 Well, thank you. Well, in a comparable format, I'm going to do a reaction to what you just heard, but I was also asked to do one other thing, which is to provide a brief overview about how coverage decisions are made, specifically about sequencing. And to do so, I'm going to give a brief overview of that. But I want to disclose that I am not a payer that may disappoint some of you, but I study payers, and I also try to engage payers in research. So with that background about my perspective, I'll begin. So I thought about what Dave shared as slides ahead of time, and I've been reading about CSER. So these are some of the key payer questions that I would think about in terms of clinical sequencing. And that's how to evaluate the benefits and risks of tests that analyze different genes and variants at one time. So this is inherently a clinical utility question. And thinking about what CSER does, what are the valid study designs for assessing clinical utility? We've heard a lot about randomized controlled trials, and payers certainly have heard that that study design, while desirable, is not feasible for clinical sequencing. So what are the alternatives? Is it large cohort studies? Is modeling, as Dave suggested, is that acceptable? What are the valid study designs? All of you certainly have the expertise to help payers with that. But fundamentally, before we talk about clinical utility, and I am going to confine my remarks to clinical utility today, because that is something that payers are uniquely asked to evaluate. How can they be assured that the analytic validity and clinical validity of the test are established? Not only for an individual test, but across laboratories, because we know that many, for example, panel-based tests are offered by different labs for the same indication. Let's say breast cancer inherited risk testing. And I share with you on the right-hand side a publication recently that I'm sure many of you saw in the New England Journal of Medicine, where, at least in the opinions of these authors, they felt that some of the commercially available gene panel tests for inherited breast cancer risk did not include some clinically validated genes for breast cancer risk. So that's a concern that payers have. And they don't necessarily feel that they are qualified in all cases to know if the tests are analytically or clinically valid. And when evaluating a test for a particular indication, why are more genes better? So, for example, why a whole exome or a whole genome sequencing-based test rather than a targeted panel? So what is the value, as Dave has just defined it, of adding additional genes? And then thinking more globally, how can they support appropriate clinical integration of these tests with providers and with patients? So as I think about how would all of you define appropriate use of the test if they do cover it? So I think it's important to put it, all of the discussions that payers are thinking about these tests as they're asked to cover them into sort of what's going on in the recent past. In general, personalized medicine tests, as all of you know, have had difficulty gaining coverage. And people have attributed that to a lack of evidence of clinical utility. And there's been certainly notable exceptions, and I put some of them up there. There's certainly a lot of predictive tests to guide drug therapy decisions. But I think the field in general has felt like there've been some important gaps in evidence generation to demonstrate the clinical benefit of a non-sequencing-based genetic test. And I think there are three potential reasons for that. There either is no benefit, that there's a lack of incentives for researchers and test developers to conduct the necessary studies. And I think that is a very important point about lack of incentives. And or the definition of benefit of clinical utility is too high of a bar and needs to change. And we heard Robert sort of discuss three different definitions of clinical utility and suggest that the most restrictive or narrow definition is the definition that payers have been using and saying that it's a very high bar. And whether or not you agree, that is indeed the definition they've been using. But I just wanna draw your attention to the present now specifically, so that's sort of the backdrop, the context for coverage of decision-making for clinical sequencing. There is recently and work funded by NHGRI outside of the CSER program recently been conducted a Delfi study where I've been collaborating with some folks in the room about what do they consider the top barriers to clinical adoption for sequencing. And of course reimbursement came to the top. And right here you can see sort of the top three challenges that they think that payers have different evidentiary standards for assessing clinical utility leading to inconsistent coverage and reimbursement policies. I think you'd all probably agree. They think that payers refuse to cover clinical sequencing because specific patient management decisions that should be informed by the testing is unclear when the test is ordered. And then basically this is the failure of the reimbursement and coverage framework to keep pace with the rate of sequencing based genomic discovery. And this was a Delfi that was conducted of many different stakeholders. There were payers in the mix, but it was a very diverse group of sort of all of the usual stakeholders. And many of these people when asked about the policy solutions felt for the top priority that different payers have different evidentiary standards that it really would require a multi-stakeholder group to help come together and develop those standards. But the idea was to get rid of the variation and have a more predictable reimbursement pathway. There's also another study funded by NHGRI to actually look empirically at how are payers making coverage policy decisions about sequencing based test. And the first look that we're taking a look at is how are payers making coverage determinations for panel based tests specifically? So we're taking a look at that in the largest payers in the United States. So we will get some answers. So I said I would talk briefly about coverage. And I think it's just worth saying that coverage is just one aspect of how you get something paid for in the United States. Many of you know this, that there's actually three parts. There's the coding piece of it as well as the payment piece of it. But I'm only gonna talk about coverage today. And really it's a more complicated picture. It's is the technology part of the defined benefit package? And what is the evidence? And then that's the first thing. So it has to be part of what the employer sponsored plan or what Congress says is a part of the defined benefit package. And then if yes, what is the evidence to support the conclusion regarding medically necessary versus experimental and investigational? So that's the framework. That is how payers are evaluating clinical sequencing. So all of you know these three things of analytic and clinical validity and clinical utility and we'll go through it. But I just highlight for you here that only payers are explicitly charged with looking or have chosen to look at clinical utility. And right now the definition that they're using, I apologize for my voice, I have a cold. Does use of the test lead to improved patient outcomes compared with the alternative? And the test has to impact clinical decision making. So it's the definition that you've heard all day today. I'm sorry to say that, but that's the definition that most payers are using. So the criteria that they're using right now on the left, I wanna point out a few things. These are the Blue Cross, Blue Shield Association criteria that they use, but many other large payers sort of follow either these criteria specifically or sort of generally. I wanna point out a couple of things. One is that it's evidence-based, that it's comparative. And that basically that those are the main two things that I wanted to point out. And that the sources of evidence that they go to is that studies have to come from the peer-reviewed literature. And that we've heard today as well that criteria have to be that the evidence it is relied on that they usually use evidence from professional or specialty society guidelines, for example, that's important, as well as typically evidence that's looked at could be FDA-labeled documentation, lots of independent technology assessments, they may use technology assessments done by EGAP. Sort of much lower down on the scale of evidence that have been looked at would be looked at would be what the test developer themselves may have produced if it's not in the peer-reviewed literature and may even look at guidelines adopted by other healthcare organizations. So Dave pointed out when you ask payers, would you cover sequencing or what evidence do you need to actually approve a technology, sequencing-based technology for coverage? And it's difficult to answer. The premise is somehow that sequencing is different. And I tried to lay out here for you and I gave you a couple of references. There's a really nice paper by Julia Trossman and colleagues about sort of all of the things that all of you in the room think are really great features of clinical sequencing and I would agree about with that, that it's efficient, that it has future utility, et cetera. It doesn't necessarily fit the coverage model that I just laid out for you very well. Payers, for example, can't really assess the efficiency of testing or future utility is medically necessary. If you get a single test with multiple results and some of those results are actionable or have clinical utility, but many of the other results do not, as we've heard all day long, and those other results the patient would better, are better seen in the research context than the entire panel is viewed by the payer as investigational and not medically necessary. And we've seen many coverage policies like that. That's just for panel-based tests. And I would agree with Dave that many payers right now then see whole exome or whole genome sequencing as experimental because there's not necessarily a specific clinical indication for that. Sorry about that. And then the whole idea that there are other studies other than randomized controlled trials. So people come and tell payers, well, there's N of one studies or basket studies or in silico modeling or other things that can show the value because it's just not practicable. They're not familiar with those methods and I would say that people in this room could actually help payers if see the value if more and more of those studies would become published in the peer reviewed literature. So this is my summary of what Dave has said that the CSER consortium is doing and I applaud you for that. Designing the clinical utility states for specific clinical applications absolutely and then using a variety of approaches with the advisory board and having site specific and consortium wide comparative studies using common data collection tools, et cetera. Exactly right. So here's my specific reaction. I think it's critical to engage payers. I think that having a payer advisory board to help the NHGRI CSER group and CSER 2.0 do priority setting would be critical. They are very familiar with this. Payers are involved around the country already in developing evidence. Some of you may have heard of groups like Optum Labs. It's a collaboration between the Mayo Clinic and United Health Group. They're already helping develop evidence. We know other payer groups are already doing that. Pares have been helping to develop coverage evidentiary standards. So they're not unfamiliar with this. They would be I think more than willing to help in priority setting and even giving advice on specific studies. I think it'd be great for NHGRI to continue to develop the infrastructure methods and evidence for demonstrating clinical utility. And I just chose a few indications of why they occurred to me. Pharmacogenomics because the variants are common and certainly for generic drugs or for tests that are already on the market, there aren't incentives for the private sector to do the studies. So I think again, there's a role for NHGRI to do those types of studies. We prenatal testing would be another area where there isn't a big investment there now, but commercially, payers right now, that's the one area that payers are paying for testing right now with non-invasive prenatal testing. And then disease risk prediction for all the reasons that we know of how to deal with the incidental findings. I think individual sites could adopt varying approaches to evaluating and help payers think through what are the other analytic methods to and help inform the field regarding how to either to interpret those studies because other than randomized controlled trials, what else could they be doing? And if you truly believe that the definition of clinical utility needs to evolve, then I think you need to actually demonstrate what is, how to actually measure and interpret those other definitions of clinical utility. I think there is gonna be an ongoing need for federally funded studies because even if there were consistent payer evidence standards, there's still gonna be major evidence gaps because the test developers are not incentivized to do the studies. We know about rigorous cost effectiveness studies that are dependent on effectiveness. So I think it was important for you to understand how coverage decisions are made and recognize the ways that clinical sequencing conflicts with the concept of medical necessity. And if you design studies recognizing sort of what the end game is, what the payer evidence needs are, I think that you're gonna definitely be a long way there. And I think it's only NHGRI and groups like this that are gonna help in understanding how sequencing achieve its disruptive potential, particularly in community-based care and in diverse, hard-to-reach populations like we heard in our last presentation. But unfortunately, even if we had consistent evidence requirements, there's still gonna ultimately be variation in how payers make coverage decisions as each insurer interprets evidence in their local context. So we're never gonna have perfect certainty at the end of the day because payers make local decisions. So thanks very much.