 Thank you, Stephen. So that will conclude our presentations for today and now we will move into our panel discussion And so for that I will turn it over to Melissa Parisi, my colleague at NICHD to be able to lead us off Thank you Anastasia and thank you to all of our speakers this afternoon. I think the presentations have been very timely Clear and in some cases provocative in a good way, and we hope that this will be helpful and stimulate the discussion that will follow I wanted to remind folks that For those of you who are listening on the live stream if you have a question that you'd like to raise to the panelists The address is listed on the screen Insight at NIH gov We are standing by and monitoring our inbox here for questions that you may have so please avail yourself of that While we're awaiting questions coming from the audience. I'd like to introduce our panel members and Make sure that everyone actually what I'll do is I'll call on the individuals and have you introduce yourselves and tell us where which institute you Recommend which institution you represent and your affiliation and then we'll go ahead into the panel discussion So I'll start with Alan Bags Thank You Melissa. I'm Alan bags. I'm PhD molecular geneticist and co-pi of the baby C study together with Dr. Robert Green at Boston Children's Hospital Great, thank you David Demek David Demek I'm a clinical and biochemical geneticist and I am a radiogenomics in San Diego Thank you Cynthia Powell I'm a clinical geneticist at UNC Chapel Hill And co-pi with Jonathan Berg on the NC nexus project Thank you Jennifer puck Hello, I'm Jennifer puck at University of California, San Francisco, and I'm a co-pi of the NBC project along with Barbara Canig and Puy quack Thank you, and we're not a Gallagher I'm a metabolic geneticist at the University of California, San Francisco, and I've been Working on the NBC project at the University of California, San Francisco with the team Thank you So in starting off our discussion today We thought it would be worth Taking us sort of a step back to when we first conceived of this program and these projects and really Redress or Think about where we've come and whether or not we've answered the three questions that were actually posed in the original request for applications That was published back in the early 2010s And specifically the three questions that we were interested in answering through this program Were as follows number one for disorders currently screened for newborns How can genomic sequencing replicate or augment known newborn screening results? number two What knowledge about conditions not currently screened for newborns could genomic sequencing of newborns provide and Thirdly, what additional clinical information could be learned from genomic sequencing relevant to the clinical care of newborns? So just wanted to ensure that those questions were on everyone's radar screen As we progress and I think one of the rewarding aspects of this program from NIH's perspective It's just how different each of the projects has been and how each of the projects has Tackled a component of each of these questions in a little bit different way with really different results So from a from the perspective of a pilot program to try to answer these really Challenging questions you all have really taken Taken charge of the issues and really been very successful in in your approaches to addressing them So at this point we'll open it up for discussion So this is Alan begs I could start out I'll just tell you that you know at first blush I think when we read the RFA and the questions NIH was asking The question in our minds was you know could genomic sequencing actually one day replace Traditional newborn sequencing or screening and I think very early on we kind of discarded that And said that you know traditional newborn screening Has certain criteria in terms of sensitivity and specificity making sure that you don't miss many cases missing making sure that you don't have false positives and It's functionally based based on biochemistry. You're analyzing the level of entap metabolite in the blood You're analyzing the end result and so you are closer to the clinical features. I think our thoughts about DNA based testing are that it's not going to achieve at least under current circumstances the same levels of sensitivity We're not going to be able to identify all the cases or as many cases and we are likely to have false positives So we're going to have reduced specificity So the question really becomes at what level do you find the optimal balance between those two competing interests? sensitivity and specificity and How do you use that really I think to augment newborn screening results? How can you use that to help potentially identify some cases that might have been missed or? To Demonstrate additional information such as information about prognosis for a child or Hopefully eventually specific information about therapies No other panelists have any thoughts Sure, I want to add That it we have learned that the DNA can't replicate the results that we get from tandem as petrometry and really that as you say it's it's different information and I Think we've seen in our project that it's really Complementary and it can help determine whether someone is affected a carrier or unaffected So it's an additional piece of information and there are multiple limitations including in some cases not knowing all Of the genes that might have caused an elevation of a particular metabolite that may be one limitation in our data set Not covering the whole gene Not regulatory regions not knowing all changes that cause disease or not knowing the effect of those changes I think one issue that's come up that I think about is when our pipeline We had a requirement for two variants and we know that we don't always know When two changes are found if they're from different chromosomes one from mom and one from dad and It was interesting to hear from that earlier presentation that that doesn't happen that often So we're it's helpful to have Many pieces of complementary information from this project. I think one of the things that we've seen dramatically Is how Sequencing can allow you to get to a definitive answer from abnormal newborn screen I think the success of California's program in looking at cystic fibrosis is is sort of a poster child for that but also Children that further you drink a look at history screen and end up having Ikari Gutierre syndrome that you pick up on genome kids with an elevated C3 that you've got 40 50 maybe 60 genes that can cause that getting to a definitive cause and being able to plot Treatment and provide families with clear answers This is what your child has and your child is on the correct or they do need this or they don't need that So it's one of things we anticipated But it's really exciting to have seen it play out over the last five or six years And this is Cindy Powell. I would certainly agree that The genome sequencing is not going to be as sensitive as traditional newborn metabolic screening But I think that you know one one case that Tamara had presented earlier the malony'll coad carboxylase deficiency child was a Premie in our newborn intensive care unit and you know as those of you have who have dealt with newborn screening and premature babies There's often a need for repeated samples and you know This child had about four samples sent to the state lab over about a month and a half period and each one kind of coming back borderline and equivocal and Although this case happened to be an infant with two well homozygous for V us's in the gene You know that we had to report as V us because it just hadn't been described before that specific variant Hadn't been described before in a patient with this very rare condition But I think as we accumulate more knowledge over the years We will be able to you know utilize the the genomic piece to help You know figure out these these difficult cases in in newborns, you know specifically with the preterm infants This is Jennifer Puck. So I think over the past five or six years. We've really seen a sea change in that newborn screening programs across the country Did not use DNA sequencing Extensively or even at all before Five years ago one of the things that actually brought DNA technology into many of these labs is the newborn screen for skid, which is based on T cell receptor excision circles and actually that is a Bio marker that isn't a sequence based test at all It's just a single PCR test for is it there or is it not and And yet in order to do that test the labs had to start making DNA from dried blood spots And then that opened the whole Area of sequencing and one of the early things that our project had to do was demonstrate that you could Do a whole exome or a whole genome sequence from a tiny punch from a dried blood spot and you could get results that were equivalent to whole fresh blood and so once that was clear then the you know the DNA platform is available and and Then we have to think what's the most effective and the best way to use it and Just as when tandem mass spec was introduced and all of a sudden there were many many different Conditions you could identify or try to identify by looking at different peaks there and some of them had better justifications than others in the realm of the traditional Reasons to do newborn screening that things have to be newborn onset urgent medical emergency treatable Those kinds of considerations Didn't necessarily apply equally to all the things that you could look at in tandem mass spec and similarly When you start doing sequencing You find that you can Detect things that aren't all in those urgent reportable treatable kinds of realms and and I think we've heard today about how you can sequence things but not report them and you do have to think carefully about What what the goals are and what the context is for the Sequencing that's being conducted And I think that's a really good point Jennifer in that five to six years ago when we were launching this program Being able to sequence from a dried blood spot was not routine or standard of care Or even feasible in most respects and that was one of the barriers that we identified when the program was first launched and in addition Genetic or DNA-based testing was not traditionally utilized in newborn screening state programs So all of those things have really evolved at the same time that the inside program has been Working and and producing its results. So I think the timeliness is is really Remarkable I wanted to follow up with you on one of the comments that was just raised which was if we think that genomic sequencing Genomic sequencing can either Augment or complement traditional metabolite based newborn screening. What do you think would be the next steps for trying to implement? Genomic sequencing in the context of newborn screening programs I'm happy to start that firestorm So, I mean, I think One of things that and I'm gonna answer a bit of question be first because I think it's relevant One of things that we've realized is that actually having a diagnosis of disorders that fall outside of our traditional Biochemical treated treatable disorders There's relevance for example seizing children that have de novo mutations in genes for which there's a different seizure medication So it shifts the paradigm of the way we think of things being treatable As to the question of how do we get there newborn screening is a system. It's not a task and To me, I think we could I mean The numbers that I think Alex campus there was there's 120 conditions that currently meet or the advisory committee Standards that you can pick up with DNA that you can't pick up with analytes Genomics England data looks like it's more like 200 to 250 conditions that fall into that category So the fundamental problem now is not can we screen for conditions and even set really stringent criteria for Avoiding false positives because that's actually the thing. We're worried about The problem is do we have actually anyone to catch the babies? And so I think The big challenge here is actually the rest of the system The test is feasible to go live tomorrow, but there's not a system in place to catch the babies Well, so this is Jennifer Puck and I just I want to push back a little bit because I think we need to To be very cognizant that that the core of newborn screening is a public health mandate And it's considered so important that we don't require Explicit parental consent to do this. We think it's so so urgent that we detect conditions like PKU very early that we Conduct newborn screening on all babies and I think this category of disorders has to be kept in mind and When we do depart from that we have to recognize that we're departing from it and and be sure of The context of what we're doing and and so I think one context is if you have a sick baby in front of you as the Rady program has shown us so well the genomic information can be life-saving and and Certainly change management, but that's when you start with a baby in front of you who is sick and who needs a better diagnosis if you are looking at newborn screening in general where There are vast numbers of nurseries full of almost all healthy babies that's not the case and then the issue of false positives and and And also just discovering things that that maybe aren't so Imperative as the core Screen conditions That is something that that I think Perhaps parents should be more involved in the decision to pursue and Those don't all necessarily have to be pursued right At two days of life when that newborn screen is obtained and and I think we could think about different ages and stages and pediatric care Is Conducted through a child's life and and what is the best age and then what's the best time to engage parents and thinking about it Yes, I think Jennifer's you're making a very good point that newborn screening traditional newborn screening Is really focused on large populations of mostly healthy infants right at birth and that there's a role for For that but that genomic sequencing really serves a much broader function I think one of the things that baby seekers try to explore is the potential risks of Getting uncertain information about a healthy child to parents and we've seen Minimal to no evidence for harm associated with that But there's certainly a concern and I think that maybe the North Carolina group would like to talk about their concept of age-based Sequencing because this is something perhaps that in other words staging The examination and reporting of genetic information over the course of a child's age To be appropriate to the risks and potential benefits that somebody of that age would experience Sure, so this is some Research that Jonathan has been doing with his postdoc Lana mollison Jonathan Berg. Yeah. Yeah, sorry Jonathan Berg and can you introduce yourself as well? Yes, my apologies. I'm Laura Milko and I presented some of this Information they published it. I think late last year and really thinking about Targeted sequencing in terms of being able to provide Digestible packets of information. I think it was either Bob or Flavia mentioned that one of the comments that that parents and and potentially clinicians had made was that if there's a choice then maybe you should say no and and these are just the kinds of of of Things that we're trying to to prevent by offering relevant panels at appropriate time stages and I mean, this is not to say that I am I'm certainly becoming more convinced by Rady's data and Steven's excellent presentation about about diagnosing Ill children but for for actually Expanding newborn screening and healthy newborns. I think we really need to think about how to Prepare parents to receive this kind of information How not to Create patience and waiting and and this is also not to say that potentially When children reach the age of consent and become adults that they couldn't opt to receive a genome or an exome But really the question is how how are we expanding current newborn screening as we know it to be useful? And I think to do it at Targeted times when it's relevant to the child's development and pediatricians and parents can really understand it's relevance at that particular moment It is imperative So so having said that and I actually like that concept quite a bit But I think another consideration we need to take into account is the in traditional medicine There's a one-to-one relationship with doctor patient relationship between these two individuals and the physician is thinking only in terms of The potential benefits and risks to their patient, but in genetics you have to think about the patient as really being the family and so in babyceak, for example, we identified Infants who carried a cancer risk allele that wasn't going to impact that individual until they became adults However, most of these conditions being inherited There's obviously the implications for parents and in several cases with parental consent We actually returned the result to the family that the parents were indeed at risk And so I think you have to consider the that with genetics. We're thinking about more than just the one individual under consideration We're not a did you want to So this is we're not a Gallagher from the University of California, San Francisco, and this is a great discussion I just wanted to address a couple of Points on answer and answer to Melissa's question to me It seems one of the things we have to decide to do first as we think about employing this is deciding about a technology So we've discussed that our project was exome slice and then maybe there would be a different way to do this I mentioned another paper in which there was a different initial technology The other is a gene list. I'm very struck as a metabolic geneticist when we look at the timeline and we think about the 2006 paper in which the ACMG proposed the initial Candidates for newborn screening. We thought of methylmalonic. Acidemia is just a related to a few genes They're actually 26 genes at least that can cause elevated C3 MMA we know there's another gene that can cause elevated phenylalanine that's only been described in the last two years Tango-2 deficiency a newly described disorder can probably be identified through this So we really have to look and vet a gene list I think we have to know have more knowledge about causative variants And this is the clean gen project is a very important part of this and it sees our project expanding information about other Populations we have to be able to interpret variants because we found that there were 30 percent novel variants We have to think about how we want to use it as a primary or secondary screen and whether we knew that might do that differently depending on the condition and Guess the last point is Personnel and workforce, which should really be the first point probably all the way through the system which Dr. David Demick mentioned there's a Small ish workforce in genetics and a lot more That we I would like to be able to do As we have more and more conditions that we can identify and as there's actually more treatment for these disorders, so it is a huge systems That has to be addressed So I wanted to sort of step back in again, you know, we have current criteria for how we add conditions to newborn screening I don't think those need reinventing. I actually think they're reasonably good there are challenges with them But there's challenges with the UK's criteria as well I Think as we think about newborn screening, we should be applying those same criteria If we have an acceptable test that is cost-effective We have an acceptable test that is cost-effective. We have to define what that is As I said, you know Minimum, there's a hundred conditions probably more like 250 that we can get off of a genome sequence that would meet both cost-effectiveness and All of the other newborn screening criteria So I don't think we need to be worrying too much about what we do with a two-year-old. We we have Children that are needlessly dying of conditions that we could detect by sequence based screening today Aldo the red fruit fructose intolerance would be a good example of this Many of those kids either end up needing a liver transplant or dying something that you can get with a Probably a three three or four penny test And so the fact that we aren't Screening for these conditions because we're scared about sequencing to meet is foolish when they meet current criteria and we could test for them cheaply and Yes, we can get more data from a genome But there's actually a lot of data we could get today that is relevant to a newborn With diseases that they're gonna get in the first 30 days of life and I think To let babies die because we're too worried about what might happen when they're two or five or 20 Misses the point, but if there's nobody to catch the babies, it doesn't matter if we make a diagnosis In thinking about oh, did you want to say something dr. Gallagher? I'm sure this is we're not a Gallagher from the University of California. I just wanted to add to that. I think as we Into new disorders the paradigm of Fennel Keaton year is always helpful to think about newborn screening Because we will we know that there are variants that we can't interpret and we need to be prepared for this And that's an additional really workforce and interpretation challenge And I think that it's important to be aware that There can be Difficulties for families and physicians with information of uncertain action abilities so Benign hyper felony Mia for example is a good example of a condition that can be identified on newborn screening in it Was not clear whether this needed treatment in some cases. We still don't have They can be into Clinic variability and what people think about how people should treat individual conditions So I think it's important to be aware that there's there will be some uncertainty going forward as most of you know the There's a process for Nominating and reviewing conditions that are proposed for addition to the recommended uniform screening panel and that process is Basically adjudicated through the advisory committee for heritable disorders in newborns and children Part of the process for reviewing nominations is really to not only ensure that there is a Validated test that there has been a pilot study done on a population basis that shows benefit But also that there is the public health impact is is reasonable and one that shows net benefit over potential Negative aspects, so I'm just wondering if any of you have some thoughts on approaches that might be valuable to address some of those concerns particularly the Challenges that a lot of the state newborn screening laboratories have around how will we deal with the Huge influx of data that could potentially come from a genomic based approach to newborn screening how will we have the personnel and staff to actually interpret and Utilize these data in a positive manner and you know the the general feasibility and impact of of the possibility of Increasing or utilizing genomic sequencing in a newborn screening laboratory setting I think the question is how much of the potential of Genome sequencing do we want to try and access and If you want to talk about trying to learn anything about everything then it's a tremendous amount of work The test itself is a single test and so with one experimental manipulation You get information about a large number of our genes so there's not additional marginal cost associated with Adding the analysis of one more gene the way there is for a mass-spec test where maybe you need to alter your bio the parameters and and have a different biochemical test but as you implied the cost comes in the interpretation and the analysis there and The the low hanging fruit. It's actually not that much work to do we've been talking about issues of sensitivity and specificity and if we're willing to accept that we're going to be identifying a Relatively smaller percentage of cases, but with a high certainty That's not so hard to do by just limiting ourselves to Genes we know a lot about and where we can identify a variant as being likely to cause disease If we limit our region of interest in that way, it's fairly straightforward We can learn a whole lot more information about a lot of genes than we currently do And but then we're going to be missing a number of cases So we just need to accept that Produced sensitivity comes along with greater utility We'll try to deal with this the phone line Somebody desperately has a question Okay, sir. I was just going to add I don't think Jennifer didn't show these slides, but we did that analysis in the NBC project of trading sensitivity and specificity looking for example only a curated variance and with a And really so we're talking about sensitivity, but also a huge issue is specificity and I don't remember the numbers I'm trying to find the slides, but What was it? So we found that if we used a highly curated list of Variants about which we were quite sure that they would be a disease associated then the sensitivity Plumeted to 55% but the specificity equaled that of mass spec pretty much So so yes, you know, we could do that So finding half the cases of a whole bunch of things that we could never test before but which for which there's benefit for finding I Think is actually a pretty useful goal David. I see you shaking your head Yeah, and I think just from a bioinformatics point of view It's really easy because you just make it a lookup table if x then y you generate a report You recognize you're going to miss kids But if you've got that data available and the kid ends up in the hospital you've got the data available So a whole lot faster than sequencing the kid to know though So I think the reality here is screening is exactly that at screening We have kids that are needlessly dying of conditions that we could detect right now with the lookup table approach and If you're sequencing the kids and you're collecting the data and you're collecting the outcomes You can then actually learn more about the genetics as you go forwards but at some point we have to buy the bullet and start sequencing the kids and I think saying we're just going to sequence three or four genes We missed the opportunity to learn more about those genes So not necessarily owe the data is available for the kid in the future, but this cohort of kids 10,000 100,000 a million kids that are sequenced you get longitudinal data You see if they die you see if they end up in hospital And then you can actually start to understand what the penetrance of things like long QT syndromes are I think a lot Of newborn screening labs are already utilizing next generation sequencing right as part of their regular Protocol for second or third-tier testing. So the idea of Expanding that I think is is definitely within the realm of possibility. You know, yes, it's going to take additional resources to do that Whether every single state newborn screening program will be able to do it I guess I envision more of a perhaps centralized way of of doing more You know genome level Screening but I do think it's it's happening I mean a lot of the new conditions, you know for MPS one and Pompeii disease XALD, you know, you're you have to utilize Sequencing or it's very beneficial, you know to to come up with a final diagnosis And to leave it within the newborn screening system is critical because many states do not Support genetic testing for medicaid patients and there's no point in screening a baby When we can't give the family a final answer and the only way for that to be equitable is if we do all of the Testing from the first screening method to the the last and if that's you know genomic testing We need to be able to do that Yeah, it's Robert Green from Boston In listening to all this We very much in Boston I think focused on item C up there what additional clinical information can be learned from genomic sequencing relevant to the clinical care of newborns and not so much on a and B but It's it's hard to avoid talking about the state mandated newborn screening system because it's the only coherent system in America Where you can actually get hold of all the children and look at them at one place in time if we if we had a more coherent system For public health in other domains You could imagine doing this in a very different way and Reserving the newborn screening system for what it was originally meant for and what it's done So well, which is the acute care of time limited urgent lifesaving kinds of situations Given the however that this is the situation It's a really interesting question. How far do you expand that mandate? And we're all aware of the different advocacy organizations that are trying to add things to the newborn screening things things things there in many cases are Outside of that acute care per view So, I don't know just to pick one familial hypercholesterolemia Would you add that to newborn screening knowing it has no relevance for Really for immediate care, but yet in childhood being aware of it Certainly could in and it might be the only chance in childhood that you're picking those up in a public health way So we can create thought experiments like that that really push us to try to understand What are the boundaries these different systems and where we're missing a system that would actually allow a more coherent elective type of of preventive kind of kind of testing So I'm gonna push back and saying that familiar hypercholesterolemia isn't relevant in childhood or Infancy because if the parent has it and they drop dead that actually really hurts the child So this is one thing we have to remember is that kids are typically born into families that are biologically related to them And so that actually is quite important information. I Actually think we've done a pretty poor job of the last little bit of see which is Clinical care of newborns. We've talked a lot about incidental and secondary findings that may be relevant in later life But we haven't really looked at things like aminoglycoside induced hearing loss And so it's been one of the interesting things when we talk a lot about secondary findings and looking for these lists and adults We haven't actually thought about pharmacogenetics pharmacogenomics of newborns. We haven't really thought about Hearing loss. We haven't thought about things like apology and how that plays with Valbroa Acid So I actually think we've done a pretty poor job of thinking about additional clinical information from a genome That is relevant to the clinical care of newborns beyond a primary diagnosis Well, just to be clear I'm very much with you on that I think I think that's clear. No, but I'm asking sort of what you know There's the systems as they're set up and are really set up along those acute lines But do we do so putting it a different way David? Do we try to do we try to push to expand or redefine a component of the existing newborn screening system to Encompass see all the things you're talking about all the things I'm talking about or Do you say that's that's not gonna work that's that's a different one's a banana one's an orange and We're gonna try to set up a different kind of system that would allow children to be Electively sequenced with their parents permission for all these other potential health benefits So showing my true newborn screening stripes for a second I Don't think many people who are in the newborn screen space went through deployment of skid screening completely unscathed Certainly There was a period of time when certain communities in Wisconsin Completely dropped newborn screening because they didn't want skid screening so they completely opted out of screening completely and that led to intellectually damaged and dead kids from the other conditions that we could detect if We do things to newborn screening that risk hyperthyroidism BKU MCAD and Maybe a couple of the other fatty acid oxidation defects We do a huge disservice to Probably four or five hundred kids a year in color in the US whose lives are saved by that I worry that it Part of our problem with vaccines is we have so many vaccines now people don't see the really vital ones from the not so vital ones and So my huge worry with adding more to newborn screening is the people actually don't get the kids screened for the conditions that we can really make a big difference about and I'm gonna be honest and say that I think Pompeii and Excellency Druna lucidistrophy are going in that direction. I'm not saying that they're And one of the interesting things is the advisory committee says a condition meets criteria or not it doesn't There's no real system for ranking what actually should be implemented first because it's more important and I know that's in the eye of the beholder of it, but I Am I'm really worried that we blow up the whole new ball screening system if we try and put too much onto it I think if you sequencing For conditions that we can treat And that data is available and parents want to elect to go back and look at it in the cold light of day with a reconsent Kind of like what they're doing in Cambridge with their acute sequencing protocol. I Can think about that I wouldn't attain that but I'm scared that people would stop screening. Hi I'm Charlotte Hobbs from Rady Children's Institute for genomic medicine in terms of the processes for newborn screening. I just think back to Before we had universal hearing screening or screening for critical congenital heart disease And when we did introduce those at a population level, then we did have to introduce new processes So although they all go to the public health departments in more states And if you look underneath the hood then in those public health departments, they are handled by very different Groups and the staffing is different and the follow-up is different. So it wouldn't if there was Not that I'm advocating for whole genomes Sequencing for at a population level But there is precedent for it to be a completely different system outside of the ones that are currently available for the metabolic screens So so I agree with you and I I also agree that The newborn screening programs we have enjoy A huge amount of public trust and we have to preserve that trust and That is what enables it to be a universal Program is that the public supports it with us And they appreciate the value of finding these urgent conditions early And and so We have a lot of buy-in For those reasons Another very important thing that I think Cindy also brought up is the idea of fairness That that Newborn screening is for everybody you don't have to be of means and I think it's fine to have sequencing Initiated and opened up to rich people first because they are already taking advantage of it. But but I don't think That That we should take our eyes off of the fairness issue in The population-based screening context and and so that means as as Dave Demak has said you need a whole system not just a test and Even if we could sequence everyone we can't now Follow up everyone and I don't think it's right to have Only Wealthy people follow it up and have the rest of our population left behind Can I just this is Ingrid from at Boston to just kind of build another reiterate that I mean thing part of it Is you're going to detect all sorts of new diseases and children you need to have health insurance They have medical care and so tell someone that they're risk for something and then don't give them the opportunity They don't have the opportunity to do something about it because they don't have the Insurance and the you know these the other infrastructure kind of in in our current kind of health system that's also kind of a matter of kind of justice and fairness also which Kind of feeds into that Richard this is a richard parrott from bergman women's in boston I just as someone who's been a newborn screener In the past life. I just wanted to point out That each of the 50 states Determines what their newborn screening panel will be And so there's no federal even the rust panel is not a mandate for a state to do anything and historically each state has based on The beliefs in that state the financial status of that state have chosen to do different things and I think One may be hopeful example is Is the story of of cystic fibrosis newborn screening which we started in 1999 in massachusetts And was actually the first dna use of dna and newborn screening And this sort of concern was raised then about being forced to do dna testing so some states chose to Not do it and some chose to use an algorithm that didn't use dna and some chose To go ahead and use dna in their testing I think the model in massachusetts has been successful because it doesn't rush right into a mandate with a new technology It introduces a new test As a pilot And the each parent gets to choose whether they want to have that pilot done on their newborn So there are potentially mechanisms for Introducing sequencing The four pilots Disorders going on in massachusetts right now all involve dna sequencing And there's about an 80 percent acceptance rate of parents to To to test for those disorders. So each state could choose to do their own thing And each parent could choose to do their own thing And then it takes time for this to all for us to generate data to show that it's useful and make people comfortable that They're not having the wool pulled over their eyes in massachusetts. It took 10 years for cf From the initiation of the pilot to converting it to a mandate So I think actually looking at those kinds of models. We could introduce sequencing into newborn screening and let The states and the parents Choose at what speed to to adopt this as we're generating more supportive information Robert Well, if we if we agree for the moment that the newborn screening program as it's written now is a sacred trust It's it's got good social justice Legs and we don't want to mess with it because we don't want to then how do we go about introducing even the top layer of potentially life-saving Genomic elective genomic information into the lives of children What what's the first step? Do we have to wait for an an expert group within the pediatrics community to say The prices come down to a hundred dollars for doing 10 genes and in good interpretation and We now recommend that every pediatrician should send to that is that It's a hundred dollars. It's still not going to be socially just for everybody, but it's a lot more affordable Do we wait until it's somehow The state systems are invented to subsidize it. I mean Somebody draw me a blueprint for how this gets started In a way that that respects the value of the information But it's also acceptable from at least a moderate social justice point of view. So so Robert there is A mechanism and I think it's pretty similar in each state Which is each of those 50 states has an advisory committee To their commissioner public health and then you need a champion. Oh, but you're talking about the newborn screening, right? If you want to work within the car I'm talking about actually in putting this outside the newborn screening system a whole new system for Screening children through genomics. I mean you've convinced me for the moment to stay away from the newborn screening because babies will die If we mess it up, right? So for one thing It's universally mandated in pediatrics that somewhere around their first birthday a child needs a hemoglobin and a lead level And and I think this is Nearly universally done Pediatricians all know to do this public clinics do it And this is a time when Babies even if they don't get a whole lot of regular medical care and checkups are going to be Encountering the medical system to get this done. I actually think that is an opportunity That's a time when the family can Think about consent in a calm, you know environment. That's not That's just having had a baby and I think we should really pursue Whether that could be a time. I'd love to have another Insight, you know batch of funding to look into Could we look at one year olds as a as a target For some of these things which might have a bit later onset We could make a huge difference. We could use the new technology and we could get buy-in Love it. That's a great idea At the risk of contradicting myself Which I often do The challenge we have is that you know when you look at this list of conditions that I think Alex talked about a couple of advisory committees ago The list that Genomics England has the the list that's on the NIH website of conditions That meet current criteria That's to say they have onset in the first 28 days of life. There's an accepted treatment that is available There's a task which is sequence-based It's all good and well talking about what we're going to do when a kid's a year old But we have kids needlessly dying right now And if we're going to sequence the kids and we're going to look for 130 conditions To be honest doing whole exome is probably actually Just about the same price So I don't think that we necessarily have to have The generation of the sequence data And the interpretation that sequence data Connected there are conditions that we need to know about In the first week or two of life There are conditions that we need to know about within the first few months of life There are conditions and so the This issue of long that you know records that are tied to you as an individual Yeah, you can solve things like bitcoin that just get crazy But most states get called by Athletes to get their sickle cell trait Status so the the newborn screening lab is actually used to this recontact from families. It's clunky. It doesn't work great But there are informatic solutions to this So I think we can sequence kids at birth look for what's relevant at birth And find ways of keeping that data tied to that child I want to I want to come back um to follow on your point david to ask them about So would you accept then a lower sensitivity? And how would we determine that for a higher specificity so that we don't have the burden of false positives, you know for pku we missed an m cad we missed five to six percent for each in our cohort And we would say that that would be unacceptable So we would do tandem eschatometry for m cad and pku because it's better than dna sequencing and You know as melissa pointed out this has done gene by gene disorder by disorder currently on the rust how would we retain the public's trust And be sure that we were working within our own You know it's sufficient knowledge to Go forward for for an individual disorder that that you might choose on that list if you have an example Can I actually quickly respond to that? I mean, Bernata? I think you're making a presumption that there is a better test available in the course of pku There is this available screening test with high sensitivity and high specificity I think a lot of the conditions we're talking about are conditions for which there is no biochemical test No other better test So then you accept lower specific lower sensitivity And your limit has to be how many false positives are you willing to accept? So that's actually exactly what i'm asking alan because your specificity and then see what your sensitivity is And you just have to accept that whatever it is That's fine But but I guess what I want to add here is just the the public trust because then you are going to sequence You are going to look for disorders knowing that you're not going to be able to identify everybody all the time now We return results negative results to families and have to explain to them Just the fact the fact that we didn't find something doesn't mean you don't have it And that it can be a difficult Concept for many people and it does need explanation to some extent But I don't see any way around that But I think again, you know, and such an important point of what we've discussed here is the difference between a Phenotype and the absence of a phenotype and newborn screening is done in the absence of a phenotype And there's no relationship really between the Screening it's done out. It's automated. And so There's no there's no physician there. There's no Context there's no counseling And the parents didn't ask for it exactly So I think there are ways to handling that You know, I think you can say, you know OTC we some states actually say they screen for it and they Miss more than they find Probably they find about one in ten kids with OTC and miss nine out of ten So we we're doing this currently. We just pretend that we're not missing kids. We miss pku when they're premature But the deal here is that if a child gets sick and you have the ability to look at the data again The public trust you have is to say this is a screening test. We're going to miss a lot of kids This is better than what we're currently doing. And if your kid ends up in the hospital We can look at the data again, then you maintain the trust that And I think getting to the social justice issue here if I can just step on that pedal again You know, the most common URI cycle defect in the world is citron deficiency. It's really common in east asians We don't screen for it Because it needs a DNA test Um, you know, there are certain conditions that are common among certain people groups that cause them a lot of problems that we don't screen for because We don't have enough cases to set standards But we know what that population is found to mutations are and they're treatable conditions. I I think The enemy Of good is better. And I think we just to miss Walter I think we have to get over the fact that we're going to miss stuff We we miss stuff on ubuild screening now It's life if we pretend that we're not missing stuff on ubuild screening. You haven't sit through enough flagged cases I think that you know, we just don't know whether this is going to be a good thing or not It might be it might be, you know, really beneficial and save lives But we just don't know especially at a statewide level You know, when we first started tandem mass spec, we had You know, at least two statewide pilots in massachusetts in north carolina that showed this was a good thing We picked up and saved the lives of you know Many children with m cad who otherwise would have would have been missed and died Um, but we had that data, you know from statewide pilots and I think and also with You know that expansion with tandem mass spec Because most states weren't offering State-level newborn screening there was a commercial lab that offered it and that was direct to consumer testing I mean grandparents were buying it for their newborn grandchildren and things like that and you know, eventually it It expanded to you know states being able to do it So, I mean, I really think Probably there will be some Innovative states that that take this on and and do pilots and provide us with the information But I would hesitate at this point to say, you know that that it would be a good thing to do on a population level Can I just ask you if I understand what you're saying correctly In in a thought experiment if you could do it cheaply And you could as people have talked about reduce your sensitivity to get very very high Specificity and using only variants that are well associated moderately highly penetrant For diseases of childhood that are actionable in even a conservative sense of the term You wouldn't apply that to a population at this moment in time No, because I think you know, we I know that's a fictional situation thought experiment Well, we we just don't know enough about it, right? I mean it we don't know what it's going to take We don't know what the penetrance is. We don't know what the You know follow-up is going to entail who's going to explain this information to families Um, I mean it it may be something that yeah, the primary care providers will be able to To do and do the appropriate follow-up and everything but you know, I really think We we need more data before we go down that path Well, the devil is in the details and there's Tremendous heterogeneity in the issue that we're grappling with I think because you have to go gene by gene But there are so many examples of really well understood Conditions and so for example, if you have a deletion of your dystrophin gene on the x chromosome We know you're going to have muscular dystrophy And we now know that there are things you can do early on they're going to that are going to help So that's a subset of findings that you could make that would lead to direct intervention We heard today that for fennel ketenuria that's not necessarily the situation so finding variants there We're not certain what the outcome is so we don't go there. We don't report those on a population level But my sense is that we've gotten to the point where there are so many knowable things that can have benefit Um that we have people have to understand there's no guarantees in life There's no hundred percent certainty on anything and that we're going to be missing things But that we still can identify and report Things where we're highly confident of the implications for what we have found And I don't see much reason to wait on that other than to establish what those things would be So I think what we learned from Pompeii what we learned from pretty much any new condition that we've added hearing loss Is that the piloting in a couple of states To to learn where you're going to stub your toes and what once again new world screening is a system. It's not a task And so you need to figure out what you need in place to catch the babies And so trying that out somewhere first And I do think you have to actually probably go gene by gene condition by condition because The individual that follows up with spinal muscular atrophy is different than the individual that follows up along qt And so getting them into clinic is is different So I do think there is an element of piloting Gene by gene but things if you're getting all of the genomic data You actually can start to understand penetrance on your historical controls that you have the genomic data on that you also have one two Three-year phenotype follow-up I think david I would agree with you that it would be ideal to have a system that that Was ready for all that complexity But I don't think we have one and so then the question becomes if if you believe as as as I do what ellen just said that There is at least a subset of genes and variants That we are fairly certain would have a great deal of positive impact on children's lives How do we give it back? How do we attest for it and how do we give it back? What what other system? And I agree it should be a system. Do we put in place? So so let's take jennifer's suggestion for the one year mark What if you just to be speculative? What if you said We're going to do a pilot where children have an opportunity To a parent their parents have an opportunity to send for a relatively inexpensive panel of How many of many genes 20 50 100 and look only for high specificity fairly high penetrance variants And that's going to that's going to go into effect with that and the information will come back to their pediatricians You're right. Some of them won't know what to do, but some of them will get figured out Could we get behind that? I think when newborn screening works is having the expert on the phone with the pediatrician who then conveys the information to the family And then then they follow up with the expert I Would you endorse that kind of pilot that kind of an approach? I would love to try this out And I think we should be trying this and saying would other people at this table We should try this as a newborn. We should try this at a year. We should try this at five years We should see what We don't know what the optimal time to return a bracket one result is There's some suggestion that if you grow up always knowing it it's less stressful than suddenly catastrophic be finding out when you're 20 But as you just pointed out, it's a lot better than for you the affected parent and know it than lose that parent So the only way to do this is to do some real science that requires follow-up So on that note Since I think we're at the five o'clock hour I want to thank all of our panelists and Participants who led to a really robust and rich discussion. Thank you for your time Thanks to those of you who phoned in and joined us on our live stream and Stay tuned for you know further results and outcomes from this very productive collaborative set of projects. Thank you