 Good afternoon, so I'm the moderator for this session and I wasn't really prepared either But it's just been a fascinating day of discussion So much is going on in my head I guess I think it's been really nice to hear from Bruce in particular about an example of How you can have a clinical program set up to kind of mirror what's going on in terms of just that? Diagnostic evaluation and maybe even that after word follow-up So I was just wondering Bruce maybe I can ask you and then we'll open it up to everyone else But I'll start with a question You mentioned it would be valuable to have access to the core facilities the model organism metabolomics Are there other aspects of the network that you would want to be able to connect to? To enhance what you have at UAV I Don't think I can think of an aspect of the network. We wouldn't want to be able to connect to if we could Not not even knowing what all the moving parts honestly are, you know The obviously we don't solve every case and we'd love to get into a situation where we could share our Unsolved cases with others and perhaps learn from them So that would be obviously one thing and all the various functional studies because you know We like everybody else often come to kind of dead ends where we we see something that we think might be meaningful But it's pretty difficult to take it to the next step to establish for sure that that's the case So does anyone want to I'll just open it up to the questions are here listed The first what are the end goals? What about the idea of broadening the Objectives that already exist for the UDN and including that mechanistic understanding and and ideally leading to treatment Maybe how are you how are you thinking about that? Or are there other objectives that should be added to the list or should there be anything that should be taken away? Yes well, I think Now I have to admit a bias in this but once we're sequencing the genomes of all newborns And querying that information, you know once Symptoms start or if there's Conditions that are actionable will start treatment, you know pre-symptomatically But you know we'll have far fewer undiagnosed diseases, you know that have a genetic basis at least But beyond that I think you know speaking as a clinician And Partly, you know given what Bruce has demonstrated with being able to do one of these at his own institution that it You know it may take some money, but it doesn't take huge amounts of resources to do this and you know as a clinician Some of our limitations are certainly the lack of insurance coverage for doing Exxomer genome sequencing and as well as you know some of the more specific areas of expertise So I think you know trying to develop a broader network where you continue to have, you know larger centers of excellence But you know to utilize all of us that are doing this every day, you know look trying to figure out diagnoses and patients And you know if we had more resources in terms of you know a place where we could send a sample for genome sequencing I think that would be you know extremely valuable and increase the amount of You know patients that could filter in then for some of these other Components, you know whether it's the basic science research or looking at You know model organisms and what have you Yes, I guess I would first I would echo what Steve said during his presentation about the concern of the mission Cree and Focus on the idea of what can the UDN uniquely do it was designed to come up and figure out ways to diagnose undiagnosed conditions and We've been working for several years, and I think we're all still trying to figure out. What's the best way to do that? So I think the next cycle should be improvement on How we do that what we're already trying to do it was with the primary focus And then thinking about how to to give the the long-term viability of it Whereas bringing in a lot of other things to it, I think we would know you know they're ten years to figure out how to Keep those viable. I think so from my perspective at least it seems like Focusing on what the charge was initially and thinking about how can we more broadly create what they did for instance in Alabama? And and how there's other models for places so then you're really exporting the methodology of the UDN making it more general But not trying to take on all these other roles which are valid and important, but Really beyond the scope? So the UDN right now has an infrastructure There is a manual of operations as I understand. I haven't read the manual of operations But I'm assuming it has a set of core functions processes activities But I also understand that the clinical sites Have developed their own implementation strategies around those core functions And so maybe what needs to be studied are the implementation strategies Even understanding why one site has opted to do X and another Y To solve the same core function or process as you say so that we can understand, you know What works given one set of resources and what works when you don't have that around So that people could have a package or toolkit to Select from based on what their local needs and resources are It sounds like I'm involved in health services and implementation research and it kind of sounds like that's where you are going Yes, Judy It seems to me that Most of what has happened have been single gene Disorders so there are a couple of people with two gene disorders because they're consanguineus But what's the plan for moving toward? Multi-factorial disorders, I mean are our is anybody looking at modifiers For severity is is there any way to think? Forwardly about I mean, maybe it's really old-fashioned to think that there is multi-factorial disease But I kind of think there probably is so as is there any one of the Groups that are doing additional modifying factors within the genome So I think that that raises an incredibly important point that There's really two activities of the UDP that maybe could be separated from one another One is making a diagnosis if a definitive diagnosis can be made, right? You know, there's enough information in the world So if you find the right clues you have the right knowledge you can make a diagnosis confidently and return that to the Participating family the other activity is taking everybody else Which is the majority of our patients as far as in our in our hands Anyway and moving the story of those families forward either by determining new mechanisms or by Sharing data so that you can so that you can generate cohorts And so I think the second of those two things is really more in the category of what the UDN can do Uniquely and falling into that category are going to be things like disease causation Mechanisms that we're not at the point yet where any combination of acute. I'm sorry astute clinical experience and Existing testing is going to prove to the point where you could tell a patient that could test a pregnancy with it So so for instance it seems to me that the whole business of some Drugs have bad effects in some people and you wouldn't be taking those drugs unless you had another disease cancer drugs for instance and At least our cancer agency is now testing everybody before they get the drugs To see if they would have a horrible reaction and there's some component of that the interaction between The genetic disease you have and the potential therapies. It seems like it should have some Potential can I ask does that get back to the argument of maybe not having a large enough Population to work with to dissect this stuff out necessarily, I mean That's how in worms and flies in yeast people have worked out all the pathways They start with single genes and look for suppressors or enhancers and so this is essentially the same thing So once you have a database where you have all the mutants And that's what they're already doing in yeast is making double and triple and quadruple combinations and seeing How dominant the leels can interact with each other And so it's a matter of funding the basic research that promotes the understanding so that when you find your gene and it's in a pathway You know what other genes to look for I completely agree with that But I'd say it's a slightly different question if you say like should the network you set up to do to try to find them in patients I think you're I think there it's you're right This is the scale is not even close to correct here that you could hope to find a second site suppressor or whatever in the patient Population you've got to do it in the model organism and then go back and see if it's relevant in some patients And if two candidates would be identified in a patient and you can test them both in a model organism you can just look for them, right? But I think it highlights the need for a lexicon that allows you to move efficiently from patient to model Bottle to patient. I think the fundamental problem is you know thick heart or big liver Doesn't translate and worm and fly in the way that we would like so we need to have much more Granular clinical phenotypes to be able to make any sense of most of what we do in medicine There is a yes and there is a no the yes is you're right, but there is also a no You know flies biologists discovered the notch pathway and and if you find a wing that's not you think hard invertebrates because you know that The notch pathways involved in hard So you know you can translate these things pretty easily if your knowledge is there And in fact there is databases now to categorize all the phenols and all the phenologic phenotypes So that you can translate quickly from one to the other No, I completely agree. I just think it's that's a very fairly modest subset of the total pathway space That's all I'm saying. Yeah, are there are there other things that the UDN can do uniquely do? Yes, so I was Knowing I was coming as racking my brain for weird wonderful new kinds of diseases and it seemed to me that organelle wear out might be one of them Or organelle mismatch in other words within my genome I can make endoplasmic Reticulum, but I like wear it out fast or something like that and what I'm wondering is With the biopsies that are being done or with cell material, is anybody kind of thinking on an organelle level? That requires transmission electron microscopy and systematic TM and I would be very useful. It's our main diagnostic to So that brings up the issue of what's on the horizon. I mean are there going to be new diagnostic Like clinical diagnostics like ultrasounds going to become even cooler and they're going to be new DNA Diagnostics, I think so is there any Subgroup that's out there looking for the potential application of new stuff This is a it's a huge industry at the moment is is new diagnostics But I think to to get back to your point and this is one of the reasons for example Our site has chosen to really go sequence first is if you have something that's in an ER pathway You're going to go and take a biopsy Specifically to collect an additional sample for transmission electron microscopy because you're going to actually want to look at that in the organs Of interest, so I think I mean it gets to the fact that route You're just having a stamp that I suppose the real question is what's the definition of deep phenotyping? You know one person's deep phenotyping is the next person's Superficial quick lookover that's the fundamental problem with all of clinical medicine and until we have some fundamental Ground truth. It's very very tough to move forward because it just depends on who you happen to see first That's the the whole of clinical medicine is completely based on who did I find first many of these diagnostic Audices would never have occurred if they'd if they'd happen to see the right clinician on the first day Yes, I Think one of the other potential deliverables is around Kalem and I were talking about this Before lunch, which is you know when we think about evidence-based medicine right now We've got you know the medical students thinking about randomized control clinical trials And that's not going to happen in the space and so I think one of the challenges is is that animal models are bad There was that editorial a few years ago about how mice have done more to hurt Drug therapy than anything else. These are the barriers were fighting against and so I think part of what can be Adopted in this is how are the physicians who are using the information making decisions? And why are they willing to use a zebrafish or a mouse or a fly to help them in the outcome? Because I think in absence of that as we've talked about earlier what we're going to have is an amazing pipeline Amazing data and no one's going to use that so I think that's another place that the clinical sites could maybe be helping on is Why are they using that information and when it's published? Is there a way to capture that as to why that choice was was valid? Obviously having a hundred other cases is ideal, but what do you do with one or three or six? I? would argue that Those of us who do both both the basic research and clinical research That's a very relevant question most of us be interested in but most clinicians are not And so that's not going to be a relevant deliverable for them I think more important for them is how do I figure out? Which patients need to be getting into this type of network infrastructure sooner? So they're not wasting a lot of time and then two is you know What is it we're going to provide back to them that's useful to them if we're going to hand the patients off? Not a simple question, but but I just think that's reality. So so that's the reality that scares me Okay, so there's 20 million undiagnosed Americans if we all Started working 24-7. There's not enough knowledge base to be able to do that if we can't translate this out to our medical Colleagues, it's a failure. I mean we're helping a handful of patients But we've got to really think about how to get this out to the masses or we can't get there So you're speaking about again kind of Implementation outcomes what the six sites are currently doing how they're getting there. What's working? What isn't working to get to a specific outcome so that we can develop a package that could be exported out Is that? Maybe I mean it's more of it's a logic series, right? What data will you use to make a clinical decision and what will you use for that information? So is what data do you need? Yeah, that's the question right so so right now the answer is it depends on the physician Right, so that would be the argument to have more standardized for for the UDN right now Perhaps over the next five years to have very standardized approaches to how patients get in what happens to them once they get in and Some kind of follow-up maybe what's currently being done is satisfactory But and everybody doesn't have to do it the same way, but just using instruments data collection So that we can look at these and Evaluate the processes is that I think well, so I don't have a better idea I think hard you mentioned it earlier I think it may have been Bruce or maybe it was David who mentioned it a Learning system is what you actually need and so that requires you to have Essentially some shared minimal data set Some sense and those minimal data sets could be things I mean I think bill is advocated very strongly and I believe it's very important those data sets could be you know The number of specialists that you've seen before and what those specialists are that information may actually be the best predictor of Coming through a network like this has any chance of efficacy whatsoever, but just having some Set of data at the very front end including I believe some limited phenotypic data. I think if there was some Standard phenotypic matrix that everybody had to fill out and that you know included just a general exam from a from the clinician That's referring them. I think that would actually be a useful start Yeah, I'd like to make just a few points our protocol allows us to do medically indicated Invasive studies and fibroblast biopsy But it doesn't allow us essentially to get a biopsy specific to a variant that we found on X-ome or genome sequencing so for example We couldn't do a liver biopsy just because we found a liver specific variant and go go after that not by our protocol Now we could change that I suppose I was imagining we would take it next if we're going to take a biopsy We'd take an extra one for EM or whatever if we knew what we were looking for at a cellular level fair enough But it needs to be medically indicated to do the invasive procedure to begin with but I like the idea of having an EM core I think that would be spectacular. It's so difficult and so expensive for us to get electron microscopy We don't have you know decent electron microscopy at the NIH that I know of hope no one's listening but And the other possible core that I've been thinking about a little bit is a stable isotope core for metabolic studies In the very special cases in which we need to find out how the whole organism Manages a particular defect that we find and boy if you had that that would revolutionize I think some elements of clinical research the way the CTSA's don't any thoughts on Outcomes from the UDN and best strategy to optimize the outcomes and I think that relates to Well, maybe maybe not the next question about The sustainability. I think I heard earlier today that in order to be sustainable We need to somehow demonstrate value of the process Thoughts on that. Yes, I just wondered Sorry, go ahead. I Wondered if we could ask Kevin to To say a little bit more about your partnership with the CTSA at UAB I mean is that if you said that you're you guys function on two hundred thousand dollars a year But you are also functioning within the context of the CTSA. So can you help us understand? Or I think that's what I understood you to say. Yeah, how does that work? So It's it's not money that they're providing us it's I guess Networking to Investigators on campus. So they have their sort of tentacles out to pretty much well and about all but a very high number of clinical and basic science investigators So we we use them some to help disseminate Things and attract other people that we can right help even identify people that we can engage to Take a finding to the next level. They also have a set of partners the way they're they're set up is there are partners in the surrounding states, so Louisiana and Mississippi and we partner closely with Hudson Alpha actually as well and So that all happens in the context of the CTS that was how we were brought together with them So we're trying I don't think this is moved as far as I'd like but we're trying to open up the opportunity for Patients to be evaluated With some of these partners in Mississippi and in Louisiana particularly no so I Hear the potential because of an example like bruises Of being able to develop a strategy wherein you could create additional centers that could Learn how to leverage some of these resources if they had access to some of the core functions that you're building They're so critical to to discovery It's really hard for one of us when we get what looks like a really causative gene To come back and say well, we're pretty sure that's what it is We don't have any way to go and find the experts who can help us with that unless you're much more Enterprising as an individual than most of us have time to be so one of the things beyond the expertise that you're developing in how you approach these problems is building a strategy by which Others might be able to access some of that information and broaden it a little bit It makes it a little bit more for the common man But that just increases your reach if at all possible So I'm not that I think all of us and now we're gonna run back and try and set up our own little mini UDNs Because we can't but if we knew that there was someone we could call and say, you know I have this great patient and has this great gene, but I don't know how to prove what it is You might be able to extend the reach just a little bit further because you'll have that much further down the road In terms of the discovery Yes, I fully agree But the resources are pretty limiting already and in fact I get calls from physician almost weekly who want to jump in and are not part of the UDN and I can't cope with the numbers of the UDN so I Invest three times almost the amount of money I get from the UDN for this project and that's because I'm a Howard Hughes Investigator I can do that, but it's impossible. It's just It's easy to do but it's labor intensive and you're gonna be able to deliver And you're speaking about the core the core support Yes, John Actually hadn't calculated it that way, but I suppose you do the arithmetic that may be right Well, it's important to know is it Scripps clinic or Scripps research I don't know I'm not familiar with No troubles is on the medical side the clinical side, so it might be a business development scheme by Scripps But well certainly we do functional genomics on things but in collaboration ours are all in the cancer space currently I mean there are half a million unique mutations described in totally in cancer patients without No, no functional significance are maybe no maybe unknown So there's a huge burden in all types of diseases that's that's there now And so it's not unique to this set of patients that there's this backlog of information that used to be accumulated But in cancer research many of the drugs have been developed because people understand the basic pathways that are leading to the cause Here we don't have that and so that should be a deliverable that after During the UDN functional genomic systematic functional genomics should be part at least for those key diseases where you can't interest the rest of the community To do basic functional and annotations and so you can start thinking about deliverables on a drug That's actually one of the was one of the aims of our proposal was to use pathways So that short of actually getting a mechanistic Gene variant you could at least identify which pathways were perturbed Well again at the risk of being redundant. I think it's crucial I mean there's lots of good ideas, but it also strikes me that particularly given the example in Birmingham and the comment earlier That when everybody's sequence to birth you one could imagine that this is this is going to devolve into different pathways Undiagnosed diseases are going to go away Because they're going to be diagnosed or that this set of patients is going to go away. Okay, probably not But maybe largely and that you mean that could be you could declare success then right this is as works You know that and secondly it's going to devolve to being run as a clinical enterprise at different places all around the country So maybe part of the strategy should be to try to engage CMMS or somebody who has the it's their responsibility to try to create funding mechanism for the clinical care of patients and You know, maybe it could be defined something like in-stage renal disease where there's a mechanism to pay for it The $2,000 a patient or whatever it's going to be but it seems to me that it's going to be a continuing effort at NIH There has to be an unequivocal research component to it now and it's you know What's the model of how to diagnose it has been established? That's no longer a search project in my opinion That's established now Or maybe there'll be some some refinement of that ultimately But you're going to have to go someplace else and I think it's this functional genomics mechanism Ultimately therapeutics potentially So so do we understand that clinical model for that diagnostic pathway so that we can have that pipeline to get you So where you would like to be with figuring out the mechanistic pathophysiology? I mean, I guess that's where some of the clinical people around the table are wondering if we haven't really quite figured that out yet or that we can't That getting reimbursed for providing those types of services has been challenging and that's why the UDP was established in the first place So we're kind of this vicious cycle So I guess I would propose I don't know if anyone else would agree that one of the maybe aims over the next five years is to really Figure out those, you know diagnostic pathways What's most efficient as we heard the previous speaker comment on and then try and get some measures of Outcomes that maybe relate to quality of life or real health outcomes avoiding Treatments that were incorrect because of a misdiagnosis Just trying to quantify all that with a program like this so it could be you know You could sell it to a University or CMS Yes, Cindy. I'm recalling a presentation that we had at one of our recent North Carolina medical genetics meetings from physician at Wake Forest University who started a complex disorders clinic That was initially funded by the parents of a child who died After being on a diagnostic odyssey and you know not figuring out what the underlying problem was until they were deceased But she brought together, you know a group of experts who just meet on a regular basis They don't have a genomics core anything they can't get exome sequencing, but just you know having the experts talk to each other They were able to you know show benefit and in making diagnoses and actually now their their hospital is Paying for that service because they were able to show, you know a cost savings to the health care system Yes At Duke we have a parallel undiagnosed clinic that's funded by the Dean and we have the UDN So I can offer a direct comparison between the two we also get some funding just like you do at UAB We can a genetic counselor that's funded the Dean subsidizes some of the sequencing if you're not able to get insurance coverage for the Sequencing so what's been the challenge there compared to the UDN is the time it takes So to get for clinical consultations for a patient could take up to two to three months sometimes because we don't have a Focused one week consultation that can happen We've not been as successful as you have been in having buy-in from all of these clinicians Some people say oh my clinics for this month can accommodate they don't do that with the UDN We get a lot more buy-in The second challenge is of course insurance coverage any of us who sees patients know that that's a problem And the third is on a shoestring budget the genetic counselor is stretched to her limit Just to coordinate things for the family to bring them back in or to you know, you need disappointment You need to go here. So that kind of thing. So if you're talking about sustainability of the UDN I loved Rizwan's idea of having you know Centers of excellence may be going forward But perhaps these kinds of challenges should be taken into account when we are talking about how we can make the sustainable Maybe there's got to be some institutional commitment and some funds elsewhere that in combination will let this work Go forward with a reasonable amount of you know less money But a reasonable amount of outcomes that are sustained for the patient the other challenge Of course, we also face is what I think one of Susan said we don't have any base of pursuing candidate genes other than Reaching out to individual investigators and begging and I think that's been there's been a lot of discussion about that Yeah So maybe as I think also Ruslan was saying right like having a white paper around best practices You know if the American College of Medical Genetics and Genomics would maybe have a their face on that as a professional society of You know how best to approach a patient with an undiagnosed disease that might go to a certain extent You know far enough to convince a health care system or maybe a payer It's better than where we are right now Eric no, I mean I have I totally agree if I've had similar experiences with My more limited field of reproductive Disorders of reproductive development when there is a consensus statement or guidelines or best practices and not talking about standard of Care's but insurances and hospital systems do buy in if it says in the consensus statement that there should be a mental health professional at Clinics with patients with the DSD hospital systems actually like okay Yeah, we don't want to be liable not to do it. So there is more it puts pressure on hospital administrators, but also on on on insurance companies, so I However imperfect those guidelines, maybe I think it would be It certainly would be a good outcome of the of the UDM Yes, so So picking up picking up on that thought people said earlier one hybrid model could be what UAB is doing and so looking the eight years down the road, maybe if Maybe the program could be configured as in a way that the cores are supported by NIH, but individual centers are Take part in it based on their institutional commitment So because the idea, you know if you look 10 years down the road or eight years down the road Maybe they're a different way to look at model organism, but we will still need model organism So the idea of there would be that we have the enough in a fantasy world, I guess that we have actually a Model organism core. We actually have sequencing core We have a metabolomic core or another core that we deem necessary, but the centers are Applied to NIH to get the designation based on their own funding and of course, they'll be working at a limited level, but Something to think about I think And that might foster some of the collaborative efforts that we heard today You know and in particular even with patients and families and I can see a lot of that being part of a center of clinical excellence around this So when we just have a couple minutes left in this session is the vision mission for the UDN Different for the future than what it has been Can I ask a question about? One do you think you'll be able to diagnose 80 or 90 percent of the patients that are I have no good feeling about the number. I never think it will happen in five years. So Maybe 15 or 20 and even then and so if the UDN is an eight-year kind of lifespan We're gonna be maybe halfway if we've grown 4% in In three years, then it's gonna take a long time before we get there So I think there are two things I think there are the undiagnosed diseases which I think like our patients Let's say with rare Noon and sphenotype and then there are unknown diseases where we have a rare variant that has never been associated with disease. So I think While we may be diagnosing a lot of undiagnosed disease We will be nowhere close to 10% in diagnosing unknown disease And I think the other element is that there are lots of diseases at the moment that are Undiagnosed and unknown that are lumped together with things that look like they're known And we love them together just simply because we have a generic treatment that vaguely works for each of them So diabetes coronary disease heart failure. So I didn't get a precise answer yet What I'm saying is it's an it's enormous is I think you're 100% right you go I think it's a it'll be a long time before we sort it all out Udn need to end in far five years or eight years or whatever it is If if it's clear that this is not gonna You know be ended by that. I guess the issue is it not so much that it may end But what how might it transform you know to continue with funded sites as they are now and or I think one of the suggestions was to maybe have these clinical centers of excellence that could do much of that diagnostic evaluation component and Then with the support of the course funded by NIH. I think that's what I heard As an option all right, so anyway, thanks for David pulled up the objectives the the current objectives just to remind everyone and I One of our questions for this session was should this change I think is this still the right Target thoughts You know, I think the point I was trying to make before is that Maybe it goes under number one But one thing I believe the program can do is to help establish the paradigm of how these things are approached And I turn I actually prefer the term evolve to devolve by the way But in any case, I think we can evolve to a situation where there are Many programs around the country that are following the lead and implementing this into day-to-day care while the cutting-edge part of it can Remain in a kind of hybrid research clinical environment I guess the other thing which maybe it got covered early this morning when I wasn't here And I suspect everybody deals with this but for everybody with an undiagnosed disease It's probably five that are sure they have an undiagnosed disease and How you actually sort out the wheat from the chaff is a I think an important question Okay So so I don't think we have to change these goals because we don't have enough experience We barely started six months ago So I think as somebody earlier mentioned I think our goal should be going forward in the next five years of the funding cycle is to actually refine These goals and really figure out where we you know where we are But I don't think we have enough data to actually now think about changing these because Like you said, every site says site has looked at a very limited number of patients And I agree with that but these are corporate goals None of these things up here tells you about the individual patient. That's in the room with you when you're examining her I think we shouldn't keep our we shouldn't forget about those Individual patients in the one-on-one and taking care of that person because all these things are Systematic, how can we do better as it as a group as a nation as a program? But we have to take care of the individual first and not let all these other things get in the way of that But I think creating a model that can be Put out through the rest of the country so I can reach more Individuals it is a goal that we could aspire to that would Is that kind of under item three or is that a separate? Yeah, I think I would just add Clinical to so I can integrate it and collaborative clinical and research community Okay, and these are the shortened versions of the objective so that they fit on a slide not all of the language That's actually in the FOAs for each of these. This is like the summarized person So the idea of this being clinical and research is in the full version. I would just follow up I guess on Christine's comment and and Anastasia's comment. I guess the idea of talking about integrated collaborative Clinical basic research whatever there is I don't really see that the charge is optimal management It seems like that's what occurs after you figure out pathophysiology and you're thinking or some point that that's not really our focus, but I'm wondering if the and care and item one could kind of encompass the optimal management, but John's telling me we're out of time. So Thank you all Okay, so thank you everyone We've made it through our six questions at this point in time We will have another 15 minute break for folks to come back here at 315 at that point in time. We will have our panelist discussions We're going to run through all five of our speakers in a row and then have a little bit of time for some more discussion Before we move on to talking about prioritizing what we've been discussing today I have heard that a number of folks have flights that they need to be able to catch leaving around 5 p.m. Or so so we are going to try and get some of these discussions bringing together the Prioritization discussion with the draft recommendations and condensing those together to try and get done by five if we possibly can So if everyone can make sure they're back here at 315 then we can try and get out of here a little bit early for people to make Their flights. Thanks