Dr. Bill Gahl — Medical Genetics and Hope for Rare Diseases

>> Hi Dr. Gahl. It's great to meet you. So when are we going to have all the diseases cured?

>> Boy, it’ll be a long time before we have all the diseases cured. I had a talk about, oh five years ago or so when next generation sequencing was just coming to the fore for clinical efforts, and someone said, “You know, in a year or two we won't need any clinicians anymore who do genetics, you know, clinical geneticists.” I thought that was really pretty crazy because there are 23,000 genes, and there are only 8000 or so diseases known to be associated with those genes, so there are bound to be many, many more discovered, and then there are variations, and so we'll always need people to discover new diseases associated with genes and new therapies and to identify the variants in those particular disorders.

>> Sure, so when you say variant, does that mean that in any given gene there could be multiple problems with it?

>> There certainly are different diseases associated with the same gene because of variants in different parts of that gene. And many of the genes have different functions too. So if you alter a particular area of the gene, you may be altering a particular function, and that may give different clinical manifestations, which is a different disease.

>> Okay. So, how many different areas are there on a gene?

>> Boy, there are areas that every gene has, like the beginning of a gene where there's regulation of that gene. In other words, how well it is expressed, and then there are the exons, which we call the areas that are essentially put together to make a protein out of that gene. And then there are the introns, which are between the exons, and they have some functions too. But not all those functions are known, and then there's the end of the gene, which also has some regulatory functions.

>> Okay. So it sounds like there's a lot of combinations. So you said, I believe you said there's 8000 diseases discovered in 23,000 human genes.

>> People believe that, that's right.

>> And so I've also heard that you're quite the jokester.

>> Sometimes.

>> Have you heard any good jokes lately?

>> Boy, let's see. Well I always talk about how happy I am to have been married for 45 years. And I say, you know, the secret to our success is that we go out to dinner twice a week.

>> Yeah, it sounds like a good thing to do.

>> It is. She goes out on Tuesdays, and I go out on Fridays.

>> I've recently got married, so maybe in a few years I'll start trying that strategy.

>> Yeah, it's probably good to wait a little bit.

>> Yeah. Cool, and I know your work is focused on some very serious issues. So you're, I believe, a medical geneticist; would be the field that you work in?

>> Yes, that's right. And perhaps, you know, my training might be a little germane, which is to say that I went to medical school and then I did a pediatrics residency and chief residency and then did a fellowship in genetics. And there are different types of genetics. I did clinical genetics, which is roughly speaking generic genetics where you take care of patients, and then I also did biochemical genetics, because I'm very interested in inborn errors of metabolism or biochemical diseases.

>> Okay. So inborn errors of metabolism. I guess that's how you use energy and from birth you've got some sort of genetic defect?

>> Yes, that's right. But it isn't really just energy, it's basically all biochemical mechanisms and pathways that the human body employs, and if any of those is disrupted, you could have an inborn error of metabolism. So, for example, a classic would be phenylketonuria for which we have newborn screening, in which case phenylalanine, this amino acid fails to be converted to tyrosine, and as a consequence, phenylalanine and its metabolic products build up and cause cognitive impairment and developmental problems. And one reason that that's interesting is because if you screen for it as newborns, you can reduce the amount of phenylalanine in the diet and prevent many of those complications.

>> Okay. So was this a, is this a rare disease or is this a common one that is a good example of one?

>> No, phenylketonuria would be a rare disease. Actually rare disease is defined in the United States as one that affects fewer than 200,000 individuals.

>> Okay.

>> And there are 320 million people in the United States. The actual incidence of phenylketonuria is recorded as about one in 14,000 live births.

>> So that's pretty rare?

>> Pretty rare I would say, yes, but certainly not among the rarest of biochemical diseases.

>> Sure. And so in 2008 you founded at the NIH the undiagnosed diseases program. Is that right?

>> Yes, that's right.

>> And you still lead that program next to your lab? You have a lab and then you also have this program. How are those related?

>> Well, so maybe I can tell you a little bit about how the program started and how it evolved. In 2007, two things came together. One of them was that Steve Graft, who headed the Office of Rare Diseases at the time, which is now called the Office of Rare Disease Research, said that he was getting calls in his office from individuals who instead of actually having a known rare disease did not have a diagnosis. Was there something we could do about this situation and help these people? And the second thing was that the directors of the institutes, of the NIH here, were interested in sort of bolstering the NIH Clinical Center, which is our research hospital, and one way that they considered that was to have what they would call sort of a mystery disease clinic. So because of both of those things, I was able to sort of incorporate myself into the committee that was going to pursue this and then made a plan to hire some people - actually, two nurse practitioners and a scheduler - and this program was then announced on May 19, 2008, by Dr. Zerhouni, and the audience was about 20-25 representatives of the press and about 90 representatives of advocacy groups. So then Dr. Zerhouni continued to provide funding out of the Office of the Director through 2012. Then we applied for money in 2013 from the Common Fund, which is a pot of money that Congress gives to the Director of the NIH for special projects that are innovative and that cut across many different medical disciplines. So that program is now the Undiagnosed Disease Network, and the Undiagnosed Diseases Program within the intramural program of the NIH is the one that I direct, and we collaborate with all the other groups within the Network.

>> So how did you originally become interested in science and biomedical research?

>> You know, I guess it goes back to one day when I was in high school and my father brought home a copy of a paperback book called The Cell by Loewy and Siekevitz. I still remember that. And it was small enough that, you know, a teenager might be able to read it.

>> Oh, was it illustrated.

>> Well, yeah, it had pictures of cells.

>> Uh-huh.

>> And it was interesting for me, and I think, you know, this might be something that I might really like to do. And then the next sort of thing that happened was when I started to take courses in biochemistry at MIT, and I realized that the pathways that I was learning about were actually functional and operative in human beings, and that with sort of an inclination towards medicine made me pursue sort of a science of medicine, which to me is the biochemical pathways. For example, the feeding of electrons into the electron transport chain to produce energy, and the fact that all the calories in our body are determined by how the substrates for the biochemical pathways produce electrons for the electron transport chain, essentially. And you can, you can actually do the calculations and know by virtue of the biochemical pathways how many calories there are going to be produced by any particular chemical. If you sort of trace it through such a pathway, you can figure out how many high-energy ATP bonds will be produced, and that's basically a measure of the calories that food corresponds to.

>> And then that would have some outcome on the way a person feels or how much energy they have?

>> Well, for example, if there is a block in the energy-producing mechanism, as you would have in a mitochondrial disease, then you wouldn't have enough energy to conduct the sort of operations of our bodies, and therefore, there could be defects in development or strength, muscle or heart function, things of that sort.

>> Okay. What are the goals of the Undiagnosed Diseases Program. Is that any different from the larger network?

>> No, really the network received its goals from the Undiagnosed Diseases Program that we established here in the intramural program. And we really have two goals. One is to help patients who have long sought a diagnosis but been unable to get a diagnosis to find a reasonable and accurate diagnosis. And the second goal would be to make some contribution to the science of medicine, which would be physiology or biochemistry or pathology or sometimes even therapies.

>> Why is the NIH intramural program well suited to doing rare disease research?

>> I would go as far as to say that this type of program probably could not be instituted anywhere else in the country or actually in the world because we have stable funding for one thing. For another thing, we have a research hospital with a clinical center that is able to do basic medical, clinical research reasonably easily. In other words, the nursing units are used to collecting 24-hour urines and things like that, that, you know, some hospitals are not, but most importantly because we have experts in rare diseases and in biochemical pathways and in the basic science that sort of feeds into those types of disorders. So we're able to consult with real experts in all sorts of fields. Neurology, immunology, malignancies, rheumatology, things of that sort, and get world experts to consult on our patients.

>> So how does it typically work, maybe from the patient's perspective, and then perhaps from yours and the doctors here. What happens when someone is going to their doctor and they've got health issues and they don't know what's going on, and the doctor can't help them. What do they do next?

>> So a typical patient will go to a physician, and if the physician can't make a diagnosis, the next step is generally to refer to either a consultant within that center or else to a tertiary medical care center, like a university-based center where they have all these consultants, and then that center will pursue let's say more unusual testing and diagnoses. But when a tertiary medical care center is unable to make a diagnosis, then it would be an appropriate time to refer to an undiagnosed disease program, you know, like ours. And the way that they would do that is to send their medical records, and incidentally, probably half or more of all the cases that come to us as applicants have their cases submitted by the patient themselves.

>> Uh-huh, why is that?

>> It's because there's enough public awareness of the program, and they are, let's say, more incentivized--

>> Yeah, I bet.

>> You know, to send us their records, but we always need a referral letter from a physician or a caregiver like a nurse practitioner or something of that sort, and that constitutes the application process, and then we review each application in a certain way. So, for the adults, I look at the charts and the medical records, and often the medical records will also include pathology slides, like biopsy slides, and very often imaging, like MRIs of the brain and stuff like that, and we'll send the package to some of our consultants in the field that the disease appears to be in, and we'll get an opinion about whether this could be an already known disease or it might be a new disease or there might be some other testing that needs to be done, things like that, and then have really sort of a back and forth with a patient over this. And about 70 percent of the time we turn down a patient, we don't accept the patient, but sometimes we offer advice about how they may pursue their particular illness.

>> Of those 70 percent, how many of them are, oh, you actually might have this already known disease, is that something that happens?

>> Yeah, it does, it does. We think that there's already evidence in the chart that they may have such and such. Or, let's say a particular disorder has not been pursued enough and another test should be done, you know, locally.

>> Uh-huh

>> And then about 30 percent of the patients we accept and the ones that we accept we invite to the NIH clinical center for a one-week inpatient stay, and there's no charge at all, not for the hospital charge or for the physicians or their consultation, and we even pay travel for the patients.

>> Oh, that's very cool.

>> Yeah, it is. That's why people like to come, among other things.

>> And they might have already racked up a lot of medical bills earlier in the process.

>> You know, if we look at some of the charts, the size of the some of the charts would be maybe, an average chart may be an inch and a half or so.

>> That seems pretty big.

>> Yeah, it is, but we probably have maybe a couple hundred charts that are as high as four or five inches, and we have some of them, you know, maybe 10 or 20 that are like a foot high.

>> Oh, wow. That's a lot of paper.

>> It's a lot. And we've had over 4000 applicants over the last 10 years to our program here in the NIH, and we've accepted about 1200 of those.

>> Wow. That's a lot of people who've come through your program.

>> It is a lot for having such rare and mysterious disorders, yes.

>> And how many, so are all 1200 of those completely unique or are there a few that kind of have the same situation?

>> It's an interesting point, that is, once you distill the undiagnosed disease patients enough to have them come to a single center, which is ours, you're going to find a lot of, not a lot, but you're going to find commonalities among some of them; 50 percent of the patients whom we see have neurological disease, for example. And some of them will have in common a very unusual neurological presentation. So just to give you an example. There's a variation of ALS, so amyotrophic lateral sclerosis, or Lou Gehrig's disease, there's a variation that starts out with sensory problems in the middle of the face. So, in other words inability to feel or know what's, or touch, in the middle of the face, and then it progresses to the motor problems, in other words inability to move, that are associated with ALS. It's called facial onset sensory motor neuropathy. So there were five cases reported in the literature as of maybe six or eight years ago, maybe even ten years ago, and we in our program have seen about eight cases. So, you see nobody else in the country is going to see eight cases of FOSM.

>> Yeah, and it sounds like there's only been 13 ever seen in the whole world?

>> Something like that, yeah. But they come to us because it's not a commonly diagnosed disease, and if a neurologist sees that constellation of findings, they'll refer to an undiagnosed diseases program, and not only that, but there's no gene known to be associated with it yet.

>> Uh-huh. So, what's been your progress with these patients? What have you learned so far?

>> Well, for one thing, it generally ends up to be a fatal disease, but for another, despite our genetic analyses, including using the parents of three of the affected individuals for genetic studies, despite all that, we do not have a gene that we think is causing this disorder, so it's still a mysterious disease.

>> And the common form of ALS, is that also generally fatal?

>> Yes. There are several different genes that are associated with ALS, but I know of no treatment for that.

>> Oh, okay. And for this particular one that you've been seeing, you haven't found the gene yet?

>> That's correct.

>> And are you still looking?

>> Yes, absolutely. You know, sometimes you get a break by, let's say seeing a special family, a family that's more instructive or illustrative than the families that we've seen so far, and that might point us to a gene, and then we would look in our other accumulated patients for variants in that gene too, but you know, they could be potential mutations.

>> So how does the family help? Is it because they all have very similar genes, so you can kind of cross reference things a little easier within the family?

>> Yeah. So let me answer that in two ways. Sometimes when there are two affected members of a family, we can look for genes that are shared by those two individuals, or I should say we can look for potential mutations that are shared by the two and not by other members of the family, as an example. But by and large we do genetic studies of two sorts of research, and it's maybe a little bit complicated, but let me give it a try.

>> Sure.

>> So we all have 3.2 billion bases. So that's 3200 million bases, and these bases are DNA bases, meaning they have a certain chemical structure, and there are four possibilities, A, T, C, and G. So that's sort of a start. And these things are like in a row but mostly on different chromosomes. And we usually have, we're supposed to have two copies of all those except for men who have X and Y's, but largely it's two copies of all the genes of our DNA. And there are two methodologies we use to try to pursue what variant might be causing the disease. One is what we call a snp, or a single nucleotide polymorphism analysis. And this, for example, is a group of about a million bases that are specifically chosen by a company to be what we call polymorphic. And polymorphic means there are either A or C or G or T or G or C or something of that sort, there are two possibilities, and those possibilities are reasonably even, in other words 50/50 or 70/30 or even 90/10 or so, but it's not real rare like 1 in 999 for that.

>> Sure.

>> So, there's a million spots, and since we have 3.2 billion bases, they are on average about 3000 bases apart. That's 3.2 billion divided by a million. So what a company can do is to take the genome of a person and determine at these 1 million spots if they have the more frequent or the less frequent of the two possibilities. The most frequent is called A, and the less frequent would be called B. So because we have two copies, we're all going to be AA, BB, or AB. Okay, so but some individuals if they have a deletion, in other words one copy missing, that person will be just A or just B.

>> Okay.

>> And if you have both copies missing, you're going to get no signal at all, and you're going to know that both copies are missing, and those bases are 3000 bases apart. You know, those polymorphic, those snips, are 3000 bases apart. So if you have a run of let's say 30 snips in a row that are missing, then you have about 90,000 bases that are missing. That's called a deletion, and because we have sequenced the human genome, you can look where that deletion is within the human genome and read down what genes are affected.

>> Okay.

>> So, now you know where you might look for a genetic cause of the problem in that person. So that's called looking at copy number variants, because basically you're supposed to have two. Instead you might have only one or zero. That gives you an idea of where to look. There's another thing that snps do, snp analysis does, and that is to determine what we call homozygosity. So the AA and the BBs are homozygous. And if there is--

>> What does that mean?

>> Well, the homo means it's the same, and zygous means egg, but in this case it means that both copies of the gene are the same.

>> Okay.

>> So if you have a run of snps that are homozygous, that's, you know, really long like let's say a million snps in a row, a million bases in a row, or let's say 10,000 snps in a row or a thousand snps in a row, something like that, you know that the region is homozygous and likely was inherited from a common ancestor. In other words, your grandparents could have been sister and brother. Your grandparents, well one grandparent on one side could have been a sister to a grandparent on the other side. In other words, there was a common ancestor. And that means that if there was a mutation in that ancestor within that region, then the person who has that run of homozygous snps is going to have both copies contain that mutation.

>> Interesting.

>> And that is what is required for a recessive disease. And recessive disease means that both copies are affected. So when we have, for example, a first cousin, they share 1/8 of their genes, so 1/8 of the entire genome is going to be homozygous. That's a place we can look for the gene variants. If a person, let's say, is a third cousin, they share 1/128 of their genes, we can look at those regions of homozygosity and look for a potential mutation, it will be in both copies, and therefore the person could have a mutation in that gene causing their disease. So those are snps. And by the way, a snp analysis only costs about $250 for us now.

>> Okay. And so that means to run someone's entire genome through the snp analysis.

>> Yeah, yeah.

>> Oh, wow.

>> So that's a, that's almost a million snp analyses that are determined to be AA, AB, or BB, and for that it only costs 250 bucks.

>> Uh-huh, and that's the kind of primary place that you start?

>> So that helps to guide us when we do our second test, which is a, let's say a more complicated test, and that test is what we call an exome. So the exome is comprised of the two percent of our 3.2 billion bases that make proteins.

>> Oh, okay.

>> So, that means it's about 60 million bases, and we get sequence of 60 million bases with coverage of about 90 percent of those bases used to be like $1000. Now it's like $700 or $500, depending upon the analysis that goes along with it. But, you know, it's fairly cheap for 60 million bases.

>> Yeah.

>> So those 60 million bases then are compared to a reference sequence and determined to be either the same as the reference or different from the reference.

>> So the reference sequence is something that almost all people have in common?

>> Actually, that's a problem, because the reference sequence is actually one human being's sequence. So we're all going to be different from that.

>> Okay.

>> And we, in fact, are different from that 20,000 times out of 60 million.

>> Oh, okay. So a lot of the people are the same as that?

>> Nobody is going to be the same as that reference because every human being is going to be different from every other human being by about 20,000.

>> Got you. Okay.

>> And that means that every one of those 20,000 variants could be the cause of the disease, and you can't deal with 20,000 possibilities, so you need to filter it down. So there are ways to filter it down, and the best way is to compare the family's exome too. Because let's say you have the same variants in two individuals, let's say two siblings, and one of them is affected, and the other one isn't. Then you can eliminate that as the cause of the disease, you know, assuming that it's a dominant disorder, in which case you only would need one variant of the two copies to be mutated. So, I guess the general point is that you can use different hypotheses on models of inheritance to evaluate a family's potential mutations and look at what's shared and what's not and who's affected in the family and who's not to determine how strong a candidate a variant is for causing the disease.

And so there are computer programs that do this, and in addition, some of the variants are considered to be what we call pathogenic, meaning they're bad stuff because they're inherited, and let's say they've evolved and they haven't changed through evolution. And, so, you could filter down from 20,000 to a reasonable number. Sometimes 20, 10, 5, etc. And that helps you to evaluate the few that are left and sometimes put a gene in association with the disease. But actually the very best way to filter down the variants is to find another human being who has the same disease and then maybe has the same or a similar variant, let's say a variant in the same gene.

>> So how would you go about finding such a rare person with the same disease?

>> So that really is sort of a key to the next step in genetics, that is, and there are organizations and databases that do this, things like GeneMatcher, for example, is the name of one. But the sharing of databases and the ability to search other groups’ databases for variants along with certain phenotypes, which we would define as manifestations of the disease. And so if you can search for a phenotype and search for a gene's potential mutations, which we call a variant, let's say for everybody around the world, that could be like a world goal for genetics. One of the issues with phenotypes is that people describe clinical or medical characteristics in different ways.

>> Yeah.

>> So we want to use what we call a common ontology or organization of names, because there are many different names for different medical terms. So, we have that, and it's called PhenoTips. There are other ones, but PhenoTips is one of the types of ontologies that allows us to use common terms so they can be searched to find a similar patient.

>> Okay. And is that, that's what researchers use that ontology. Do like regular family doctors also use that, or are they kind of in a separate [inaudible] system, I guess.

>> Well, so there's putting the data into the system, and then there's trying to get it out. So for putting data into the system, family doctors are not likely to do that. It would be geneticists who would make those decisions about what terms to use and then enter the data into one of these databases along with the genetic data, you know, the sequencing and things of that sort. Anybody can try to pull the data out, and I think that physicians, practicing physicians are becoming more astute at this, but I think that as a profession we have a long way to go. And part of the reason is because this type of genetics is just now entering into the medical curriculum, you know, for medical students. And so there are physicians who probably are not going to go out and learn it.

>> Sure, yeah, if they've been doing medicine for decades already, it's maybe unlikely.

>> Yeah. I remember one of the medical students was shadowing someone, and he said, yeah, I mentioned something to the person that I was shadowing, it was like a regular practitioner, and the regular practitioner said, no, I never heard of that. I think that occurred after I went to medical school.

>> Oh yeah.

>> And, you know, that's pretty much the case in many times.

>> And I imagine with the power or the potential power of genetics to diagnosis things, especially once we kind of know almost everything about the genome maybe someday and how genetics works, maybe that'll be a very significant part of the medical education, maybe everyone will be like you when they're a doctor.

>> Well--

>> Maybe not but--

>> I hope nobody, I hope not everyone is like me. But yes, I think, let's put it another way. Genetics is currently considered a discipline, but genetics is pervading many of the other disciplines as well.

>> Yeah.

>> So, for example, cardiologists are looking at genetic causes of cardiomyopathy, and the neurologists are looking at genetic causes of hereditary spastic paraplegia or spinal cerebellar ataxia or, you know, some categories of diseases. And the different specialties have incorporated genetics into their diagnostic armamentarium [phonetic]. And so I suppose at some point there will, there will always be a genetics course that you take to learn the principles, but in every medical discipline, genetics is coming more and more to the forefront, and part of that, a reason for that is because there are other family members that may be affected, and you may be able to do prophylaxis or at least anticipation in your surveillance.

>> Are there any conditions that you may have seen where it was actually an environmental cause that affected the genome in some way?

>> Well, let's say the Undiagnosed Diseases Program doesn't see a lot of those individuals. But there are classic cases where, for example, in galactosemia, you have to be exposed to galactose to have symptoms.

>> What's galactose? It sounds like a sugar.

>> Yeah, yeah it is. Galactose is a sugar, and the problem with galactosemia is you can't break down that sugar, so its metabolites accumulate and cause problems. So, again, you can treat it by avoiding galactose, and so that is the environmental exposure, you know, to eating galactose. We did also have a patient come in early in the course of this, probably about 2008 or 2009, who had taken nutritional supplements, and I think they came from another country. And they had 1000 times what was on the label in the amount of chromium and selenium.

>> Oh, that's not good.

>> It was, it was bad. This woman was severely affected by that, but actually a couple people died because of that exposure. So, I guess you might call that an environmental exposure.

>> Okay.

>> So it's self-inflicted.

>> Yeah. In addition to all the genetic testing and investigation that you do when a patient comes to the UDP, what else happens when they get to the Clinical Center? Can you kind of walk through their stay, their week long stay here?

>> Sure. The first thing they do is they go to admissions and sign a lot of consent forms and a lot of paperwork. I think we all do that. And we always get routine bloods, but the main issue starts with a history and a physical examination by either an internist or a neurologist or whoever the attending physician is along with the nurse practitioner. And we already know some of the history from the medical records that we've gotten. So, there's probably a list of things that are planned, but the list differs for every person.

>> Definitely, yeah.

>> Yeah. So sometimes it could be an EMG, and a nerve conduction study, which is a measure of how the muscles respond to stimulation and how fast the impulses go through nerves and things of that sort. CT scans or MRIs, especially of the brain. Sometimes a spinal tap to examine the cerebrospinal fluid. Always certain blood tests and always getting DNA and sending it out for specific testing or else doing it in house and doing the snps and the exomes that I mentioned. And urine collections and analysis, and then specialized tests, consultations. For example, the eye clinic does a wonderful job of examining the retinas and the optic nerves. Rehabilitation medicine will generally be involved for people who have neurological problems, and then we get specialty consultations from pulmonology, cardiology, rheumatology, you know, endocrinology, dermatology.

>> So a very busy week.

>> Yes. So they actually are tired at the end of the week, and you know, there's a lot to do, especially for children. This is really what I was getting at. Because a lot of, some would say that, many of the tests in children can't be done because they can't cooperate unless they're under anesthesia. And so Dr. Tifft runs a program in which they bundle the tests that need anesthesia and then have a fairly prolonged anesthesia for like a few hours and do all those tests like on a Thursday morning.

And that would include sometimes an eye exam, a skin biopsy, sometimes some neurological testing, like recordings. And if they need to hold still, for example, an MRI, and you know, the anesthesiologists take real good care of those kids.

>> And it seems like that would be impossible to do at a regular doctor or probably even most hospitals to do that all in one day?

>> Yes. I guess the way that I would put it is they don't. So that sort of tells you how difficult it is. I mean, for one thing, in hospitals outside of the NIH, there's third-party reimbursement, which means--

>> Like insurance.

>> Insurance. So in other words, the insurance has to approve such and such a test. Well, practically speaking, they might approve a test for looking at the eyes, but the timing of that isn't always the same as approving the test for the MRI or the test for some other examination, etc. Whereas we don't need approval because we're not charging you anything. The insurance company is not involved, so we can get that done.

>> Yeah. And you maybe sometimes don't even quite know what you're looking for, and so I imagine an insurance company is not going to approve certain tests.

>> Well, that's true, and they're also not going to approve likely a skin biopsy, which is sometimes for a specific purpose or, because we can grow the cells and study them in the future when we have a candidate gene to determine if it's dysfunctional.

>> Yeah. I guess a regular doctor is not going to be doing that.

>> Well, I would say never in regular practice. Sometimes at a university-based center where they're still pursuing the research portion of the diagnosis.

>> And then they finish up their week and you've got all these tests and some results. Maybe you're still waiting for some results.

>> Yeah.

>> And then what happens from there.

>> Well, there's a wrap-up where we tell them what we know and what we don't know yet and we get back to them when we do have those results in. Sometimes it takes, you know, a couple, a few months to get back the genetic studies. If we find something, then we tell the family. But largely the followup is with a local physician who referred the patient.

>> You send information back to their doctor and--

>> Yeah.

>> And they go from there.

>> Yes. Certainly there's a discharge summary, and if we make a diagnosis, you know, we'll speak with a referring physician, and sometimes there are special places that take care of this unusual disease or that unusual disease and we can refer them to that place as well.

>> That makes sense. Do you remember the first patient that came through the UDP?

>> Oh, my gosh. I think, I think I do. Actually I have it written down somewhere, but I know that either the first or the second patient had a serious central nervous system problem and needed to be biopsied and had a brother with the same thing and basically didn't do well. But that's what I remember about it. It was very sad. But--

>> And how about some of the later patients that, you know, actually maybe by stroke of luck or just had a good hunch, and it worked out really well. What's maybe one or two of the more memorable things that come to mind for a patient you were able to help?

>> Well, I've actually spoken in talks about this one case, a woman who had increased amount of muscle all over on her body, so she looked like she was a weight lifter and worked out, but she wasn't that. And she wasn't taking steroids or testosterone or anything of that sort. This was her disease, to have just huge amount of muscle. And she wasn't stronger, she was actually weaker, so the muscle that she had was weak.

>> Interesting.

>> And she'd had a biopsy, a muscle biopsy that didn't show anything significant. That was about a year before we saw her. And we repeated the biopsy, and it turns out she had amyloid there, which is an accumulation of protein that roughly speaking shouldn't be there.

>> Sure, that happens in people's brains, right. But it also happens in the muscle?

>> Well, it can happen all over the body, and amyloid is a very general or generic term.

>> Oh, okay.

>> So it can be all different sorts of protein. Some of the amyloid that's in the brain when you have dementia will be apolipoprotein E, E4 I think. Anyway, it can be that type of protein, but there's another type of protein that it could be, and that could be, for example, immune globulins. So we thought, well, she might have immune globulins causing her amyloid, so we did a bone marrow to look for the plasma cells, which produce immune globulins. And she had ten percent plasma cells, which is way too much, so that made the diagnosis of multiple myeloma.

>> Okay.

>> So, she carried that diagnosis then, and we referred her to the Mayo Clinic where they have multiple myeloma and amyloidosis experts, and they did a stem cell transplant, and that was about eight, nine years ago. Yeah, it was nine years ago.

>> Have you heard from her since?

>> I hear from her all the time.

>> Really.

>> yeah.

>> That's very cool.

>> Yeah, I remember a couple years after she had this, she told me what she did that day, you know, because she would essentially be dead without a diagnosis. The reason I say that is because the amyloid had invaded the atria of her heart, and when the amyloid invades the ventricles of the heart, it's irreversible. So, she would be predicted to have died, you know, shortly after that, but she was able to get the stem cell transplant to stop all this process. And so she's, I mean I think you could say she's cured after nine years of not having a relapse.

>> Yeah, wow.

>> So, that's pretty memorable.

>> Uh-huh. I bet. That probably feels pretty good.

>> Yeah. You know, the Sammie awards that you mentioned before, I invited her to that. So she was at the ceremony downtown in 2011, this patient that I'm talking about.

>> So in 2011 you were awarded I think it's Samuel J. Heyman Award for Public Service, or Service to American Medal.

>> Yes, yes. In something like science and technology or something like that.

>> Yes.

>> At any rate, the point was, it was for the Undiagnosed Diseases Program, which is basically the NIH's program, and then this particular patient was there.

>> Oh, very cool.

>> Yeah.

>> Did she have a good time.

>> Oh, yeah. I think we might have even had a drink.

>> Uh-huh.

>> And there's another case too that's more recent. There was a 65-year-old male who for the last let's say few years had had episodes of terrible headaches that were debilitating and also was losing the ability to concentrate and some cognitive function. He had had lumbar punctures to look at his cerebrospinal fluid, and there were white blood cells there. So he had an inflammatory central nervous system disease, which would be like a meningitis.

>> Okay.

>> But they were never able to culture anything from his cerebrospinal fluid. So this would be called an aseptic meningitis, meaning they couldn't identify an organism.

>> No bacteria or viruses or anything.

>> Right. Although, let's say often when you say a person has an aseptic meningitis you mean it was likely viral.

>> Okay.

>> Because the viruses are tough to culture.

>> Yeah.

>> So in that case aseptic means they couldn't find bacteria. But in this case, it wasn't a bacterium or a virus. In this case, it was an auto-inflammatory disease. This person had an NLRP3 mutation, a genetic disease in which there was let's say “always on” inflammation. So one of those biochemical pathways that we're so interested in is within the realm of rheumatology or inflammation or immunology, etc., and so the inflammatory reaction was always turned on by this particular gain of function mutation. So this person would have an inflammatory reaction that manifested with the headache and the increased white blood cells, etc.

>> It was primarily their nervous system that was affected.

>> Yes. So it's an interesting thing that the rest of the person's body was not affected at all. And so, you know, Dr. Kastner, who, you know, just recently won this Sammie award--

>> Yeah.

>> He was involved previously in the development of a drug that inhibits the interferon-1 alpha receptor. And so we were able to give him that drug, which is called Anakinra, and within hours he was able to get out of his wheelchair and walk.

>> Wow. And he hadn't really been able to do that at all before or he would have flareups?

>> Well, during the flareups he wasn't able to get out of his wheelchair, and the flareups were pretty common. I mean he'd had several lumbar punctures in many of these episodes. During this particular episode, he was actually here in the Clinical Center, and a few hours after getting the Anakinra, you know, he had been confined to his wheelchair, and he got up, and his wife, he said, you know, it was like a miracle. It didn't hurt him anymore, and so they were, they were very appreciative. So, that's the type of thing that could be memorable.

>> Yeah, yeah, that's very awesome. And then so what did he have to do after that? Did he have to continue taking that drug or--

>> Yeah, yeah.

>> Okay.

>> He still had to, you know, take the drug, but it came out of the NIH.

>> Are there any real, real tricky mysteries that you're working on right now that might be interesting to talk about?

>> Well, there's a, I would say it's still a tricky mystery, but I think we have a solution. We had a little girl who was about I think 14 months, 16 months or so, who had a constellation of findings that we hadn't seen before. Roughly speaking three things. One is she was hypopigmented. So she was like almost albino. You might call her albino, but she actually had some pigment in her eyes, so it's a little different. Secondly, she had development delays and, you know, wasn't walking or even sitting up. And the third thing was that she had storage in some of her organs. So she had a big liver. When the cells of a certain organ are unable to break down a large molecule, for example, then that molecule accumulates. It causes the cells to get big, and then actually the organ gets big too.

>> Oh, okay.

>> So her liver was large. And we did some genetics and found what we thought was a candidate gene, which is called CLCN7. And then this other patient came along through the basic researcher, Joe Mindell, that we were collaborating with, and that second patient had the same type of disorder. In other words, the albinism, the storage, and the delays. So now we had two patients, and they both had the same mutation in CLCN7. So now then we did the research, the basic research, why would they have those problems and found out pretty much the answer. And the answer is that CLCN7 encodes a transporter of chloride in the lysosome, a certain compartment within the cell. It's a vesicle, so it's got a membrane around it. And inside the vesicle it's very acidic. The reason it's acidic is because protons or hydrogen ions are pumped into it to make it acidic. You can only pump a certain number of those protons in before the process stops because the inside of the membrane of that lysosome becomes positively charged. So, in order for that lysosome to become more acidic, you have to pump in chloride ions.

>> Okay.

>> And CLCN7 pumps in those chloride ions.

>> Ah, um-hum.

>> So, for example, if you didn't have enough CLCN7, you wouldn't have enough acidity inside the lysosome, and in fact there's a disease associated with that, which has to do with inability to break down your bone and reformulate the bone. It's called osteopetrosis. But in this case, these two little kids had too much chloride being pumped in. Therefore, they had too much hydrogen ion being pumped in. Therefore, they're lysosomes were too acidic, and as a consequence of that, several things happened. One was that their pigment formation was disrupted because the pigment-forming vesicles in melanocytes, our pigment cells, those vesicles called melanosomes were too acidic to make the melanin pigment.

>> Wow.

>> The second thing was that the lysosomes of the liver or the kidney were too acidic for the enzymes to break down big molecules into small molecules, so the big molecules were being stored. That's why their liver and for the other child the kidney was big. And then the third thing is that there were developmental problems, and we don't know the exact relationship between hyperacidic lysosomes and developmental delays, but, you know, one thing was we were able to give a drug called chloroquine to lower the acidity of the lysosomes. And we know about one of the two patients that received it who actually got better, started to. You know, development was a little better and the kidney size got smaller, some things like that.

>> Excellent.

>> That was a, it's just one of the success stories within a large group of roughly speaking failures that sort of keeps us going.

>> But even in those failures, you're probably learning things that, what do you do with that information? Is it useful in some way?

>> Well, so first of all, I've just mentioned that for the patients even though we don't have a definitive diagnosis, there are ancillary benefits to being enrolled in this program. One is the consultants often offer symptomatic relief, like in rehabilitation medicine or the ophthalmologist does this or that, etc. And the other thing is that there's a lot of hope offered to individuals because they know that people are working on their disorder.

>> Yeah. I imagine that has to make them feel a lot better just knowing someone's trying to help.

>> Yeah, you know, somebody cares. That's basically what we look for as human beings. And for us as physicians and scientists, we remember cases so that if another comes along we can sort of put two and two together. And we learn about mechanisms, mechanisms of action, that hadn't been known before, and when we find a new gene that's associated with a disease, we learn more about the function of that gene. And when we publish this stuff, we share it, and other people have different thoughts. And then we also have, you know, rounds in which patients present their cases, and there are young people at the rounds. So, the next generation is getting some benefit from learning about these new diseases as well.

>> When you're looking for, when you have someone with an undiagnosed disease and you're trying to find similar people who might have similar conditions, did you say that you use databases primarily? Are there other things, I think I heard you talk about social media one time being a tool that maybe patients organize themselves to find people like them.

>> Yeah. So there are formal databases that we use, and there are sort of rules for sharing and how much information you can give and things like that that are governed by the consents of individuals. Then we have colleagues that we know of around the world where we can ask them, do you have a case with such and such. And when we publish a paper, that's a formal way of describing a case, and the readers will sometimes contact us and say, you know, we have such and such a case. Like I can tell you an example of that if you want.

>> Sure.

>> The social media is now playing a greater and greater role in this. I think it's really axiomatic that patients and families that have a new disease want to find other individuals who have a similar or the same disease, that want to become a community. And they're not really shy about sharing their information, even publicly on social media.

>> Yeah, I would guess if you feel like you might not have a good quality of life for a long period of time, that problem takes away some of your fear of being embarrassed about certain things or privacy concerns. You just want to, you want to connect with people and figure things out.

>> Yeah, that's right. I think there's a certain sense of “what have you got to lose”.

>> Yeah.

>> And there's a good example of that in a fellow named Matt Mike who put his son's story on the social media, and in fact, the title of his piece was Find My Son's Killer. And there's a picture of Liam Nelson, because I think that was the--

>> The actor?

>> Yeah. Because he was in a movie where his son was kidnapped or something like that.

>> Oh yeah.

>> So it was really a takeoff from that movie, and based upon that presentation online, he found about 13 other individuals who had the same genetic basis for their disease. It's called NGLY1 deficiency. And now there are over 50 individuals. And, well actually my group has seen about 14 of them at the NIH, and published about the natural history of that. You know, that's another issue with really rare diseases, we need to know the natural history because if there's ever a time when there's an intervention with a drug or a therapy of some sort, the FDA is going to want to know if you have modified the natural history of the disease. And if you don't know the natural history, the FDA could easily say how do you know that the patient wouldn't have gotten better on his own because the natural history is to just to get better.

>> Yeah, um-hum.

>> So, having a community is necessary for defining that natural history. I can tell you about this other case where just by putting things out in the literature we got not only a second case but a bunch more cases.

>> Um-hum. Sure.

>> So some of the information that we had about disease phenotype and let's say candidate disease-causing gene variants, we made public, in a de-identified fashion, you know, so that the patient would never be identified. And we had this list on our website for a couple years or so, and then one day about two years ago I got a call from a doctor who runs a dystonia clinic, dystonia is a type of neurological muscle disease, which they're sort of clenching.

>> Okay.

>> And let's say involuntary movements of muscles. And this neurologist called me and said I saw the gene on your list, and I saw the phenotype or the clinical presentation that was associated with it, and I just want you to know, I have 25 patients with that, exactly the same thing.

>> Oh, uh-huh.

>> So she had already made the discovery of the cause of this disease, and it was in a particular gene.

>> But she had not published it or--

>> Hadn't quite published it yet.

>> Oh, okay.

>> So she offered to have our patient or our case join hers in the publication, which was subsequently in I think Nature Medicine. It might have been Nature Genetics. But it was one of those nature journals. And it just goes to point out that however you get to your colleges and your collaborators, it's incredibly beneficial.

>> Yeah. Do you have any idea of how many papers have come out of the Undiagnosed Diseases Program?

>> Well, let's say out of our program within the intramural program about 90. But many more are probably a score or two scores more have come out of the entire network including the centers around the country.

>> So you've seen about 1200 patients here at the NIH?

>> Yeah.

>> And how many would you say in the larger network at--so that's at universities and medical centers around the country, the network?

>> Yes, the network. Let's see, there is a study section that chose the centers among those that had applied. So they're--

>> Oh, okay.

>> Maybe there are 30 or 35 centers that applied in 2013 for this, and they chose six.

>> Okay.

>> Now, for the second group they chose 11. They're the original six plus five more plus ours.

>> Got you.

>> So there are actually 12 now. And of those, throughout that there have been maybe 20 or 30 or so more papers, and several hundred patients seen. I don't have the exact numbers, but several hundred.

>> And they have been, they've been operational for over 3 years, so it's really increased the capacity of this larger undiagnosed diseases network beyond what--

>> Yes.

>> You and, you and your colleagues here can do as a small group.

>> Yes. I think, again, the first patients were seen in August of 2015, so it's only really been about three years, but in that time, the capacity has been tripled. So, yeah, that's about right.

>> And do you see patients from around the world from other countries, do they come here to the NIH or to the network?

>> They do but not real, real often. So, in other words, we're able to accept them. We don't charge. They have to be good cases. We do ask them to pay for their travel to the United States. But one issue is that, let's see, in many other countries, it's difficult to get the extent of the workup that's required to have eliminated all known diseases. So we ask that of the patients who come from U.S. medical institutions. Sometimes it's difficult from other institutions, you know, outside the country.

>> Um-hum. Yep, that makes sense. Before you started the Undiagnosed Diseases Program how long were you an active researcher in the intramural program?

>> Well, I came here in 1981.

>> Okay.

>> And the program started in 2008, so it's 27 years, I guess, at that time.

>> And you still have that lab going as well?

>> I do. We study specific inborn errors that our lab has been interested in for really two, three decades.

>> Inborn errors of metabolism or other errors also?

>> No, pretty much biochemical diseases, like inborn errors of metabolism. Things like cystinosis, Hermansky-Pudlak syndrome, alkaptonuria, Chediak-Higashi disease.

>> Are these rare diseases or are they maybe less rare or how common are they? Some of those?

>> They're pretty darn rare.

>> Yeah.

>> So, cystinosis, 1 in 100,000 to 200,000 individuals.

>> And what is that? Cystinosis?

>> This is a lysosomal storage disease, so in other words--

>> Similar to the other one you were talking about.

>> Well, similar but a little different.

>> Yeah.

>> It is. Most lysosomal storage diseases have a large molecule stored inside the lysosome because it can't be broken down. There's an enzyme missing. But in cystinosis, it's a small molecule that's stored, cystine, which is a disulfide amino acid. So it's small. And it's not an enzyme defect. It's a transport defect. The cystine is stored because it can't be transported out of the lysosome, and then in the lysosomes of many of the cells, it crystalizes. In fact, there are crystals in the cornea of the eye.

>> Uh-huh. Does that affect the vision? Is it painful?

>> It's painful. It doesn't affect the vision until very late. So an occasional patient will have a huge number of crystals, and the cornea will appear hazy, but even that doesn't affect the vision until there's a reaction in the cornea that causes there to be calcium that forms over the cornea, and then that calcium is opaque, and you can't see through it. It's called a band keratopathy. But interestingly, the drug that we use orally to treat cystinosis, which is basically developed around the country through NIH support and through which, and for which we saw, you know, more patients than any center in the world, that drug is called cysteamine, and if you put it on the corneas, it dissolves those crystals. It's pretty darn amazing to dissolve a crystal in a human body.

>> Yeah, and to have it help them.

>> And it helps them because they no longer have the photophobia or the pain in their eyes.

>> Was it your group or the UDP that developed an FDA-approved therapy for nephropathic cystinosis?

>> Yeah, I think there are a lot of, so let's say people are involved in this.

>> Yeah.

>> So, I've told you what cystinosis is. It's this lysosomal storage disease. Somehow we want to try to get cystine out of the lysosomes, in other words, and deplete the cells of cystine because that's what's causing the problem. And the problem is really largely a kidney problem. Kids lose their kidneys at age nine or ten, so--and then there are a whole bunch of other complications because the cystine destroys the muscle, sometimes the retina, the pancreas, the thyroid, etc. So, you would like to get the cystine out of the lysosomes. And in 1976, a fellow named Dr. Jess Thoene, now at Michigan, studied with Jerry Schneider the use of cysteamine, which is a very small molecule.

It's just Sulfur-Carbon-Carbon-Nitrogen, that can get into the lysosome and chemically interact with cystine to form two products that can actually get out of the lysosome whereas cystine itself can't. So he demonstrated this in cultured cells, fibroblasts from cystinosis patients, and then in 1978 there was begun a protocol to treat patients with cysteamine. That was essentially world-wide but mainly in the United States, a few in France, but let's say the NIH was able to bring back the patients every three months and do very close followup. So those patients had much better followup than patients in most other places in that protocol. And we published the results in 1987 to show that the cysteamine was beneficial with respect to kidney function. We published another paper out of my lab in 1993, and we went to the FDA in 1994 in like July, and it was approved on August 15, 1994, for human use, and cysteamine is the drug of choice for this disease. And so it was an Investigational New Drug. In other words, the Food and Drug Administration had to provide us with an investigational new drug exemption, you know, so that we could use this under a certain protocol.

>> So what is, what is your typical day like, because it sounds like you're very busy. You've got a lab, and then you've got the Undiagnosed Diseases Program. What is a typical day like for you? How do you start your day and make sure it's going to be a productive day?

>> Well, making it a productive day doesn't start with the morning. It starts with the night before, roughly speaking.

>> Oh, what do you do?

>> Well, I bring work home, and a lot of the work that I do now has to do with writing papers or revising papers or thinking about what projects should be done or where we should go at such and such a direction, etc. So, that's work that I find is virtually impossible to do when I'm in my office. Because people knock on the door.

>> Yep.

>> And so I don't get anything done then. So preparation for a day means the night before I will revise a paper using a pen, you know, rather than a computer, because actually I find it easier to be able to, you know, page back and then make an arrow, you know, X, and that's going to go there, stuff like that. In other words, edit.

>> Yeah. It probably helps you sleep rather than staring at your computer too.

>> Yeah. So, I will do that on my couch, and that means the next day I will have a paper that I can go through and make the changes in. Because that type of work can be interrupted, you know, it doesn't require too much thinking.

>> Yeah, you're just kind of transcribing your notes.

>> Yeah. I think maybe that's the best way to put it, that I divide my life into thinking, which I do at home when things are quiet and then actually sort of doing, and you know, not checking things off and getting things done. I think people talk about this in terms of emails as well.

>> Yeah.

>> You know, email can ruin your life.

>> Yeah, it can take your whole day if you let it.

>> If you let it, yeah. I get 200 emails a day. Most of them are junk, but if I get, and I probably do, maybe 50, 60 emails that I have to answer.

>> Wow.

>> That's, and some of them are only a minute, but, you know, some of them are, you know, a lot longer at their tasks.

>> Yeah.

>> They're data calls and stuff like that. That's the stuff, type of stuff that I can and should do in my office. By the time I get in, Europe is like a few hours into their day. So there will always be Europe emails.

>> And you've got a lot of collaborators there.

>> Well, not just collaborators, but there are people who ask questions that I don't even know. You know, people write emails from all over the world about either our lab's area of expertise or else they have a completely undiagnosed disease themselves and they--

>> Okay.

>> So, I'll look through that, and I'll see if there's anything like really important, try and knock those things off, and then I'll start going through it more systematically.

>> Cool. And so if you, if you weren't a researcher, what do you think you would be doing?

>> Oh, my gosh. Yeah, well there are so many things that I'm not good enough to be, that I'd like to be.

>> Yeah.

>> A little bit like to be.

>> Like what?

>> Well, a baseball player.

>> We played baseball every day for all the years I was growing up. I still play softball. Maybe writing. Maybe even something like woodwork for example. Wood is beautiful.

>> Yeah, that'd be a blast, right?

>> Yeah, and you actually end up with something at the end. You know, when we do sort of this esoteric and academic stuff, you know, our product is thoughts, and sometimes it's nice to, you know, have something concrete.

>> Do you do that in your free time on the weekends or anything? Any woodworking?

>> No.

>> No.

>> You know, it's pretty time consuming, so my life is--my one avocation would be softball. And of course I follow the Milwaukee Brewers and, you know, with football I follow the Packers because I'm from Wisconsin. But, you know, anyway, so I follow some of that. But occasionally I read. I read the paper every morning. You know, and now I'm actually following politics because--

>> Uh-oh. [laughter] Don't get too sucked in.

>> Yeah, actually I have to be careful what I say.

>> So if you were a writer, what do you think you'd be writing about, like fiction, nonfiction, any idea what you'd pursue as a writer?

>> Well, you know, my experience with the Undiagnosed Disease Program has made me wonder about, you know, what it's like to be a human, so sort of humanity. Because I've seen up close how wonderful humanity is in terms of people who sacrifice their own lives for their children or their family members, etc. They give up all that, and they never give up hope. So, those are really profound attributes of humanity and how difficult it can be to endure those things. So, I mean there are people who actually have talent at this. I doubt that I would have that much talent, but I think one could, you know, learn how to convey those types of thoughts by reading the classics. Because that's what they do. I mean a writer is actually a thinker. In order to put anything down on paper, you have to have thoughts first. So, yeah, I would like to accomplish those types of things.

>> Well, what would you say is the most difficult thing about what you do as a researcher at the NIH?

>> Well, sort of the research part at the bench really isn't that difficult. Everything I do that's difficult has to do with interactions with other human beings. You know, so, and I suppose occasionally there are employee-type issues, but the main issue is how difficult it is to deal with patients whom, roughly speaking, we fail, and by that I mean we don't make a diagnosis. We don't have a directed therapy. You know, there's some consolation of being able to provide some hope, but many times we're left with, you know, saying goodbye, and you know, that's difficult. And it's not something that you actually get used to, but probably there are ways to cope with that. I think maybe the way is to think about the successes that you have, the people that you've helped, and not dwell on the failures and the people that you can't help. But, yeah, I would say that's the most difficult part of my job.

>> And what are you looking forward to most in the future in your position?

>> Well, for one thing, lightening up a little bit, of course.

>> Yeah.

>> Yeah. I'm going to try to divest myself of some of the paperwork associated with it. But I think professionally speaking I think that the best thing would be if the genetics community of the world were able to develop a system by which they share the information that they have. And by that I mean the genetic information and also the phenotypic information, because that would benefit so much the patients.

>> Yeah. I imagine a lot of that never comes out to the public if it doesn't result in a journal article, so it just kind of disappears unless they have a reason to put it into the system or a system like you're talking about.

>> Yeah, and one of the issues is that there are very parochial systems, databases that, you know, are not shared.

>> Oh.

>> And, you know, some of those have to do with, let's see, a country's politics, and some of them have to do with peoples, you know, individuals, individual scientists or physician's politics.

>> And intellectual property maybe and--

>> Yes, yes. But, you know, sharing of intellectual property is of benefit to the patients and actually the whole world.

>> Yeah, and that's one of the great things about what you do at the NIH, because I believe all the research that happens here in the intramural program is sort of public intellectual property, and it's shared freely with the world to benefit from.

>> Yeah, with some limitations. I think there could be some embargoes and things of that sort, but yeah.

>> Sure. Yeah, well thank you for doing that. You're helping a lot of people, and you're trying to help a lot of people, and I wish you a lot of luck.

>> Yeah, well, we're trying. In fact, that's what my wife says about me. She says, I'm trying.

>> That's all you can do, right?

>> Yeah.

>> Try your best.

>> She says I'm very trying.

>> You are a trying individual, is that what she says?

>> That's what she means.

>> Is there anything else you might like to talk about with your research or your life or the NIH or anything else that I didn't cover?

>> Maybe just that it's really a privilege to work at an institution like the National Institute of Health where people care about other people, and they show it.