 What I'm going to do is review today. So Terry started the day today reviewing yesterday, so I'm going to review today. It ought to be really pretty fresh in your minds, but if I say something completely stupid let me know so that we can capture the events of the day this way. So Paul started with a discussion of barriers to getting tests implemented. I thought one of the things I took away from his talk was that there is a link between what he is trying to do in the implementation space and the basic science that underpins that. One of the lessons that he taught us was that it is possible increasingly to work with industry and we heard other examples of that in the discussion and that guidelines which we spent a lot of time talking about and how important they are in terms of moving things to practice tend to really lag. I thought the big focus was on Bencherus' clinician tensions, that the sort of almost he said, she said kind of thing, the clinicians want the basic scientists to sort of focus on what they think is important. The basic scientists think the clinicians are all doofuses and obviously somewhere in the middle there is a sense that the word personalized medicine may be really offensive to people who are in practice. Sorry, doofy. You know, seven years of Latin and I still didn't get it right. I mean, I sort of, what can I say? Already practice, so there is this notion, and it could be the fourth declension and then it could be female or male, but then the plural I think is dooferey, doofus dooferey. Somebody help me. I can't remember. Doofus dooferey, opus operey. Dan, you're on tape. Okay, I know. I'm just, I know I thought of that about 30 seconds ago. So this idea that practitioners who fight the fight every day and see one patient every 10 minutes are practicing personalized medicine and take offense at the idea that they're being told that, you know, here's how to practice personalized medicine. So we probably ought to think about the semantics of this a little bit. I like, I had to type, put down the quote from Pearl soccer as the game of the future and always will be. So the suggestions, how do people decide what to do, need assessments from clinicians. I don't know what these words actually mean. For the system, need to assess patient expectations, need to encourage translation and interactions between the basic and the clinical scientists very early on, maximizing opportunities to make sure sort of genomic data are included in large trials. For example, then David Letterbetter made the point, and I think it was made by a number of people afterwards, but I think he was the first one to make it this morning was that there is an opportunity to train a cadre of PhDs to be clinician scientists, and that would include informaticians and mathematicians and basic scientists of many stripes. So there's this idea that we're manpower limited or person power limited in this space and that there are a number of, there are lots of PhDs who are underemployed, so there may be a way to make those two communities meet. Howard and the person, the pharmacogenomics summary, so I'm summarizing the summary of the summary, I think. So one question was do sequencing platforms add value to pharmacogenetics and there's some issues around technical aspects of variant calling, and then how to get to genotype-phenotype relations, particularly for the rare variant. So you have CYP2C19 star 87, what does that mean? Mark retains idea of a meta-meta analysis, which is the wrong term, but the idea of looking at meta-analyses and deciding what the metrics ought to be for performing meta-analyses in pharmacogenetics, so do you include things like dose and indication and the kinds of things that have raised controversy around some of these analyses of pharmacogenetics outcomes and which have been subject to criticism by the pharmacogenetics community. There is a need to compare platforms, and then there's this issue of sort of who is the advocate for pharmacogenomics implementation, the comparison to the imaging community, where the imaging has been very, very rapidly adopted, and whether that's driven by radiologists or by manufacturers and who are the sort of corresponding advocates in the pharmacogenetics community was something that we talked about. And then I think a really important point, that's why there are the little two asterisks, is what are the right endpoints? We focus on predicting the right dose of warfarin or predicting intracerebral hemorrhage with clopidogrel, but maybe there are other endpoints that are more interested to payers, like do patients come to work more, do they stay on their statin, do they have fewer heart attacks, do they get out of hospital sooner, do they use less ICU beds, those kinds of things. And then we were told to make sure that we pay attention to developing those evidence for those outcomes, whatever those outcomes are, but it's important to think about different kinds of outcomes. I put this back in again, Rex's barriers, and I'm not going to read them again because I just wrote them down from a slide that you all saw within the last 45 minutes, but I did add the idea that biology is important. I actually believe that. Paul Rittger made that point. Cholesterol crystals is the driver of the inflammatory process. I think that you can't help but do better if you understand the underlying biology. So the three updates from this morning, Howard obviously had a big and raucous meeting, and I hope I captured some of this. The sequencing group looked at return of reports and focused on qualities. There was this idea that you should build a standard set and use that as a metric for how different laboratories perform and perhaps set up standards for that. There may be different metrics for reporting variants in so-called clinical genes, the 3,000 clinical genes compared to the 17,000 genes of less significance. Best practices are needed. Cap and clear, it says here we'll do this, are doing this, but some of the best practices should include that coverage is annotated and demonstrated for target regions, that they're uniform QC metrics that just sample, the standard sample sets need to be generated. There need to be metrics for comparing platforms. There need to be metrics for developing and comparing data analysis and other kinds of tools. There need to be metrics for deciding which platform is best, not for a whole genome, but for which particular region it may be that there are certain approaches that are better for certain regions of the genome than others, at least for the time being. And the long-term target, and I'm not sure where long-term is, is to figure out a way of getting to the point where you don't need to do a validation step with next generation sequencing. For the time being, that's going to be an individual institutional decision, and it's not going to be mandated centrally. And then who curates and how those data are displayed in the EMR is a point that Mary raised, that we're not even at the point where if somebody has an HLA variant, the name of the gene may not even be displayed in the genetic report. We probably need some uniformity around that. The family history EMR integration isn't yet standardized, and there are a number of approaches that people are thinking about in terms of standardizing and expanding the EMR integration of family history information. These are the things that I captured during Jeff's discussion. And this idea of using Facebook or something, the next Facebook, to capture more family history in a better way. And then the idea of developing a demonstration project to develop and validate tools to inform any system on how to acquire and display family history in the electronic record, you would implement that at sites that don't do it now and evaluate performance, and the performance metrics might be that are the data valid or the data used, are there changes in behavior, and some of the other issues were to create a Spanish language version. In Nashville we would want a Kurdish language version as well. So when there were elections in Iraq, there were four sites where people can vote in the United States and one of them is Nashville, Tennessee, because there's a Kurdish population. So there are other populations besides Spanish populations, and then the idea of creating guest accounts so that people could look at MeTree and the Intermountain site to get a sense of what the tools look like. And then I really like the idea of sort of developing this kind of approach to a demonstration project, approach to evaluating what was going to happen, and then substituting the word pharmacogenetics or substituting the word sequence data for the word family history, and then developing approaches to validating those kinds of tools as well. And then the microbiome, Murray has been encouraged to extend the studies of the relationship of the microbiome and human phenotypes to other phenotypes like inflammatory phenotypes and there's a cancer phenotype as well, and then he continues to look for centers that are interested in this. So that's the update from this, oh, and then we had an education discussion, a spirited education discussion just now. So the points that I took away from this were that subspecialties march through this area at different paces. Infectious disease is way ahead. The HIV guys are way, way ahead in the infectious disease space. The professionals associations are the in into practitioners because they are the ones who tell practitioners what kind of education they need to be getting, they're the ones who drive recertification, they're the ones who drive guidelines, they don't want to get into this space too early, they don't want to get into this space too late. There's a lot of genuine interest in the field, but there's apprehension as well. Gene talked about the New England Journal of Medicine series. I personally think those sort of come, they get ripped out, they get looked at maybe or not, they go into a stack somewhere, but I don't think that's a durable approach. There's this NHGRI TV, which, that wasn't what it, genome.gov TV, I actually, so there's some things on there that I've actually, sorry, there's some things on there that are fun to look at. So it's a site that actually, if you have a free five or 10 minutes you should look at because there are some interesting things on there. I thought, I found these statistics sort of pretty interesting that there's 850,000 licensed MDs, 625,000 of whom do patient care, so all the MDs who are in this room are obviously part of the 225,000 who are not full-time patient care, and then about a third of those are in primary care. There's this real sense that pharmacogenetics is going to be the first place where genomics enters the clinical workflow, but there are other examples as well. And Jean was a big fan of, and I must say I resonate with that as well, the idea of using cases as educational tools, those are, that's really fun. Other issues that were raised in the discussion, so there's a lot of interest in partnerships and education, partnerships with professional societies, partnerships with even payers. The idea of, once MDs have had genetic testing on themselves for whatever, for MEN2 or for 23andMe, or for whatever, they tend to be more interested in using that for their patients. So the drivers of MD interest really are CME recertification, and then when their patients come to them and say, what do you think of this? And they have to figure out what to do with it. The idea of short courses on the web, and then there is this downside though. I was, I'm going to go through an exercise one day, and I'm going to figure out how many times in a course of a week do I enter my Vanderbilt password to get to a site to do something that will only take me two minutes to do. So there's this idea of web alert fatigue, and then I thought I heard the beginnings of an action item of sort of see how this is, I can't remember what this even says and it wasn't written within the last 15 minutes. So if anybody can figure out what I meant, I'm not sure. But it's see how it's done across, I think across these kinds of short course of sites. Josh advocated using informatics approaches, Pearl pointed out that there are these, there's this idea of, there's an educational spectrum and that there are educational requirements that are different at different stages. The idea of genetic testing on yourself as an educational experience was raised by a number of people, and then the idea of training cardiologists or infectious disease specialists or oncologists who have a specific subsubspecialty in genetics related to their particular subspecialty is another way of increasing the physician pool and interest, and I think that that's an important thing to do, and I'm not sure how to push that forward, but one of the things that physicians have to do as they do that is to learn about uncertainty. And then I put down here that one idea would be a T32 mechanism to provide support for those training programs. If the institute wants to hear suggestions, there's a suggestion. So that was the summary of the morning. I hope I captured that, and are there other, are there other things that I should. So Terry, should I just go on? Right, no, I'm actually looking at lunch over there. I had so much stuff at the snack, I'm not sure I need lunch right now. Okay, so I'm going to give you three slides on ideas for Genomic Medicine 4. This is Genomic Medicine, Genome Medicine 3. So Genome Medicine 4, the planning group met last night in the back of the gas-lit bar, and it was just a real sort of experience for many reasons. So one of the action items was that we all needed to have showers this morning just to wash the garlic off ourselves because we were right next to the kitchen. So we didn't have anything to eat, but it was just permeated. So it was pretty clear that there was the session that really caught people's imagination and attention was the session with the payers yesterday. And I think we came away from that, at least I came away from that. And I think many people came away from that with the idea that the payers are really interested in figuring out how to do this. And we were challenged by the payers to create partnerships and to go ahead and look at outcomes in a very, very broad way. And the way that Howard had suggested, look at outcomes like ICU, bad use, or whether people stay on statins for a year, the kinds of outcomes that we hadn't thought about. The thing that resonated with me was Reed Tuxin's statement that he's perfectly willing to pay for a very, very expensive leukemia drug if it keeps people out of the emergency room and keeps people out of the hospital. So I think that that's the kind of challenge that we have to sort of come up with. And how do we frame the design of a study like that is sort of the thing that I think we're being challenged to do. We thought about what the next step might be. And this part of the presentation is meant to be a provoking discussion and not just me standing here. But one idea was to work with the Center for Medical Technology Policy, Sean Tunis' organization as Reed Tuxin suggested. Another was for NHGRI to take the lead in convening a small working group to figure out what the next step would be. So thinking about how to do this rather than sort of whining about why the payers aren't paying for it, one question that we had that we sort of left open was that we weren't sure that other NIH institutes and centers had done this yet. So and there are many partners that you could imagine, HRQ, the Foundation, the Patient Center Research Institute, and then we need to engage health care economists at a number of different levels to do this. So those are the kinds of things that we sort of thought about as the theme around which this next meeting might coalesce. So one discussion that was very practical was there seems to be momentum in this initiative so we could try to plan the next meeting for September. Or September seems like it's really, really close. I mean, it's really sort of four months away. Melissa, I don't know whose problem it is, but it's not yours and mine. But planning is a really pretty intensive exercise. So there's this idea that we might plan later. And the focus would be on standards, documenting, and we would also include in the meeting a focus on the standards for sequencing, reporting, documenting the eOMR. That might be just a subgroup, but the big focus would be on this outcomes effort. So at the broadest stage, continuing to engage stakeholders. So that was the theme for this meeting. And we certainly got engagement by a number of stakeholder groups. But there are a number of groups that we managed not to have at the table here and that we would like to engage as well, I think. So professional organizations, there are lots and lots of them listed here. So we need to hear from professional organizations, obviously not all of these, but some of them, the relevance of genomics to their field, the levels of evidence that they find convincing, how they make decisions on guidelines, what their professional education initiatives are. We need to engage the patient care-oriented research initiative and AHRQ. We need to engage patient groups, like patients like me or the Genetic Alliance. We need to continue to work, I think, with EMRs. And then I think we've had engagement from the FDA at this meeting, and I think that's been very, very useful, both in terms of sort of making sure that we understand what it is that the FDA regulates and doesn't regulate and how we can work with them to move the field. And then the big debate was where to have the meeting. If we engage lots of professional societies, many of them have their headquarters in the Washington area, so that's an argument to do it in Washington, D.C. The argument to do it in Seattle is that we've been in the middle of the country, we've been on the East Coast, so we should do it on the West Coast, so people don't have to wake up at 4 o'clock in the morning, their time to come to the meeting. And then Dallas was floated as an idea because at least one of the professional organizations, the American Heart Association, is in Dallas. It's in the middle of the country. It's not Chicago. And doing Chicago in December is risky. So we thought Dallas in December might be a little less risky. And there is an airport hotel in Dallas, OK? So that will make Mary happy. Actually, there are two, the high end one and the really high end one. And I think it would just depend. You're a tourist to me if you know what they are. The high end, the high end grand. The grand high end and the grand high end, come on, Mary. Those are, those of us, those of us. Those of us who fly bankrupt American Airlines like Howard obviously continues to do, know that. So that's where we are. Comments on this particular slide. I think this is the key to thinking about the next steps in this initiative. No comments. So we have Kate. So one organization you don't have list is ASCO. And this is basically the number one priority of the ASCO board this year. And certainly a very, very large professional organization for whom they realize they really need to be educating their patient population. So I would put that as an organization that you really need to engage in this. I must say I don't, I'm not sure how these names got on here because I don't recognize some of them. I think this is sort of, I mean, AHA I recognize, ACC I recognize, ACMG I recognize. I'm not sure what AAFP, the family practitioners, AAP, Patti Pediatrics. Okay, so somebody recognizes each one of these. So ASCO I've put at the beginning. And there must be an infectious disease society. We just don't. What? IDSA. IDSA. So I'm going to put them down as well, right? I mean, it's sort of neurology. So I think we did just hear the suggestion what about pharmacists and other groups. I think, you know, we're talking about a day and a half meeting and we probably want to focus maybe one on physicians, forgive us, and then another one on non-physicians because it's a big space to try to fill. I mean, there's, you know, there's tremendous interest in the nursing community. The pharmacists are really, really important to this effort. I mean, as we move towards implementation, we found this as well. Nitchpeg covers the other health professionals. Sorry? Nitchpeg covers the other health professionals. I can't hear. NHGRI is in active conversations with the pharmacists and physician's assistants and other groups. So maybe we could focus just on the physicians this time. Okay. When you focus on physicians, make sure we also reach out to the DOs because they've just stressed an interest in the last few years and I've done one or two lectures for them. Yeah, and most of the osteopaths actually are members of the professional societies that we would be, you know, we'd be involving. Actually some aren't at all. Yeah, I understand. And there's a big push now for them to have to do CME inside the DO field. So we've had a push. We have a lot of DOs at Geisinger and a lot of them are dropping the general professionalizations because they need the DO CME, so just keep that in mind. Oh, that's a great point. Okay. So, but the theme of continuing to engage sort of the external community seems like a reasonable pathway forward and we, it's a way of all of us to learn what's going on in the greater community because I think it's very difficult to implement unless you know what's going on out there. Rex. Lest any of the work groups think they're off the hook, I think the idea would also be to make sure the work groups did presentations. Absolutely. Yeah, so, so, right, right. I mean, that's, I'm going to put that in here as well. Mike, did you have a comment? I mean, there's a, I'm sorry, as far as location and timing, the clinical sequencing consortium is meeting in Houston in the beginning of October. I don't know if putting it up against that makes sense for people. We had, we were aware of that and figuring out some way of engaging some of this room with that meeting was something that we had discussed and we leave that with Terry and Eric and Brad, I'm sorry, and Brad. We leave up those guys. Other comments? Dan, maybe in addition to the four bullets you have there for what we would expect the organizations to do would be to highlight some case, have them highlight some case studies if they have some that are relevant to this, to this topic where they have either made significant headway in genetics and genomics guideline use or otherwise. I just think it's something very more practical instead of some philosophical discussion or theoretical discussion. When you say case studies, you mean, you mean studies of how to do this, not, not, not, not cases. Not, not, not clinical cases, but a case of a particular genetic test and how they dealt with it and either rejected it or accepted it. At the back of your CYP-2C-19. Okay. I think that, John, oh, something about, something about small people? No. This is about technology. And so this meeting is concentrated on sequencing and SNPs, but there's lots of interesting things happening in transcriptome and epigenetics that we haven't talked about at all. The, it's obvious that just dealing with the sequence is overwhelming, much less trying to talk about the, trying to talk about expression and how to interpret that and how to teach people about that or about the epigenetics which is even more complicated or what the, there was one mention of the mass spectroscopy and what it can do. And so I, you know, I don't know what your, what your space is that you operate in, but it looks like we're concentrating on a little small corner of it by, by restricting it to the just genetics. I saw Howard running out, running out of the room as you do. Sorry, my initial, my initial sense was that, that we could sort of assign that to the sequencing working group. The sequencing working group has so much on its plate right now that, that seems unfair to me and they, they focus on sequencing, as you, as you say. So whether there should be something in the sort of new technology, a work group around new technologies and how they impact genome medicine. Can, can we ask, are those technologies that really are about to be employed clinically or are being employed clinically? Well, so there's, expression is. If you, if you want to, so, so, Terry, I'm sure you've heard the term, if you haven't, I'll introduce it to you. Have you heard about the Snyderome? The Narcissome. So, so, so there are some people who think that, you know, the systems biology and the Snyderome is sort of like just over the horizon. Now, you know, you can sort of decide whether that's, I mean, but, but, so there's, there's, there's a sense that integrating many data sets, acquiring and integrating many data sets, and that may be beyond the sequencing working group. I don't, I don't know. Right, so it sounded like the answer was yes. So, so given that the answer is, is yes, then maybe another working group would be apropos. Right, so. Who would like to lead that, John? I thought Scott was going to volunteer. Yeah, I mean, it's sort of, right. Systems. So it seems like also among that list up on the second line there, that there's sort of two groups, there's, you know, and from, from the viewpoint of developing sequencing standards, I think it would be great to have CAP or maybe the medical genetics people or the CDC. You know, because I think if we have a, if we're going to go to this trouble, it would be nice if what we come up with is embraced by these people or that they would have input so we wouldn't just, you know, be working in different ways. At the risk of offending my, my friend from the FDA, I'll put, I'll put that on the same line as the FDA, sort of the regulatory superstructure around the implementation efforts. So that would be CAP and CLIA and FDA and other regulators. Is that, is that fair? And actually Deborah Leonard was the CAP representative and I think she's interested now in intrigue. So we'll definitely have her or others back. And both David and I, you know, are part of ACMG. I chair the next gen standards working group. So I think there are probably some of us that, that, and Mike Murray that can be liaisons easily to ACMG. Okay, other comments? So you did have presentations from public health. So I think adding CDC or public health, which I noticed was not on there. Well, actually that's what Toby was doing, but, but having more of that was probably would probably be worth it. Yeah, I think the concept. Well, he was here, so I just didn't see it, show up on the slide, so. I just put it next to FDA. I'm not sure whether to set up. That's perfect. Our other public health activities. So local or state public health departments, maybe the implementation arm and that's asked though, I'm not making that up, association of state health officers. And so there are other ways to engage the public health community beyond just CDC. So, so, so say that acronym again. ASTHO, yeah, pass though. How'd you like to answer the phone that way? Hi, I'm from ASTHO. It's a T though, not a P. A T. T. Health officers, yeah. Territorial. So the web will, we'll, with the web we'll let us find the rest. But, but I think this group is separate from the standards, the regulatory. So you might want to have a separate line for public health. Professional organizations. ASTHO is a professional organization. Where's that? Okay. But I would be cautious in moving CDC because they're in charge of CLIAC, which is a scientific oversight for CLIAC. And a lot of people don't realize that. Congress put them at CDC. And there's a whole department that looks at these kind of issues is actually working closely with CAP and with FDA. And I think that's an important group to think about. I'll just leave it that like that. Okay. I just, I wasn't sure where to put CDC. I guess, I guess, I guess I'm not the only person who isn't sure where to put, oh, sorry. Yes, that was a little, okay. Okay. We're wordsmithing, but this is good. This is good. Yes. So another organization that you may want to add and you have payers here and you were discussing a lot about CLIA is actually CLIA or CMS. Right, so I have. I don't know whether you have it on there or not. Because on the previous slide, when you were talking about sequencing, what's put there as a fact is that CLIA and CAP will take care of best practices. But, you know, all we are talking about is what to do and how to do sequencing. So if you're putting that as a fact there, I don't think that ACMG or other organizations are gonna be very happy with that. And we're not there yet. And there, CLIA has not said anything about developing best practices actually. Blame Howard. Other comments? Yes. So NCI went through a very expensive, I guess learning with standards issues through the CAB. And it may be interesting to bring someone that was in the advisory panel of the CAB review and maybe can prevent us from making some of the same mistakes. Brad, is that in your space? Is that a, so should I write that down? CA, I'm gonna have to make the print smaller on this once again, which is good, okay. David, if you, no matter how loud you talk, it won't be captured on the video, right? So I guess one of the other, now I've got two. One of the other questions that I think we need to address is just whether or not we're going to include things like infectious disease sequencing as part of genome medicine or not. I think we either need to make a conscious decision that it is or it isn't. If it is, then we need to consider how we're gonna approach the implementation of that and really how much of that wants to come from infectious disease and how much of that wants to come from, or whether we want to decide that it's just a whole nother area that is not part of genome medicine. So a little different than kind of the microbiome questions really, the infectious disease, testing, and those issues. So there's, it seems to me there's two or three issues. I'll start the discussion. One is the business of sequencing germs and whether that's sort of a whole other effort that would just sort of dilute what we do, but there are all kinds of issues around, for example, pharmacogenetics in infectious disease, using markers like CD4 counts in infectious disease. One of the reasons that HIV doctors adopted a back of ears because they're used to sort of changing the way they think because of a lab test as opposed to other people. So I'd hate the idea of saying, we're gonna do everything except infectious disease. So we have to figure out a way of saying, if we're gonna do that, we have to think of a way of saying it, but still embrace that community because they're one of the big users of this kind of technologies in many other ways. I think what you're saying is whether you really want to include resequencing of microbes as part of this effort. And I don't have a strong feeling about that. I'd let somebody else decide that. Pearl, did you have something to say? Is that a different thing? Well, or so any other comments about the infectious disease? So if I'm just gonna write that down and let as we sort of think about planning, how to have to make a conscious decision about whether we include that in the sequencing, I think that's part of it. Can I just add something for the microbes? It's not gonna be always just resequencing. It's gonna be sequencing too, because you have new resistant genes and new species coming up all the time, so. Oh, I'll just leave it like that. Pearl? With some trepidation, but one of the elephants in the room is direct to consumer genetic testing. And I'm not sure if it goes on this slide, but in terms of someone active in this space, I think we can't ignore it, but I don't know where to put it, so I just thought I'd give it to you so you could handle it. Is that, it's almost like patient groups, but it's not. I mean, some of those companies are doing some really pretty interesting things in terms of just sort of using questionnaires to get phenotypes now that they have a gazillion people genotyped, and they are sort of doing GWAS on scale because they can. Michael? I should put that down somewhere, I'm not sure where. Well, it might fit into a bigger category of private genome companies, including Nome and some others that are trying to impact this area in lots of different ways. Well, yeah, I didn't say that right, but private genome efforts. Of course, I don't think we're sitting here trying to say, well, this is entirely government space and you're not allowed to. Andy has a comment while you're typing. Just one comment related to that, I was at the personal genome meeting last week at Harvard and I think all of this is focused around disease and disease prevention and there's a whole another effort that's looking for positive traits and other things and I think when you're talking about the DTC part, you're probably gonna see as much in that space as you are in the disease space and those are a lot of the early adopters people using it for that reason and it might be interesting to have someone from that group come and talk about when you find positive things in a genome that you need to share with someone. Great, we probably should wind down. Yes, no, I'm sitting here. Or have we already. I think this has been, you know, this is really fun but at some point we ought to stop and I think that this is gonna, so before we stop, I need to once again thank, is Robin in the room? Robin has left the room again. Okay, so I need to thank Robin who was out of the room the last time we thanked her. Lauren and Melissa for putting this meeting together. Thanks. And we'll have lunch and we're adjourned. See you all in Bethesda, Seattle or Dallas in September or December.