 Thanks, Ebony, and thank you all for listening to this presentation. So just to give you a brief background, this workshop grew out of a part of our Ignite project, which Ebony mentioned earlier on monogenic diabetes. And one of the things that the RFA asked for was an economic or payer perspective, and I was fortunate to have on campus Health Economist Daniel Mullins, who's here. And when we sort of talked about how we should look at this, he said that it's not just about looking at cost effectiveness, which is sort of how sometimes people think about economic issues in healthcare, but that we really needed to engage with the payers because they're all different. They have all different perspectives. They're all different models. So we really need to have conversations with them in order to then be able to take action to see what needs to be done to make it possible to get genomic medicine reimbursed in a sustainable way. So we have a panel, a small panel of payers who we've been working with to get input to guide our study. Out of this grew, we were awarded a supplement to have a larger payer meeting across. It was originally going to be an across Ignite meeting, and then it was really an across NHGRI meeting because there's so much interest in this topic. And so the title of the workshop ended up being unifying the evaluation and implementation of genomic medicine, but I have also known as the payer engagement workshop. That's what's listed on today's agenda. And that's because the purpose of this was to engage payers, but we also, as we were coming up with the title and the agenda for the workshop, we really wanted to make it clear that we're having a conversation and that everybody has an agenda in this and we need to kind of sort out what our agendas are and we need to talk about it and communicate with each other. And so it wasn't, it is about engaging payers and that's a really important component and that's ultimately a huge key, but it's really all of us working together, figuring out a way to work together in a sustainable way. So after several months of calls and group calls and individual calls with various people interested in this, this culminated in this meeting that was just less than two weeks ago, two weeks ago Thursday on August 18th, we had about 100 people were involved in the meeting either at a very large conference table or calling in remotely. And so we, the group consisted of NHGRI program staff and attendees and patients, a couple of patients who were impacted by genomic medicine, both care providers and technology providers, researchers and of course the payers. And our panel of payers, we had six different payers, although a few of them are from the blues, they're from different parts of the blues, which have very different perspectives so you can see we had, we can see the list here that we had including Deborah Smith who was also on our planning committee from the federal employee program, Blue Cross Blue Shield. So just to the format of the meeting was, it was organized into case studies and also into roundtable discussions. So there was first, we had a brief introduction from Eric Green from the about on Ignite and genomic medicine as he mentioned. He just did something similar a couple of weeks ago. And then Daniel and I talked about the objectives which were developed with Ebony and with the committee and you know because you know it's really again it's a matter of talk and then action. And so these were the objectives we came up with. We want to build a process. We know there have been meetings like this before and we wanted to figure out how can we start to build a process that's going to be ongoing that engages all the stakeholders in evaluating and implementing genomic medicine to understand what evidence is needed and how it should be disseminated both to the payers and then ultimately to the patients. And then to identify what protocols both in research and in the clinic should be used to generate the evidence. And so we had, first we had a keynote speech from two people from Vermont, from the CEO of the medical center and Health Network which is gearing up to be a single payer, provide single payer health for the entire state of Vermont. And the main points that Dr. Rumpstead made was that an investment in genomics is a powerful tool to improve patient care and control the cost that can be linked to quality parameters obviously if done in a strategic manner and that they're actually looking very seriously at in the not too distant future a statewide whole genome sequencing plan in order to assist in developing lifelong care plans for their patients. And along with that was with him was Dr. Plavin, medical director of Blue Cross Blue Shield of Vermont. And they're currently piloting kind of as a kind of experiment, as a real time experiment to look at the effectiveness of using genomics on that widespread level of universal for cancer patients, universal solid tumor sequencing using a particular panel. And reimbursement has been approved for that. So yeah, so then so we had three case studies and the case studies that we presented in order to kind of stimulate the conversation went from kind of simpler types of genetic tests and more complex type of genetic tests that we started with single variant, targeted variant testing. And we use as an example a pharmacogenetics example that's that's been studied by, looked at by a lot of people and implemented by a lot of scientists which is the use of of CYP2C19 variant genotyping for people who are undergoing stenting. And so Laurie Cavallieri, who's here today, presented data both from their group at the University of Florida and also talked about work that's being done across the entire network with both network sites plus affiliates across nine sites to pool together data. To look at the, to look very pragmatically at the utility of CYP2C19 genotyping for predicting response to clupridigwell. And as you can see, you know, this is, this is, here's the mouse. Okay, this is, you can't see the mouse. But as you can see in red, that's the data from people who, the these, the MI event free survival rate. And you can see that the people who are in red, who had loss of function variants, but were put on clupridigwell. But because they have a loss of function, they cannot convert pro drug to drug and had a much, much lower event free rate. And that data, again, as I said, it's been replicated. It's now being pulled together to produce what we think is going to be important evidence for, for, for helping to guide what is happening. That was presented. And, and then there was a discussion. Most of the, the, the actual presentations were very short in this meeting and most of the emphasis was on discussion. So as Ebony said, this is very preliminary. We're currently working with a professional writer to, to develop the proceedings for this conference and, and review them. But this is kind of some, some initial, some initial output from, from notes that, that were taken at the meeting. So, so a lot, so some of the discussion focused on the idea of preemptive versus reactive pharmacogenetics testing. So most of that testing, although it was, it took place before the stent was, before the medication was prescribed, it's actually considered reactive because it was done in response to a certain event. So sort of thinking about, well, what if we did this at the beginning before, you know, before there was an event and it was in the EHR. So that generated conversations about the challenges in keeping the genomic information in the EHR, particularly in patients change insurance providers, what kind of clinical utility data is needed. You might not be able to get this, this amount of data from all examples, from every drug gene pair, you know, looking at, you know, is genotype testing, how is it better, how is it worse than actual phenotype testing, which is, which has a, a time aspect to it. What constitutes, and we got into kind of broader questions of what constitutes medical necessity, how to define who and when a patient gets tested. And, you know, and then lots of conversations about clinical utility and clinical validity, and then, you know, really important is the communication results because if you generate a result and it's just in the chart, it's not being used, then you might as well have not generated at all, it's like the proverbial tree in the forest, so, so, so really importance of communication results and, and can that, and can that be reimbursed? And then discussions about variants that are not related to the particular prescribed medication, but also can perform other medications in other areas of medicine, thinking about how to develop and discuss gene-related treatment plans, and then, you know, again, this is just sort of recurring of how to, how to, how to transfer information between systems and, and payers. So for the second case, we moved up a little bit in complexity to talk about targeted sequencing panels. And this example, this is, this is the project that, that I've been, my group is doing for Ignite, and, and I know, you know, one of the concerns that's been expressed about panels is kind of making sure that the panels, that the actual genes on the panel are actionable concerns about even if the panel is, you know, even if it doesn't cost more to add more genes, then there may be less enthusiasm about covering genes where the action is not as clear. So, so as the example for panels, I talked about the project that we're doing at the University of Maryland, which is, which, in which we are looking among people, among all people with diabetes to identify people who might have highly penetrant genetic forms or monogenic forms of diabetes, most notably, MODI is, is the one that most people are familiar with. And so we, we've developed a process that we've, that we're implementing and we're also disseminating in a private practice and another integrated health system to identify these patients with some very simple questions but are not always, that are not necessarily always in the patient chart, followed by some basic medical tests and some medical tests that can be used to distinguish, again, with this idea that currently, anyway, the genetic testing is very expensive, but if you, if you can screen out just like you do in any screen, in newborn screening, if you identify people who can most benefit from it, then, then it can be cost-effective and valuable. And, and then we are, so when we identify these individuals, we're then sequencing several genes on a, on a panel and, and then if we find something, we are confirming and disclosing it and, and also doing research to follow up variants of unknown significance. And we, this is just, we, we've diagnosed several patients with monogenic diabetes. This is just one that benefits in kind of a way that you don't really think, you don't typically think about when you think about genomic medicine, we gave her a diagnosis that told her that actually she wasn't sick when she thought she might be getting sick. She was told since her twenties that she was prediabetic, she was told in her late twenties that she needed, she, her blood sugar is high, she needed to really up her activity level, she needed to, to improve her diet and she just didn't know how she could do this because she was a kickboxer on a very intensive training program and, and on a very strict diet. And so she ended up coming into our study and we found that she had glucokinase deficiency in which rather than thinking people have prediabetes or mild diabetes and keep needing to be followed up to see if they'll get complications, they find out that they have a lifelong, a lifelong deficiency of, of the enzyme glucokinase that causes a lifelong modest, modestly elevated blood sugar that doesn't lead to complications and doesn't require constant surveillance and so this is actually a lot of interest to payers because it actually is a, is what a, is a rare, possibly rare example of a cost-saving finding. And so in discussing, in discussing panel-based testing, then these are some of the concerns that came up during the discussion that, you know, one of the things we're doing is we're tracking, you know, we want to know how, you know, how much is this costing, is it, is it saving it worth it and that one of the things that came up is rather than looking maybe at costs, it might be better to just look at resource utilization and then you can transfer that into costs for different institutions, comes up, you know, that particularly when you get into these panels with these rare variants, it's hard to convert that to a randomized control trial and, you know, and then with these panels and the increasing complexity, the information available to players can be, you know, can be limited or unclear or, you know, it has to be presented well because this is actually, this particular case that I presented is actually somewhat of a complex case but it can be presented in a straightforward manner so some of it has to do with the presentation. But regardless, you know, just difficult to make an informed decision without a molecular technology background. There's also just kind of assorted issues that are, you know, that are just more kind of communication and policy issues and just, you know, things need different areas needing to keep up with other areas which is the, which is coding issues, there's CPT codes that stand for similar things and when you're talking to an insurer, you're not just talking to an insurer, you're talking to an individual and each individual may have different information and then just lots of talk about, you know, how can we use existing, how can we use outcomes and knowledge from these to inform future decisions and then again emphasizing that economic data is not the whole picture, it adds weight but they were very clear, all of the payers, although they have different perspectives, it doesn't solely drive medical policy decision making. So then we, our last case was looking at genome wide methods. We talked about actually a couple of examples of genome wide methods. Mark Williams I think is going to talk a little bit more about the work being done with, about the success in getting coverage for exome sequencing for developmental delay at Geisinger and then we also had Debra Leonard also from the University of Vermont talk more about their genome wide sequencing plan with the idea of kind of a different paradigm of thinking about medicine and genomic medicine with the idea that you know, that, that thinking about how, you know, basically that each individual you know, has an, has a lot of influences including the genome and that leads to a phenotype and then how that, how that you know, and then the importance of the interaction with the patient and the healthcare provider but you know really emphasizing the idea as of the genome as an ongoing resource. And so then next we heard from, we heard from Greg Merhar who had been experiencing chronic pain, chronic attacks of abdominal pain since his teens. So for over 30 years and as part of a kind of fun project as part of the Understand Your Genome Project he chose to have his genome sequenced and having his genome sequenced actually came, yielded a variant. It was actually not even a clinically, it was not a what was, what would be classified as clinically actionable. It was a variant, not a pathogenic or a likely pathogenic variant but it was actually a variant called a, it was a notable variant or noteworthy variant of unknown significance. There wasn't quite enough evidence to call it that but it was suspected to be an important variant in familial mediterranean fever and he turned out to respond. Once the, once he found out that that's what it was and talked to a few doctors was finally was convinced to put him on colchicine and his symptoms ameliorated within 24 hours. And so this is a really dramatic example but nevertheless a true example of of how different the model of having the genome as a resource would be in the future versus kind of being guided by symptoms and phenotypes. It was noted that he has blond hair and blue eyes and nobody ever suspected him of having familial mediterranean fever. And so the discussion points around genome-wide methods focused on questions of should genome and exome sequencing be a first resort? Should it be done on everybody? Should it be a last resort? They that there needs to be a time element linked to clinical utility. There are questions about, asked about how many sequences do you actually have to get to get a significant clinical result. That's obviously very dramatic. This is a little bit different but at Geisinger it was noted that about 3% of healthy patients have a clinically actionable variant in one of the genes that they consider actionable. The but it's also noted that the cost of genome sequencing chemistry doesn't capture the cost associated with the interpretation and the therapy changes which can sometimes result in higher costs, not necessarily lower costs. The mechanism for there needs to be a mechanism for maintenance in the EHR and also payment for reinterpretation of genomic sequence at both not just the variant level not just reclassifying a variant but are there other genes and then when other phenotypes come up, can you then get an interpretation out of data that's already been done? Is that data accessible and is there a mechanism to pay for that interpretation? So a lot is talked about this future ideal that genome sequencing could be done once to inform future care but it was also stated that the current reality is that the technology is evolving, the coverage is not perfect and so we're not necessarily quite at the point where we can do it once and be done so that's sort of just the conflicting realities that we're dealing with right now. So we then reported back from the lunch roundtables, so during lunchtime we had a working lunch where there were about six different roundtables addressing questions related to the objectives of the meeting and so one of the roundtables focused on the idea of building some kind of coalition to address all these issues on an ongoing basis and we talked about we actually need to really clarify exactly what we're trying to accomplish and also we need to really know what the existing organizations are, which we know some of them the various roundtables and what can we build on them, can we work with them, we need to clearly identify the stakeholders we've started to do and there was also some question about whether payers could drive the coalition and it was pretty agreed upon that they probably couldn't a lot of them are interested in being active participants but they're not really able to have the resources so that those resources would have to come from elsewhere. In terms of evidence needed, some of the things that were discussed were the need for evidence on downstream costs looking at guideline setting organizations and separating out the organizations guidelines and how does the organizations agenda enter into it the really importance of getting things out into peer reviewed publications, particularly in the higher impact journals is most likely, more likely to make it into practice that the large payers gather and analyze their own evidence to make decisions there was some thought and this could go probably for the coalition too but the diagnosis manufacturers support studies to generate evidence and then just really needing to more define what we mean by change in care you can see there's a lot of overlap in these topics but the third set of round tables focused on designing research and clinical protocols again, randomized clinical trials are often not possible the need to focus on high clinical value conditions or diseases rather than really focusing on gaining reimbursement in terms of what to study it's challenging to get the payers and the manufacturers to share the risk the idea of using economic modeling studies before starting a study so modeling the different costs and trying to figure out a model and then doing the actual study that you need both the modeling studies don't do everything but you need that first and along with that really working closely with the payers upfront at the beginning and designing the study which is what we're trying to do again, economic analysis not the most critical factor in study design which should be considered and then there was some disagreement about this because for some people the diagnosis is an end it's an end to a diagnostic odyssey but is the diagnosis enough or how much is it important how it affects patient care and then finally the groups on disseminating the evidence talked about interest in having some kind of central database for evidence talked about the idea of having some kind of clearing house for letters of medical necessity and again emphasized that this is how payers generally make decisions there's also input taken for some from patients in advocacy groups the idea of some kind of newsletter was proposed and seemed to have some appeal if we could figure out how to centralize that and so again we're still going through many hours of audio footage and putting all this together but the kind of preliminary summary of the key recommendations are that there needs to be some kind of we're looking for a standardized process for getting the clinical utility information out and that communication of genetic results across different levels not just to the patients but to the payers to the providers and the communication has kind of both bioinformatic and people aspects to it so we need to take better advantage of bioinformatics tools for getting things incorporated into clinical decision support but then we also need to figure out to what extent we need specialists to communicate very specific results and then this idea that that we see in the future that a whole genome sequence at some point it does look like it would be an ongoing resource and so thinking about coverage and also storing of data needs to be looking at that and how can we make sure that the data is available and how can we make sure that there's funding available to interpret and actually run an in silico test where the test is separated from the actual laboratory procedure and so that's our preliminary report and I just want to thank everybody that this was a really big effort we had a very dedicated planning committee as you can see we had a large turnout to the meeting and we were pleased to get a lot of a lot of very frank discussion about some of the challenges faced and so we're looking forward to the next steps which as I said we will be publishing the proceedings in the next few months at the ignite site we also will be interpreting developing an author group and submitting a manuscript based on the important findings from this meeting to a peer review journal and then we'll be developing a strategic plan for how to continue this engagement and and obtain resources so again I just want to thank everybody who was involved and thank you all and I'm happy to take questions so we do have time for a couple of quick questions Mark yeah thank you Tony I think one of the things that I heard out of the group that I hadn't heard previously was at least a couple of the payers had indicated interest and opportunity to actually participate as part of study teams to address this issue which is I think a much more productive way to approach developing the evidence because it's very difficult to put payers in front of you to say well tell us what evidence you need because they don't really necessarily know what evidence they need but having them involved as part of the study team can help them to say well this is a question that if I were evaluating this as a medical director would be really I'd need to know the answer to that I know Howard's had some experience with that that's been relatively positive and I think the more we can try and encourage that within whatever projects we have that are funded by defining that as part of the desired investigator team would be good and then we have Levi do you have any discussion of the role of the Moldy X and Palmetto discussions in your sessions? Oh, whether of Palmetto we unfortunately were unable to have a representative from Palmetto at the meeting but what was your I guess what was could you clarify your question? Yeah, I'm just curious about whether you'd had any input from a one of the national I think has huge impact on utilization of genomics and paying for it and that is the Moldy X operation which is based in Palmetto as part of Medicare. On this round we were unable to engage them at this point. Yeah. It was a great summary. My question is someone related to that one because when you have about a month ago there was a similar meeting in the Canteraside a think tank that had multiple stakeholders at the table including several private payers and it was a very interesting discussion but it also makes one mindful of the kind of competing agendas that can be at play on both sides. So obviously on the side of the people who are doing genome sequencing of course there's an agenda to get payers to cover the cost of doing that because it's not always easy to have that come by other means. On the payers side though there's often an agenda to avoid having to pay for it. Not so much the testing per se but the cascade that may come after that may encumber their books. And in fact when we had the cancer discussion it was very interesting because there were kind of questions what kind of evidence and some of the payers would kind of lob out these studies that theoretically were interesting but functionally would be impossible to do and would take a decade. So what sort of came out of some of those discussions afterwards was well maybe the middle ground is actually to work with Medicare which theoretically is sort of has less skin in the game in terms of the financial encumbrance. Their goal is to figure out what coverage makes sense and so maybe the way to move forward is you have the experts in particular disease areas where genomic medicine has the potential to be relevant who can help really carve out feasible intermediate studies that get to the issue of clinical utility and could be done sort of with consortia like this etc and working with Medicare to within populations where there could be sort of there could be studies that where whatever the results are ideally those will be less biased by one side or the other but on the other hand it may be that the private pairs will follow so if Medicare decides this is something that we see and we think it's worth it then the private pairs may likely follow as opposed to sort of putting on them the onus to sort of come up with their plans because one reasonable point that they might make as well we have stakeholders we can't put shareholders money into research which is not an unreasonable perspective and accounts and stuff like that so it may well be that Palmetto or Medicare in general in collaboration with consortia like this is the right intermediate step and obviously the payers we keep them involved but we don't put it too much on them to try to figure out how to get coverage. That's a really good point very important to engage Medicare in this process. I just want to comment on something that you said earlier and I think one of the things we're trying to do with these meetings is you sort of talked about well there's people who the payers don't want to pay for it and the people doing sequencing want to do it and I think part of the idea is that of actually having these meetings is that you can kind of get a little bit more a little bit more dimensional than that because I think that is kind of maybe one of the barriers is that we're kind of thinking is that kind of the thought is like something doesn't get covered oh payers don't want to pay for it but just learning more about because yes the Medicare because they're going to work differently than the payers but the payers in the end still their objective is not I don't think can be simplified to we don't want to pay for it it's a matter of that their perspective is just different and so if they can understand what the purpose is and I think also the importance of having because there's no there's no like representative example of genomic medicine and I think that's part of it so the importance of the ongoing conversations and hearing the specific examples and as you said having people experts in the specific diseases present and it really has to be an ongoing process just a couple more quick questions like Moon and then Roger in the back thank you Tony this is a conversation that has been occurring for the last two decades more or less as genomics has matured from research to practice the question I have specifically did the conversation include things like coverage with evidence development because there are situations where clinical validity is established you need just a little bit of a push for clinical utility that the right trial may or may not have been done this kind of partnership that Mark was talking about with designing studies for clinical utility or even covering for them until you get the final results I know this came up with CMS for a while but I'm wondering if that was part of the conversation oh certainly yes certainly in terms of the importance of designing studies for clinical utility but I think one of the points about that is captured in this idea I mean it's kind of grossly the randomized clinical trial doesn't work for everything but the thing with genomics for example I talked about an example in our study where we identified a child who had an insulin mutation that caused her diabetes and that was hugely helpful to the family but that's very rare and so to achieve the clinical utility at the highest level is not possible because you have all these uncommon variants with all these uncommon conditions and so again the importance of really having the different people involve the patients and the providers to communicate you know why is that clinically useful even though it you know the particular study design doesn't necessarily lend itself to that so I don't know if that answers your question but last question in reference to the panel testing well one question that always comes up in addition and I think there's not a lot of consensus and actually quite a bit of controversy I'm wondering about the process that you went through to select the genes on your panel that's a great question and that was a question the payers were interested in as well we started with the genes that are considered the maturity onset diabetes of the young genes the modi genes and but where you know where most of them are rare so it's really only three of them that have you know very clearly international guidelines for clinical utility but then we also looked at syndromes that could also be useful to diagnose because they would tell you about other symptoms so we looked so we really focused primarily on conditions that could be missed and thought to be ordinary diabetes but we also there definitely was kind of an area where it got gray where you know where you know there's like there's two genes on the panel that are involved in familial hyperinsulinemia and have not been reported as diabetes genes but theoretically variants in the reverse direction could be associated with diabetes and just have never been caught never been looked for so I think and that was the concern of addressed at our initial payer consult meeting for just our study was that was kind of well you know what am I you know and because we had some for monogenic obesity and kind of you know what are you going to do with that and that was the exact conversation we had we said well you know the clinical utility might be there but you need more cases in order to know what it is and they said again you know we don't want to pay for research but yet there needs to be some way that we can gather that that evidence not just from people who are consenting to a research study in order to figure out what to do with it so. Thank you Tony and you'll be hearing more about it we'll have meeting proceedings as well as hopefully a paper coming out. Just some housekeeping things first the meeting format for the rest of the day we're going to have four different sessions the sessions will start out with two talks on the state of science and the gaps and then the night investigator will highlight some things that we've done within that area as well as opportunities that ignite maybe particularly good to actually tackle just let you know the talks will be 15 minutes each and please hold your questions for the discussion part there will be 30 minutes at the end of each session for discussion and it's going to be moderated and we'll have discussions actually leading that discussion so if you could please hold your questions and comments until the end of each session I want to thank the planning committee that took a lot of time we met every week actually coming up with the agenda they gave us a lot of feedback on not only the background document but how to make this meeting useful Katrina Goddard line card and Chris shoot Howard McLeod in case they over be wonderful and really made this meeting what it is also the let you know the microphones please use your microphones they're at the table if you don't have a microphone near you there is a microphone at the back that you can walk up to we have a background document in your packet that was put together by our administrative coordinating center Duke University led by Jeff Ginsburg and Lori Orlando this was also emailed to you if you didn't have a chance to read it please this is a great document it really tells you what ignite has done this far and I'm going to give it over we're going to have pictures at 1115 at the 11 o'clock break so if everybody could gather for pictures before you leave the room that will be wonderful and I am going to give it to Colette Hoppe Fletcher Hoppe who I introduced when she was out of the room but she is actually the heartbeat of the network and so Colette has some comments as well. Hi so this is just a quick notice that during the 355 p.m. review of recommendations and the 420 prioritizing future opportunities we hope to engage audience participation through a poll so during lunch I'm going to circulate a trial version of the poll you're welcome to participate via webinar via the internet or via text message so I'll be circulating a link to all of the meeting attendees please look for an email from Colette Fletcher Hoppe at the National Human Genome Research Institute and for those of you participating via text I'll include some further instructions essentially to join the poll via your phone you need to text a code to a specific number and then we can move through the different poll questions one at a time unfortunately if you're participating via text you can't really skip around the different poll questions and please remember to text the word leave in all capital letters when you're done with the poll. Thank you Colette we will go ahead and start with our first session genomic medicine implementation in diverse healthcare settings and populations if the speakers moderators as well as discussants will come to the front of the room and if at the end of each session I know we have a break after this one but for the rest of the day at the end of the session if the next session speakers moderators and discussants could kind of come forward at the end that would be great so it will save some time and as well as the note takers sorry thank you