 I want to have a chance to update you about the genomic medicine working group, just to give you a sense of a reminder about why it is that this working group was formed. You'll recall that the strategic plan has this right, the so-called right-hand side. We keep talking about moving to the right. That's not a political statement, but it is on this diagram moving towards advancing the science of medicine and improving the effectiveness of healthcare. And so there's a number of programs that the NHGRI has established that are initial early foils into this process. And what we wanted to use this working group to do was to think about how we actually are going to achieve the goals of moving in that direction, what is actually going on in the field, and to try to figure out what role NHGRI should play to support and accelerate this process. So the goals of the genomic medicine effort were to identify research directions and priorities, promote collaboration amongst existing groups, stimulate investigator-initiated efforts, and issue funding solicitations as needed. You'll hear more about that in a minute. Establish the working group as a subcommittee of council, which includes a rotating membership, at least one council member, and with the key element of reporting back to council. And so did this just stop working? All right, I guess I'll step close. And so what I'm going to do today is give you one of those reports back to council. So the charge was to plan a few genomic medicine meetings, to provide guidance in areas such as ongoing infrastructure needs, identifying needs related to future collaborations, and reviewing process and the overall implementation. This is the current working group. You can see it includes Pearl and myself, current council members, former council member, Jeff Ginsburg, and I guess you'll probably need to add a new council member to make sure you have somebody actively on that. It's been a very active and engaged group, and we've been doing a lot of phone calls and meeting activities. One of the things that they have been involved in is trying to help come up with a definition for what genomic medicine is. It's actually helpful to have a definition, and it's remarkable just how much debate the group has had about trying to come up with a definition. The definition that we have agreed to, and I should note that for those of you that want to take a look, it's for the council members. This is available in the electronic council book, so you can go take a look at that. But the current definition that we're proposing is an emerging medical discipline that involves using genomic information by an individual as part of their clinical care, for example, for diagnostic or therapeutic decision making, and the other implications of that clinical use. So there was, like I said, enormous amount of debate amongst the group about this definition. Some thought it was too narrow, but it was decided that it should be purposefully narrow. By genomic, NHGRI means direct information about DNA or RNA. Downstream products are outside the immediate view. NHGRI recognizes a dominant portion of its current portfolio appropriately supports the foundational research that will ultimately contribute to this. There's an enormous amount of activity that's going on in the strategic areas one, two, and three, which contribute to understanding the role of the genes play in disease and then ultimately to identify genes that may actually be predictive or diagnostic of disease. Fourth and fifth, NHGRI strategic plan domains capture the research on umbrella that really fall under the domain of genomic medicine. And metaphorically, they view this as a destination for attaining the mission of improving health through genomics research. So it really is about getting to the right-hand side of that strategic vision diagram. So what are the activities? Well, there have been a slew of them, and just taking a few from the website, you can see the schedule here. There was a genomic medicine symposium that kicked this whole process off in June of 2011, and I'll give you a little bit more detail about this in the future. There was a characterizing and displaying genetic variance for clinical action workshop that actually led to an RFA that's currently outstanding. There was a Genomic Medicine 2 meeting that was held here in, or held in Bethesda. There was a Genomic Medicine 3 meeting that was held back in Chicago, and there's a Genomic Medicine 4 meeting that is currently in planning for June, for January. So the goal for Genomic Medicine 1 was just to get people together who were doing things in that space to identify common areas of interest that they had, to define demonstration projects, to think about if there were collaborative activities that could begin to study how we moved to that side of the strategic plan, and to stimulate development of consortium for conducting translational research. And I think it's fair to say that everybody was really quite surprised by the amount of material that was available at the Genomic Medicine 1 meeting. And it was an interesting meeting because it was an on-your-own-die meeting. It brought together a large collection of people who were so passionately committed to this area that they were actually willing to pay and come to the meeting just to talk about it. And one of the things, we learned a lot of things, but one of the things that we really learned was that there were a very large number of institutions that have used institutional funds to make a huge commitment to this area. So there were really a large group of people that were actively involved, and we spent some time talking about the challenges. And the key challenges were limited evidence and consensus on which genomic variants are still medically relevant. There was a lot of debate about clinical utility, clinical validity, actionability, the definition of each of those words. And I must say, I think it's still fair to say that we're still having some of that debate. One of the issues that came up regularly was that as you think about moving into clinical care, one of the things that always drives clinical care in this country is mechanisms of reimbursement. And so that was something that surprisingly came up very early in that initial discussion. And then there was also a lot of discussion about the burden to patients and clinicians, and especially, I think, to clinicians of assay intervening and following up in genomic findings. It's clear that a lot of the physicians that are involved in or likely to become involved in genomic medicine are ill-prepared to do that. We're already seeing a flood of physicians reporting people showing up in their office with their 23andMe or similar report and saying, hey, doc, what does this mean for me? And they're not well-prepared to explain in many cases. So this is really an area that we need to think about. And again, how that fits into the normal workflow where a physician might have seven minutes to spend with their typical patient, it creates a real burden. There's a summary manuscript and an implementation roadmap that's now under review that will report the outcomes of this meeting. Genomic Medicine 2 then was to build on that and to develop ideas for multi-center collaborative projects to expand the group of people that was present at the first meeting to find even additional ones. And there were quite a few that we learned about to identify infrastructure needs and to establish mechanisms for sharing of best practices. And we heard from some institutional leaders some really important features about what it takes to make genomic medicine successful at your place. We heard that you need to make genomic medicine part of the institutional strategic plan. It can't be just an add-on, it has to really be part of the plan. You need to demonstrate value, especially in the cost arena. But actually one of the things that we learned in a subsequent meeting is it's not just about cost, it may be about adoption of a medication or of a test that you could demonstrate value. And I thought it was really nice to see that genomic medicine from the standpoint of a leader, actually of St. Jude's Medical Center, said this needs to be part of the cake, not just the icing on the cake. It really needs to be fully baked into the healthcare system if it's going to be successful. One of the main things that came out of genomic medicine too was a series of task forces. You can see them listed here, but there was one in the area of cancers, one in the area of family history, periodontal microbiome, clinical research interface, pharmacogenomics and genome sequencing. And these groups have been very active, a large number of phone calls. They met again at the Genome Medicine 3 work group in breakouts and continued to do some of their work. I'll just note one here that Pearl O'Rourke is leading, which emphasizes an interesting problem that this whole area of genomic medicine is going to create for us. And that is it, is that the interface between what's clinical activity and what's research activity and how does that interface become fuzzier and fuzzier as we start to integrate these things more and more. And I think that's really an important topic for us as we think about genomic medicine going forward. Genomic Medicine 3 then reviewed the early stage deliverables from each of these task forces, started to identify barriers and focus on them, particularly with this whole question of reimbursement and some of the other stakeholders. And to develop additional approaches for collaborating amongst the genomic medicine centers. You see this one comes up at virtually every meeting. What are the barriers that were identified? Well, the lack of evidence for benefit and value, and hence we need to think about how we get some of that evidence. Institution and physician acceptance, one of the real challenges everybody reported was that even at a major academic medical center, there is a wide range of willingness to become involved in these kinds of activities. Education of patients and physicians is a key element that needs to be worried about consents. Are our consents appropriately constructed to really allow us to take advantage of this? There are issues about sample availability and the breadth and extent of those. And there are also a lot of issues around the recruitment of participants in genetic studies. These kinds of genomic medicine activities create a real opportunity to do that, but they also create some social and ethical issues, I think that need to be well thought through before we get too far down that path. So the conclusions from Genomic Medicine 3 were to take up the challenge of partnering with payers to identify evidence for outcomes. And this actually is going to be the topic of another meeting that will be held in the Natuelson future. There's a lot of discussion about whether we need a center for medical technology policy and how does that get implemented. We thought a lot about how to design study and metrics to do the right things for patients in these space from the payers rather than trying to understand why they are paying. So it needs to be part of the process of why do we need to have this as part of the medical process. And then there was a lot of discussions about needing to broaden this beyond NHGRI and look at other ICs. How have they partnered with folks to make sure that new discoveries in their domains actually get integrated into healthcare? And there are a lot of obvious partners you could imagine meeting with. And then finally, as comes up over and over again, and in Eric's report about the IOM Roundtable, the economic discussions are a key driver of this and need to be an important part of our efforts going forward. Currently under development are plans for Genomic Medicine 4. It's going to be held in January. And the goal there is to bring professional organizations and societies who actually already have experience about establishing clinical practice guidelines and how do you integrate new discoveries into that. So we want to learn what their approaches are to do that, understand the process for establishing new guidelines, and to explore how to facilitate integration of Genomic Medicine into best practices where it makes sense. Finally, there was a meeting that happened in December, which was on characterizing and displaying genetic variants for clinical action. This now, as you know, is outstanding as an RFA. This was one of the things that came out of this group. The meeting gathered perspectives on the need for such a resource. The group was unanimous that a resource such as this was needed. The current approaches for identifying variants could be greatly improved, and we needed a place to capture and be able to annotate those kinds of variants and how they link to phenotype. We recognize that there's a translation loop here as phenotype-genotype correlations are identified. They will constantly be changing in terms of the evidence that supports them, and so it was clear that there's a translation loop that needs to be put in place so that as new evidence is uncovered, as identified, that that new evidence gets put back into this process and either upgrades or downgrades any recommendations that you might have for thinking about the clinical utility, clinical validity, and then how to think about translating this into practice-based medicine. There were a variety of recommendations that came out of that. The key elements being we need to capture these variants and the clinical annotations that go with them. We need to promote annotation of both germline and somatic variants. We need to capture all variants and associated phenotypes because as people uncover new variants, one of the questions they keep asking is, well, has anybody else seen this before and what does it mean? We need to support research into this whole question of clinical utility, clinical validity. We need to facilitate the developments of best practice guidelines, one of the topics for future genomic medicine meeting. We need to encourage dissemination of support logic. All of this only works if the physicians that are actually using it in the clinic have decision support tools that enable them to have additional information available to them to help make their decisions. And then to develop mechanisms to update this new data so that this translational loop that I described can actually be implemented. So to conclude my part of the presentation, I'll just say that it's thinking that it was a year and a half ago that the first group met in Chicago. There's been an amazingly rapid evolution of this process. We've gone from really realizing that there actually is a lot of activity out there. We need to capture some of its very, very exciting and that all of this has been funded on really local institutional commitments. That healthcare providers actually care about genomic medicine. That was a clear lesson that we learned from the Genomic Medicine 2 meeting. At the Genomic Medicine 3 meeting, we actually had a few payers and we learned that they too also care a great deal about this process. So we've gone from finding out what's out there to actually moving towards a meeting that will actually help us think about, okay, well what does it actually take from payers and providers to put this into practice? So I would say that in a year and a half, that's been a pretty rapid evolution and Terry used this slide in the presentation that she and Jeff Ginsburg gave a year ago where she said we were going to try to avoid meeting how, I have to say, we failed. But I think all of us that have been involved in the process think that it's been well worth it. So, Terry. And I might note, yeah, we solved the problem of the cold coffee by not giving any at all. There are solutions. Okay. So I thought in addition to talking about the meetings we've held, and I don't think it's even been a year and a half, Rex. It's been, you know, since June of 2011. Talk a little bit about, and actually you've heard these from Eric, a number of funding opportunities, the pilot demonstration projects, the clinically relevant variance resource formerly known as ClinAction, and the genomic sequencing and newborn screening disorders. So a number of things out there and waiting to come back in. Last fall, we also told you about a potential partnership with the Pharmacogenomics Research Network that's led by NIGMS, but we participate in with a number of other NIH institutes. And this was mainly building on there what they call the very important pharmacogenomic gene sequencing platform, which was something that they had funded and we contributed to in a minor way, developed to identify rare sequence variants in 85 pharmacogenetically important genes, and these were identified within that consortium. And the Emerge Network, our premier genomic medicine program, really is in a position now to apply this validated VIP array for discovery and clinical care in about 9,000 patients. So really, you know, a dramatic provision of increased information about the platform as well as its use in care. One of the advantages of this platform, many advantages, are that it can be exported to other CLIA-certified labs so that there isn't a mechanism developed to not only assay these variants, but to get them CLIA-certified, permit genotyping of common and rare variants, as well as discovery of new ones, and use the PGRNs, what they call the CPIC group, the Clinical Pharmacogenomics Implementation Consortium, which develops guidelines on how one can actually use these, and these are published in peer review. So a very useful resource, and so we're partnering with them on that as well. And so what we see, the PGRN contributing is a state-of-the-art pharmacogenomics array, the ability to update that array because a number of high-octane sequencing centers are involved in developing that. They also have the gene guidelines to help us with, and CLIA standards and quality control while Emerge can provide less pharmacogenomically focused labs. So really kind of take this out a little bit more into the community, not, you know, totally away from the academic setting, but at least less of an emphasis there. A very large patient base with electronic phenotyping and a strong emphasis on privacy concerns and other issues related to consent and agreement. So this was endorsed last year as a very viable way to move forward for Emerge and for the Institute. And we might just note that partnerships like this permit rapid responses to pressing clinical questions, and you may have heard of this very sad report of actually multiple fatalities after tonsillectomy in children given coding who had functional gene duplications for the CYP2D6, which is the major gene that basically metabolized or responsible for metabolizing close to 25% of prescribed drugs. And it caused significantly greater proportion of production of potent morphine from its parent drug coding. This too happens in adults, but when it happens in a small person the effect is obviously much, much greater and has led to some deaths. This actually led to the FDA issuing a press release just a few weeks ago with a black box warning about use in some children following surgeries could lead to death. So we see this as an opportunity then to really be able to jump on things rapidly. Wouldn't it be wonderful, for instance, if we could survey 9,000 people and find out all the variants or at least all the variants seen in those 9,000 people that might be associated with reduced, sorry, with increased production of morphine and have that information available to be used and shared and explored more in a clinical setting. Just to put all of our efforts kind of in some kind of a context kind of listed up here, our existing efforts, the Emerge Project and Emerge PGX as I've mentioned, the Clinical Sequencing Exploratory Research Program, the Phoenix Project, looking at developing phenotypes and the genomic medicine meetings and then the sort of dotted line separates those that are kind of in solicitation or in the idea stage right now. And one can consider this of the areas that our genomic medicine working group as well as the genomic medicine meetings have identified as kind of a continuum series of progressions. Variant and Association Discovery really I think we've defined as not being part of genomic medicine, it's preparatory too, but certainly our major two programs are doing that. But then, you know, do we have transportable phenotypes so that we can share this information across studies and actually gather among the rare variants who might be having similar phenotype exposures. Emerge is heavily involved in developing phenotypes and the Phoenix Project is totally devoted to developing transportable phenotypes. Evidence generation variants, the clinical implications of these variants are also very important steps. Consent and community concerns need to be explored. Reporting of these variants to patients in their use in clinical care and I should mention the Return of Results Consortium, obviously working very heavily in this space. Clinician and Patient Education Policy Development. And you can see that as I mentioned previously, Emerge is probably our premier genomic medicine program currently and it's involved in all of these areas to a greater or lesser extent. Cesar is also heavily involved in these areas. The Phoenix I mentioned in the phenotype realm, the genomic medicine meetings have addressed several of these and are giving us additional ideas for how best to pursue them. And then we can anticipate, although we don't know what's going to come in in applications and what will eventually be awarded, but the clinically relevant variants for search obviously is squarely in looking at the clinical implications of these variants. To some degree also, clinician education. The Genomic Medicine Demonstration Projects cover a large number of these newborn sequencing, obviously a large number as well. Family history I should have mentioned, there should be a Consent and Community Family Concerns check here, but also evidence generation and patient education and physician education will be heavily involved there as well. So we see this as a moving target, but an overall whole that we hope to make even more whole as time goes by. Rex's very optimistic slide that I stole is that we might actually get there before 2020 in terms of getting more to the right. And just like to recognize the contributions of many people supporting this effort, most particular the people who participate in these meetings and who come many of them at their own expense to provide their ideas, to look over our ideas and to give us input. And especially the NHGRI folks who provide the infrastructure, particularly our web team that is now video casting or videotaping all of these meetings and posting them on our website for anyone to be able to look at. And I would just note the professional societies meeting that Rex mentioned will be in Dallas in the end of January. So I'll stop there. Yes. Terry, there's a lot of activity in the community around the creation of a large database of genotype, phenotype variant, phenotype information. And there was also a big meeting about a public-private effort in this regard with pharma a while ago. So I'm just trying to understand in the larger picture how these things leverage each other, fit together, don't fit together. Sure. Well, and I think much of that database and association of variants that's going on is in the sort of the clinically relevant variance initiative that we're bringing forward, the relationship of that. We saw that, you may remember in the two discussions that we held of that, we saw that as being the step that comes after one identifies what the associations are in the consensus process building on those databases that we are delighted that are being put together, but we're not actively participating in those, but interacting with them. So you think of this as further downstream? I wouldn't say much, but further downstream. Yes. Terry, I wanted to ask if you could comment on the piece that I didn't see talked about here and that is decision support and engagement with some of the electronic medical record providers in particular. Yeah, and Rex probably should comment on that as well. The Emerge project is heavily involved, particularly with Epic and Sturners, and so those are the two main ones that are working, not to mention any proprietary names unduly, but those are the two that are heavily involved with Emerge. So that is going on as well as some discussion about should we bring, we invite them to our meetings and should we bring them and specifically focus on the electronic medical records specifically. Rex? Yeah, I'll just mention that one of the goals of the current Emerge group, we have one of our working groups is focused on integration of genomic information back into electronic health records. We have a very strong presence, as you might imagine, of groups that are using the Epic platform, and so there's been a real focus on trying to think about how best to go about doing that. One of the deliverables I think Emerge will be providing is an approach for both how do you develop a logic for doing decision support, and then how do you actually physically integrate that with an electronic health record system. And it's early days, but I think the thinking is that models are probably likely to look fairly similar to the way radiology deals with having very large external data sets that then integrate into an electronic health record. It seems like a pretty good model that's attractive. We did have actually all of the three big three electronic health records providers attend an Emerge meeting, and I think it would be fair to say that what we learned from them is that they were looking to us as much to provide them guidance about where they should be headed as the other way around. I was going to say, but my experience with that is that they're waiting to hear that their customers really want it. And if they do. And so I think they will be hearing at least from the Emerge project some good ideas. Lon. A quick question on your definition of scope there. You said genomic information about an individual. So is that including bacterial viral genomes or even somatic variation in cancer? So as you saw, the goal is to include somatic variation as well as germline variation. I think it's early days about whether you'd actually have human microbiome samples for any of them yet. We just need to get the human part of that down first, but that's obviously an important next step that we've learned a lot about from some of the NHGRI work that's happened in the last six months. So related to the somatic piece then, have there been many discussions in the payer and reimbursement space about learning the lessons that are ongoing right now from oncology in cancer? There have been some. You notice the genomic medicine working group has a cancer, genomic medicine working group has a cancer working group. And that's really been an important topic of discussion for them is to think about not only best practices and lessons that we've learned from that, but also to think about specific projects where we can focus and get new data about some of the changes that we see in genome versus, in the genome, germline versus somatic. Obviously TCGA has a lot to offer us in that. So thinking about how to integrate with TCGA is a good opportunity for the future, but it has not been an area of focus yet. Yeah. So Rex, it sounds like almost everything or indeed maybe everything is a limit variation and not the sort of readout. You said RNA may be okay but nothing downstream. Is that, is like measuring RNA is certainly in our experience and I'm sure Rick Wilson would say this too that combining other readouts, especially when you integrate them is actually really valuable. Epigenetic changes to DNA methylation stuff like that. Is that just put aside for a while? It's fine if it is. I just wasn't sure if that's what you meant when you said. Because you're focusing on the definition of genomic medicine that was presented. We could revisit the whole debate that we've had in the group about what the breadth of that should be. The goal was just to focus initially on primarily DNA. I don't even think we could honestly say we've gotten very far towards the RNA piece of this. As you know RNA complicates systems dramatically because you need to capture the right tissue from the right time so I think right now the focus is really pretty much squarely on DNA, primarily on germline but I think ultimately there will be expanding to somatic. I don't know if you want to add to that. Well, I mean all of you were provided, we aim for a one sentence definition which ended up being a one pager for how many bullets. It was because we needed to fill out so many caveats to just make sure we were an overly narrow, overly broad. I will say, I mean it's a second bullet there. What I believe here is that by genomic, NHRI means direct information about DNA and RNA, putting the study of more downstream products derived from the genome, proteomics, glycomyx, metabolomics, as outside our immediate focus view of genomics and therefore genomic medicine. All the time that might be of interest but I would include epigenomics, I would include RNA analysis. I think biting all that off right now when they're thinking of the more clinical application might be a bit much but I still think it should be within our near-term view. Let me tell you why I'm bringing it up because we do a fair amount of this, not delivering to the patients but we do it in our research and we're finding out by integrating DNA methylation, RNA, microRNA, a whole lot more than we are by sequencing the genome. It's complicated but it's also reducing the genome. We're not looking at everything. In the case of cancer, for example, I think there's a lot of thoughts about cancer, diagnostics of the future, not just be only looking at genome sequencing of tumors, a much broader view. So I think we should be very open to that. Technical one, the PRGN array, that's really sequencing though, right? Yeah. Okay, this array meaning array of genes. Oh, I'm sorry, yes. Yeah, no, we're going to get all, we'll get the rare variants that are there from that process. Okay, thank you. We are painfully on time. And everything's going so well. Well, I'll fall apart after lunch. So we're going to break for lunch now then as scheduled and I think we should just reconvene at one o'clock. It just seems to give everybody a chance to also talk. I know everybody likes to catch up and various things. So let's just have an hour break but let's sharply reconvene here at one o'clock.