 So, this is going to be moderated by Pat Deverka and she'll introduce the panel. Okay. Good morning. Welcome back, everyone. Hopefully you've noticed that we have a slightly different format that we hope is really conducive to a lively discussion for this morning. And as you just heard, you're at the panel of what evidence is needed for implementation of genomic medicine and when is there enough? I just want to point out that we heard examples yesterday of health systems such as Geisinger, North Shore, and Intermountain deciding that in specific clinical context, there is sufficient evidence for implementation. We also heard about how they're building the processes and infrastructure to support this implementation while also evaluating both implementation outcomes such as feasibility and fidelity and health and economic outcomes such as survival and direct health care costs. And we also heard that there are implementation frameworks and toolboxes that have been customized for genomic medicine and used in the precision medicine context. These are used to support implementation research as well as program planning and evaluation. Today we're going to turn to explicitly address the question that has permeated each session's discussion yesterday and today about what evidence is needed for implementation and when is there enough? And this all posit is a complicated question that often results in the answer, it depends. And what we've heard so far is that the answer depends on the type of test, whether it's predictive, prognostic, screening, or diagnostic. The clinical context, for example, would we think about the evidentiary requirements for seriously ill patients such as patients with advanced cancer or in the same way that we would think about screening of healthy individuals. What about the setting, the academic institution or the community-based health system or ideally as we've heard a high functioning learning health care system? What about the presence or absence of supportive organizational leadership that has a clear vision of the value of personalized medicine? We certainly heard about that, it depends on that. And it also ultimately depends on the specific stakeholders such as payers, both public and private, as well as practicing clinicians and of course patients. And so today we have an excellent panel brought together to address these issues from diverse perspectives. I'm going to introduce the panel and then have each of them lead off with sharing their perspectives. So let's see, on the far end we have Donald Kearns, who's president and CEO of Rady Children's Hospital and Health Center and he's going to be sharing the Children's Hospital CEO perspective. A similarly sort of executive level health system perspective is going to be shared by Keith Stewart, who's director at the Mayo Clinic Center for Individualized Medicine. Next we have Jay Wogelmuth, who's the chief medical officer of employee health and wellness for quest diagnostics, who's going to be talking about population-based precision care from the employer employee perspectives. And finally we have Bob Nussbaum, who's the chief medical officer of Invite, a genetic information and diagnostic company, and he's going to be sharing that particular genomic testing company perspective. I think it's worth mentioning on the panel that today that that empty chair over there was going to have been filled by the payer perspective, so it's just worth pointing out that we did want to have the payer perspective. But unfortunately John Wilson of Optum was called away and has a professional conflict that prevented him from being here. So the process that we're going to use today is that each speaker is going to share their perspective for three to five minutes on sort of that broad overarching question. And then I'm going to lead off with a question that I actually shared with the panel to prepare for today and get each speaker a chance to respond and then we're going to open up for discussion with all of you. Okay. So Donald, do you want to lead us? Sure, absolutely. And we all have a sense of humor here, right? Yes. The fact that the payers didn't show up, is there a message there? So I think my take on this is that it depends as, of course, absolutely accurate. And we are, we're the children's hospital here in San Diego. We're the largest children's hospital in California, 551 beds. And our passion for genomics was partly due to the fact that Mr. Rady, we're Rady Children's Hospital, so he's an important guy, by the way, woke up one day and gave us $120 million to open up a pediatric genomics institute. And I will tell you that the person that ran the genomics institute for the first year or so had a hard time spelling genomics. And that person is me. And I spent a year trying to make sure that we could get the Mick Jagger of genomics, Dr. King's more involved. And I will tell you for that year, as we created this genomics institute, we were heading in a very different direction than we ended up. And my very dear friend who is our chief financial officer, as we were creating our business plan, turned to me and said, Donald, I promise you that this business plan and the genomics institute are going to be completely unrelated. In other words, we're going to do something completely different. And when Stephen arrived, he talked to us about what our plans were, and he said, well, we're not going to do that. We're going to do this instead. And what we've created in our institution is a passion for treating the sickest babies. And we've, Stephen, spent a fair amount of time making sure that our doctors were on board before we opened this project up. And our concentration has been in the neonatal intensive care units and in the pediatric intensive care units. The challenge children's hospitals have is that we all live on a razor-thin margin. And the only way we were ever going to be able to do something like this and take this on was via philanthropy. And so we've been very successful in creating more and more philanthropy to help support this. But the challenge all of us have is trying to figure out how to convince our friends, our payers, that this has become the standard of care for medical reimbursement. And I think that what you're doing here today, what we spend a lot of time doing, is talking to folks about how this is going to change care. And what we've discovered over time is that we actually make a significant difference in the kids in whom we are treating. So we've recently published some data that has demonstrated that if you take the long view of this, we're actually going to save a tremendous amount of money for payers. And in addition to that, we're going to do the right thing for these very, very sick kids. So as we move forward, what we are attempting to do is to try and spread that gospel. We've just recently received funding from the state of California to be able to sequence Medi-Cal, Medicaid for you and not in California, babies here in California. This is a big step forward. We just received a $2 million check to sequence these kids. And we think that this is going to be a big part in helping third-party payers to understand that this is actually the future for them. One of the things that Rady does, which is a little unique, is we have a very large managed care plan. So we take risk for close to 400,000 children in our region. And that allows us over time to actually understand what this looks like and start to allow our kids that we take risk for to allow those kids to be sequenced also. But what we've chosen to do, where we've chosen to get started, is around the sickest kids in the NICUs and PICUs. And then we'll extend from there. So we have recently received philanthropy for brain tumors. We think that's a big part of our future also. But our real passion right now is in the sickest kids. And we think that we're going to be able to prove that that makes a difference in a very short period of time. So my answer, when do you have the evidence to be able to do this? I would suggest in a lot of instances we have the evidence now. The challenge is getting our partners on the payer side to understand how important this is as we move forward. Great, thank you. Keith. Thank you very much. Thanks. I start by introducing myself, I think. And then I want to do three things. I want to tell you a little bit about the Mayo Clinic Center for Individualized Medicine, which will give you an understanding of the foundation that we're coming from. Talk a little bit about when is enough evidence present and then briefly about our strategies and thinking around that. So I'm a hematologist by training. I treat multiple myeloma, it's a rare blood cancer, and I've been engaged in the sort of genomics of that disease for many years. After the last three years, it's been my privilege to lead a Mayo Clinic Center for Individualized Medicine. I'll tell you a bit about that in a second and probably relevance to this audience. Last year I joined the board of Genomics England so I can maybe give you some insights into thinking there since I don't think there's anybody else representing them today. So the Center for Individualized Medicine, as you heard yesterday from Geisinger in North Shore, I think part of the success has been that it was a strategic decision by the institution and at Roddy Chilbham Hospital, it sounds like, to start an Individualized Medicine program. So it came from the top down, which I think as we've heard from many of the systems has been part of the reason we're able to move quickly. Our job is to deploy genomic medicine across all of the specialties in all of the Mayo Clinic campuses, and we are given substantial funding with which to do that. We're left alone beyond that, so it gives us flexibility. We're encouraged to look at external partnership and business development so that we have a more sustainable model. With respect to what constitutes an off evidence, I think you took the words out of my mouth, it depends. I think in Mayo Clinic we believe that the future of payment is more about value than it is about volume. We think that that's how we will end up being paid for our services, and precision medicine obviously sits at the center of value. It's both the promise, but also the Achilles heel of the technology. Our value equation is quality over cost, so that's patient outcomes, safety, and satisfaction over the cost of care over time. So if we look at precision medicine in, for example, targeted therapy and lung cancer, we have a cheap, very inexpensive test. We've proven value in terms of extending survival, and so the value equation is very obvious. We will look at sequencing populations of people with a healthy genome test. We still have a relatively low cost, but we run the risk of increasing healthcare costs rather than decreasing them over time, and we still have not proven the value, so the value equation there is still to be determined. I'd say that in terms of our strategy, in terms of when we accept enough being enough, is based on perceived, proven, or payment. So perceived value, proven value, or payment for the services we offer. I'll give you some examples of perceived value. Today, the use of cell-free DNA for monitoring or detecting cancer early while it's still curable, everybody in the room would say that that's a perceived value. We don't yet have the evidence, so our strategy is clinical testing. Proven value, and I would say also, and I'll get in trouble for this, pharmacogenomics I would say is still in the perceived value realm. We all think it should be helpful. There's not a lot of hard, concrete proof that it actually changes healthcare outcomes yet. So it's somewhere in the middle where we're doing clinical studies, but we're also deploying. Proven value would be something like hereditary cancer. Through the work of MSK Impact, we now know that 10 to 20 percent of all solid tumor patients have an inherited predisposition to develop malignancy, and we also know that 50 percent of those are missed using standard screening criteria. We think that's enough value for us to start to deploy that in our practice, and we do that through practice improvement projects. Usually pilots, we're doing one with NVT today in 3,000 solid tumor patients to prove not just the scientific value, but also the downstream economic value of such testing. And then finally, payment, and this is where we get to the tried and tested, and things like targeted therapy with small gene panels and cancer we now think is standard of care, and our job is to disseminate that throughout our practice and have all of our physicians benefiting our patients using that. So I'll stop there and hand it over to the next speaker. Thank you. Thank you. Jay. All right. I just wanted to say before I start that the payer is here because it's actually employers that pay for health care in the United States. And actually my story, I'm going to outline for you a little bit on my journey around coming from working with payers to get coverage policy to realizing that in a self-insured employer environment, it's not the payer that decides it's the employer who decides what the policy is. And so my quick story, and I'll get to the genetics part in a moment, I'm Jay Wolgemuth. I'm our Chief Medical Officer at Quest. About eight years ago, I took over R&D in medical. And so I was working to do innovation, working with health systems and hospitals, and working on genetics, and basically spending a lot of time with Etna and John Wilson, my friend, and working on coverage policy, which is very, very difficult to achieve through the payer system. And I think we know kind of what the standards are. I have a good understanding of what is needed to gain a coverage policy. But then I left Quest for a year. I came back in 2016, and my boss, who's our CEO, said, hey, we have a self-insured health plan. We have 60,000 members in it. Two-thirds of those are frontline employees that make $40,000 to $50,000 a year. Some of those people pay 25% of their take-home pay on health care. Our costs of health care, $400 million, increasing by 6% every year. And we have no understanding of it. And by the way, Etna and UHC, they're not managing it like it's their money and their population because it's not their money and it's not their population. Right? It's our money we pay. Okay? And also we have employees who have been with us for 50 years. Quest is unique that way, okay? But we also, on average, keep people for 10 years. So am I incentivized to improve population health and deliver new genomic precision medicine that we believe works? Absolutely. Much more so than in any other environment I've seen. And so we started a journey about two years ago. We put in place a very aggressive employee population health program that says, I have all the claims data. I have all the screening data. I have biometrics. And now I can begin to offer services to those people based on their needs and direct them directly into care. And if they need a nephrologist, we get them to a nephrologist. And if they need a BRCA test, we get them a BRCA test and cut through a lot of friction in health care. And through that, we have lowered the cost of health care for both employees and for the company for the first time in 10 years. And we've also gained the trust of the employees. In 2015, the number one complaint at the Quest Employee Engagement Survey was health care is too expensive and too complicated. In 2017, same survey. The answer is thank you, Quest. You're investing in health care. It's not great, but we're trying. We're offering services to them. And so in this environment now, just a couple of things about it. Number one, we never experiment with our employees. However, if there's a proven, and I'll say proven, theoretically proven, benefit of a certain service, like a diabetes prevention program or getting on a statin for your cholesterol or having FH testing when you're syndicated, I don't care what the coverage policy says. And I know that it's good medicine. And the real question we're trying to answer is can we effectively implement it in a population? So it's all about implementation science. So I believe actually that pharmacogenetics is valuable if it's implemented in a way where you actually can head off prescriptions before they're made. And if it's implemented in a way where, you know, the information is actually used to change decision making. And so in our environment, we're able to say, okay, I'm going to offer to Quest employees the following. I'm going to have the claims data and all the information to study the clinical and health economic impact. And again, I'm not experimenting, I'm offering them a service. And then some of them work really well and some of them don't. Last part of this then is again on the, you know, sort of the research side of it. You know, we make our decisions and we'll talk about what evidence is enough. We make our own decisions about what to offer. We try to influence the payers on their policy. But again, from a implementation standpoint and kind of showing the evidence that not only the science is there, that this would be beneficial theoretically, I forgot what terms you use, but that you can actually implement it in a population and get people to benefit from it. We're able to now do that translational work and publish data in that regard and are trying to also get other self-insured employers. I see my friend from Stanford here, we're talking about getting other employers in a region to do the same thing, to get much more proactive and offer services and demonstrate that you can move the needle. So that's kind of my base camp or description of where I'm coming from. And I think as we go, I'll talk about then, specifically how do we view genetic testing and how can we make decisions on evidence? Okay. Thank you, Jay, Bob. So, talk a little, I can talk a little bit about myself and about the company. So, I practiced medical genetics for almost 40 years and then in academic setting as well as at NIH. And then three years ago, I joined Invite, which is a relatively new company, really open to stores for business about five years ago. And the reason I joined and what I found particularly compelling about the mission at the company was that we really believe that genomic medicine needs to be integrated into routine medical care and that there were many obstacles to that. It's a multifactorial problem, but that one of the problems was that the cost of genetic testing was just too high. And so part of our goal is to drive the cost of genetic testing down to the level where people stop arguing about it as if it's some kind of esoteric brain surgery. And in terms of its cost, and instead recognize that genetic information can be obtained and delivered at a low cost, useful information. What I've learned from the last couple of years that I've been working in Invite is a couple of things. First of all, the question of how much evidence is enough. I don't have an answer to that question, but what I have learned is even having evidence doesn't mean that things are gonna change. So that's the first important thing. So let me just give you an anecdote. I wrote to the American College of Obstetrics and Gynecology about their practice guidelines on carrier screening, which I think are outdated. And the answer I got was, I'll quote, it's important to understand that the primary audience for our clinical guidance are practicing obstetricians and gynecologists. At this point, the applicability of sequencing for clinical care provision is still too early for non-genetics-focused providers to have the familiarity with the information in order to address all the complex issues appropriately. In other words, from my point of view, every time I've seen a child with a devastating autosomal or excessive condition that could have been screened for, and the only way that the parents know that they're at risk for having such a child is because they've had one, to me, is a tragedy. And there's no reason for it. And so I was attempting to try, try to get the practice guidelines changed off of the very ethnically-based or very narrowly-based carrier screening. And that is how the professional organization responded to my attempts. I think that what I've learned from this meeting so far here is that a company like Invitae can partner with medical systems that follow patients, see patients, have data on patients. But that medical system really has to consider itself a laboratory for innovation and be willing to make the investment. And that's certainly what I've heard from Mayo and I've heard from Rady and from Geisinger and from a variety of other places. That's essential. Without that, nothing's gonna happen. But also what I've observed is that for many medical systems, it's a way of discriminating themselves from others. It sort of sets their system apart and puts them ahead of the competition because like it or not, the medical care system is a highly, highly competitive industry. You talked about narrow margins, they are narrow. And so when you have this combination of a leadership that believes in the ideal of trying to innovate but also recognizes that it can be helpful to them as an institution in terms of setting the level of competitiveness, that sets up a situation where they are open to implementation of innovative ideas and genetics. And finally I should just say that I have not seen a system work where you gather data, you gather data, you gather data in the sort of the classical academic mode and then the amount of data becomes overwhelming and then somebody implements it. That doesn't work. I think the way this has to work is that leadership has to say we're going to do implementation as a research problem, that the medical center, the hospital, the outpatient clinics are gonna be our laboratory in which we're going to study how to implement this and what the outcome is going to be. And you're constantly learning it's what everyone's heard about from clinical learning systems and that when that sort of sweet spot develops in terms of leadership, the desire to be innovative, the desire to have their institution be a laboratory that a place like in VTEC can really collaborate with those folks and move things forward. Great, thank you. All right, so as I promised, I'm gonna pose a question and all of you start writing down your questions. We'll use the same process of turning your card on the side and I'm gonna go ahead and join after I ask my question, go ahead and join the panelists. So I'll do my best to see all of you and call on you but if some of the people on the perimeters want to shout out and point to where I might have missed people, I'd appreciate it. So we've already heard on our panel today that there is variable interpretation about whether there's sufficient evidence to implement pharmacogenetic testing, right? You heard it right here today but many people in this room would be on the affirmative side. So one solution that's been posed to this in many different settings but we've touched on it here today is harmonization of evidence requirements. If test developers and clinicians would actually understand, researchers they should say would understand where the goalpost is then that would lead to easier evidence generation and ultimately the goal of evidence-based implementation. So what I'm gonna ask the panelists is do you see a pathway to evidence harmonization or do you think that we will always have different goalpost and in terms of predictability for evidence requirements? So I'll just let you guys choose which order you want to go in but I will call on all of you. So I might as well start off. I think that one of the biggest challenges we all have everybody sitting here is you're all pretty unique. You're way over here in regards to this knowledge base. There's a whole lot of catching up to do over here. I was a surgeon for 25 years. I can tell you that there are not too many surgeons around who understand much of what's being discussed here. So there is a massive education that needs to occur. I don't know how that occurs. I don't know who funds it. I don't know in what way that occurs but when we talk about a harmonization of where we're gonna end up we gotta all be sort of on the same football field and I see a big, big, big knowledge gap that needs to be addressed. And I think before we can actually even start to think about creating the same goals we have to create the same definitions. We need to change the way that people are being educated about this. Folks that are my age who are fighting using electronic medical records and all of a sudden have a pop-up that's gonna tell them you need to order genetic testing here. How are they gonna respond to that? I don't know. So I think that a lot of thought needs to go into how that education is going to occur. So my take on it is, is I see that we have a lot of work to do before we get to that point. So everybody in this room believes that pharmacogenomics will be eventually successful. I had my own pharmacogenome sequenced. I'm never gonna take another drug in my life without looking at the list quite honestly. But what I'm responding to is the skepticism I meet when I go to our primary care providers, our surgeons, and say, why aren't you doing pharmacogenomic testing before surgery? You'll have better nausea outcomes. You'll have less pain issues after surgery. And they're skeptical. So I tell them, well, why don't you help us create the evidence? Help us create the value equation that we need that it will become proven. So we're doing that in a number of ways. We're number one. We have Dick Wensherborn with us today, who you all know very well. Dick has pioneered a project to put 10,000 pharmacogenomes sequenced at Baylor into the electronic health record and with about three months support, I run clinical decision support around that and get a real life experience of what this looks like. And to measure, we have 19 different groups are gonna study the value of that in terms of emergency room visits, switching antidepressants, pain in the surgery, length of stay after surgery. The second thing we've decided to do many years ago was to run the world's largest phase three clinical trial of Plavix. A lot of comments yesterday were, well, Plavix has a proven value. Well, apparently the cardiologists don't think that. So we've enrolled 5,000 patients on a phase three trial with very hard end points. Myocardium fraction stroke and death. The NIH eventually funded it to help us get over the finish line. That will finish a crew in the next couple of months and one year from now we'll have a result. I'm a bit nervous about it because if it comes back as another negative study, much like the Coumadin ones, I don't know what that's going to do for the adoption of pharmacogenomics around the world and certainly nationally. We are having an impact because we're running the phase three trial because we're doing these 10,000 patients because we're educating and using our pharmacists by the way are a very key element of this. We are seeing uptake. 200 patients per month are now having primary care physicians order pharmacogenomics at Mayo Clinic. So 2,500 a year and that number is going up month after month. So I think we're trying all kinds of strategies. We're trying to create the value that will result in some kind of harmonized opinion about whether this is valuable or not. So I'm a cardiologist by the way and it's a pain in the ass group to try to convince of anything. But in years ago when CYP2C19 came into Vogue, we actually did a implementation study with Dr. Topol's group here. And of course he's all over it and is zealot about it. And it was a failure because you're not going to genotype someone before they come in for a catheterization in a one-off sort of way. Even if you believe in it. We do, it's done in the cath lab. Yeah, but the workflow in order to say I'm going to convince every physician type that they should order a specific test every time there's an indication for pharmacogenetics. It's not very scalable in my view. And I think in the evidence, the question was about evidence. There are certain parts of pharmacogenetics. It's not even a question. You don't put someone with the, on a back of ear that has the HLA mutation where they have a fatal reaction. It's not a question. There are other areas that in psychiatry where I think it's pretty well clear that you should not use that SSRI in this person. And then there are fringe areas where the evidence may be debatable. I actually think the debate is much more about implementation. Again, implementation science, which is what is an effective way to implement this? And my view is if it would be that an individual like an employee in our plan has, which we're doing now has an opportunity to have a comprehensive genotyping, which has a potential impact across 300 drugs. And there's a great range of evidence grades there, but there's some that are no question. And now as they go forward, when a prescription is made, that information is integrated into the pharmacy benefit and a flag goes up and a pharmacist has an interaction with the doctor. So now I pay once for what is a very cheap test to do SNP chip for this, very cheap. And then that, every time a prescription's made for this person, they get value, right? And then you pick, there's some things that are obvious, like don't take hydrocodone or you're gonna die. And then there's a question around Plavix, which maybe you'll answer of should we use the information on Plavix that's within this at the time the person goes into the cath lab. But so my view on it, it's not about evidence standardization necessarily. I think there are frameworks there. But it's much more about implementation and what is an effective way to actually implement pharmacogenetics that will make sense. And if you pay once for a relatively cheap test and now I can use that information for the next 10 years, I think that's a strong case. I agree completely with your comments and I think it's expandable beyond pharmacogenetics also to cancer risk, cardiovascular risk and other areas where in a relatively inexpensive way, one can have a general genetic health check essentially and then use that information when needed at different stages going forward in the medical care system. I see the goalpost as being maybe not the analogy I'd like to use, I'd like to use it more like a croquet game where there's a first hoop you gotta get under and that might be demonstrating that there is some value in knowing and having this information. Then there's another hoop which is that that information is actually used and is usable and so I think the point for me the biggest problem with pharmacogenetics all along has been this difficulty of expecting people to use that test at the point of care when it needs to be done in advance. And then secondly is the lack of decision support widespread. I mean, Mary and other people here have been really at the forefront of getting that in front of people. I mean, does anybody at St. Jude use a six per capita purine based drug without knowing TPMT in advance? No. No more than they wouldn't give a flu vaccine when our hospital CEO says you have to use the flu vaccine. I mean, we insist on some medical procedures being done whether the surgeons are convinced or not. So exactly. So I was at a morbidity mortality rounds about three weeks ago about a patient who came in on New Year's Eve. All the residents were out at their party so everything was covered by fellows and the ER was covered by fellows and this woman had no white blood cells and she'd been given ASA thioprene. And did she have a TPMT test ever? No. So this is the problem. This woman should have never suffered that severe reaction. It wouldn't have happened if it had been a child of St. Jude's getting the stroke. Well, I think there's lots of proponents here of preemptive pharmacogenomic testing and for all the reasons that you pointed out. Okay, now as promised, I'm gonna open it up to the panel. Oh, okay, Carla, well, we have a process here. I'm sorry. I'll be a stickler. Yeah, so I saw that Mark Williams has his card up first but just flip your card as if you're at a Brazilian steakhouse. And Carlos, no, dance like a monkey. Now reset. And I'll call on you next. Okay, go ahead, Mark. So I'm gonna make a comment and then ask a specific question. The comment is just in response to Keith's comments that I think that for the most part the physician pushed back on evidence as a canard. They just don't wanna do it. They don't wanna change. If you ask that roomful of cardiologists how many of you adjust the dose of clopidogrel if the patient's on a proton pump inhibitor they all raise their hands. There's a 10th of the evidence for the impact of that interaction but they're familiar with drug-drug interactions and so they're comfortable with that. They're not comfortable with pharmacogenomics. So I'm much more in Bob's camp of this is something we're doing, folks. And if you can prove to me how somebody can magically convert a pro-drug into a drug if they have zero enzyme I'm happy to listen. But we've known this for 40 years, you know? It's not magic. This is biochemistry and you have learned this. So anyway, that's my rant. But the question is to Jay because and Jay and I have talked about this about the role of the self-insured and the fact that this is really, you know, where we need to be going that we really need to, you know, with the exception of payers that are associated with healthcare systems like ours where we have some advantages in terms of interaction that we really need to be thinking about this from an employer perspective because that is really where the biggest return on in the investment comes from and I think you're developing evidence that's beginning to show that. So the question that comes up with our genomic medicine working group is, you know, we can operate as a convener to bring people together. And I think examples like yours of forward thinking companies that, you know, could be in a room with a number of other large self-insured employers that say, hey, this really makes sense from our perspective. What's the venue where we could convene that group and really have a substantive discussion to really allow the employers to try and move this forward? That's terrific. And I'm asking myself that same question and there's several possibilities. But first, the actual venue, what you would do and what the employers want is number one, I want a basic genetics formulary that tells me what I should and should not provide to my members amongst all the genetics that's out there. And if Etna has a certain policy and I don't agree with it and I want to offer this, I can offer it even though Etna doesn't cover it. So I think there's kind of like a basic genetics 101 formulary for an employer and they're all asking for it right now. So they would flock to a room of experts to help get a baseline suggestion of what should and should not be provided to employees as a base. And then in beyond that, you'd say, okay, for progressive employers, do you want to try to implement pharmacogenetics in your formulary? Do you want to try to, you know, get involved with special programs around genetic testing? And I'll tell you, there'll be a whole bunch of employers that would also want to be involved with that. So the question is the venue. Now, one venue is that we're the largest provider of health and wellness services in the United States and we are trying to convene employers around this topic. And we're thinking about doing it regionally, we're thinking about partnering with medical, academic medical centers who want to do that. The second is there are employer consortia popping up. The HTA is 40 companies, about 7 million lives. The Berkshire Hathaway, Amazon, that group will come together slowly. They're moving like a snail's pace. But there are other of these consortia coming together of employers that are saying, okay, we want to band together. So you could either use one of those groups and invite them as a group to come in to a setting to talk about genetics. Or we could work with this group to create an environment to bring in the employers. And I'll tell you, they would come. They would absolutely flock to that sort of environment. Quick follow up. And I know one of the issues with convening payers is that they are very sensitive to the idea of antitrust and collusion. And that if they all agree on something, then there's that risk. Does that apply to self-funded employers as well under that type of aggregation? That's probably not the right person. Yeah, no, I mean, I probably am the right person because I think that on these HTA collaboratives, I don't think there's been any problem there of saying that employers will band together to have purchasing power, which is what that's all about. And then what I'm thinking about here is, I think it's much more of a opt-in type program where now all the employers have been educated on here's an opportunity to implement genetic testing. And we're not all agreeing to do the same thing, but we have an opportunity to raise our hand and participate. So I really don't see that that would stop anything of this nature. Should we just make one other comment? And that is that I really have observed and feel very strongly that patient-initiated movement in this area is powerful and growing. And there's a much closer relationship between employers and their employees who are the patients than there is between covered lives and insurers. And so I think what we're going to see is a confluence of movement among consumers and among employers to try to get this right. This is exactly correct. And it's actually a different than the payer view on genetics, which is much more defensive. The employers, our employees come to us and they are learning about genetics and 23andMe and Ancestry and it's all out there. And now they're saying, what about this? What about that? And usually the health plans are run by an HR leader, nothing against that. But most of the employers are saying, I'm not necessarily saying I want to fend off genetics. I just want to know what I should do. And I'm happy to pay for a cancer genomic profile in my X percent of people who have cancer. And it's part of it is, I don't even care about the health economics because I have X number of employees who have cancer. So am I willing to pay for a profile? You do the math and you say, even if it's all lost money, I'm going to do it because it's a very important dynamic within the employee population. So that's a real good pickup, but it's different than the way the payer is viewing it, which is pretty defensive still. Whereas the employers are saying, gosh, I just want a genetics program to give to my employees. Like tell me what to do, sort of thing. So I'm just going to take the prerogative of the chair. I'll get to you in a second. And just ask sort of two evidence related questions. So we are in the solutions mode. We think the groups like this, genomic medicines work group could convene purchasers. But one of the things that we've talked about is, and you've talked about the value of a basic genetic formulary and we could help in terms of defining that. But we've heard a lot about value and purchasers I sure want to optimize value. And it was already defined by, one of you, I think it was you Keith, as outcomes relative to costs. So one of the things that we talked about here yesterday was a core set of common outcome measures. Do you feel like that would also help to advance evidence generation in support of value-based purchasing decisions? Just from this, to finish out on the employer discussion, this is where these, the quests and the other aggressive employers on these issues, we have all our biometrics data. We have all our population data. We have all our claims data. And we can then do an analysis of what happened and what were the outcomes and what was the health economics over five years of this decision. So we're really well positioned to generate that type of information. I don't know about common goals. I think the common goal is, in our view, it'd be I have at least improved the healthcare of the population and I'm cost-neutral or saving money. So I can, you know, you sort of, then can I prove that through these employer groups and get data that actually shows that it's either improving care and lowering costs in long haul or, you know, in the worst case, it's improving care and cost-neutral. So, but we can do it better than probably a payer could. So as I'm gonna channel John Wilson because we all know him well. And John was here. He would say, that's the problem is there's such, let's take cancer testing as an example. He would tell you that they're dealing with two or 300 different labs of incredibly variable quality and incredibly variable cost. So you can get mom and pop shops doing a very poor product at a very high price and you could get Quest and Labcore and Invitae doing a very nice product which is high quality and standardized. And they're still trying to sort that part out. I think what you'll see them doing in my guess is they will pick labs around the country where they feel the quality and cost makes sense and they'll start using them to pay for things like cancer testing. And that may be part of this genetics formulary concept it has to do with what testing in what population and where you shouldn't pay $3,000 for something. I think it comes into the same discussion. Okay, great, thanks. All right, thank you for waiting, Carlos. I appreciate it. Thank you so much. This is an incredibly exciting panel. Jay, you and I have talked a lot about the 70% of Americans receive their care through their employer. So it's absolutely the place where this is gonna be coming from. Furthermore, I would argue that the consumer movement for health and wellness is absolutely going to be overwhelming. We live in an aging population. This is not a problem just in the United States. It's global. And there is technology coming afoot faster than we can even imagine. So we're sitting around here worrying about when should we be paying for genetic testing? That's the wrong question. The consumer will walk into the clinic with their entire data stream on their watch in less than a decade, absolutely no doubt. With the, we heard about fire today. Apple already has a far better EHR visualization tool for the consumer on the watch today. It's already here. It already exists. So it's an implementation science question, but far more about how do we build out systems that will make use of that data for decision making and keep people healthy? I mean, it's an absolute no-brainer that the way to avoid the future cost is to keep people healthy. 90% of FH patients aren't diagnosed today. That's just dumb, right? We gotta fix that, get them on the right therapy because 70% of them aren't getting reimbursed for PCSK9 inhibitors, right? The healthcare economics of PCSK9 inhibitors are that at 14 grand, it doesn't make sense for anybody, right? You gotta get the price down to four grand which is what the UK actually pays today. So in a world where the only group that can actually bend that cost curve is pharma but they need like a 20 to 30 year timeline to put people on prophylactics, it's actually a data question around who's going to own the benefit of how that data is mined. And I actually think there's a movement of foot, there's tons of folks talking about how that data, particularly from rare disease patients, right? You're taking these data from rare disease patients and then turning around and trying to sell them a $750,000 drug, right? That world is untenable. So I think we're actually asking the wrong questions. It's really far more about how do we go and prepare for the world and all that data is going to be there. The data streams are going to be there. You're going to know as an employer what every person costs, even if it's anonymized to you, right? So using that information in ways that can actually benefit everybody is a much tougher question than trying to fiddle around with, oh, should I be paying this much for CIP2D6 testing, this much for this other kind of testing, because that's the wrong question. Like that information will exist and you'll have it in less than a decade and nobody's prepared for that. I also completely agree with you. It's all about education, right? We're sitting at the top 0.1% of people. Most docs don't understand it. Most of the general public doesn't understand it. So if you don't want them to revolt against us, right? Because it's, I mean, it's serious. It's absolutely serious, right? People get crippled in this country economically because of the healthcare costs. We've got to do a far better job of engaging today with the consumers. And I think the employers are at the helm of this because they realize that insurance today is bad software for managing your federal obligation. It's not really insurance in any sense of the word. And so I predict insurance, as we know, is going to collapse in the next decade in this country. And we've got to be at the forefront of far better ways of thinking about this problem because we're all part of the dinosaurs that are going to go extinct, right? Healthcare systems are, I'm not kidding, right? I sound like a complete lunatic, but I will tell you that the smoke is there. And if we don't get ahead of this, much of what we're talking about, people are just going to say it's fake news, I don't need it. And at the same time, the data's coming. 60 million people will be genotyped in less than a decade. Those numbers are out there. So how are you guys preparing for that world? I mean, no, that's the very serious question. As the leaders of this country at the helm of this, how are you prepared? So it's very different than the genome England experience. Genome England has demonstrated the clinical utility. That's why they cut the pilot early and commissioned genomes directly into the NHS budget, right? They're starting with rare disease. Lisa gave a beautiful talk, we know 25% of Mendelian disease is already undiagnosed in this country. Again, dumb mistake, we got to fix that, right? But it's ultimately about keeping millennials and our kids healthy. And what's our strategy there? Because that's where the real savings have to come or we're headed to 30, 40% of GDP devoted to healthcare, full stop. So I don't know who this guy is, but hire him. That was brilliant. And I think that one of the challenges is that folks in my position have not heard this before. So I will tell you, I am vaguely involved with our genomics institute. I just happen to be around running a children's hospital. But it is obvious to me that you're absolutely correct. The insurance, think about what insurance companies do now. They print the little cards. That's about it. Everybody else is taking risk for them. And so that's what they're providing now. When you look at the long-term consequences of the insurance company getting out of this market and they're gonna go kicking and screaming. Let me just tell you, this is not gonna be an easy fight. But when you think about what they've done is mainly get in the way of progress. And so as more systems get involved with speaking to employers and basically getting the insurance companies out of the way, this is going to be a huge change. The challenge we have as a children's hospital is 55% of our patients are Medi-Cal. Let me tell you about that insurance plan. It doesn't get a whole lot worse than that. And so my big challenge, and you guys are absolutely accurate when you talk about the patients driving these changes, is that's what's changing in the government here in California, is patients are actually having conversations with their representatives. That is going to be a big change for us here taking care of kids. But I agree with you. I think that I wouldn't be making any long bets in the insurance business because I think that it's gonna be forward thinking health plans, health systems, and employers working together. And that is a challenge that we're taking on. A lot of smaller systems like us are developing health plans. And I mean, you've heard from Mayo and you've heard from Geisinger. These are massive systems. We're a pretty small player, but small players are going to be able to move forward with that. I think you're absolutely accurate and I can't wait to hear more of what you think. So I'm just gonna ask Keith as the director of an individualized medicine department, how would you view the role of genomic medicine in advancing the vision that we heard from Carlos, and do you agree with that and the rapidity of it in the next five to 10 years? I'm a 100% believer we should all have a genome sequence that should be in our electronic health record and it should be requiried every time we want to query. It's absolutely, no doubt in my mind, that's the future. I think I'm gonna contrast genomics England with the US health system. The problem that you're highlighting is we have incredibly fragmented system in the United States. The reason genomics England is successful is when you go to meet with them, the chief medical officer of England is sitting there, the chairman of the National Health Services sitting there, three or four genomics experts are sitting in the room with vision and they can get things done because everybody's there. That is impossible in the United States today. So we have the most advanced technology on the planet. We're leading the way in many things but we can't do what they can do because of that fragmentation. They have a business model, right? They're planning to use that data and resell it into pharma. I mean, the beautiful irony of this is that genomic England- We don't use the word resell, we're gonna grant access to. Fine, genomic England could turn a profit for the British people as a wholly owned subsidiary of the Ministry of Health, right? So, you know. Just to be clear, there's a lot of sensitivity around that. That's a very important point that selling this to pharma is not the business model. It's having data that's successful to people. And they'll negotiate great pricing because it'll be discovered there. I mean, it's a no-brainer. I just wanted to make the observation that the fraction of our testing that's self-pay is just going up and up and up and that most patients and the providers feel that they've had handcuffs taken off because they don't have to worry about pre-offs and letters of medical necessity or any of that stuff. And that's only because the price has been driven down to a point where it's within many people's grasp. It still bothers me a lot that there's still a socioeconomic disparity that exists even with the price as low as it is. But for many patients now, their self-pay is far less than they would have to pay as part of their insurance company co-pay. Just to respond to one part, Carlos, of what you were talking about, this consumerism. It's absolutely happening. It's obviously going to happen that people are going to have all their information and be navigating on their own. And one of the things with employer plans last, maybe go back 10 years, maybe go back even three years, it was a no-no to the employer getting between the patient and their doctor, which is a myth, by the way. Most of my employees don't have a doctor. The women go for well checks once a year. The men are invincible and they have a doctor when they have a heart attack, you know? And so the reality of the situation is that what we're now looking at, which I think is partially responsive to the question, it's that we need to be okay as employers in helping people connect where they need to go. So if they walk in with a genome, we could have genetic counseling services, and hey, you don't have a doctor, but you may need a bracket test and we're gonna direct you where you need to go into care and it's really just directing the traffic in a way, but cutting through, you definitely don't then send the employee back with the information and say, now, go figure it out and make an appointment and navigate, because then they do nothing, nothing happens. But when they walk in with data in the future, our stance is if you have a doctor, we're going to help you understand what you need to do and get you back to that person. If you don't, we're gonna connect you to exactly where you need to go in the system. And it's not like from my standpoint, it's very straightforward. It's not like there's a lot of judgment in that on our side, the employer side, we just say we know what our network is, we know what you need, and we're gonna get you to that place. So I think the employer side is changing. I mean, we're very aggressive around it, but employers are more and more saying, oh, again, united and you, etna, they're not actually managing the health system. It's actually my plan and now I have to make sure the people get to where they need to go. And so that's the change that's coming together. Employers are getting more, not paternalistic, but involved in steering people. And then the consumers are now ready to be steered as opposed to go to a doctor's appointment. So if I may follow up with respect to consumers, I don't know if you'll chase me from the room for this or cheer me, but I wasn't sure whether to bring up or not, but we have decided at Mayo Clinic to enter the consumer space. And sometime this month, we'll launch Mayo Clinic gene guide. And the goal of Mayo gene guide is to become a trusted partner for consumers who want to learn about their genome and to direct them to appropriate healthcare if it becomes appropriate. So I think we see that there's, this is a toe in the water if you want of how you reach beyond your own walls and try and engage a larger audience. Great. All right, now, Bob, I know you've been patiently waiting with your card up. And then after you, Robert, I had seen your card up. Did you still wanna, you're okay? All right, then I think we'll turn to Rex and Terry. And I'm sorry, there's somebody, is it Mary with her? Okay, Dan. Okay, Dan. Okay, great. That'll be the order. Bob, you first. So I wanted to comment on the, what level of evidence is sufficient? What's that, is that better? Okay. I wanted to comment on what level of evidence is sufficient and come at it from the perspective of me learning about pharmacogenomics and interacting with some of the providers and the service leaders in my institution around that. And almost uniformly, they're very skeptical. And when I try to dig into what's behind that and try to understand that, it comes down to really two things. One is that they've been burned before with therapies that were marketed to them or sent to them and said, yes, you need to do this and then it didn't work. Okay, the other side is that they are taking it from the standpoint of there is no evidence right now that this works. And so you need to prove it to me in the clinical context. And that's really not what the situation is. I mean, since drugs were first started to be used, we've known that people differed in how they responded to drugs and the science has been done to show that there's, that is largely or significantly related to your genes and now we know what those are. So we have a lot of validity valid studies moving up to this and the only thing that hasn't been done is the clinical utility studies. The problem with that is that you have to do it for every single drug gene pair and you have to do it for PCI, you have to do it for other anticoagulation applications, you have to do it in psychiatry, you have to do it in surgery in all these different fields. That's a lot of clinical utility studies, right? And that's not gonna happen. So the question is really, when you have the biologic evidence and some clinical evidence that it is beneficial, can you rely on that biological evidence to say, yes, we're gonna go forward with this? We're gonna accept that. Let me, I wanna just start cause I'm all over this one on this, like these, when I look at pharmacogenetics, I would boil it down to, I don't wanna talk about the 300 drugs. I'll tell you a baccavir, if you're gonna die. And hydrocodone, you're gonna die, okay? So it's not a debate around these couple and then if you then move the model from onesie twosiege pharmacogenetics to, okay, there is valid science here and you do wanna know before TPMT, azathioprine, you absolutely wanna know. So there's science that proves that and now I've moved to a model of one time, one test that's cheap. And now if you're gonna be prescribed hydrocodone in the future or a baccavir, it's totally obvious that you should not do that or the person shows up in the ER with no blood cells. And then you can debate should I use the information for Plavix, should I use the information for Coumadin? But it changes the game. It says no, there's not really a debate. Yes, there are certain drug gene combos that are absolutely you would want to know and no physician would debate you on it. And so if you could have that information, would you use it before giving a baccavir? Yes. And so then I go back to the model of trying to convince doctors to order it every time they see a situation, it's never gonna happen. But if you convince a system or an employer that it should be inherent in the population, every member has a profile, and every time they come forth, they get to use the information and tap into the highly validated gene drug pairs, I think it's beautiful. And I could convince anyone of that. If it comes with the person and you paid once. So I think it's more a model of do you do it onesie twosie by indication or do you convince people that it should be done one time in a population, in a person, and then you have a mechanism to follow it over time? I just make one other comment, which you'd be surprised how many people think that if you're gonna do a pharmacogenetic panel that TPMT costs you this much, and then if you add HLA it costs you this much and it's this much, as if we're still doing one gene at a time. The entire concept that you have a whole panel that can be done for a 10th of what it used to cost to do a genetic test a few years ago, even that simple fact has not filtered in. You know again we're all preaching to each other here but what we've found, I started off by saying we have that skepticism. The old surgery adage of watch one do one, teach one applies with pharmacogenomics. What we've found is you show somebody the report, you say tell you what, we're gonna let you order that in 50 of your patients at $250 a person for 450 drugs. And they do once or twice, then all of a sudden they're adopters and they're doing it. And that's why we see this, now we're getting 200, 250 orders a month. That's a small number at Mayo Clinic. We're seeing a million patients a year but it's growing fast. So I think that there are ways past that skepticism and Mark started off by saying it's just the physicians don't wanna deal with it. I think it's a bit more complicated than that. I think they need to experience it I think to become believers. And implementation science and the frameworks have the appropriate tools to actually understand all those barriers which we heard yesterday. So Terry, were you next or Rex? Okay. So this has been really eye-opening for me. I think we've been focused on the wrong thing. I think thinking about the payers and you guys are making a really compelling case that we need to be focused on the consumers and on the companies that are providing the healthcare for people. So I have a two-part question. One, obviously you guys are thought leaders and are out ahead of the game hopefully not out over your skis and obviously you're completely persuaded you're not. What does it take to get us to the tipping point where everybody thinks about it like you? And maybe to build on the UK Biobank or the UK system, Genomics England situation there is another fundamental difference and that is right now because your employees, the people that get healthcare from you may not get healthcare from you or may not be your employees in five years but if you're in the UK you're still gonna be under that health system. So how do you feel about the fact that you and I guess this is really for Jay that you might be paying now and somebody else might be reaping the benefit? Well, I think we bake that into our models. On average, we keep employees for 10 years. That's an average, right? And I know the attrition rates and so we actually do the modeling and say when we implement a renal program we're going to prevent X number of end stage renal disease and here's the health economics of it and that takes into account that we're going to lose employees and they're gonna move on but we're much more incented than a commercial plan because commercial plans re-enroll their populations every year or every two years. So I would say we're pretty much in the long game around it and you would think that Medicare and Medicaid would operate that way but we know it doesn't because it's too political and it's difficult to make rational decisions. So Medicare should be in the long game and incented for the life of the person but it's not that easy but I'd say 10 years is long enough for most interventions to say heck yeah, we should be doing it but we also model it out and determine if it's going to be effective in our population. I think there's a real possibility that this could actually be a differentiator when you think about what the market is for employees and how much it costs us to lose employees and then go out and hire new employees. Would it be fabulous if this is part of that differentiation that people say you know Quest has really gone out on a limb and is investing in our healthcare and is a big part of moving this whole genomic science forward. It's a company that I want to stay involved with. So I think there's a real opportunity there to be a differentiator in the future. Yeah and I missed that point absolutely particularly on our frontline employees, our phlebotomists, specimen processors, logistics the turnover rates there are unbelievable in general and we're seeing that this is part of the retention and part of the benefit that hey, my company's investing in me and I want to stay. So I think that's one of our end points that we're trying to get to. Could I add one thing which is that Kaiser began exactly under this model, right? Kaiser began providing healthcare on the job to the steel workers and the factory makers in the 40s. The labor unions then argued for it as part of benefit and then it became a member service. So I mean I agree with you at the same time healthcare already is a golden handcuffs, right? People don't switch jobs because of health insurance. People don't move out of state because of health insurance. You actually need subscription health systems that people can sign up for and are affordable with some kind of reinsurance for catastrophic events that then is pulled or something along those lines but you could build out. I mean Kaiser's a one stop shop. Kaiser owns the pharmacies, they own the docs, they own the hospitals, right? So there are models out there that could be scaled out if the right incentives are created and they have to be the right economic incentives. They can't be sort of top down government decisions as we've seen catastrophically failed the last time this was tried but there are economic ways to make it happen if the right incentives are put in place and the long-term risk is held collectively and reinsured. So the other part of my question was how do we get to a tipping point? How do we get to the tipping point? Yes. You know, I was gonna comment, there's hardly a week goes by where I don't got a call saying from a smaller hospital somewhere in the country or even abroad saying we'd love to have a precision medicine program, the capital infrastructure that's required is outside of our scope. Can Mayo Clinic or whoever else come in and help us do that? And we just haven't been able to do it because of bandwidth. We haven't, we're busy enough doing our own business that we just don't have them and there's across state line licensing issues that come up very quickly. So there is a huge appetite out there for dissemination even to smaller networks, smaller hospitals outside of Gasinger and North Shore and Raddy and Mayo and other Cleveland Clinic, other places. There's a, there used to be, we would say, the tipping point would be when we convince Etna and Optum UHC to have a national coverage policy and get U.S. preventative services, task force A and B and, oh, you know, life's too short. And so we keep trying to do all those things but that can't be the tipping point because it's, you can, you're never gonna win. Sorry to be negative. And John's a great guy. Okay, so it's not his problem, but you know, if you try to convince Etna to have a global coverage policy change on population genetics and oh my lord, right? So I think the tipping point would be in my world, what I'm talking about would be peer review articles that show that an employer implemented a population health program and here's what the outcomes were. And our employees did better and we saved money for the plan. So we're publishing those sorts of articles but I don't want it to all come out of quest diagnostics so this idea of getting other employers to open up and run programs and then publish results and we have to do it carefully so that our employees know we don't know their identities and it's all compliant but we started publishing. We did one last month on a diabetes, a screening to OMADA prevention program and it was a very minor result but it got a huge amount of attention and I think if we can get other employers to do that and we publish these results, the employer community is gonna flock to it and that would be a tipping point as far as more aggressive programs within employers. If I can just add on one thing to that which I know is something you're also working on Jay, it's the economic studies from the perspective of the employer. Most of the academic economic work always takes a societal perspective which particularly in the United States is insane since we don't have a societal perspective for healthcare but that employer perspective that really demonstrates the value proposition on the economic side as well as the health side which are generally coincident I think would also be very compelling and so additional investment in that type of work is also needed. Maybe just to share then one of the projects that's going on is the personalized medicine coalition has a project around doing clinical and economic outcomes of whole exome sequencing and when we started it, it was around the more traditional model of societal view and a health system view which is fine, a geisinger view which is fine but then we realized along the way if we insert in this also an analysis that is from an employer health plan and we have all the claims and we see everything, it's very powerful data set actually and it is a totally different view where you can say okay, go to the other 90 million people that are in the plans, this would be the outcome for the actual payer for those 90 million people. So I totally agree, I just think not a lot has been done there yet from the employer view. Okay, so Terry was next. Do you have something that follows directly? One second Terry. Yeah, this actually does follow directly. It seems like we've gotten a little bit away from the harmonization of evidence requirement discussion and I guess what worries me is that I believe in this vision and I think most of us in this room believe in this vision being scientifically applied but there's also a parallel growth among industry of all sorts of applications of genomics and other omics that may not be necessarily applied with the same scientific rigor but that are being pitched with the same articulation with the same vocabulary with the same value proposition. So unless we can figure out and I think it's gonna have to be more than just each employer or self-insured entity creating its own set of reports. I think we're gonna have to have some kind of standard that we agree, maybe it's less than the US Preventive Health Services standard but there's some standard that we agree we can call a threshold so that we can take not just sequencing but the repeated waves of omics and multi-omics information that are in the offering and we can start to say which things trip over which thresholds at which point because as Carlos and I agree we're gonna have data streams from all directions and people arguing that these data streams are critical to our health. So how are we gonna make the distinction between what we believe really works and what we believe and what is really a process of technology and marketing and data gone wild. Great point. Does someone on the panel wanna respond to that? Yeah, I'd like to make an important distinction between clinical validity and clinical utility. I think a lot of the stuff that you're talking about Robert is stuff that actually doesn't meet even a clinical validity level. And I've had some conversations with Julie Egginton for example who started a nonprofit system that she's calling Elevate which is going to be an attempt to go beyond CLIA and CAP to really dig into what laboratories are providing and what their systems are for providing to be able to give people who are making decisions like the one that you described some information that they can use to decide whether this proposed laboratory test does not meet essential standards for clinical validity. For example, I can't tell you how many payers were in the process of getting into contract with Theranos. There were many of them. So that's a good example. Now clinical utility is another story. And there I think once you've established that at least the laboratory can do what it says it can do and that the information means what it means I really feel very strongly that that kind of utility evidence has to grow out of implementation research that's occurring in medical systems including employer based systems that are laboratories for testing this. I think that's where the evidence is gonna come. And getting that funded I think requires input from philanthropy, from systems who really want to demonstrate that they are at the forefront, that they're innovative, that they're thinking about the health of their employees, that they wanna retain employees. There's many incentives. But I don't have a perfect answer to how that can be done but it's quite obvious that there are a number of people sitting in this room now that are working in institutions that have come to that decision and are pushing forward on it. I like the clinical validity and utility framework. I think at a minimum it could be useful to say that there's an understanding of what is and is not clinically validated. As far as new molecular diagnostics and solutions. But then what I said at the start is I don't experiment with my employees but if I have a, we have an AFib genetic risk score we were working on. And I know in an individual case, in an individual person if the data is there and it's used appropriately it can be very useful. So I'd say it's clinically validated but it doesn't have clinical utility. So therefore I'm willing to do the experiment of I would offer it to my employees and the clinical utility is do we actually get uptake and do people actually make decisions based on it? But it's above the clinical validity threshold where I'm willing now to do that. And then we're climbing the mountain once you've proven clinical utility now you're going up toward guidelines and etna and UHC. So that's kinda my working framework for how we think about it. I would never offer our employees something that is not clinically validated but I certainly would offer them something that is not, what's the term? Clinical utility does not have proven clinical utility cause that's all about implementation science. And it's all about can I actually get this in the population, get people to use it and make decisions on it? And the only way to do that is put it in a system and see if you can get it implemented. Do you worry about any potential harms of putting people in that kind of stream without clinical utility being proven and with some potential for downstream harms based on non-utile... Yes, we definitely worry and think about that and look at the opportunity for errors being made and bad decision making. But one of the frameworks and I think we've talked about a little bit here is we wanna make sure that whatever we provide to our employee population we're not just giving it to them and sending them out into the community. We're saying when we give this to you you will have an opportunity to discuss it with a physician to get interpretation and referral into care or a genetic counselor. And so my safeguard is whenever we do these programs there's always an opportunity for an interaction and steering with a healthcare professional and we never do any of these programs where you just throw it over to them, hey, here's a new genetic test, go do it and go see what your doctor thinks cause that's where all the chaos ensues, right? So I think having a telemedicine piece, having a telegenetic counseling piece is actually kind of a safeguard for us. Okay, great, thanks, Terry. Excellent. So this harkens back about an hour or so but when the point was made about could we develop a prioritized list? I mean, I do agree with others that within 10 years everybody's gonna have their sequence available. Unfortunately, I myself have been saying that for the past seven years or so. So it's getting a little long in the tooth. And I think we need to recognize it's not going to happen in most medical care systems that all of a sudden everybody's gonna make that leap. And so this idea of having sort of a formula or essentially a prioritized list is really neat. We do have a genomic medicine working group there sitting at this end of the table. NHGRI doesn't have a stake in this. We just wanna see the best kind of research be done and particularly if we can work with employers who will gather that data that will actually prove to us does this work, does it make a difference or not. So I think that's something that we could really work with employers on and bring you all together. Probably we'd need to start with those who are sort of medically sophisticated and not, although we might wanna bring in you know, act me tool and die and just see what their perspective is. But I would offer that as something that I think we could do in a good way in a partnership with you guys. Thank you and I think you would have the same interest in this. And so just to put icing on that one, I would say employers have been, because I run our health and wellness business, they've been coming to us saying, hey, quest diagnostics, you're doing our employer health and wellness. Please tell us what the hell to do in genetics. Now we also sell genetic testing so I'm conflicted. And I tell them that they still want our advice, okay? But then having just a third party group like yours that says basically for a basic population, here are what we think should absolutely be done in all cases, which is tier one our U.S. Preventive Services Task Force. And here's where we think the role of other genetic and genomic solutions are. And I think that we could bring in some very large employers and I agree it would be nice to have sophisticated healthcare oriented, but then also to bring in like a Home Depot or a giant company who's been asking us for this. And then have some level of alignment that says here's a basic building block for you to take into your healthcare, into your self-insured program. They're all asking for it and it's not coming from anywhere. There's no place to go. So I don't, and quest is not a solution because we're conflicted. So I, you know, we'll follow up but I think there might be something that could really be done there. Okay, great. I think the next questioner was Dan. So I want to take the conversation back an hour and a half or a decade and a half. I hate to speak after Terry and I especially hate to speak after Carlos because I want to bring it down to sort of a very, very sort of fundamental basic pharmacogenetics question. So everybody talks about pharmacogenetics as the low-hanging fruit. There are these common variants that drive variability and drug action. And the problem it seems to me with the Coumadin trials, the potential with the very large Mayo-Taylor PCI trial is that they take heterozygates and homozygates and mix them all together. So one way of looking at pharmacogenetics is that there is variability in drug action, some of which is genetically driven, some of which is contributed to by people who are heterozygous for some odd loss of function variant. But the real action, the real people who are at risk, Jay, for dropping dead with hydrocodone or with a baccadir are not the people with subtle variation. Those are the people who are a small minority who are homozygous for CYP2C9-3 or homozygous for CYP2C19-2. And those are one to 2% of a population, still a huge number. But when you start to mix in all those other people, then the cardiology community displays skepticism because in fact, for many of those people, being a heterozygate for CYP2C19-2 won't make a difference to all the other things that go on in their lives that determine action. So I'm with Mark that there's very strong biology that if you can't biotransforms clopidogrel from its prodrug status into a real drug, you're not gonna get drug action and that happens in the homozygous. In the heterozygous, all bets are off. So I think that we need to take one step back and think about pharmacogenetics, not as the low hanging fruit because they're all these common variants, but as an interesting example of a spectrum of responses, some of which are the ones that are gonna really drive terrible drug reactions, but that's in a small minority of patients. And again, the obvious answer is to do this panel testing, which we and many other people have done. So I think it's an important point to think about as we sort of think about how to implement pharmacogenomics. There's this idea that it's all black and white and I'm fond of saying it's not black and white. It's all about 50 shades of gray, but I didn't say that. Do you wanna start? Well, I guess I'll agree 100%. I mean, I think that's why, and I shouldn't be the one speaking that Gwenshawbong should be, but that's why we're trying different approaches because I'm not sure the phase three trial is appropriate in this N of one world. So this is why we've got the 10,000 genomes. It's why we're just telling our, it's amazing to me where the uptake has been. It's not where I expected it to be. It's not in cardiology. It's in anesthesiology. They love it. It's in primary care that for them, it's polypharmacy and understanding. Our pharmacists are incredibly enamored with it. They see it as a career path to be the pharmacogenomics expert. So it's in pockets of meds and I would never have predicted where this is actually approving to be most successful. No, I think if there's an implementation lesson, an early implementation lesson, it's that the people around this end of the table and around that end of the table aren't the ones who should decide what community should get pharmacogenetics. It's the users and the users are the patients and the specific disciplines. And it's interesting to me that at Vanderbilt, it's the anesthesiologists. It's the psychiatrists who are, and the kidney transplant people who are really sort of on board with this and the cardiologists could care less. Another example is, I should have said that in public. Another example you would never have thought of is a pediatric gastroenterology for kids with chronic heartburn. Who would have thought that that was where they found a huge, just looking at meprosal, they found huge advantages of doing routine testing. They've now adopted it as part of their standard of practice for workup. One summary point though is I'd say it's very important to educate and get various physician groups on board, but I would say I don't think the model is to convince every single practitioner of when they wanna order a pharmacogenetic test. I'd say it's more assume that the employer has sponsored pharmacogenetic testing or the person got it on the street and they're walking in and so you're a psychiatrist, so you're gonna be taught how to use that information to use an SSRI. Or you're gonna be taught about when to not use a back of your in the HIV side. So there are certain segments that are definitely getting more, like physician segments that are more into it, but I like to think about like let's not try to convince each physician group to order it every time because it's kind of silliness. It's just assume it's gonna be there to Carlos's point in the future. So how are you gonna use the information? And then employers ultimately will pay to have pharmacogenetics done and so the physicians need to understand how to deal with it, not how to order it necessarily, but. There's an implementation challenge here because what we also hear is we thought executive health was a great place to start this because these are people with discretionary income and they're usually curious and they've flown across the country to come to get their work up. And yet when we spoke to their physicians, they said, are you kidding me? We have half an hour to talk to them about their diet, their exercise stress test, their cholesterol level, the medication they're taking, their cancer prevention and now you want me to talk about 450 drugs and 23 genes? I mean, get out of here. So there's a real implementation challenge there as well. That's where we found pharmacists to be really particularly helpful. Okay, I'm just noticing the time and we have just seconds left. Should we stop on time? Rex and Terry? I think we could go a few minutes longer. I mean, lunch will, we're not such a large group that it'll take us that long to get through the lunchline. So if you wanted to go another five minutes or so, is that okay with you guys? Okay, so we've talked a lot about purchasers and private payers and I wanted to specifically follow up Donald on the comment that you had made, that you got funding from MediCal. We haven't really talked much about public payers to sequence infants. I think the figure was something like $2 million. Could you talk about what was the evidence that persuaded them to make that decision? So welcome to my world, it wasn't evidence. It was politics. And I think that we've touched on a lot of the black and white of evidence and convincing our fellow physicians and practitioners, but that was completely a political move. And it was, we pitched a lot of the story about the long game and how we were gonna make tremendous changes in people's lives. But in the end, it was politics. And that's something that may sound a little offensive, but it's the real world that we live in. And we understood from the beginning that this was an opportunity. And I'll tell you that radio's actually going to be sequencing babies in three or four different communities throughout California and they just happened to be in the communities that the politicians represented. So part of this is understanding that to get to that end point, it's gonna require us being all in and doing whatever needs to be done. Great. Also I think another part of it is that California is really working hard to differentiate itself from what's going on in the East Coast, particularly what's going on in Washington. That is true. Very hard. That is true. And so if you look at the stem cell initiative, it's also a part of it. So I think this is a piece with that. Not quite so, the politics aren't quite so crass. Almost assuredly, they are even crasser perhaps, but I think you're right. The California is using this as an opportunity to look very different. I think we're the sixth largest economy now in the world and I think it's using it an opportunity to beat its own drum. Gino Minglin, by the way, was also a political move by the Cameron government. They made a decision one day to spend 100 million pounds without telling anybody. Realized they needed 200 million pounds, but then they were able to do something really good with it. So all kinds of political decisions get made that get turned into great science. There's a sunny side to that too. No, I think that's a great point and there's nothing nefarious about this. This is just the way it ended up happening and it wasn't a tremendous discussion around in a group like this. This was not the way it ended up happening. It was, would it be fabulous if we could offer this? This is pretty unique to these children with MediCal and we now actually have MediCal covering proton therapy for kids with brain tumors. Same sort of story in that proton is not available for anybody else, adults aren't able to get it and you can argue about the utility of it or not, but in kids with brain tumors, it seems to make a difference. Okay, oh Jeff, you wanna make a comment? I would like to just ask a question, seeing all the wisdom at that end of the room. And I'd like to know where you think NHGRI should be investing in this area. Since we have limited resources, we're putting our bets right now in a number of networks that you're familiar with and actually some of your institutions are participating in. But given what we heard today about sort of the somewhat political and whimsical nature of this field and then aware you're placing your bets with employers and with your own system, what's the role of the National Institutes of Health to move this field forward from your perspective? Well, I'm obviously a little biased, so I would say in pediatrics, if you can take care of kids, who cares about the adults? They'll be adults, sorry. But they'll be healthy adults, won't they be? I think a couple things, when you read the New England Journal of Medicine, all the original articles are still almost always about new technologies and new monoclonal antibodies and expensive new therapies. And then all the editorials are around population health and how are we gonna implement and how are we gonna lower cost and improve quality in the health system? So there's still a gap in implementation research efforts that really demonstrate that, hey, we've known about LDL forever and FH forever, but yet there's 90% of the people aren't identified and how do we do real good implementation science and fund it? So that's just a little plug for that field being undervalued, I would say. But the two things that we got very tangible on today is I actually think that we could do something around, hey, just a basic genetics formulary for employers to understand what you should and shouldn't do. And it's just saying, this is validated and advisable and this is guideline supported and this stuff is experimental but interesting. I think there could be a lot of value to that. And then secondly, if we did have employers who got charged up around it and then said, hey, we're willing to do a consortium where we're going to implement FH testing or we're gonna implement some other genetic testing and then share our data and do the claims analysis but it needs to be brokered by a group like you all because again, who else is gonna do this? So the employers that are progressive wanna do it but they have no place to get it done, to share the data. So I can convince the employers to get in the program and share the data but maybe there would be some kind of an effort around, hey, let's actually do an implementation program and have the employers put the data together and then have the academic community analyze and publish. So the data would be forthcoming but there needs to be some organization around how do you pull that out of the employers and actually publish results from implementation. Anyone else wanna? I guess I would comment and I think the sorts of networks that you've been supporting up until now have been terrific. I have trouble imagining what the current world would look like without the input of CSER and some of the other networks. I also feel the same way about CPIC. I mean, if you're talking about a formulary for genetics, CPIC is in essence a pharmacogenetic formulary. I mean, people go to it to find out what's the real skinny on this gene and that variant and that drug. So I think expanding the CPIC idea beyond just pharmacogenetics would be a good thing. And then finally, I think that if you required that applicants for some of your larger networks have the backing of and real skin in the game from the medical institutions where these researchers are working, that their academic institutions, their medical centers, their hospitals are going to commit at least sweat equity, if not funding to really make this a real life implementation problem to get. I mean, I'm not trying to be pejorative here, but to get things out of an ivory tower kind of appearance and more in the real world by requiring that sort of relationship. I think some of your networks have that and have it clearly and I know that, but it's sort of expanding on that and having more skin in the game from the institutions. And also, I would love to see more networks of this kind where companies that do high quality software engineering for their livelihood, that's what they do. Have them involved also in these collaborations. If for no other reason to try to get more of the great work that's being done, moved into a sector where there's a commercial enterprise that will support it, will maintain it, and will get it out to wider use. Something like the GATK I think has been revolutionary. I mean, there's basically no first class NGS laboratory that doesn't have at least some component of its pipeline. GATK is part of it. That I think is a very good model for I'd like to see much more of the software that's developed as a result of your network funding moved out into the community in a way that's more powerful. I would hesitate to tell you what to do because I don't know enough about all of your programs but I'm gonna take a page from Donald's book and I hear all these great tools that you've built and I'm a big fan of the All of Us initiative but it seems to me if you are kneeling around something where the evidence is strong, such as sequencing babies in the neonatal intensive care unit and you focus on disseminating that not just from one children's hospital in the country but throughout the country, that would be a big service to adoption and public understanding of what we're trying to do. Okay, great. I'm seeing them setting up our lunch outside. So in lieu of a summary, I'm going to just ask all of you to join me in thanking our panelists right now for a great discussion. Great, and thank you Pat for an excellent panel and panel members. Let me point out we're going to lunch, we're coming back at 1.30, please do come back because what we do in the past day and a half is all leading up to all right, where does NHGRI which is a major funder of this kind of research, where do we go from here? So here's your opportunity to build on the conversation we just had in the last few minutes and tell us really get based on this past day and a half and with our whole strategic planning process that's going on, we really need your input. So please stay with us, come back at 1.30 and we promise we will be finished by 3.30 if not perhaps a little bit earlier. Thank you all very much, lunch is right out there.