All right. Hey, hey everyone. Thanks for being here. Small but hearty crowd. Appreciate having you here. I am Eve Burstein, not related to Bernstein, different spelling, but I cover U.S. life science tools and diagnostics. And it's my pleasure to have Mike Brophy here with us today. Mike joined Natera in 2015. You served as VP of Corporate Development and Investor Relations and SVP of Finance and Investor Relations before starting as CFO in 2017. Is that right? Yeah. I mean, I'll give you a quick story if you want. I mean, yeah, so this is. I just celebrated my 10th anniversary. That's amazing. That's a great story. It was very fun. Before working at Natera, I had been an investment banker for a long time at Morgan Stanley. And Natera was actually an IPO client of mine. So I met these guys in like 2011 when it was a 30-person startup in Redwood City. And you literally had to drive across some railroad tracks in Redwood City to get to the office. And if you took a picture of me with the guys at that very first meeting, it took a picture of us at our Board meeting that we'll have here in a couple of weeks. It's the exact same people. It's very little turnover. It's the same core technology. And it's really the same core strategy. We're just a little bit further along. So it's been a wonderful journey. It's been a lot of fun to work with this group. Just for example, our co-founders are still on the Board: Matt Rabinowitz, Jonathan Sheena. Matt is our former CEO, still our Chairman, highly involved. Steve Chapman, our CEO, was the first commercial employee that was ever hired here in 2011 and has built the entire operation. He's run both commercial and operations here for a long time. So it's been a fun run with the same kind of group of people. That's special. You don't always find that. Yeah. I like that a lot, well, all right, so you've got a lot of history with the company, and what I was hoping to do, especially since you and your company have been gracious enough to do a lot of these types of chats recently this month, is think about the next 10 years, so a little bit less this quarter, a little bit more long-term strategic stuff, and I obviously have a bunch of questions myself, but to all of you guys, reminder, we'd love to hear from you. Please, we've got the Pigeonhole app open. There's also small group. Just raise your hand. Chime in, I want to make sure that we're addressing the stuff that's of interest to you guys, so we'll start with MRD penetration, so here, you've said in the past that you think MRD is low single-digit penetration from where it can be. But can we start by dimensionalizing this number for today a little bit? So if you think about penetration within the indications where you have reimbursement today, what do you think that number actually is? First, let me just kind of give you a sense of how I think about addressable market calculations. I think investors are just as capable as we are as kind of doing this math. I'm not trying to spin you on any particular number. One approach that I think is relevant is to calculate the addressable market as a function of the incidence rate of cancer. Okay? That means it's a function of the folks who do get cancer in a given time for a given year, for example. Just writ large, there's something like 2 million cancer diagnoses every year in the United States. Then you've got to make a choice about what fraction of those cancer diagnoses are relevant to MRD and recurrence monitoring. Pick a number. If you said 1.5 million of them are relevant, maybe the very early stage ones, there are certain indications where there are some pretty good technologies kind of out there already. So 1.5 million new cancer patients per year. And now you've got to put a multiple on that cancer patient to get at a volume testing TAM because obviously this is a repeat monitoring test. And so I generally think about just multiply that by 10. And you kind of get to 15 million tests annually as a TAM. I've had people come to me with very thoughtful models that had 12.5 million or 20 million. And so again, it's kind of up to your own diligence. So that's kind of the annual volume testing kind of opportunity. And then to put dollars on it, I mean, you just need to assign pricing. We've talked about it. We can talk about it in the chat. I mean, I think that a conservative estimate for where ASPs can go is something like $2,000, and so obviously that I would get you it's my math right here. I'd get you like a $30 billion TAM. I mean, I think that's roughly right, and I think in terms of us making decisions, that's plenty big enough for us to make all the decisions we need to make around just having conviction that this is an enormous market opportunity and we're very early stages. I mean, just for reference, we're probably circa $1 billion in Signatera revenue this year, I think is what the guide implies, maybe a little less than that, so that just kind of gives you a sense of where we are in terms of the innings. In terms of the reimbursed tumor types, something like 70%-75% of our volume is coming to us in tumor types where we at least have a Medicare reimbursed beachhead. In some tumor types, that's incredibly important because cancer happens to happen to those folks when they're older. And so the Medicare population is really quite common. And some like breast cancer, the patients are younger. So we do have a good kind of starting point here for unit economics that we're able to now kind of turn the crank and make the investments to scale. ASPs at $1,175, blended COGS in that kind of mid-300s range, we're able to now make all the investments in clinical trials and commercial execution to make sure that patients kind of get access to the technology. Whereas there are a lot of players, and we were in this position a few years ago where, look, if your COGS exceed your revenues per test, there's just no way to make the investments. There's no way to kind of drive the volume. It's just too capital-intensive to do that. We've kind of crossed that chasm where we're able to sustainably grow this volume and really help a lot of patients. So when you're talking about the TAM, you called this a critical number. And I think this is the critical number. It's sort of what's the multiple? How many tests does each patient get? And I think it's probably fair to say this, although push back if you don't think so, that the evidence base for clinical utility in the surveillance setting is not quite as well developed as in the adjuvant setting at this point. And even, in fact, historically, there have been trials and things like breast cancer that didn't necessarily show incremental life years gained from more intensive surveillance. So maybe just to press on this idea of surveillance, two things. First, your DARE trial, actually in breast cancer, is looking at treat on molecular recurrence or tumor. And so that's super exciting. That's a great step in the right direction. Can you talk about when and how that trial moves from phase II, phase III, how that builds the evidence base, and some of the other trials you have in the works looking at clinical utility in the surveillance setting? Yeah. Well, breast cancer is a particularly interesting example where rewind to the late 1990s. There was a lot of effort and work done with the technology available at the time to monitor breast cancer patients that were in remission, identify relapses as early as what was possible at the time, and then treat with the therapies that were available at the time. And what those studies showed is that there was not a meaningful survival benefit associated with all that kind of extra effort. And those studies were viewed as definitive. Okay? And so as a result, 25 years later, the follow-up for breast cancer patients in remission is incredibly schlocky. I mean, it's kind of all over the board. I mean, you stop going to see your oncologist. You go back to your primary care physician. People live a long time in remission from breast cancer because there has been a lot of progress. But what that means is that maybe you get a mammogram as your follow-up. And a lot of the recurrences that you're going to have of this breast cancer are not going to be in the breast. They're going to be distant recurrences. You're going to have metastases in your liver or your lungs, for example. And so the mammogram really doesn't capture that at all. So it's been a quarter century. There's been a lot of progress. There's a consensus that we should be trying again, and we definitely are. Now you can monitor for recurrence with Signatera, which has sustainably shown kind of across tumor types that you can identify recurrences nine months to a year before you could have otherwise identified those recurrences, for example, like on a CT scan. That's a huge head start. That's an incredibly valuable golden year of time you have to intervene. To say nothing of the incredible wave of therapeutics that have come out on the market. You mentioned DARE, CDK4/6 inhibitors, for example, that are now available that you can now treat proactively with these patients. But because those data sets existed in the late 1990s, it's absolutely appropriate that we've got to show prospectively that, hey, you identify these breast cancer patients that are in remission. You identify when they are relapsing. You treat them early. Then they got to do better than the placebo arm, the people who didn't have that intervention. That's great. That's happening. Over the next couple of years, you will have not just DARE, but you'll have a wave of trials specifically in breast cancer further elucidating that point in favor of Signatera. More broadly on this topic of treatment on molecular recurrence, I think we're really at the beginning stages of that. And the data that we're showing is just incredibly exciting. For example, coming up at a conference here, likely in the near term, we are going to be presenting the full evidence set from the second phase III clinical trial in muscle-invasive bladder cancer in partnership with Roche IMvigor for atezolizumab in MIBC. This is the second phase III clinical trial we ran with them. Okay? So the first clinical trial started before Signatera had even been invented. Okay? But the goal was to give Tecentriq to MIBC patients. And the idea was that the primary endpoint was kind of an all-comers readout. But before they read out the study, they pre-specified an endpoint to ask, hey, what's the treatment response for just the Signatera-positive patients? Just the patients for whom we're seeing cell-free DNA in the plasma from their tumor. And unfortunately, the drug did not work on all-comers, but it worked incredibly well, something like a 40% treatment effect among the Signatera-positive patients. So that's an amazing result. That's a study that would have been a complete failure. And actually, if you're just able to triage the patients correctly, you're able to get this drug to the right people and have a huge effect for these people. So Roche spun around very quickly, enrolled a second phase III clinical trial. This was like the 2021 timeframe. And this has now been a while ago. And now we are getting the readout for the second phase III clinical trial. So what we showed, and we press released qualitatively at a high level in the study, is that you had tremendous results once again. Actually got an overall survival signal in addition to a disease-free survival signal. Study was only powered for DFS. But what specifically are we doing in the study? Patients have MIBC. Okay? They're getting serially monitored with Signatera. When they're Signatera negative, they just continue to get monitored. When they turn positive, then they get randomized plus and minus atezolizumab. And then we see how they do. And these are the results we're getting. So, to your point about what is the utility of continuously monitoring these patients, I think this data set is a particularly relevant prospective interventional trial that has taken six, seven years really to kind of get to this point. But now you're starting to see a drumbeat of these data sets coming on a more regular basis. And it's a very exciting time for patients and really for cancer care. So I don't want to put words in your mouth. It sounds like you are quite excited about the results here. And to your point, there are going to be more studies like this reading out. But this is so important to the addressable market, to use in the future. How do you mitigate or protect against potential readouts that are not as positive? I don't think there's no kind of effort to avoid certain readouts. I mean, I think this particular story, I think, shows you just one of the use cases for Signatera is you have a slate of drugs that clearly have some activity that everyone thinks ought to work. And then you give the drug to an entire population of cancer patients. And then it just doesn't work well enough on everyone to get to a p-v alue so that you can give the drug to everyone. Okay? And that's very frustrating for everyone involved. And so running these clinical trials with Signatera to show, hey, like a drug maybe didn't work on everyone, but could it work if you just dialed in on the Signatera-positive patients? That is worth doing. And sometimes they're not going to work. Sometimes the drug, despite expectations, it just didn't quite have enough activity. And that's completely okay. We're going to shoot our shot consistent with the mission of the company. I don't think that that actually harms us in terms of our adoption because patients and physicians understand that this is worth trying. So maybe I'll just give you one more example. In colorectal cancer, for Stage III colorectal cancer patients, there's kind of a broad consensus, understanding, expectation that a lot of patients with stage III CRC are getting this polyp development in their colon because they've got some kind of inflammatory response that's just kind of getting out of control. Okay? And so what you'd love to do is knock down that inflammatory response with an NSAID like Celebrex. Okay? But then you run. This is going to sound familiar. You run the clinical trial and you give everyone Celebrex and it doesn't work. Unfortunately, cancer is just too multifactorial. And clearly, inflammation is driving some part of this disease, but enough people have another kind of core mechanism happening here that it doesn't work on all-comers. And so since it does work on all-comers, you can't just give everyone Celebrex because you're going to create bleeds. There's all kinds of negative consequences associated with that. Very frustrating to patients and physicians because you would love to give this very, very inexpensive, very safe therapy to people who need it. Okay? Enter Signatera. Run that clinical trial, but just look at the results for the Signatera-positive patients. Enormous treatment effect among the Signatera-positive patients. So you've got to be, we've got to be willing to run those kind of clinical trials, understanding that we can't bat a thousand. I mean, they can't always work. They often work incredibly well. And when they do, that drives a huge amount of adoption for us. And it makes a huge difference for the field. So this type of use case that you're talking about, you're really talking about using Signatera as a version of tumor profiling, right? Like a way to hone in on the right patient population. And I want to bring up competition here because when you think about true tumor profiling tests, those are now, at this point, there are companion diagnostic statuses. You can become a companion diagnostic for a specific drug. But Medicare has actually broadened things to say, well, you know what? You can kind of use any tumor profiling test that you want for an indication that has companion diagnostic status. You don't have to use exactly the same one where the data was created. So I want to bring that up for you guys because competition comes up over and over and over again when we talk about Signatera. How do you think about protecting your moat? Could there be a world in the future where the tests are seen as more interchangeable? You need to show concordance or similar results to Signatera to then be allowed to be used. But you don't have to sort of replicate the same amount of clinical utility data on your own, almost the way you guys did with Signatera genome. How do you think about that world and how do you protect against it? Well, so one, I mean, our only strategy as it relates to competition is just to start with the unmet need that the patient and the physician is facing and then work backward to try and solve that. If you're solving those big problems, then the volumes, the competition, the reimbursement, all those things can kind of layered in with excellent execution, those things can kind of take care of themselves. So that's rather than worrying about this or that competitor or protect against some outcome. We're just trying to just solve these problems. And it's healthy to have a lot of different companies out there also trying to solve this problem now that we've shown that it is solvable. It's heartening to see, having walked around at investor conferences begging people, hat in hand for money so that we could actually create the category. Now it's like it's well established that there's a category. Now the controversy is like, "Oh my gosh, there's going to be other players. What are you going to do?" Welcome the idea of having a bunch of other high-quality companies continue to put up data that move the consensus toward, hey, everyone should get one of these tests. Everyone should get an MRD test. Everyone should get a recurrence monitor test. That's completely fine. Having said that, there's a key distinction between the kind of first wave liquid biopsy therapy selection panels and what Signatera does. What those panels are doing is that they're measuring you specifically for a mutation that would make you eligible for a targeted therapy. Okay? So it's kind of akin to like BRCA testing. I mean, if you're BRCA positive, then there's like a known set of follow-ups that you do and interventions that you do. And whether or not you use the first company that kind of came up with that or you use another company, multiple companies can measure whether or not you have this BRCA mutation. And you've seen this in the therapy selection space. There's a bunch of different companies that can measure whether or not you have a mutation that would make you a good candidate for a PARP inhibitor. And the underlying technology is kind of getting you to the same answer. Signatera is qualitatively different in that using our algorithm and our chemistry, you're Signatera positive means that you're very likely to relapse. Okay? And almost 100% of the time, people do relapse. If you're Signatera negative, what we've shown over and over again in our clinical trials is that you're in a very, very good place. Every MRD company is going to have to publish their own data to show their own trade-offs between sensitivity and specificity. That's a very different thing qualitatively than saying, "Do you have this mutation for the CDK6 inhibitor? Yes or no?" Totally different. That's a yes or no. This is, "What is your current status? Are you in remission still or are you now likely to recur?" Very different thing. And it's much more dependent on kind of a proprietary readout from the Signatera test. And we took a long time. It took a huge amount of data to develop and hone to get to this level. Other companies will be able to do that. But in order to really make inroads, they're going to have to do these big prospective clinical trials that we've also done. And I wish them luck. Do you think they're going to have to do as much evidence generation as you've done? It's hard to tell. I mean, I think that there will be. I think you're going to have to prospectively characterize what is your sensitivity, what are the outcomes for patients when they use your proprietary MRD test and they're treated accordingly. It's going to. The field has moved away from, "Hey, here's a series of spiked samples and we detected two pieces of tumor DNA in this huge tube of blood. Therefore, our test is going to work as well as Signatera." We're way, way past that. The vast majority of cancer patients are treated by physicians that just want to rely on outcome studies. They just want to say, "Hey, you want me to use Signatera? Okay. You've got 60 seconds. Talk to me about the clinical trial that if I deploy the exact same protocol in my clinic, my patient is going to do better. The trial proves that my patient is going to do better." Okay. Boom. The Celebrex pitch. Boom. You've got to be able to make these arguments to circulate data. Those types of outcome studies are going to be what is kind of entry-level requirements. Fair enough. I'm going to ask you another competition question because I am with you. I think you're right. We're doing the right things for patients, and there is a big pie to go around, but I still have to ask. I love it. And so I want to talk about it specifically within tumor-naive testing. So you're working on a tumor-naive test. We can discuss, you have talked about before the size of the need for those tests. But what I actually want to focus on is the share within that category, however big or small it ends up being. And I ask about this and want to understand your expectations because there are some cases where a new entrant to a category can come in and dominate. And I think you guys in NIPT is probably a really good example of that. You were not first to market, but you came in and dominated. But then going back to the tumor profiling example, you've got a case where there was a tissue-based test, a blood-based test, and the incumbent in each of those ended up launching a test in the other category that never really dominated, right? Right. So how do you think about the share or the portion of that tumor-naive market that you can capture, however big it is? Yeah. I mean, I think it's good to. It's It's interesting to have this debate. This is sort of the second or third generation of this debate. One of the other large tumor-naive players launched it about the same time that we launched Signatera. And at the time, most of the questions and the pushbacks we got about Signatera were around the logistics. There's no way you're going to be able. I mean, it had never been done before. A test that's personalized to each individual patient's tumor, you're not going to be able to execute that in the lab. Oh, you'll never. Oh, you need tissue. You'll never be able to get that from the path lab. This is just going to be too finicky. It's going to be too hard. The COGS are going to be too high. You'll never get reimbursed for it. The clinical trial data will be equivocal. It's just going to be easier to do this on a tumor-naive basis on all those different metrics. And it turned out not to be the case at all. It turned out that one can efficiently offer a plasma test that is tailored to each individual patient's tumor. And if you're able to do it, then I think the consensus now clinically is clear. You'd rather have a tumor-naive test just because of the strength of the data that we've been able to deliver across a broad swath of tumor types. So given that the tissue is broadly available for most of these patients, I think the debate around tumor-naive versus tumor-informed has now kind of moved to a different level where kind of tumor-informed really is the consensus. Having said that, not everyone has tissue available. Not everyone has a tissue sample that's viable when we get it. And so that does represent an unmet need. And so we've launched a tumor-naive test. Again, kind of rather than sweating specific competitors, it's more about like, "Hey, you have existing Signatera customers that love to use Signatera, but then an 85-year-old lung cancer patient shows up in the doctor's clinic and she's just not going to stick a needle into that guy's chest to get a piece of tissue out in order to run a Signatera test. But she'd like to offer him something." And so for that patient, a tumor-naive offering is the right one. So that's why we've launched the Latitude test. It's very hard to know, is that 2% or 5% or 10% of the overall MRD market. It kind of depends on a lot of variables on how the market evolves. But the point is that this is a similar kind of discussion on the other end of the spectrum where you've got a set of players that are kind of trying to offer Signatera, but they're trying to amplify up a lot more variants and things like that. We've also launched a genome backbone assay for that same reason. Just to have that available as part of the menu, I think, gives you kind of a holistic solution that physicians appreciate. So how do you think the competitive dynamics in that category play out? I'm going to press you on that. Oh, well, I mean, you have five years of the competitive dynamics playing out. So it's not like this is some unknown. We have no idea how the competitive dynamics - I mean, they have played out pretty clearly over the last five years. Over the next five years, who knows? There's a lot of good, a lot of extremely accomplished companies that are going to come up with new ideas and new solutions. So we got to see how it goes. I think that, again, it does come back, particularly oncology, to the quality of your prospective outcomes data. And so I think when we anchor the Signatera franchise around that outcomes data, I think we can be very competitive with our tumor -naive offering or with our genome backbone offering as an additional offering around the core Signatera product. Others, I think, will likely need to replicate that to gain more share. Fair enough. Maybe as you think about the role of that category over time, or actually, it's probably even a broader question, as MRD becomes a bigger and bigger line item in the budget, I think, as you mentioned, there's a lot of room for ASP to increase as reimbursement becomes more consistent. But then eventually, you probably start seeing pressure on the other end as it becomes more commonplace. And that's probably a good situation to be in because it means that patients are getting good care and getting the test. But could you see that world pushing more people to a tumor-naive test, which is cheaper the first time, although not necessarily the subsequent times? Yeah. How do you think about what that pressure could be over time? I don't think the cost of goods sold is going to be a major differentiator. I mean, I think the blended COGS for Signatera are competitive with any tumor-naive test and certainly quite superior to some of the tumor-informed offerings that are out there. One does have to look at it on a blended basis because one is going to offer. You've got a certain number of units you deliver, then you've got a total amount of money that you spend on cost of goods sold in the lab and things like that. You just divide one by the other, right? That is the right way to do the math. I think COGS are extremely competitive with Signatera. As it relates to the evolution of pricing, it's always an interesting topic in diagnostics. How does pricing evolve in this space? What one typically sees is in return for broad coverage of a test, payers will demand more kind of in-network discounts than they did initially. And I think that's kind of another thing that's quite healthy for the system and totally normal. It's not different from any other industry that you'd look at. One case study that I think is we don't have anywhere near the levels of kind of barriers to entry around data or technology in NIPT as we do with Signatera. But NIPT is a kind of interesting case study. I mean, we went public, as you mentioned, 10 years ago. At the time, the ASP for Signatera, the average selling price for, sorry, for Panorama, the NIPT test was about $350. So we realized about $350 in payments per NIPT test back then. And now it's a little higher than that. Okay. Why is that? Pricing has come down. I mean, the contracted rates we have had with payers have come down in sympathy with the dynamic I just described. I think that's been healthier for patients, but just the fraction of time that we actually get reimbursed for the test has gone up, right? As NIPTs have become kind of boring and everyone gets one, we're able to kind of get paid 80% of the time, so I think there's a win-win there, right, for patients, payers, and physicians where the realized pricing can improve over time, and still, you have that kind of price reduction going on at the contracted rate level. Cool. Reminder, we've got about 10 minutes left. So if anyone has any questions that they want me to touch on, please raise your hand or send them in. I would like to turn to early cancer detection for a minute. So here you've said near-term focus is advanced adenoma. You've got a readout coming, probably Q4. Given recent results from both Shield V2 and Exact's internally developed tests, it seems like blood-based advanced adenoma detection is pretty hard. So do you see a credible path to improving AA performance over what we've seen from other people out there? What are your expectations at this point? And then why read out that data first before broader sensitivity data on a larger scale? Are you using that trial to make a go, no-go decision or kind of otherwise inform your investment magnitude and level? Yeah. Well, I mean, at first, I think this market, kind of the blood-based early cancer detection screening market, particularly for colorectal cancer, meets all of the criteria that we've been talking about here in our chat in terms of this is just a huge unmet need. I'm 45. I got a colonoscopy for the first time this year. That was like an unbelievable ordeal. I might not have gotten it, but for the fact that I work at this company and I've interacted with—I hear the patient stories pretty frequently. And so I'm more on it maybe than I would have been if I was the CFO of a defense contractor or something like that. I might have skipped it, honestly, because it was terrible. So there's a set of people that will go that are responsible about getting their colonoscopy, and I would obviously recommend that. There's a huge chunk of people that will not do that. They will not send in a box, a mail-in stool to Wisconsin. And they're just going to go without screening of any kind. And for them, being able to screen that population with a plasma test has a chance to do enormous good and save a ton of cost for the system as well. So it's an enormous unmet need, and it's an opportunity that needs to get solved. You've seen some of the challenges with some of the companies in the recent consolidation. There's not going to be 10 of these companies. There's going to be kind of two or three of these players, I think, that have a viable test. And we think that we can be one of them, okay? There's a lot of technological and commercial adjacencies that we bring to bear here by dint of the fact that we have a very large primary care channel already. We have, obviously, a large oncology channel at this point. A lot of the technology that we actually developed on the path to having a tumor-naive MRD is quite relevant to the product development here. So tons of adjacencies kind of lead us to this point. As it relates to why read this data out? Well, we have it. So this data will be advanced adenoma data that is just collected and read out on the exact same protocol as the FDA enabling study will read out. So the goal there, and it can only be a goal. It can't be a guarantee. The goal there is to try and get ahead of some of these issues that you've seen with prior readouts where the initial readout was one number, and then there was a lot of degradation when you got the FDA enabling study. Obviously, the FDA enabling study, that's the number. Otherwise, you wouldn't need the study to begin with. But we wanted to read that out because we'll have it, one. I think that's relevant for investors. And just in the interest of transparency, we feel like that's not a number that I want to be sitting on for another year and a half while the whole data set reads out. And two, it does bear on the decision. I mean, I think if our presumption is wrong and it's not a viable approach that we have, I think we'd have to—that would give us pause. But I fully expect this to be a positive readout for us. We'll see how it goes, though. And that's kind of the point of doing the science is you actually have to do the science. You have to report the results. And so in that spirit, that's why we want to read this data out to you. You said we want to be sure it's a viable approach. What is viable to you? Is it at the level of other tests? Is it better? Is it worse? What are you looking for? It's hard to tell. I mean, I think you'd have to have a clinical consensus that this is a competitive product. Fair enough. And then just to push on your statement about commercial adjacencies, I definitely understand that you've got strong relationships with OBs, and many OBs are also practicing GYNs. Not all of them, though. And I would imagine probably your stronger relationships are on the OB side where those patient relationships are nine months every few years as opposed to what you really want, which is the GYN relationship, which is every year, someone who could prescribe a test. So how large, what portion of that base do you really cover? And how realistic do you think it really is to do this in-house versus with a partner? I think it is realistic to do in-house. I mean, I think we've shown an ability to scale these very large market molecular diagnostics tests over the last 10 years. We've got some very capable sales channels. And it's not just, I mean, I referenced the Women's Health Channel. Part of what we've had to build over the last 10 years is a very strong kind of hospital and enterprise commercial channel as well. So there's a lot of kind of pre-existing work that we have to have already done anyway where a product like this would be extremely relevant, and we could efficiently get it to get this to patients. We don't have some religious opinion about partnering, about engaging with commercial partners. I think our own experience has been that commercial partnerships are inherently more challenging than just having our own commercial operation be fully responsible for the outcome. But that's not a decision we got to make right now. We have a lot of wood to chop before we make final decisions on commercialization. Fair enough. All right. We've got about a minute left. So let me give you one last one. When I talk to investors about Natera, the pushback I get most often is, "The company is great. They've got fantastic technology. They're in the right markets. They execute really well." But everyone knows it. And it's hard to make an argument for something that's still underappreciated at this point. So I'll turn it over to you. What, if anything, do you think is still underappreciated by Natera? Yeah. I think you might have asked me this question at a previous conference when those shares were like $60 or $70. I will have that question come up and say, "Well, we missed it, right?" You haven't missed it. I don't think in terms of the market, I don't think anyone has a price-in of any kind of adoption of the Signatera TAM that we walked through at the top of the call. Nowhere close. I think that is coming. So I think that's just one piece of it. The data that we're going to have here coming up at a near-term academic conference is just a perfect example. Really, this type of interventional data has just never been delivered before. And so we really are in kind of the initial stages of adoption, particularly for MRD and recurrence monitoring. All right. Well, we'll leave it there. Thank you so much. Thank you. Appreciate it.