 Okay, so thank you everyone for joining us again after the break. So we have one more presentation for you from the Insight program and then we'll have our panel discussion. So now we have for you Stephen Kingsmore from the Rady Children's Institute for Genomic Medicine. Thank you Anastasia and thank you to Robert for raising the bar. So I'm going to try and trump him. What does Insight taught us about treatment of infants in intensive care units? So infants in intensive care units with genetic diseases and what's left to do pretty straightforward. So I'd like to thank those of you, the many thousands who are listening online. And in case you don't know what a genome is, this is one of my favorite videos. Genome is the equivalent of a book 425 feet tall. It's a four letter language. It's written right to left, left to right. And that's the full instructions for a human being. And that's what our four Insight projects set about to decode in newborns and to try to understand what that information could do in terms of healthcare delivery. So the first genome sequence was a rather slow affair. It took 13 years, $2.7 billion. The first medical genome, somebody will probably disagree with me about this, was from Seoul National University and took a year and a half and cost about $2 million. And that really was a genome that was eye-opening because many of the things which Robert talked about for which there's now abundant evidence were first seen in that genome. Right before Insight started, we published this manuscript in 2012. So Insight started in September 2013. And we were able to reduce the time from getting a blood sample to delivering a diagnosis in a seriously ill infant down to 50 hours, which back then seemed miraculous. And it was costly. And then we have the Insight era. We've all been doing this together for six years now. And we both had a technology aim, which was to make genome sequencing faster and more robust. And we were able to set a Guinness World Record twice, $8,500 a time. That's the going rate. And we also did a number of clinical studies. And so now we're at the end of Insight. And in 2019, really, what was research is now moving over into quality improvement projects that I'll talk about at the end. So this is the fundamental problem that we were focused on. It wasn't newborn screening. It wasn't healthy babies. It was seriously ill infants in neonatal and pediatric intensive care units. And what we now understand, based on our most recent study, which is called Insight 2, is that in our NICU, over half of babies at time of admission, the etiology of their condition will be unknown. And that's kind of the subset of infants for which a genome might be indicated. And so when a baby is admitted, there's a very rapid search for an etiologic diagnosis. And before Insight started, we were in an era where getting genome information was completely irrelevant. We did microarray tests. We did some single gene tests. But by and large, the baby would either go home or go on palliative care or die without a genetic diagnosis. And so empiric treatment was the norm for infants with simple genetic diseases. And there are now 14,000 simple genetic diseases if we add together those that are associated with structural variations and single gene diseases. So at the outset, this study started at Children's Mercy Hospital in Kansas City. And the study on the left is really the one which changed my life and the trajectory of the science that I've done ever since. It was the first 35 babies to receive a rapid-hold genome sequence. So these were babies in whom it was suspected. They might have a genetic disease. And we both gave them all standard tests and this newly invented technology called rapid-hold genome sequencing. And the results flabbergasted us. We couldn't believe it. 57% received a diagnosis of a genetic disease. 31% had a change in management. 4% or 11% had a favorable change in outcome. And one baby with HLH, hemophagocytic lymphoesthetosis, their life was saved. So that really changed the world. We realized that this was not a small effect size in the grander schema things. And that if substantiated, this would usher in a new era of genomic medicine. And really all that genomic medicine is using that genome information, the 400-foot-high book, to inform management. So the first study we did was shown on the right-hand side after that cohort study where we randomized children either to get standard testing or standard testing with rapid genome sequencing. And that recapitulated the findings in the first 35. And this was now about 65 kids. And we found that rapid genomes were both faster and gave higher rates of diagnosis. Not quite as profound as the first study, but very similar in terms of magnitude. At that juncture, the team moved. I moved, Julie moved with me. We formed a new team, got some new friends. And we continued the insight study at Rady Children's Hospital. And really at Rady, I had a lot more resource availability to be able to really push on this hard and fast. And so on the left-hand side is the first study that we did. Again, it was a cohort study, 42 infants, the first 42 babies at Rady, to get rapid whole genome sequencing. And the numbers are exactly the same as in the Willig study from a couple of years earlier, that 43 percent received a diagnosis, 31 percent change in management, 25 percent, 26 percent change in outcome. And that stood in stark contrast to the standard types of genetic tests. Then more recently, David talked about this this morning. We haven't published this yet. We've just finished, in October, a study of 213 infants. And they either received a rapid genome sequence or a rapid exome sequence, which is about 2 percent of the genome. Or if they were really ill, we didn't randomize them. And instead they got ultra rapid genomes. And over the space of the last six years, our technology has gotten better and better, faster and faster. So these days, an ultra rapid genome is typically a 36-hour proposition. And the same phenomenon is shown again. It starts to get boring that we have high rates of diagnosis, high rates of actionability. And so this insight study, for me, is quite remarkable. This is the world's literature in terms of the utility of genomic medicines. That's medicine, which is informed by genome information in infants in intensive care units. There are, I think, 12 studies here. And the ones in yellow were all funded by insight, either at Rady Children's or at Children's Mercy. And together they build a very comprehensive base in terms of our understanding of the utility of this technology in this very peculiar setting. And now the work that we started has been recapitulated by other groups inside the US, in England, in Canada, and in Australia. And the rates of diagnosis and the clinical utility are now fairly standard, and we're all finding pretty much the same thing, with the one exception that the tighter the enrollment criteria, the higher the diagnostic rate, the broader the enrollment criteria, the lower the diagnostic rate. So about a third get a diagnosis, about a quarter have a change in management as a result of that diagnosis, and about one in five have a change in outcome. It's quite amazing to think that that all took place within the scope of a U award. So like others, we were into questionnaires and touchy-feely stuff. I personally wasn't into that, but others in my team were. I won't go down that rabbit hole, very tempting to do so though. So this was asking physicians whether they acutely, at return of results, thought genome testing was useful. And you can see that almost 80% said yes. And that the group who were most seriously ill, and who received the most fast turnaround time were the group with the highest positive rate in terms of utility. Parents felt the same way. And the backdrop to both of these slides is a diagnostic rate in the low 20s. So what this indicated to us, which again was a bit of an epiphany, was that many parents who were getting a negative result, many physicians who were getting a negative report, still felt that the test was useful, speaking to the negative predictive value of genome sequencing. And most parents felt that the child did a lot of good. We've seen this across all of the studies now, that there were a lot less problems with stigmatization, decisional regret, and all the big words than we all had feared when we first got together six years ago and had lots and lots of ethical and legal dialogue about whether this was a wise idea or not. It turns out that really it's very difficult to see that we're doing any type of harm and instead we're doing a great deal of good. One other thing that we showed during the six years was cost-effectiveness. We're still doing this analysis on our new study, which is 213 babies, so it'll be much more powerful than this study, which was 42 infants. But if you just look at the bottom line there, bottom right, that there were net healthcare savings in six cases that more than paid for the entire cohort to be sequenced. And some of these cases now are well known. The KCNQ2 patient, for example, is somebody whose life was saved as a result of this. So at this point at the end of insight, we have basically a completely new way of managing patients who are admitted to an ICU. So again, over a half will have a disease of unknown etiology at admission, and we favor a rapid genome sequence done as soon to admission as possible, and that these days you can expect a result, a provisional result, in about 36 hours if the baby is really sick, and that that shortens this period of empirical treatment. With 14,000 genetic diseases, empirical treatment is not optimal treatment. It means the baby's getting a generic treatment course for a very specific illness if they have a genetic disease. So I'm not sure where the 46% comes from, but we'll just go with that. A proportion will get a genetic disease diagnosis. And another third, roughly, there won't be variants of uncertain significance that may have explained the child's condition, essentially in that group. To some extent, we have ruled out genetic disease, and where there are specific genetic conditions that the clinical team is worried about, we can do a deep dive on those genes and actually rule those out, and that has turned out to be very important in decision making as well because it refines differential diagnosis. And then many of the children will get genomic medicine, and many will have an improvement in outcome. So that's now the world today, six years in. These are two of my favorite ladies. On the left is Sebastiana, and on the right is Riley. So Sebastiana is now two and a half years old. She was one of the first children to receive a rapid genome sequence. She was in status epilepticus. We diagnosed her with Otahara syndrome, so the gene is KCNQ2, a potassium channel. She's been seizure-free now for almost two and a half years. She's off all medications, and she's doing really well. She's a little bit developmentally delayed, but she's catching up on her milestones. And then on the right hand side is Riley, who's also two and a half years old from the same town in California, and Riley was diagnosed with infantile Parkinsonism. And now that she's been on L-dopa, she's also hyperkinetic. And so really, life after insight has been about making what these two little ladies experienced become normative for children all around North America, in fact, all around the world. So let's talk a little bit about that and how we're going to accomplish that. So we have a lot of barriers to adoption, and we're moving gradually from NIH funding and philanthropic support over to this being a reimbursed normal part of medicine, where Medicaid and private payors will recoup the costs of genome sequencing. So we've knocked off some things during the six years. So at the start of insight, it really wasn't feasible to think about sequencing 10,000 or 100,000 children's genomes and doing it in time to change their management as inpatients. Now it is. It's not all due to us by any means. Illumina have a new sequencer, the NovaSeq, the Dragon technology gets the compute piece down to 30 minutes or so. And we can now realistically think about 60,000, 80,000 infants in North America getting a rapid genome sequence as part of their workup. It's really just a matter of setting up a factory. But the other bottlenecks are proving much, much more stubborn. By far, the biggest bottleneck to deploying this all over the country and all around the world is that we know more than pediatricians do. Most pediatricians can't read a genomic report. They don't really understand the whole concept behind genomic medicine. They think in terms of formulating a differential diagnosis, giving empiric management, and then gradually working through that and refining their thinking. And instead we have a system where a genome addresses all conditions. And so they don't need to think. It also comes with an associated therapy. And this is kind of intuitive for pediatricians. And if we're not careful, can meet with stubborn resistance. And furthermore, the group who might be best able to help usher in this new age are all dying out, apart from Robert Green. We just don't have enough medical geneticists. We certainly don't have enough genetic counselors. Most children's hospitals with a NICU don't have a medical geneticist. So this is a profound barrier that we need frontline pediatricians to be trained in genomic medicine in some manner if we're to do this. The evidence of cost effectiveness is getting better and better. And so the good news is that we're gradually winning over payors, and they're starting to reimburse this. And we now have people in our hospital who will call up insurance companies every day on children who need a genome, and they're starting to get to yes. And then the last point is that we haven't solved the underpinning problem of bad outcomes. That yes, we've solved a bottleneck, which was we were never really aware before that genetic diseases were so common in intensive care units, and we weren't diagnosing them. Now we are, and it points out to us, that we have very primitive treatments for many of these conditions. This slide just underscores this. This isn't our complete data. We haven't audited all of our data from our most recent study, but this is 45 children who died as infants that were enrolled in our studies over the last couple of years. And what you can see there is the profound genetic heterogeneity that's associated with infant mortality, and the fact that 47% of inpatient infant mortality is associated with a genetic disease. I think there may be three huge surprises for insight for me. Number one is this, that this has provided the pans out on a larger number, the leading cause of infant mortality in North America. It will rewrite our national vital statistics. Many of these conditions don't have effective therapies yet. So a huge need there, a huge need to build a knowledge base so that we can pick these things up, and a huge need to have interventions for these. Second thing which really has struck me, and this again resonated with what Robert said, we found 15% incidence of genetic disease in a regional NICU, not a birthing NICU because that's populated largely by premature babies, but a referral NICU where babies are transferred in 15%. And as David spoke to this morning, that's just surveying about 47% of admissions. Undoubtedly we're still undercalling genetic diseases. Then the third thing which really has been mind blowing for us is the rates of change in acute management. So again, back to after insight. How do we proceed? Well, we've got to be highly inventive. We still need a great deal of money to push this through. It takes on average, believe it or not, 17 years for any new medical technology, major medical technology to go from proof of concept to being standard of care. I don't have that long to wait. So with the help of Illumina lobbying at our state legislature, we got a grant given to Rady Children's Hospital, whereby we're going to enroll about 140 babies from six NICUs, they're mentioned on the left hand side. On the right are the actual zip codes of the 90 or so enrollees to date. And you can see the interim statistics. The statistics just keep being recapitulated. 52% rate of diagnosis of genetic diseases, 41% rate of change in management. Now this study is a quality assurance project. It's not research, our politicians said we don't want to pay for research in babies. But this is supposed to, this data set will go back to the Department of Health and will help them make a policy decision for coverage of medical babies. And all across the country, somewhere between 40 and 60, maybe even 65% of the babies in a NICU will be paid for with the federal pay order. And so just winning over this one group will be hugely influential. Another thing which is new for me, haven't had to do this ever before in my life, is this idea of educating pediatricians and starting to get up to speed on how you actually do educate neonatologists. Physicians are very busy. They don't have very long attention spans. And they are only really interested in on-the-fly learning, where they get just in time information to help them manage a patient that's directly relevant to their practice. And so we have a new partnership that we kicked off last year with Vermont Oxford Network. They are the largest quality assurance organization in terms of NICUs in the world. They have 1200 members. And we now have two 30 minutes once a month webinar segments, which are cases. And we try to plant into that enough genetics and genomics so that people are starting to get over the hump in terms of the mystical language that we all use. And the word cloud there is actually from our first one and just speaks to sort of the quite amazing level of interaction that there is inside this network. So in summary, insight from my perspective has been maybe the most successful NIH program ever. We can clearly point to children's lives that have been saved. I don't know of any other NIH grant I've ever heard of that resulted in lives saved. But even more important than the individual lives is that we've really demonstrated something that's now ready for implementation in the care of seriously ill infants. We estimate the number to be somewhere between 60 and 100,000 infants per year in the US. What we find is that rapid turnaround is really what it's all about that neonatologists just don't have time to wait for results. They're going to go ahead and make their decisions anyway. And so if you can turn around a result in a day, they will take that information into account in their medical interventions. And many of these are surgical. They're not all medical. Many of them are in disposition and in counseling. And so it's a very broad swath. It's not all what we traditionally think of as treatments. So now we're in a new era of implementation science and quality improvement where we are trying to do this now outside of the research environment. This is actually very, very challenging. When we ran the insight study, we had genetic counselors who would go in and talk to people. We had program managers who would sort out all the details. And we had nurses who would go get the blood. Suddenly and now we're confronted with real medicine where you don't have any of the sports staff and you've got a busy neonatologist trying to make that decision about whether or not this baby might benefit and then getting on the phone to try to get insurance to reimburse this. So there's quite a lot of work that needs to be done to actually make this user friendly. It's been very interesting. We now are doing this in 23 children's hospitals around North America. And it's almost like evangelism that you go in and the NICU is cynical and then they have a case with a life that's saved and suddenly everybody's a believer. It's the most extraordinary phenomenon, but it's not easy for individual NICUs. It takes six months to switch on a NICU and have them go from another test to where's the genome. And then lastly, and I hope the program managers, especially listening to this, we need a huge amount of work to be done to fill in the background knowledge base about infant genetic disease. This is something that all of our studies have really unearthed, that there has not been a lot of work done at looking at newborns in terms of the burden of genetic disease. And all of us had to make quite tough decisions as we designed our studies and shortcuts and some of them paid off and some of them didn't. And so we still don't know whether it would be worthwhile to decode the genome a very premature baby. That's a great question to answer. We talked a little bit this morning about hypoxic ischemic encephalopathy. Thankfully, David just wrote a grant, which hopefully will go under review and be favorably scored to study exactly that group. You know, they get brain cooling and that decision is made relatively rapidly and it kind of bypasses the idea of a genome. And yet encephalopathies are maybe the single indication where a genome can do the greatest good. These are where we save lives. And then there are all of the community NICUs, right now our experience is really limited to level for regional or quaternary neonatal intensive care units and pediatric intensive care units. And what we're seeing is true of all races and ethnicities, but we don't know how that this translates to birthing hospitals and community NICUs. And the vast majority of babies when they're born, they're not going to have access to a regional NICU. If they're sick, they're probably not going to get transferred. And so we've got to redesign what we do for these other centers where there aren't abundant subspecialists and medical geneticists. So as with everybody, across six years, you accumulate quite a heck of a team of people who contribute to the work. And over on the right, I don't want to forget my colleagues at Children's Mercy Hospital, who really helped kick all of this off. And I also would like to point out something that I hope you all are doing, which is reaching out to the public, that one of the things which helped us a lot in terms of funding was to get support from the community, from parents to help add to our NIH funds. So thank you very much for your attention.