 Thank you. We are now open for questions. I don't know if Steve, if you have any responses. Yeah, that was a great story. I think our community has been poor at advocating for articulating and gaining evidence for the importance of a diagnosis. One of the things as an internist that I'm aware of is that achieving a diagnosis can be important in one slide, especially with the cost, if you leave out the post-doc cost, speaks to that. And I think that, you know, with a view towards what Cesar can do going forward, I think that perhaps an effort to document the value of making diagnoses in general, both economically and from a patient perspective, could be very important because we run into all the time from the payer standpoint. Expand that, which is, it's just not the diagnosis. What we're seeing is that having the diagnosis leads to, in some cases at least, new treatments being discovered. And so it starts a pathway that is much more useful than when we say it just provides a diagnosis. Right, it can, but I think we have to be very careful there because that message can dilute, it can dilute the inherent substantial value of a diagnosis. So the success stories where you actually get treatments, you know, we've had them too, they're wonderful, right? But I think that what we have to articulate as a community in addition to that is that simply getting a diagnosis is important. And we don't want to dilute that message by implying that it's only the home runs where you have a treatment that are worthwhile. Yeah. Is there a role for NHGRI in sort of formalizing, organizing, helping, facilitating the crowd sourcing of the phenotyping? That's question number one. And my other question is there are 265 variants in NGLY1 listed on the Exome aggregation consortium server. How many of those have any kind of phenotype at all? And of course that's sort of a loaded question because the answer maybe you don't know, but maybe Caesar or Emerge or somebody at NHGRI can figure that out. So I think there's two really good questions. So obviously I think there's tons of research left to be done on how best to utilize the internet to interpret variants. I think there's obviously a lot of potential for it. But where I say it's very early days. I'm doing a lot of this on a very ad hoc patient by patient basis. But there's certainly some good science to be done here. In terms of interpreting NGLY1 variants, I mean we have a pretty good genotypic database at this point. But of the 265 that could be pathogenic so far, obviously we've only seen at most two times 39 of them, which in fact is far less than that. So it's really a case by case basis when it comes to interpreting pathogenicity in NGLY1. One of the other lessons from this is in the conversation this morning about learning healthcare infrastructures. We need to think about how to make that sort of patient and family centered or patient and family partnered infrastructures and not simply clinician or health system driven infrastructures. Sharon and Gail. Yeah, I just wanted to say that thank you for the talk. And I think that we should... I think you've only shown us a small piece of it. So I'm a type 1 diabetic and there's a parent group called We Are Not Waiting for the NightScout Project. And I follow them on Facebook and they've literally hijacked the data from glucose sensors. They've created on the cloud databases. And the parents who technically don't know how to do this, there's a group of computer savvy parents who then every day on Facebook help these parents set up much, much better, let me tell you, much, much better web page showing their children's data than the current FDA approved version. And so I do think that, and their logo is We Are Not Waiting, and so I do think that we should really think in Caesar 2.0 more about how to engage the parents and the families than we have in Caesar 1 when it was all very new because it is just amazing that wealth of energy and activity that I don't think we included in our projects. Thank you. Boy, this better be good. You've had to wait for 25 seconds. So I really want to congratulate the panelists and also the comments that came out, especially Sharon's, had some similarity to what I wanted to say. I've been sort of waiting for this moment in our discussion to ask people to seriously consider how information is not under the control anymore of researchers and clinicians who feel that they need to exercise a very strong moral obligation to return or not return because of professional norms and guidelines, et cetera, et cetera, that the world we're in now, and there are these terms, citizen scientist is one, but then there are a lot more kind of offbeat punk, science punk, results punk. I mean, there's a bunch of stuff that's going out there that, believe me, I can't even understand. I don't tweet, I don't go on Facebook. I mean, I'm really old school, but I've heard talks about it. And the issue I think is really important for Caesar 2 is to address issues about who owns this information now, who controls it because I know a lot of folks who are participating in George Church's studies for example, and who are on the Internet, who are parents, who are disease-focused groups, et cetera, aren't waiting and are taking charge. And I think this is a big challenge to NIH and to Caesar to get on board and to really understand what would be good evidence and not, and to really help people. And so I applaud your helping, and I guess I just think this is a challenge for all of us. Katrina Armstrong from MGH. I just wanted to pick up, I think, on two points that I agree offer, I would say an amazing opportunity for the next version of Caesar. So one is just to echo the sense of the importance of diagnosis, and in some ways, the opportunity for the Genomic Medicine Program to lead in reintroducing that into medicine altogether. So I will say that one of the things that I found incredibly distressing was about two years I discovered that internal medicine residents are now graded on a million different things. There's some 20 competencies in particular, but the ability to make a diagnosis doesn't show up on there at all. It's not one of the competencies that we consider fundamental to internal medicine. And so I think the concept that that is, but I think the key piece is that those competencies were actually driven by a belief that we were focusing more on our patients and that actually making a diagnosis was kind of a doctor-focused thing, and that actually I think your story brought, I think, home, the concept that having a diagnosis is an incredibly patient-centered moment, and being able to frame that in genomics could be an incredibly powerful thing, I think, for pulling the whole field back. The other thing I would just say is I think your story also brought up this key point for Caesar of how critical it is to engage the communities, whether we define the communities as an area or a community and a population of an area or a community with a disease, and the opportunity, I think, for Caesar 2.0 to really focus on how that could become part of this triad of the tools, the delivery, and then the community. Apilard, you had a comment? Yeah, so a couple things. One is I wanted to agree again on the importance of diagnosis, and we've seen this at our own institution that we have three major healthcare payers, one of whom just says basically we won't pay for genetics except in these very, very narrow contexts, and so this argument about the importance of diagnosis is something that our practitioners are trying to develop and make back to payers, and they're kind of gathering their own data, but it would be really useful to have additional data. At the same time, I also think that it's very important to actually study the impacts of the kinds of activities that we heard described today because, for instance, we've heard some success stories. I'm sure not everything is a success story, and for sure not every patient who gets back genetic information manages to find other people who have similar mutations and manages to find actually medically beneficial interventions, right? And so having some actual data as opposed to... I love the really great stories. I have a student who actually did something similar, but I just think actually having some data would be important, and then my final point is that I also think not... there's a lot of variability in how people respond to this kind of information, how much they want to put themselves and their kid out there on the web. So again, we shouldn't assume that some people are doing that, that everybody would want to do that, which I think complicates our job in thinking about how our policies should relate to the communities out there, right? Just to follow up on that, I think one of the things that CSER has been very successful at is presenting to patients what do they want, what are the important parameters when they're making a decision? Is it how long you'll live? Is it pediatric versus adult onset? What are the categories you want? You can't ask a patient about thousands of conditions one by one. Do you want this? Do you want to do this? So how do you put those together in a way Katrina's done a really nice job with her carrier screening for classification? Other studies are collecting really good data on what proportion of patients would want VUSs versus would not want them. That's some really important data because we hear from the people who want everything but we don't hear much from the people who want nothing and they're out there. Actually, Gail, I'm going to follow up on your comment, which I thought was a very useful comment, that right now we've been talking about this in the context of people who have some phenotype or some problem. I think your comment is even more salient if we think about healthy individuals who don't have a problem and think about secondary findings. Bob, okay. I wanted to ask you to reflect a little bit on the following question, which is there's been a lot of data now, as Gail mentioned and others, on how patients respond to particular information or what kind of information would they want and how would they like to return to them. Are we done now with that question and can we now go on? Or is this something which is going to change over time as society changes and as generations age and do we need to ask these questions again? I don't think we're done with the answers to this question. I think from Caesar over the next year or two years you're going to see a lot more information about how people actually respond when they get the information back. And there may be some surprises in there, like people may be very interested in getting a piece of information related to their phenotype, but then they may lose interest in getting the other pieces of information that are not related to their phenotype but that are more secondary. I think implicit in your question is is there a generational effect of older people thinking about it one way and when the younger generation comes along they'll think about it very differently. In our experience in our Dana-Farber project with a mostly older adult population is great enthusiasm about getting results back, at least when asked initially for preferences. I have no reason to believe that younger generations are going to be less enthusiastic about it, but there's not a lot of room to go up, if you will. But I guess we'll see. I also think that there's a like many other areas of science it's iterative in the sense that once we answer certain questions about this topic we'll find new questions and new nuances to explore further. To add one thing, I'm not yet sure that we've figured out how to phrase the question directly to patients or families or people about what it is that you want back. We may be asking the question in a very primitive way. Yeah, I would underscore that. My sense is that certainly in our project that there's so many things that have moved and the way that we ask questions now, I mean the field changes so quickly it would be great to have an opportunity in 2.0 to reformulate the questions. So the technology has kind of advanced that even the way of understanding when you say would you like the information we can ask about uncertainty and different ways of phrasing that we can ask about how it might impact their course of care and the use of there's a whole different series of ways that we're going to want to ask the questions and then there's the diversity so when we go to populations that haven't really been part of CESA 1.0 and then how do we even ask those questions and pursue that there. So I think there's a lot to do and in fact it's crucially important to maintain that because I don't know again I mentioned this before but I don't know that there's another funding mechanism that allows us to ask those questions in parallel with pushing the latest out there. Amy McGuire from Baylor College of Medicine. So I just wanted to I find it very interesting the tension that's being discussed because on the one hand I think we're all acknowledging that there are individual differences in how people feel about these things and I think we're seeing that in our CESA projects and I think one of the major challenges that we've had within the CESA consortium is to account for individual and group differences and trying to distinguish sort of how people hang together. You know it's based on their particular clinical presentation phenotype it might be based there are certain people who are inherently more enthusiastic about this or less concerned about their privacy or things like that and so I guess it's kind of a question kind of a comment but that's in tension I think with our need and desire to have sort of standard policies and sort of guidelines in place for how we treat all patients and so there's been a very robust discussion about sort of not everybody wants to have their kid out on the internet some people do and how do we from a policy perspective address those differences and I personally think that CESA and CESA 2.0 is really well poised to try to figure out how we categorize people and we can't do it on an of one individual basis right we've got to have some unifying categories but how do we account for those differences among different groups and what are the relevant groups and how do we adjust our policies to take that into consideration so I just was wondering I don't know if Steve or anybody else has thoughts on how we can best address that in the next phase I mean I think we're addressing it in this phase because in some sense we're talking about bringing genomics into medicine and so it's inherently individualized right you might have some defaults in ways that you approach populations in some places where there are sort of policies that are enforced across all population but it's when it's one clinician with one patient or one family there's inevitably going to be a component of individualization to it and I think that's one of the places where CESA has or the work that CESA and the CESA investigators have done has had a lot to say because we're learning how to individualize that in individual interactions with particular patients so how do we make some of the decisions that are to be made here what you might call preference sensitive decisions and how do you make preference sensitive decisions together with families at the same time for example what are the defaults that you might have in place so if somebody does you know what are the sort of presumptions about the ways that you might go and if a family wants to do something differently then they go a different way so I think CESA is informing both of those things how to sort of negotiate those individual preferences with individual patients and at the same time what sorts of defaults make sense across a clinic across a health system etc so I hate to be totally provocative but I'm from Vermont where healthcare reform is raging and so I think that a lot of the research that's been done by CESA comes from a medical genetics perspective which comes from a background of looking at genetic information as exceptional as should be protected and I wonder if we wouldn't have a different approach to this so my ultimate question is how much choice should families, individuals patients be given in knowing or not knowing information that's medically important and could save money within the health system and provide better healthcare I think your question is potentially one of the further areas for research in the next in CESA 2.0 in other words not only the empirical issues of how does this information impact the health system but how should it and I think that so I think I don't think there's a simple answer to your question but I think that's one of the areas for further work right and thinking about like a public health perspective of people don't have choices about sexually transmitted diseases being reported and follow-up being done Heidi so I've just been thinking about Bob's question about do we know enough about the choices patients have and what they want back and are we done with that and you know this question of some patients want back VUS is others don't you know we've figured if Matt had been given not just two variants that were almost known but instead a VCF file with thousands of VUS that were probably in his son's genome he might not have gotten as quickly to where he is today and so the point is that not all VUS are created equal and I think that one physician getting a report with a VUS back on it and I've seen this happen all the time oh I'm sure it's pathogenic and they just don't want to make the call to the other physician who probably rightly says most VUS will end up being benign which is actually what we understand today including the list of 100 that Dan just found on EXACT so I think we need to do a better job of trying to figure out how how to return variants that are uncertain you know and what do we do with that and try and figure out how we'll make them certain in the future and try to contextualize and figure out what to do from there two really quick comments Jim and Debbie then we have to turn to the end I just want to echo what Heidi says I think it would be an abdication of our responsibility where we to not make any judgments we have to make some judgments and there will be things that are more actionable than others and I think we have to figure out ways of adjudicating those things Debbie but I just wanted to bring up too that it's obvious that the public is interested in engaging and perhaps in the next phase of CESAR we could think of some way to accomplish this in the Mendelian Centers we're starting to develop a website called mygene2 where people can come to us with their clinical sequencing data and it can become incorporated in Mendelian findings like Matt if he got a thousand variants or 500 variants could have come and his data could be put together with lots of people's data so if they want to engage in this they can, right and I think more and more people will use social networking as a way to get information that's really important to their families and to their diagnosis so I think that CESAR should think about using this in novel ways too because there are obviously people who want to engage and maybe this is a way we'll identify other cases of that variant that we're not sure of and could get more clinical phenotypes associated with that. Well thank you I want to thank Steve and Matt as well as a group for this great conversation I'm going to turn this over to Shanita to make some summary comments now