Natera, Inc. (NTRA) Earnings Call Transcript & Summary

We'll go and get started here. Thanks, everyone, for joining us on Mason care [indiscernible], I'm the diagnostics and in genomics analyst at Stephens today. We have Mike Brophy of Natera with us. Mike, thanks for joining us.

[indiscernible].

If at any point, anyone in the audience has a question, just feel free to jump in and ask. Maybe we'll kick it off with the quarter. You guys beat on the top line, made additional progress on gross margin had a pretty significant step down in cash burn. You've called out a number of COGS projects in recent quarters. Could you just talk about what has driven the margin improvement recently? What COGS projects are you benefiting from? And maybe what projects are still ahead that could benefit 2024?

Yes. No, thanks for having me. Thanks for the question. I think just more generally, I was pretty encouraged by the Q3 results. I think it shows progress we are making on a strategy that we articulated a time ago and have executed once really in the women's health arena in terms of building a first-class business, getting scale in that business and becoming cash flow positive. And now we're kind of repeating that once again kind of at a whole company level, really through the oncology channel now, right? So we made a big investment to launch Signatera in a first-class way, develop the evidence in the right way. And now we're going to rapidly move into [indiscernible] Signatera as the standard of care that is rapidly becoming just that. And while that's happening, we've been able to hold, hold this investment stable and see the volumes and revenues continue to ramp and now also get some gross margin traction such that you kind of see that -- see that cash burn go down, and we've got this target that we established a while ago to get to a cash flow breakeven quarter in 2024, and we felt like in light of the Q3 results, we're more confident than ever that we can get there. Specifically, on the gross margins, one of the key components of our strategy over the last decade is to make the investments in R&D to continually reduce the cost of goods sold per unit. So a case study from deeper in our history is that I think when we went public in 2015, the NIPT assay, the COGS were something like $300, $350. Now that's much lower, it's circa half of that today. And that was through a lot of informatics work, a lot of -- lot more flow improvement, so on and so forth. We've made similar investments and have the same kind of success in carrier screening as well. Launch Signatera we kind of see about executing on a playbook. So going back, even in just a couple of years, I mean, the cost -- blended cost of goods sold per unit for Signatera was well above $500, they've been even like $600 per unit, something like that. Now as you're starting to get scale, you've been able to launch some workflows in the lab, that cost of goods sold per unit has come down pretty dramatically. And something there's significant room to run there. So what's launched so far is we've launched upfront exome capability in our San Carlos lab in California. We're slated to launch another exome workflow in our Austin lab next year that will give us beyond the COGS just gives us a lot of kind of control of quality, all the kind of simple things that you want, but it will also help to drive cost on sold per unit down even further. So that, combined with the ASPs continue to mature oncology as a business as it's ramped so dramatically on the volume of the revenue line. It's been a drag on corporate gross margins over the last year. And I think we now have reasonable line of sight to see that actually flip next year where we feel like we've got line of sight just on current reimbursement dynamics to kind of have the ASP, be above $1,000 and the cost of goods sold will be kind of in that $450 range. So that's hopefully not done there. There's a lot more to do on the ASPs, but that's an important milestone for the company. So I know that's more than what you asked for, but that's kind of a quick summary of everything.

Yes. No, that's helpful. And maybe on the leverage you're seeing on the OpEx line. I think what you guys said in Q3 was the plan is to keep rep count pretty stable next year. I'm not sure how to ask this question, but maybe how much leverage do you have from a rep productivity standpoint, essentially in terms of like the peak number of accounts that an oncology rep can handle. Where is the average oncology reps stand today?

Yes. We got some leverage we think is still available in the current sales channel. There's capacity to kind of do the same kind of growth rates that we've achieved this year on the current team. I also would point out that this is not -- the rep counts, I wouldn't think about that as some very strict number where once you get to a certain volume level, you then need a step function up in your SG&A to support that next incremental unit. It's obviously, it's much more linear than that. One of the advantages that we've deployed, I think, to great success in the Signatera launch is we've been able to leverage all of the lessons and all of the successes we had in a very competitive commercial environment in NIPT, and we've been able to kind of bring those lessons and bring those management techniques to the commercial operation in Signatera and that comes -- hits on a couple of different vectors. I mean, one is a very, very serious focus on the customer experience. It also comes with a lot of lessons around how to manage and encourage and empower a commercial operation, okay? So for example, what we found in the NIPT space is that we have -- we would have these reps that we have these amazingly successful territories and their volumes would get to a level where they could spend all day just trying to just manage their existing book of business, but that's suboptimal because these are folks that are the best -- these are your best reps. These are people that are great at winning business and so what we learned to do via trial in era was to add a talented junior person to a territory, let them learn under the wing of some of your most successful reps -- and their job is to manage while you free up that top rep to go win new business. And over time, these more junior folks, they really learn all of the things they need to learn from these top reps. And then they themselves become -- they get their own they themselves become a top rep. So when I meet with sales reps now having been here, this is year 9 for me, it's actually incredibly rewarding to go and do a sales meeting particularly the women's health team because I know these people, I feel old because these people that were kind of [indiscernible] to your people are now some of the best reps in the space and they're homegrown, right? They've learned through our ecosystem of the way, the right way to care for patients and for physicians. And so you just repeat that here in oncology. You make incremental adds to huge territories, and you just bump up your capacity at relatively low incremental costs.

Yes. That makes sense. Maybe a higher-level question here on Signatera. A couple of years ago, it seemed like everyone was developing an MRD test and this space was going to be crowded, a lot of competition. That hasn't really played out. I mean, Signatera is the clear leader in MRD. What has really enabled that? Why is Signa is so successful? And I know you, guys, point to clinical data, which is pretty clear, but could you talk about some of maybe the more underappreciated parts of that, that could potentially be hurdles to competition or things like scaling infrastructure, building a commercial team, optimizing the workflow turnaround times. How do you think about those competitive advantages?

Well, I think it's helpful to understand a little bit of the kind of the deeper history of how we got into MRD and recurrence monitoring. And we didn't -- we started with this core technology that can detect cell-free DNA with exquisite sensitivity and relatively low cost. And the core idea of the strategy of the business is that there's a lot of way. There's a lot of unmet needs to healthcare that can be served with that core technology, particularly when you pair it with an excellent commercial operation that we've already talked about. So the first application for that was in noninvasive prenatal testing. And then the middle of the last decade, we wanted to find a way to serve patients in the oncology space. And we did not hire a consultant to go identify what are the big markets in oncology. We thought about it much more organically and much more pervasive-centered approach, which was what are the unmet needs that our core technology is perhaps uniquely or very well suited to serving that's not being addressed right now. And we wrote these MRD and recurrence monitoring was something that these are now turns in the field. These were words we wrote on a white board in 2014 and 2015. So you start with that. You start with, we already had a massive head start from a technology perspective when we first started developing the product because we chose the space based on what our technology is really good at, okay? The contrast with other players, now that we have some success, we generate some data, which show that it is actually an enormous market opportunity in addition to being something very important for patients. Other players want to get into that but they are also kind of path dependent on whatever technology they've been building in their labs and whatever products that they've been building. So what you see from other players that is, okay, let's try and take what we do well and retrofit it on to serve this new population. And inevitably, it just hasn't worked out as well, right? So we're a few years in now to the commercial launch, and we really haven't seen a significant competitive entrants because it's hard to do. I think that's maybe step one. Now in addition to that, there's an enormous amount of commercial and kind of lab execution, I'd say, required to offer a personalized plasma test that's offered on a repeat basis. I mean you can just start to kind of think about some of the challenges associated with that. You've got to have seamless interaction with the pathology labs. You got to have, obviously, perfect or very good execution in terms of designing the assay and building these personalized assays very efficiently and at relatively low cost. And then you've got to attract the patients. I mean, you've got just a repeat monitoring assay presents a unique execution challenges as it relates to helping the patient get the test when they're supposed to get it, getting the results back to both patient and physician in a clear and comprehensible way. All of these things are things that you've got to just learn. There's no way to kind of teach it ahead of time. So those are just a couple of the, what I would say, are the challenges for fast followers. Then beyond that, I mean, you've got the data, okay? So again, kind of leveraging the core strategy of the business is to make the right investments in landmark clinical trial data that demonstrate the efficacy of the test and the use case for the patient. So we've now -- I mean we just presented 24-month prospective follow-up data in the CIRCULATE trial for colorectal cancer, okay? So we first started working on that trial. I think in like 2017 or 2018, it's taken a very long time to get to now where patients had colorectal cancer. They got enrolled in the study. They had their surgery done. They got one plasma test post the surgery. And now we sit and wait for 2 years to see how are they doing on their journey, some who were Signatera-positive on the strength of that one point somewhere Signatera-negative. And what we showed was, hey, people who were Signatera-positive and got chemotherapy are achieving disease-free survival at circa twice the rate as people who are Signatera-positive and avoid -- did not get chemotherapy. The reverse was also true. So people who were Signatera-negative at that one time point post surgery and went on to get chemotherapy seemed to get no DFS benefit at the 2-year mark versus people that would it -- takes a long time to get that data. But now when we talk to patients and physicians and they asked to say, like, oh, how does the test work? Like what's the test performance like? Competitors are going to come out with, hey, we have this sensitivity or these spikes samples, we had this period [indiscernible] fraction. And the customers are not -- these people are -- they care for patients, right? And so what we're able to go in and say is, "Hey, when people in this study, people that were Signatera-positive and got chemotherapy, had a DFS that's twice as good as people who didn't get chemotherapy." Okay, so that's data that's framed in a way that is actually useful to a clinician. It gets very hard to rerun those studies once we run them, okay? And so that's enormous, I think, differentiator for us.

Yes, that's helpful. So the path to ASP above 1,000. Can you just maybe unpack that a little bit in terms of what's required to get to that target? How you're thinking about the potential timeline to get there?

I mean I think that, that -- well, I think we've framed as a near-term opportunity. I think that there's a path to getting to 1,000 by the end of next year. Just based on the current stack reimbursed tumor types and our current kind of reimbursement paradigm. That does not require footnote inclusion in the guidelines or some substantial change in the reimbursement environment that we're not experiencing today that requires us to continue to get reimbursed on covered tumor types like we should get reimbursed.

Got it. Okay. And then. The mix shift to monitoring and the ADLT rate seems like a big opportunity. I mean, how much is that played in the recent ASP expansion? It seems like you guys are still adding a lot of new patients every quarter. So has that mix stayed relatively stable over the past few quarters? And I guess how much has that benefited you or how much of it is still ahead?

Yes. I mean, I think that the one of the things to model about this is that there is a kind of a waterfall effect where you have when you first launched Signatera, everyone is getting an exome and everyone is getting their kind of first plasma test. But then here we are a couple of years out, we still have patients who got -- who started with us in 2020 who are now getting recurrence monitoring tests, and we continue to win obviously, new patients are coming on all the time. That is inevitable that the mix of volumes is going to continue to move toward people who got their exome a long time ago. That's continued to happen here in '23, and it's going to continue in a fairly in linear way, I'd expect going forward. There's still a substantial portion of our volume are people that have either gotten an exome and this is their first test or they are in a -- they're in their [ adjuvant ] treatment window because we are still really in this kind of rapid launch phase where we're kind of going from 0 to now something like 1/3 or more of oncologist ordered Signatera test in the U.S. in the last quarter, right? So that's not, I think, it's well above our expectations going back a few years, but it's not the 50th percentile, which is where we think that ought to be well above that, right? So there's a lot of room to run in order to drive broad adoption. And that's why the data that's going to read out over time is going to be critically important.

And on the state biomarker bills, there's been progress there. We've seen some commercial payers adopt coverage of Signatera. How material is that opportunity through the biomarker bills? And ultimately, how do you think about the timeline until those actually become a material tailwind?

Yes. I don't -- I mean I wouldn't ascribe any of the progress that's kind of been our base plan to exogenous factors like biomarker bills or things like that. I mean that's a very interesting development. It's been led by the American Cancer Society has done a lot of work. It's got a lot of bipartisan support has not really been something that has been something that we've observed with interest. I think it remains to be seen what impact that's going to have on the space? I mean I suspect that obviously can be positive. I think ultimately, where we want, where the product needs to be is it just needs to be kind of broadly covered by commercial payers, okay? So how do you get there? You've got to deliver the right level of data, and we think we're well on the path there. You've got it then -- that data has got to feel, it's going to be compelling to the relevant professional societies. And once you have professional society support with the right data sets then that's a time where commercial payers will generally cover the test. So that's kind of the standard playbook. I think one potential additional factor is this idea that it's challenging for a payer to cover Signatera in California, but then not in Arizona. I think that's just like administrative and systems challenge for payers that may influence their decision making, but that's kind of all in the future and not really something that when I guide, I'm not going to have some piece of this that this is because of the biomarker bill. It's one of these things that it's part of a list of factors. Yes.

You talked about CIRCULATE. That study was published, CIRCULATE-Japan earlier this year. How do you think about the opportunity for Signatera as a footnote near terms of upcoming guidelines? And could you kind of break down what type of an impact footnote in collusion could have on Signatera from a volume perspective as well as potential ASP tailwind?

Yes. I mean I think it could be yet another one of several hope for catalysts to just continue to drive adoption for Signatera. I think the reality is that both on the ASP front and on the adoption front, we're seeing progress here with or without that [footnote] inclusion because the data has been so compelling. And I think that's been a critical but the most critical piece of that. The footnote inclusion, I mean, the background for this is you've got kind of Level 1 guidelines that are basically like a decision tree for treatment. And then at each point of those decision tree, there's tons of these footnotes and then you can go in the back there's pages and pages of additional discussion about, hey, different options and different choices, patients and physicians might make at various points in the decision tree. Just given the results that was kind of referencing, I was referencing 24-month data, which a lot of us looked at even a little bit better than the 18-month data, but the 18-month data was sufficiently important to make the cover of Nature Medicine earlier in the year. We submitted that for consideration to NCCN. And we'd like to be considered for the inclusion in the footnote that just list out factors that are prognostic factors for cancer recurrence, okay? We feel like the data that we've published and submitted well supports conclusion in that guideline. Whether or not you get it, if it's included this year or next year is obviously not for us to say. We obviously have respect for the committee and the process. And regardless, if we get it, fantastic. If we don't get it, that's also okay we're going to be able to meet all of our [ poles ], and there's going to be more data coming. So just by middle of next year, we will have kind of the next level, have the final level of data in an escalation arms for ALTAIR, which will be the same study. But similar to what I just described previously, but now patients that are Signatera-positive are going to get randomized plus and minus state of care chemotherapy -- sorry, plus or minus chemotherapy plus TAS-102 where the control arm would just be standard care chemotherapy, right? So if you can show a survival benefit from a more aggressive chemotherapy regimen. We view that as really compelling data for future guideline inclusion.

And with the readout of ALTAIR and potential publication, do you expect that to be published ahead of the next NCCN meeting in 2024? Is it -- what is it they meet in August? Is that right?

Yes. It's hard to say. I mean I don't know exactly like when they're going to be next year. The Altera arm is fully enrolled. And so now it's just -- you just have to wait to have a sufficient number of patients to have their cancer actually recur such that the study has sufficient power to make conclusions, okay? So we're just kind of waiting for recurrences. Would love to have top line data out first half of next year, but obviously, that's dependent on the PIs and how the study evolves. And the same goes for timing to get from top line data to publication also something that you've got to let the PIs do what they do. And then is that enough time to get submitted time for evidence development, evidence collection process? That's hard to tell now. So we have to see.

And then again, I think this is probably an unknown, but going from guidelines published, let's say, theoretically, Signatera is included. How quickly do commercial payers move after that?

Yes, a good question. I mean, I think our experience is that there's a couple of quarters there where you hit, there's an initial movement that happens relatively quickly. And then a couple of quarters out, there's enough experience where you can actually see the results of better reimbursement that comes from that. The variables matter. I mean, I think the wording of the guideline matters as well. But I would expect that to be relative to the timeline.

Okay. And I'll stop here to see if there's any questions.

[indiscernible].

Yes. So the question is, do you just -- would the goal be to just forever kind of be in the footnotes? Or would there be an aspiration to kind of get into the -- like the actual decision tree? And it's very much the latter. It's very much as more randomized data is produced, our aspiration would be to be tried and fully in the standard of care for these patients. So it's -- the footnote would be kind of an interrupt step from our perspective.

The biomarker bills just like [indiscernible].

Yes. So the details will vary state by state. But I think the spirit of the -- of most of these biomarker bills is that Medicare covers a service [ poor ] patients in a state and the commercial payers in that state need to provide that same kind of level of coverage. And so that's kind of the spear behind it.

And then maybe one last one here on Signatera and CRC. As we think about what guidelines could look like, the impact to Signatera. You guys have talked about CRC. Maybe it's a couple of quarters ago or a conference earlier this year. Something around 50% to 55% of volumes. And is the breakdown between Medicare versus commercial, something like 50-50? Or what's the split?

No, it's more like roughly 1/3 of the volume is Medicare volume. The balance is commercial and Medicaid. So obviously, there's all of these discussions that we've just been having about kind of the unit economics are kind of based on the Medicare reimbursement, which is obviously critical, but it should give you some sense of how important it is to keep or [why it still] works for us to keep deliberate great results in these big randomized trials because if you want to have the [SAP] part of the standard of care and really help patients across the spectrum, you've got to be in the guidelines in order to affect more and more patients. The majority of these patients are -- they've got commercial insurance.

And then maybe last one here in oncology. The FoundationOne tracker, how are you thinking about the incremental volume opportunity for Signatera IO volumes? How should we even think about the impact from like a margin standpoint or anything like that?

Yes. I mean, look, I think this is a chance to really -- from a -- if you -- like we do, if you think about how do you solve problems for patients, I feel like this is a really elegant solution for a set of patients that have already gotten a FoundationOne panel for therapy selection. And now they've likely -- they've been matched to put, for example, a targeted therapy. And now without any additional effort or time from them really, they can now get the FoundationOne tracker to see how are they doing on that [indiscernible] inhibitor, the accounts going up or down and so on and so forth. That's quite valuable in the way that works between us and the foundation is, the FoundationOne panel, it works as a stand-in for the exome. We get that data from a foundation and then we build a personalized panel for those patients [ and at ] today, that's -- we care for them just as though they were a Signatera patient at that point. So very excited about that, very excited about the ability to really help these patients and to partner with a first-class organization like Roche Foundation Medicine is incredibly exciting for us. It's been a long time in the works in we signed a deal a few years ago. So really excited to kind of get to a launch. We've got Medicare reimbursement, they've got some good data that's been published. So we're ready to go there. I think as it relates to our guidance, it's going to be something that we leave outside of the guidance. We'll leave it as upside because -- it's not something that -- the ramp of this is not something that's going to be within our control, right? Is something that it's going to be driven through the Roche field team, and we need to get some actuals under our belt before we start kind of prognostic thinking about the specific contribution within a given calendar year. But over time, I think it is a good opportunity for us. I mean, I think, the foundation does a very high volume of therapy section panels or one of the market leaders. So if you think about even a fraction of those patients does it make sense for them to get back [ H1 ] on track or, I think absolutely. So I think over time, it can be a nice contributor for the business as well as helping the patients.

And then maybe moving to the women's health business. You guys have made a lot of progress there. The California testing program was a headwind. That seems like it's now largely behind us. So could you just talk about kind of some of the progress you've made there from even an ASP front, what outside of like, let's say, guidelines. How you can continue to drive that ASP higher drive margins expand margins from here as well?

Yes. So we -- for many of you in the audience and maybe some of you online, they have followed us for a period of time. You know that we kind of the longer story. California launched a [ penal ] testing program, requiring kind of a stripped down version of the NIPT result compared to what we typically offer commercially at a price point that is basically impossible for anyone to actually [ service ] the population. So that they presented the labs with a very difficult choice of either, continue to serve these patients at a loss or just exit the space, which is a branching choice, right? So we made the choice of stay in California on the idea that the mission of the company is to serve these patients and we can, somehow, we'll find a way, we'll interact with our partners in California to make this is a sustainable situation. As it turns out, a consortium of the labs that chose left to state actually won an injunction. And took some of the requirements of the mandates of the everyone is foresees stay program we're no longer in force. And so it's kind of more or less back to normal in terms of patient choice. And so as a result, the majority of our -- majority of our patients in California have opted to return to the broader kind of Panorama panel. So some patients that run through the -- run the state program, which we're happy to serve. And so to your point, that's now. You can see a little bit of that kind of in the P&L and the gross margins over Q4, Q1 to now. I think that's now largely kind of we're kind of dead neutral to that program, which is I think is a good outcome. In terms of maybe more broadly women's health ASPs from here, I'm actually cautiously optimistic. I tend to be very cautious in terms of guiding on ASPs improving in diagnostics space, but the trends, I have to say and we said this on call, the underlying trends for reimbursement look reasonably healthy in women's health space. For example, we had guided to continuing erosion in the carriers [indiscernible] for ASPs as part of our guide for this year and one of the reasons why we've outperformed is that, that hasn't come to pass, okay? So the carrier screening in ASPs have actually been modestly improved. So do the NIPT ASP is going to modestly improve. And these are not the kind of moves we've seen at Signatera, but the volumes are sufficiently large here where modest improvements really now, they really show up in the gross margins and then the cash burn for the business. We made a strategic choice about a year ago to really put in more focused effort on getting reimbursed for covered services kind of across the business. And a lot of that focus was just required because we had a new set of payers and its Medicare advantage, payers for Signatera they have certain requirements and certain documentation needs so you've got to meet those but also in the women's health space, there's just a lot of volumes that we run -- and we -- there's a coverage policy in place and the layman would say, "Well, you're going to pay on those 100% of the time and for a particular payer, and you're getting paid 30% of the time. So there's a lot of kind of people and systems and management time and effort required to understand what are the disconnects, fix those disconnects and just get paid on covered services, right? And so that's an effort that we're going to continue to really drive on it.

Got it. And then just touching on guidelines within women's health, microdeletion specifically. Could you kind of walk us through like your expectations of the economics around microdeletions. I mean as the attachment rate stayed pretty similar to where it was previously? And what's a realistic contracted rate for [22Q]?

Yes. I mean, look, the -- I think just more broadly, I think the use case for [ 22q ] is extremely compelling, okay? So in 2014 or thereabouts ACOG put out a guideline that described [ discrete ] for [22q] as experimental because it was unclear and no one has really run the study to prospectively define how common 22q is in the population, even though it was well understood that if you could add [ neonates ] with 22q that there's a substantial intervention that you would take versus just standard delivery care for that infant. Someone needed to run that study to show how common 22q is in the population. And so we ran that study, okay? So we ran a 6-year 20,000-patient prospective clinical trial. What we showed was excellent sensitivity. PPV of about 50%, which is well in excess of the PPV that you'd see for, I think, a broad swath of standard of care diagnostic tests. And I think a significant contribution of new science to the field. So now with that data now in hand, I think the -- what we respectfully hope for from ACOG is updated guideline that just reflects the reality of that data that do science that's been published. ACMG, the American Cost of Medical Genetics, updated their guideline in the wake of that data last December. And so I think that's a good case study for how data can affect guidelines, and we hope to see that continue. I do think that the impact to us can be positive on the P&L front. But really, our focus is on the benefit that we can deliver to patients with that test. And we've seen that kind of in the field. So for every 100 [indiscernible] that we run, about 75 [22q] orders come in the door as well. So the patients and the physicians have largely spoken that they view this as favorable. And I think it's reasonable to expect that we could get some reimbursement for the future. And I think that would be a meaningful benefit to us as it should be.

[indiscernible].

So the -- my understanding of the process was to ACOG did their annual meeting in May. And there were two publicly disclosed working perceptions. One in the summer and then one at [ last ] weekend in September that was on the website. And mind I say that, that process was that's kind of a guideline working group session. And so from here, this is why guidelines don't work as catalysts for this business and why I don't plan for them. I plan to be successful whenever they -- we've got a plan to be robust whenever they come. Oftentimes, post these meetings, you see kind of a 4- to 5-month lag before you get kind of an updated guideline and it could come faster, could come longer and it's really up to the committee. So that's about as tight of a [ Windows ] I can give and even [indiscernible] that, obviously.

[indiscernible] sort of in process now versus waiting [indiscernible].

Well, yes, I think it's hard to -- I think what I've shared is kind of what we know and so from there is kind of just speculation.

All right. Well, finishing up here, free cash flow or cash flow from operations breakeven quarter in 2024. What needs to happen in order for you guys to achieve that from a reimbursement perspective from continuing to just collect? And what do you think is like the most material hurdle or risk to getting there?

Yes, we have to keep executing the strategy. It's going to require good execution. We're going to have to on a fairly stable commercial operation. We're going to have to substantially grow the units again next year. Now we've done that for years on that now, and we feel like we can do that [indiscernible] doing that. But it's a -- the reps are going to have to grow these units while managing a bigger book of business and see above, like we've got experience with how to deal with that. So that's kind of -- that's the volumes. The ASPs need to, I think, need to remain kind of stable in this health side, and we need to see some continued improvement on the Signatera side. And we feel like we do have line of sight to both of those things now. Can we be delayed on those things? Of course, but status quo, we feel very good about where we are there. And then -- so you got -- that gives you kind of growing revenue on a stable expense base, okay? So that's obviously critically important to running any eliminate stand. You need to have a fixed spend on the stand and then you need to have more customers coming through. And then the leverage that we're seeing on gross margin, I think, needs to continue. But to get to cash flow breakeven, it doesn't need to be some incredible weep upward from here. I mean so you see that we burn about $38 million on 45% gross margins in Q3. Cash burn is going to slosh around guys, like as we've said many times before, I mean, cash burn kind of a strict metric in certain ways because you're beholden to like working capital dynamics, CapEx dynamics and things like that. Your collection cycle will affect that quarter-to-quarter, if it doesn't affect it over -- average out over time. So you should see cash burn come up and down, but I think that's a marker to say, hey, like a 45% margin, they burn $38 million, it's possible for them to kind of -- if they can grow into like the high 40s gross margins, they can get to a cash flow breaking quarter. I think it's very consistent with what we've talked about for a long time. So that's why I feel very good about kind of getting to that target, and we're very focused on trying to get there.

And then one last one here on early cancer screening. You talked about $50 million in spend next year on this program. That will give you two data readouts. You talked about the strength of that data ultimately dictates where you guys go from there as well as -- if you're hitting your cash flow goal. So I guess, question is data versus cash generation, two considerations, if you happen to get guidelines from whether it's microdeletions or expanded carrier screening or something like that. How does it look the data hurdle internally for you to continue to invest there? Does it change your thinking in pursuing that opportunity at all?

We really should have said we're going to spend $14 million on it because it's hard to hear on the audio. I think we say [ 1.5, 15 ] or did we say 50? And I think there was some confusion even in like well attention, but like there was even a note to that had 50, 50, what we said was 1, 5 $15 million spend. I hardly hear from some of the sell side that they thought [ 50 ] was a reasonable number. I said I was -- I got a few internal e-mail saying, "Hey, let's ramp it like." But anyway, so we're not going to do that, right? So the idea there, I think the spirit of what we're saying is that we're very excited about the core technology related to early cancer detection. That core technology is not sole purpose of only delivers in early cancer detection assay. You can also deliver, for example, a tumor naive MRD assay, which we would -- launching that would be consistent with our strategy again over -- sorry to keep repeating this theme, but the lessons we've learned over the last decade plus, is to just offer a menu, if there's alternatives out there, there's going to be some physicians and some subsets of patients that just want an alternative. And so you just -- give your better served by just planning on having a menu and being able to kind of serve the patient and the physician holistically. So that's something that could come out of that program in the near term. As it relates to ECD itself, I mean what -- the spirit of what we saw on the call is that we understand that it's critically important that we kind of get to this cash flow break-even target. That's over the pool. We fully expect data on ECD to be fantastic, and we're super excited about sharing with you. So we're going to have one day readout probably early next year, late this year, early next year and then another day to readout in the spring with kind of different cuts of data, which we think can be quite exciting. But we're not going to then like watch into some level of spend that will put us -- force us into losses for years and years on where there's -- you just can't see the end of it. We're going to be -- we're going to be very disciplined and phase gated in our investment approach there with the understanding of being cash flow breakeven is important to us and to our Board and to shareholders.

All right. Last chance for questions. All right. Mike, thanks. I appreciate it.

Thanks for having me. It was good to see you.