point ultimately it was decided to offer people a possibility of opting out that's the recommendation the experience of labs anecdotally has been very very few people do opt out but there are some and so be it there's a small proportion who just don't want to know in spite of the medical action ability of these things these were not genes where you learn the name of the thing you're someday going to get and can't do anything about it anyway enough said about this but it's an example and I'm sure there'll be others of ethical issues that arise as we have new tools that really in many ways change the medical paradigm alright what about preconceptional screening so now Laura is married she and her husband are considering starting a family they meet with their OBGYN they're of northern European ancestry and among the things offered to them is carrier screening for cystic fibrosis this has been practiced over this last maybe what 15 years in the late 90's there was a consensus conference at the NIH recommended that all couples be offered a panel of tests for the common CF mutations that comprised the frequencies of 1% or greater of the CF mutations and this therefore it's couples to learn that they're both carriers therefore at risk and then incorporate that into their family planning otherwise couples only usually learn of their risk after their first child is born and sometimes they don't even figure that out until more than one child is born because it takes some while sometimes for the diagnosis to be made so ethnicity based screening has been practiced certainly in the US for a long time the most common are for individuals of Ashkenazi Jewish descent this is a partial list I think a commonly used list now is about 19 conditions that are due to founder effects in that population that lead to a relatively high frequency of being a carrier for these generally devastating conditions individuals at risk of various hemoglobinopathies and then pan-ethnic screening carried out for CF and now also for spinal muscular atrophy well what about doing this at a genomic level and there are companies now that are beginning to do just this well I quote the same paper I quoted a few minutes ago a few years back and show you a different slide pulled from their paper and this is for the 400 in some conditions they looked at so this is not looking across the whole genome how often did they find an individual to be a carrier for one of those there were a few individuals who had no mutation for any of these genes but indeed it was very few the majority I guess the kind of mode here was to have be a carrier for three different conditions you see some people were carriers for many more I think it's a fair statement if you look across the entire genome that we are all carriers for something that if our partner happens to be a carrier for the same thing would have implications for our offspring and we're finding more and more that couples are interested in a sort of pan-ethnic screen I think a lot of people are not clear on their ethnicity these days information may not be so obvious to them and hence going back a few generations may be difficult and the thinking is even for very rare conditions that it wouldn't be practical in a kind of enzyme based system or even a one gene at a time system to test for all of those but on the other hand now the incremental cost of adding another mutation to a screen is so close to zero that even if it's exceedingly rare disorder why should couples not have at least the information to use for their decision making if in fact they are at risk because obviously some are that's how those conditions happen does raise an interesting question if the carrier burden if you want to think of it that way is that high everybody is going to be found to be a carrier for something and probably want to have a conversation about it and it raises an issue in terms of manpower or person power to counseling that needs to accompany this and I'll return to that at the end what about prenatal testing so in fact they are CF carriers and they like to have a prenatal diagnosis it's done by amniocentesis and indeed the child is found to be a carrier but not to be affected so prenatal diagnostic approaches have been used now for many many years amniocentesis is the kind of initial paradigm corionic sampling this is shown trans-servically it can also be done trans-abdominally offers a somewhat earlier diagnosis many couples are opting for pre-implantation testing where you would do in vitro fertilization of multiple eggs carry them typically to about the eight cell stage biopsy a single blastomere and tested for the condition for which that couple is at risk and implant those embryos back into the mother not found not inexpensive approach but one that many couples find sort of attractive as an alternative well for a long time there has been a kind of selective screening in the early days it was on the basis of maternal age knowing that there is an increased risk of trisomy especially trisomy 21 associated with Down syndrome over the years it's evolved into a combination of ultrasound and various biochemical measures but what about the possibility of doing this using a genomic approach in this past few years that has really seen the light of day using cell-free DNA that is circulating in mother's circulation and a proportion of that is a fetal origin I think from trophoblast cells mainly and if one does sequencing of the DNA that is found in the mother's blood of course most of it's maternal but enough of it is fetal that you can quantify the proportion of chromosomes in this case 13, 18 and 21 these are the three associated with live born trisomy syndromes and can show that in each of these trisomies there is a substantial increased representation of sequences from the relevant chromosomes this chart pulled from a paper on this topic there's a lot of literature on it now demonstrations of clinical utility it's increasingly used there are a variety of companies that are offering it very possibly it will move to a sort of hospital based test at some point but increasingly this is a genomic test that is in routine use could you sequence the fetal genome from this cell-free DNA to as you can not a trivial process this particular exercise required essentially sequencing the parental genomes and the fetal genome and you can infer the fetal genome from a combination of the data that you get from all of that it was used as a kind of proof of principle of the idea of doing a genome sequence prenatally and raises all the incidental finding questions only become more complicated when you think of it from a prenatal diagnostic perspective realizing usually the target is some immediately medically significant phenotype not a phenotype that would be delayed by decades but at least in principle this can happen and it reaches a point where it does to any substantial degree it's going to raise some really interesting questions about how to handle the incidental findings actually participated in a perspective in the New England Journal a few months ago a couple of colleagues were bioethicists asked if I would work with them mainly to add a medical genetic perspective and so the concept here was not whether it would be a good thing for people to do sequencing of a prenatal genome but rather whether parents who want to do it should be permitted to do it and the argument here is essentially a kind of autonomy based argument that they should be parents typically make decisions on behalf of their either fetus or their child I expected another firestorm from this didn't seem to take place though either means it was ignored or people agreed with it alright so what about pre-symptomatic testing so now Laura is 45 she's learned that she has a sister who is few years older diagnosed now with breast cancer and that raises the concern about the risks that might apply to other members of the family this is not their family it actually is a kind of pedigree that we have seen in our own clinic the pro-band may not be easy to see here is this individual and she came to attention or came to clinic because her sister who is still alive has breast cancer and you see there is also a paternal aunt with breast cancer a cousin a grandparent and then there is an uncle with prostate cancer and another aunt who died of ovarian cancer this cancer on both sides of the family actually these two individuals cancer is not rare everybody has a family history of some kind of cancer but this was enough family history to raise concern about whether there was a genetic predisposition we actually organized to do testing of the sister first this is a paradigm that is not that uncommon in genetics but is a little bit unusual in medicine which is the person that comes to clinic isn't the person you recommend first be tested and the rationale here is that if this family history is accounted for by a mutation then the most likely person who would carry one is an affected individual so if we had tested our patient first and she had come back negative we wouldn't know if that negative test was because she didn't inherit a gene mutation that is in the family or if this family history was accounted for by some other risk factor besides a mutation in the various bracket genes by testing the sister improving she actually did carry a mutation we were able then to test our patient and she indeed turned out to be negative and that was very reassuring because there was no issue about whether either some other gene was involved or some undetectable mutation in either of these genes might be involved so we could tell her that this family history was not relevant to her anymore that we can account for it and she didn't inherit that risk factor and therefore it was quite a useful thing I alluded to this earlier but I think it bears emphasizing that this is an example of a medically actionable genetic diagnosis because you know in the early days which is in the what I guess mid 90s when these genes were discovered it was a huge sort of accomplishment and it offered insights into the pathophysiology of breast and ovarian cancer so it really was a landmark and very soon it became obvious you could offer testing many families were interested in doing it but there was a lot of debate about whether you were doing them a favor part of the concern was you could tell them that they were at risk you might not be quite sure exactly what the magnitude of the risk was because data sets were perhaps for one thing not that many and secondly they were biased probably because the families that had been studied were families that had a fairly high penetrance and the question is is that typical good data to show that the prophylactic approaches surgery, chemo prevention surveillance were effective because it wasn't really enough time to know that a cohort had been followed where you had done these things and could demonstrate benefit there was some sort of common sense benefit surgery a fairly extreme approach if the target tissue has been removed at least you believe mostly removed it should reduce the risk but that hadn't been proven and there was concern that there might be residual tissue this was a paper published a few years back showing the benefit of salpingo opherectomy in individuals who are carriers of mutations and you can see compared to surveillance as these pots go down that means more people get cancer so there was a substantial protective effect and I think this is really pretty standard in the field right now I don't think anybody would argue that doing this test is of academic interest only or that it just tells you something that you can't do anything about you may or may not choose to do all the things you could do but you at least have choices before you and I can tell you in our center we have a very substantial demand I think in most centers that offer this there's typically a backlog even of people interested in being assessed sometimes because they're self-referred because of concern about family history or young age of onset of cancer other times referred by either primary care doctors or community doctors but nevertheless substantial demand by the way the outcome of evaluation is not always testing sometimes it's the opposite we often see people whose risk based on family history isn't as high as they think it is and we can reassure them so not everybody who comes through gets tested and the testing landscape as you probably know has really evolved in the last year or so the company that owned the patent on these genes that's been overturned there's still some legal stuff going on but there are other companies now that are offering it so it's a fairly available test genetic testing has increasingly informed therapeutics not just diagnostics I won't walk you through every image here I realized too small to see most people view the story of imatinib as a kind of poster case of designing a drug that is a target a particular genetic change originally in chronic myelogenous leukemia the BCR able oncogene and has made a very big difference in the treatment of that condition and actually others as well that respond to imatinib this is showing the her to new amplification that occurs in breast cancer for which there are specific drugs this is unreadable but it shows a BRAF mutation in melanoma that also is relevant and so this concept of tailoring a drug to a particular diagnosis which we hear a lot about has already become a reality at least in the world of oncology and I think there are increasingly many instances where genome sequencing is being applied in particular cancers you get these kind of dramatic pictures you can create lists of genes that are mutated and in some cases use that as a basis for therapeutic decision making what's depicted over here is a kind of different paradigm namely that of pharmacogenetics the idea that individuals metabolize drugs or respond to drugs in individual ways on the basis of genotypic is that mostly have to do with the polymorphisms in metabolic enzymes or in particular targets of these drugs and this is showing the excretion of or the targeting of warfarin very commonly used blood thinner with a very narrow therapeutic window you give too much and persons at risk of hemorrhage you give too little and the outcome is they haven't been adequately protected against coagulation I don't think anybody really debates whether this is an effective way of predicting the optimal dose for a particular individual there is however a fair amount of debate as to whether it can be done in a timely way since the decision to anticoagulate needs sometimes to be made very quickly and ultimately whether it's cost effective whether you actually improve outcomes over a kind of long haul and can justify the cost of testing although as the cost of testing goes down and maybe asymptotically approaches zero as more and more genomic testing is done it may shift that equation and there's a lot of work on going to try to demonstrate clinical utility so clinical validity does this test predict a particular response to this drug is pretty good clinical utility is the area that has been mostly debated something that I'm not sure I would have predicted as a medical geneticist but has been a particularly exciting area is that a lot of the conditions that we kind of grew up thinking of as not really treatable have evolved to be in fact quite treatable I wouldn't argue that CF would be a case in point of not treatable because there have been lots of treatments offered chest PT antibiotic use things to thin mucus so it's not that it wasn't treatable and a lot of progress has been made over the years in improving outcomes but the fundamental problem is this chloride channel that is ultimately put together in the Golgi finds its way to the cell membrane which is mutated either is degraded in the proteasome in some instances of mutation or it gets to the membrane but just doesn't work and so it's generated the interest in either so-called potentiators that would restore function to a mutated protein or correctors that would resurrect the or prevent the destruction I guess is a better word of the mutant protein and there is now an FDA approved drug Ivacaftor which targets a very specific mutation this G551D mutation and does function as a potentiator and so this is the forced expiratory volume measure of pulmonary function in individuals treated versus placebo a very substantial improvement was seen fairly quickly and now as I said is an FDA approved drug and this concept of defining the mutation and searching for a small molecule that restores function to the protein or potentially there are other approaches that will restore function to the gene is an area of research that certainly one that couldn't have been imagined before the era of genetic and now genomic testing and this plus other small molecule approaches once you understand pathophysiology you heard my interest is in neurofibromatosis we have right now multiple clinical trials ongoing for a condition that when I started seeing patients 30 odd years ago with NF1 all we could do is make a diagnosis and send people to surgery now we can offer definitive treatments but we can enroll people in clinical trials using medications that target the particular pathophysiology of their condition and that is a huge huge step forward alright finally what about predispositional testing so Laura is now 60 and she's been well but then she and her husband have heard that there is the possibility of doing genomic testing and they go on the internet or whatever the internet looks like in 60 years and they look at what testing can be done the paradigm here is we're all born with a genetic liability of some sort sometimes it's overwhelming and that's how you wind up with neurofibromatosis or cystic fibrosis or something but in most of us it's not however over the course of time in the exposure to environmental factors you sort of cross from a so called presymptomatic to a disease state and the question is if we knew what those risk factors were maybe we could help avoid some of these environmental effects or whatever else drives the progression towards disease or if you did cross this line find better treatments than currently exist so that's what I think what pushes a lot of the efforts to find genetic factors that are associated with common disease and you'll all know well here some of the progress has been made let me just make a comment about some of the clinical applications and so here's a screenshot which is increasingly becoming historical of the concept of direct to consumer testing this was taken a few years ago now and these three companies were offering this testing this one and this one best I know are completely out of the business this one has been more or less well not shut down but FDA has now put a stop on their reporting any clinical interpretations at least pending their satisfying the FDA that they're compliant with the FDA rules on medical testing well I'll tell you so I actually did this before that happened and so therefore I did get this information and I'll tell you remind you actually that I did not need to do any financial exposure of this because I paid for this the paradigm here is that you go online you give them your credit card number which I did they send you a tube to spit into up to a line which is not as easy as it sounds by the way you mail it back to them bar coded so it's associated with you and then you get an email that says well sort of like your genome is ready of course it's not your genome yet anyway it's like a million snips that they and you get this kind of thing at least I got this kind of thing if you were to do this today I don't think you would although anybody who had done it before the FDA clamp down still can see this stuff so this is showing my risk of type 2 diabetes and they give these pretty graphic ways of showing it and you see my risk was about the same as the population risk that's actually fundamentally what I learned from this that I was at average risk for everything you've ever heard of and I kind of decided that was okay there's been a lot of concern in the medical community about this or at least in some parts of the medical community the idea that well alright so you're at average risk what happens if you're found to be at low risk does that mean that you would ignore the advice that you should lose weight and exercise more which by the way is the advice given for every single medical condition practically that you'll see on this but that kind of cynical interpretation aside the concern might be somebody would attribute more significance to these results than they really deserve because although there's a lot of actually pretty good quality information about the interpretation of this who's going to read the fine print and perhaps not realize that first of all the degree of heritability is limited and the contribution of the various markers they're looking at towards that heritability is even further limited and who knows if the data comes from a population that resembles your background and whether those are accurate or not so there's a lot of caveats here and they're all in there but whether everybody will see them is another matter so there's been a fair amount of concern about clinical utility once again and even clinical validity in this case by the way you get other stuff that's kind of interesting you learn about for example can you smell asparagus in your urine which is a genetic polymorphism and some of you know what I'm talking about and some of you have no idea but you know some people call it recreational genomics and it sort of is it's worth a few minutes of cocktail party conversation not too many you also learn about ancestry and then you have the option of turning on a function that allows others with the same haplotypes on their Y chromosome or mitochondria to contact you and it's gone for a few minutes my email box started getting deluged with you might be my relative and I decided it wasn't too productive because most of these if they were related there were so many generations back I wasn't sure what to do with it anyhow you also do get pharmacogenetic data I'll show you in a minute but before let me just point out this study done a few years ago where five individuals sent their samples to two companies when there were and if the arrows point in the same direction it means they got concordant results but if they point in opposite directions and you notice here many examples where that is the case they got discordant results why? it's not because they disagreed about what the genotypes were they generally did agree the analytic validity was very high it was usually because different algorithms were used to calculate risk based on different combinations of markers and other information that they might have incorporated into the risk figure so it is or at least was very much a kind of buyer beware scenario this is the pharmacogenetic data you get this is mine I'm at increased sensitivity to warfarin I showed this slide anywhere I can where there might be physicians in the audience because I figure if I'm ever in a situation where I need to be administered warfarin I don't want to be the only one that knows that I'm at increased risk in fact I gave my primary care doctor a printout of this and I got the predicted response which is what am I supposed to do with this and I actually know what he did with it it was put in a pile and I suspect nothing happened with it after that but if anything did it was then maybe scanned into the medical record which is pretty close to useless because it would be buried in there it's not flagged as being anything that would be important to know about to an emergency room doctor which might be the context where the actual prescription would actually be written and so there's a systems issue which is not unique to our health system I think it's a pretty common one which is this is data that here it is doesn't cost anything and it could in principle be used it is a clear approved lab by the way and yet it's going to get buried at best and therefore not available and why not use it I would like to know I don't if I ever got warfarin and I was at increased risk of hemorrhage having trained us a neurologist and seeing people with hemorrhages after falling when they were on warfarin it's not a good thing so anyway I think we're a ways away from having not so much the data as the systems to incorporate pharmacogenetic information into day to day care alright so in the last few minutes let me just make some comments about kind of where all this may be going you often hear the metaphor of the genome as the book of life I prefer to think of it as the library of life because if you conceptualize anyone gene as a book the genome is really the collection of all of these now put yourself in the shoes of a non-geneticist a primary care doctor say or somebody you know a specialist outside the realm of genetics and this is what you're faced with and you could think back if you like to your medical school or wherever you learn this and you might remember that the genetic code is a three letter code and so you could say to yourself so what's an example of a book that uses three letter words and here's one that comes to mind and so it's possible to come to the conclusion how hard could this be fairly simple code and I guess my perception is that if you are going to use a literary metaphor for the genome a better one is this book I think you could probably get a first grader or second grader to read Ulysses if you mean by that to read the words but I doubt too many of them would have any idea what the book is about it's notoriously challenging to read you have to have first of all sophisticated knowledge of the meaning of words you have to read between the lines see hidden meanings that are sometimes very obscure and I think the genome actually behaves much more like Ulysses than it does the cat in the hat and if you want to strain a literary metaphor one more bit it doesn't take long to go through the looking glass where the rules you thought applied as to how genes are expressed when they're turned on when they're turned off the role of epigenetics a lot of the rules are much more subtle than what might have been previously appreciated so how will the kind of average non-genetics physician use genetics and genomics tell you this quick story this is one that happened in our clinic this child who is 18 months old and there's a family history through this child's mother of multiple endocrine neoplasia with these various tumors that individuals and the family are at risk for and this child actually presented with diarrhea which arguably could be a GI tumor symptom it was a pretty young child to have that kind of tumor that was the possibility and the endocrinologist it was who saw him sent a genetic test it went out to a lab that does testing for this disorder and they found a variant of unknown significance this is kind of my depiction they had at the one end of the spectrum whether it was pathogenic in the other end it was benign and somehow they could tell you it was right smack in the middle and so it was not clear whether it was truly pathogenic however the child got labeled in the record as having multiple endocrine neoplasia based on this finding that's actually a frighteningly common occurrence I see a lot of neurofibromatosis patients as you've heard happens our lab does a lot of NF testing recently I saw somebody where our lab had done a test when this child lived in California actually found a variant of unknown significance had lots of instructions on what to do next to try to validate it looking to see if it segregates in the family for example none of which was done referring doctor labeled the child as having NF they moved as it happens to the southeast we saw them in clinic this is not a child I would have diagnosed with NF had some Caffeol-A spots but nothing else and on the other hand it turned out had a different condition it causes Caffeol-A spots again misinterpreting a variant of unknown significance back to this story that's what this child was found to have well the mother then came to attention separately and was sent actually to genetics clinic for testing by an adult endocrinologist testing was sent to a different lab and a known pathogenic mutation was found that is to say one that had been well described clearly pathogenic whoops and let's tell you from here it turned out whoops from here it turned out that when we tested this child again he did not have his mother's pathogenic mutation so this whole thing was a complete red herring and he does not have that condition and by the way his diarrhea resolved it was just a gastroenteritis when all was said and done there were people where if you kind of take the perspective of this child's position they knew enough to know that there was an inheritance pattern here and that the child was at risk 50% risk in fact they knew enough to know that there was genetic testing and even to send it but stumbled when it came to interpreting it actually this was the next thing I was going to show you which this issue of how do you train to appropriately use genetic and genomic tests has been the subject now of initiative that NHGRI has spearheaded the Inner Society Coordinating Committee which is a group of professional societies representing many different specialties that have been convened now on multiple occasions to try to coordinate efforts to educate their various constituencies about genetics and genomics a working group that I was involved in tried to set up, did set up a framework for competencies that we felt could be used by different groups to define the competencies in genetics and genomics that would be appropriate for particular specialties realizing they're quite different from one specialty to another this is a work in progress this recently was published and our hope is to partner with these different specialties now to try to help them actually to achieve this all right, so let me end with some speculation about will a day come when everybody has their genome sequence lots of people talk about it and it probably, if it doesn't happen it won't be because it costs too much I think you're all aware of how quickly the price is coming down I could believe it will intersect the X axis at some point in time we've come pretty close to doing that it'll become so cheap it'll be hard to put a number on it I used to say that we'll have back when the thousand dollar genome was the holy grail that we'd have a thousand dollar genome but a million dollar interpretation and I think that's even not proving to be true now as improvements are being made in the informatics approaches but you could ask the question when in life would you want to be sequenced and where would the information live could be done prenatally we spoke about that before lots of opportunity to intervene but then lots of questions as to whether you want to know some of the things that you could know before a child is born we talked about newborn screening some of the same issues come up but as you heard it's being tested could be done in children at least then there's still a possibility of intervention for early onset conditions we'll have to give informed consent which is probably ideal except for the fact that they've already kind of foreclosed some possibilities of intervention perhaps that apply to some earlier onset conditions you could ask where would the information live in the electronic health records any of you who have worked with one will I'm sure laugh at the notion because they can barely store ordinary clinical information maybe the right place to store genomic information hard to know I will say however even if they could and even if they do they're not particularly good at talking to each other so if you have your information in one health system what's going to ensure that it will be available to the next one if you're traveling or for that matter hardly anybody I would guess these days is born lives and dies associated with just one health institution people do move around a lot and is it really practical to keep it in one place maybe it could be stored in the cloud I'm sure some already do that it does raise issues of trust and privacy but maybe those are surmountable you can load it on a personal device as long as you don't lose it or put it in the washing machine it will be there this is my favorite the most efficient place to store genetic information for years to be the cell nucleus and nobody ever forgets to bring their cell nuclei to the doctor so I actually could find it possible to imagine the day will come when it will be cheaper to just re-sequence you while you're in the waiting room waiting to be seen use whatever is going to be used that day and then throw it away because it will be easier to have it done again than to bother storing it anywhere I'm not sure obviously that will be the case but the question often gets asked when is this really going to happen and as I said earlier I don't think the limiting factor is going to be the cost of doing it and here's the metaphor that I would offer you can't see this well I'm sure but it's a screenshot of iTunes and why to show this is do you think back what I guess would be 14 years something like that what was it like with music because they already had before iTunes the possibility of downloading mp3 files and you remember the days when people would go on and use these sharing programs that are essentially violations of copyright but the only way to get stuff is either to rip it off of a CD or to go online and I think a lot of people resorted to criminal activity as a consequence but it wasn't really a practical alternative and I can remember a few early situations where they tried to put in a capability of paying for things and it worked so poorly eventually gave up it just wasn't worth the trouble and this is I think what changed that there are many other examples of the same thing now but people started using this on a regular basis because there was a workable system and I guess I would argue that there is the sort of iTunes of medical records or whatever metaphor you want to use a clean fun user friendly effective way to store medical and genomic information that's the day you're going to see genomics become sort of mainstream in day to day clinical practice and I think that's actually going to be more rate limiting honestly than education because I do believe we need to educate health professionals but I also think that health professionals have adapted to new technologies over the last century or more when they were mature enough and ready enough and when the systems were in place and I think it's the systems issue that is actually the bigger challenge and not the educational issue people do figure things out when they're ready to be used so I'm not arguing against education it's been a pillar of my career actually I think we have to put it in its place and I really believe there's been a lot less attention to systems than there needs to be I'm going to end with this quote it's become a little trite now which maybe because it's so good but this idea the best way to predict the future is to invent it that's what all of you are in the business of doing I think it's what makes this really the most exciting time surely in the history of medicine medical research or medical practice and obviously lots, lots more opportunity for all of us, thank you