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

Hello. Thank you for standing by, and welcome to the ASCO GI Conference data review conference call. [Operator Instructions] Please be advised that today's conference may be recorded. [Operator Instructions] I would now like to hand the conference over to your speaker today, Mike Brophy, Chief Financial Officer. Please go ahead.

Thanks, operator, and thanks, everyone, for joining us to review the data presented at ASCO GI on Saturday. Joining me on the call today is Steve Chapman, our CEO; Solomon Moshkevich, GM of Oncology; and Alex Aleshin, who is our Head of Medical Affairs and is a practicing oncologists. We'll be presenting slides today that are available on the webcast and they'll also be available at investor.natera.com and there will be a replay of this call available at investor.natera.com. Let's go to Slide 2 for the safe harbor. I'm not going to read this entire safe harbor, but I'd encourage you to review that carefully. And with that, let me move to Slide 3 and hand it over to Steve. Steve?

Great. Thanks, Mike. Since we launched Signatera, we've seen incredible growth. Despite the success, we are still very early in penetrating what is an enormous market opportunity in solid tumor MRD testing. We've estimated in stage II and stage III colorectal cancer alone, there may be approximately 1 million tests per year once the market is fully penetrated. And that represents only a fraction of the pan-cancer opportunity. Today, we will be sharing data that may drive increased utilization, deeper penetration and may set us on a path towards achieving NCCN guidelines in colorectal cancer. I'm excited to share new predictive data from the CIRCULATE-Japan study, the largest prospective MRD study. There are 3 key predictive findings that have never been reported on before. First, Signatera-positive patients post-surgery benefit significantly from adjuvant chemotherapy; second, Signatera-negative patients post-surgery received no significant benefit from adjuvant chemo; and third, Signatera's ctDNA clearance is predictive of treatment efficacy. In addition, we will be showing exceptional sensitivity from a single time point post-surgery. This is very meaningful given the size of the study and the number of recurrences. Finally, I'm pleased to announce Natera's expansion into pancreatic and anal cancers and review the implications of this data for the broader field. With that, let me turn it over to Solomon Moshkevich, the General Manager of Oncology. Solomon?

Thanks, Steve. Just before we dive into the data itself on Slide 4, I want to review why this predictive data matters so much for the field. Fundamentally, the lack of predictive data has been a barrier to adoption of Signatera into the guidelines and in the clinical practice. Even though our growth has been very strong to date, and when I'm out speaking to physicians, community or academic oncologists, everybody values the prognostic nature of MRD testing, but they always ask, look, are my MRD-positive patients just destined to relapse no matter what or can treatment help? And we know that fundamentally, that predictive evidence will change the way that doctors speak to their patients, not only about the risk of their disease, but as they counsel their patients on the risk benefit of actually going through with toxic chemotherapy. Furthermore, just last month, the ASCO guidelines committee released an update to their treatment guidelines being early-stage colon cancer, saying the following: ctDNA was identified as an emerging potential predictive factor. However, insufficient evidence of predictive value of chemotherapy was available to warrant its inclusion in the main recommendation. The expert panel anticipates that data on ctDNA will be forthcoming through prospective clinical trials and included in a future version of this guideline. So this just underscores how much the field has been waiting for predictive evidence of the nature I'm going to present. So on Slide 5, a quick overview of the difference between what we're presenting now versus what was presented last summer at ESMO GI 2021. Last summer, we presented analysis of about 800 patients where outcomes were disease-free survival stratified by MRD test status. And at that time, the mean clinical follow-up was about 6 months. This past weekend, what was presented was over 1,000 patients analyzed. And this time, the disease-free survival was stratified not only by MRD status, which is the result of Signatera, but also now by treatment status. And that's the key new element that enables the predictive claims. In addition, clinical follow-up was extended now up to 12 months instead of 6 months. So let's dive into the data. On Slide 6, this is data that looks quite similar to what was presented before. This is now disease-free survival from a single time point at 4 weeks post-surgery, and it is only stratified by MRD status. So the green line are the patients who tested MRD negative at 4 weeks post-surgery and the blue line are the patients who tested MRD positive, again, just at that single time point post-surgery. And a couple of key things to point back here. First of all, the 12-month disease-free survival was 95% for patients who are MRD-negative regardless of whether they receive treatment or not, and we'll dive deeper into that in a moment. And the MRD-positive patients had disease-free survival at 12 months of 55.5%. And again, we'll dive further into that in a moment. The other key thing I wanted to point out was that our ability to detect recurrence with a single draw at 4 weeks post-surgery was 67.6%. So this is a pretty significant finding because previously, we presented data, for example, in the Reinert study, where detection with a single time point was approximately half and now we're showing a much higher number in also a much larger cohort. We had over 5x more recurrences and more than 5x more patients analyzed in this setting. This is also a prospective study, where we're actually receiving blood samples and reporting our results in real-time. And finally, I think it's an indication that the test performance has actually improved since the analysis we did with the Reinert study back in 2018. So this data looks strong. And on Page 7 now, the question is, well, let's dive further in. So for the MRD-positive patients, do we know if they actually benefited from chemotherapy? And was that benefit observed across different stages of disease. On Page 8, the finding was very clear, MRD-positive patients do benefit significantly from adjuvant chemotherapy from ACT. You can see it on this slide, the results are divided up between high-risk Stage II on the left, Stage III in the middle and Stage IV resectable disease on the right. As a reminder, all the -- the data points shown on this slide are all MRD-positive patients. So the green lines represent the patients who received adjuvant chemotherapy and the blue lines represent the patients who did not receive treatment. And you'll notice that the p values in each case are significant at less than 0.05 and a pretty strong distinction between the curves in each setting. So we think this is a very significant finding. This has never been shown before in early-stage CRC. And I'd like to ask Dr. Aleshin for a clinical perspective on how this might impact clinical practice.

Thanks, Solomon. We believe the potential implications of this data for clinical practice are significant. The central question in the management of early-stage CRC has always been if adjuvant chemotherapy could benefit CTA-positive patients or are these patients destined to recur regardless of treatment rendered. This data strongly suggests that the former is the case across all of the stages of disease presented here. The benefit of -- benefit shown is also striking, even when adjusting for all known confounding factors such as age and performance status. For example, the presented adjusted hazard ratio of 8.8 in Stage III patients suggest that individuals who are CTA positive and undergo only observation are more than 8x more likely to recur versus the patients who received adjuvant chemotherapy. As an oncologist, I think this data will change how MRD-positive patients are treated. It is hard for me to imagine a scenario where I would offer observation to CTA positive patient who is otherwise a candidate for adjuvant chemotherapy. With that, Solomon, back to you.

Great. On Slide 9, now let's do a similar analysis on the patients who were MRD negative. So again, the question is, did any of these MRD-negative patients benefit from chemotherapy? On Slide 10, again, we answered the question based on the data presented last weekend, there was no significant treatment benefit observed in the MRD-negative population. Now the data you're seeing here now, again, as a reminder, these are all patients who tested MRD-negative at that single time point 4 weeks post-surgery. The green represents patients who receive chemotherapy. The blue represents patients who did not receive chemotherapy. At 12 months post-surgery, 96.2% of patients who got treatment were disease-free compared to 94.7% for patients who did not receive chemotherapy. That's an absolute difference of 1.5%, which was not a statistically significant difference in this study. So this is a significant finding. This is why you see 2 curves truly superimposed upon one another. And we expect this also is going to make a significant impact on the field and is a data that people have been waiting to see for a long time. Alex, anything you'd like to comment?

Yes, definitely, Solomon. So when we think about assays, we can think about assays as either being positively or negatively predictive. In the setting of Signatera-negative patients shown here, the clinically relevant predictive nature of the test would be of the ctDNA-negative individuals did not significantly benefit from adjuvant chemotherapy. These results are striking in that they suggest that this indeed may be the case across the most clinically relevant stages of disease. Avoiding unnecessary chemotherapy is really the holy grail for improving the management of early-stage CRC. Chemotherapy is not only toxic but is associated with a non-insignificant risk of mortality. While this data will continue to mature, today these results suggest that CTA-negative patients have exceedingly good prognosis regardless of adjuvant chemotherapy use in this setting. This data may inform decisions about deescalating therapy from, for example, 6 months to 3 months of FOLFOX or deescalating from a platinum-containing regimen to a purely fluoropyrimidine-based regimen, especially in situations where these difficult clinical decisions are already being considered. Solomon, back to you.

Thanks, Alex. So we just presented the data from that single 4-week time point post-surgery. But on Slide 11, we're illustrating here that, that was not the only time point taken. So patients are being monitored regularly specifically patients after surgery are tested at 4 weeks, then if they're going to get treatment, chemotherapy started just before that 8-week time point. And then patients are tested again with Signatera at 12 weeks. So one of the interesting questions that was investigated was does the MRD status change for a patient between 4 weeks to the 12-week time point between positive or negative? And does that change correlate with outcome? And can ctDNA clearance actually predict treatment benefit. So that data was also presented this past weekend on Slide 12, which you see a summary of that. Fundamentally, the finding was that ctDNA clearance is very important and predictive. So here, you're seeing 4 different categories of patients, 4 different lines, representing patients who started out positive at 4 weeks and stayed positive, patients who started out negative stayed negative and then patients who started out in one category and flipped to the other. So I think the most important line to zoom in on here is the orange. So these are the patients who started out positive at 4 weeks and then tested negative at 12 weeks, presumably because they are benefiting from chemotherapy. And the results here are significant. So for those patients, disease-free survival at 6 months was 100%. These patients in the orange essentially fared just as well as the patients who tested negative to begin with. So that's a pretty significant finding, and it's something that nobody has ever seen before in early-stage CRC. Alex, could you give your perspective on the clinical significance here?

Absolutely. So CTA clearance is important to further stratify the risk of Signatera-positive patients receiving adjuvant chemotherapy. And here, really, for the first time, we show definitive data that Signatera can resolve patients who may be benefiting versus non benefiting from adjuvant chemotherapy. While Solomon discussed patients who cleared their ctDNA, it's also important for us to understand the implications of patients who have persistent ctDNA during adjuvant treatment. So focusing in on the hazard ratio of 15.8. The way to interpret this number is that individuals who did not clear their ctDNA had an over 15-fold increased risk of recurrence compared to patients who cleared their ctDNA at a 12-week time point. Clinically, this suggests that really for the first time, we're able to monitor benefit of adjuvant therapy real time. Today, clinicians can use this to better tailor adjuvant therapy to patients who may or may not be clearing their ctDNA. So for example, we can imagine a scenario where a patient who does not clear their ctDNA at 12 weeks has earlier imaging to hopefully find an oligometastatic site of recurrence or possibly have their adjuvant regimen escalated, for example, going from a 5-FU monotherapy to a FOLFOX-based regimen. Furthermore, the implications for drug development from this finding are also profound with this data really allowing for more efficient adjuvant trials to be designed and run in the future. Solomon, back to you.

Thanks. Very exciting findings. On Slide 13, we've summarized the key takeaways from the CIRCULATE-Japan data that was presented this weekend. On the left side, the key prognostic results, on the right side, the key predictive claims. We're excited about this, really having an impact for many thousands of patients. I'd also like to acknowledge on Slide 14 the incredible leadership and contributions from our collaborators in Japan, in particular, Dr. Yoshino and Dr. Kotaka, who presented the data. We are really honored to be collaborating with this group on this very successful clinical trial. With that, I'd like to go into new data that's been presented on Slide 15 and 16 in pancreatic cancer and in anal cancer. We had an oral presentation also by Dr. Greg Botta from UCSD in pancreatic cancer. Pancreatic cancer is a real scourge. About 60,000 new cases per year, just under half of which are early stage and a very high rate of recurrence in pancreatic cancer, 85% after surgery infinitive therapy. Surveillance for early recurrence detection is already recommended by the guidelines, but current biomarkers are not very accurate. CA 19-9 is the serum-based biomarker that's in most common use, but it lacks sensitivity and specificity. And in fact, a significant number of patients do not express CA 19-9 at all. In this study, we presented 116 patients with 450 time points across all stages of pancreatic cancer. The results were very strong. Simply speaking, Signatera status was significantly associated with disease relapse and outcome while CA 19-9 was not. And we think this has potential to have a significant impact on the clinical care for patients with pancreatic cancer. On Slide 16, a summary of what we presented in anal cancer. This is about 9,000 new diagnoses per year, but a really significant unmet need. So there are no established circulating biomarkers in anal cancer. Imaging, furthermore, is difficult to interpret. Surveillance tools tend to be invasive. And for the 10% to 30% of patients who do recur locally or regionally there's really something that you can offer these patients. Many of these patients if caught early enough, are candidates for salvage surgery or can be treated early with other points of therapy. So in this study, we presented 25 patients with over 140 time points, again, across all stages of disease. And the results were very strong in the early stage cohort. There were 3 relapses, all of which were detected by Signatera. And in the MRD-negative cohort, we observed no relapses. So very strong data with both of these new indications, which we have not previously presented significant data on. We look forward to publishing this data and then having discussions with Medicare and MolDx about expanding our coverage. Alex, anything you'd like to add from a clinical perspective?

I think that's a very good summary, Solomon. I agree. I think this is the first step in bringing Signatera into the arsenal of tools to help improve clinical management of these 2 diseases. I'd like to point out that both pancreatic as well as anal cancer have already strong established guideline recommendations from the NCCN for frequent surveillance scans and visits in the perioperative or post definitive chemoradiation settings. However, even with these recommendations, pancreatic cancer is frequently diagnosed as recurring too late with many patients presenting with an organ dysfunction, which prevents effective next-line systemic therapy from being delivered. Furthermore, for anal cancer, where detecting recurrence early is imperative, especially since these patients may be offered salvage surgery options. However, with established methods, this is not always possible. So as such, we really believe this data is very exciting and provides an opportunity to really bring early cancer detection to many of these patients hopefully, in the not-too-distant future. With that, Solomon, turning over to you for the conclusions.

Thanks. In conclusion, we were very excited and proud of the data that was presented this past weekend. It demonstrated that Signatera is both prognostic and predictive, of treatment benefit in colorectal cancer with exceptional sensitivity at that first time point post-surgery. This data increases the likelihood of adoption of Signatera into the guidelines and expansion of private payer coverage. It demonstrates Natera's expanding data leadership in colorectal cancer, but also in other GI cancers. And finally, I think, again, it emphasizes the strength of the tumor-informed and personalized approach that Natera has taken, which has produced consistent great performance in [ curing ] cancer. So with that, I'd like to open for Q&A.

[Operator Instructions] Our first question comes from Puneet Souda with SVB Leerink.

Thanks for the excellent scientific deep dive here. Exciting data for sure. So I wanted to bring it to a bit of a higher level and really ask the question. First of all, the prognostic value of Signatera was already being proven in the prior data sets. And this data set is clearly much larger than what you have seen, I mean, shown before. But now with this data and the role of adjuvant chemo, the key question here is what does this data do for adoption of Signatera both on the sort of the clinical side and the academic and the community setting as well as what does it do to the biopharma partners? I mean what sort of inflection, if I could ask, if you could quantify that, what sort of inflection should we expect here given this data set at this point in time in the journey of Signatera as an assay?

Yes, that's a great question. And I'll tell you like having been out in the field, met with many physicians to talk about Signatera and talk to the sales team, the #1 objection that we've been getting for the past 2 years is that we don't have predictive data. Now we've seen excellent growth and continue to do exceptionally well, outperform our internal models just based on the prognostic data from Reinert alone. So I absolutely think this new predictive data and the size and the real-world prospective nature of the study and the performance that we're seeing will fuel the next level and next wave of utilization. Exactly how far that's going to go is going to be hard to predict. But certainly, with the sales force out there armed with this very strong data with the response that we've seen on Twitter from many of the top KOLs and from our individual conversations with KOLs, I mean we're feeling very, very strong about this.

Got it. And then maybe a question for Alex. How soon do you think the ctDNA clearance starts to become a reasonable key mark or endpoint for versus something like DFS or OS? And just wanted to see if you are -- if there are any biopharma trials that are currently employing that strategy. And also, if I could ask just briefly on NCCN. This data, does this get too closer to NCCN guidelines? Or is this enough for NCCN guidelines? Or what would be needed to see and guideline inclusion on the NCCN?

Puneet, let me comment on NCCN and then I'll turn it over to Alex and Solomon to talk about the surrogate endpoint use in biopharma. So for NCCN, we think this is enough to get guidelines changed. Now we have to go submit the data, get it published, and we think we can do that by this summer. I think the PIs are working really hard right now to get the paper drafted and get it submitted. So the NCCN committee meets a couple of times a year. We think there's a meeting coming up this summer. So the goal is to have the paper submitted for publication and published ahead of the meeting. And if that happens, then we can submit it for consideration at that summertime meeting. So based on what we're hearing right now, all the KOLs are positive about the data that we're presenting, the strength and the quality of the predictive data that we're seeing. So certainly, we're going to be pushing for guideline change in this meeting coming up the summer if we can get everything operationally executed ahead of them. Of course, we don't control exactly what the guideline committee is going to do. But certainly, we're going to be pushing hard for it, and we're going to try to do all the operational things to put ourselves in a position for that. Alex or Solomon, do you want to comment on the surrogate endpoint utilization or the use in biopharma studies?

Yes.

Yes. Go on Alex.

I can definitely provide some commentary here and then Solomon can fill in any gaps. So already, we've seen that for certain Phase I and Phase II studies, CTA clearance can be used as either the primary or secondary end point. I think the bigger question is when can CTA clearance be used as a surrogate endpoint for registrational Phase III study. I think that's still an open question. I think on a one-off basis, the agency may consider that, especially for indications that historically have not been studied, for example, or where progress has been slow. But I think broad adoption and acceptance of CTA clearance as a surrogate endpoint for registrational studies, it is still a year or 2 away, and there are multiple efforts ongoing to try to bring that into practice even sooner.

I'll just add one other piece. Yes, if I can have just one other comment to your prior question about clinical adoption based on this predictive data, I think what we're seeing again here is a pattern, a strong pattern of performance. We previously published very strong predictive evidence in journal Nature last year in bladder cancer, where we saw incredible performance, both in the MRD-negative and MRD-positive arms. And now we're showing it in a prospective trial in a second cancer type in colorectal cancer. And I think as we've seen before that physicians and academics recognize that pattern effect and it's going to have a domino effect not only in colorectal cancer but in other histologies as well.

Our next question comes from Julia Qin from JPMorgan.

I know you talked about having confidence that this is going to be enough to push for guideline changes. Just curious because I -- this data is based on a Japanese sample. So would this be sufficient? Or do you think guideline bodies would need to see U.S. data such as BESPOKE registry trial that I know you also have undergoing. And then just to confirm, in terms of the upcoming guideline update, do you expect that to be about both adjuvant escalation and de-escalation? Or do you think escalation is going to be more likely than de-escalation in the near term?

Yes, that's a great question. So Solomon, do you want to comment or Alex, on Japanese cohort and then we can take the escalation question after that?

Sure. Yes. There's -- we have no reason to believe that the data coming out of a Japanese trial could not be used for guidelines changes in the U.S. or elsewhere. This is a very highly respected group of researchers and physicians. The data is being collected in a very robust manner. But all that being said, as we've announced previously, we were very pleased to have been selected for the U.S. arm of the CIRCULATE trial that we look forward to kicking off over the next few months in early to mid-2022. Steve, do you want to comment on the other question?

Yes. Well, yes, also just say these Japanese partners that we're collaborating with, I mean, they're really part of the sort of global who's who in the colorectal cancer. And Japan is an area where colorectal cancer is very common. It's actually -- in fact, there's as many cases in colorectal cancer in Japan as there are in the United States roughly. So they're considered to be really world's experts in this field, and I think they bring a lot of respect. And they're part of that same community as the sort of top-tier clinicians in the United States. I think, Alex, do you want to comment on NCCN and sort of escalation versus de-escalation?

Yes. Just to kind of close out the prior question. So we've seen Japanese and U.S. data be combined in studies like IDEA. So we do believe that at least in the scientific and clinical communities, the 2 could be interchangeable. In terms of the guidelines short-term question, we definitely believe that escalation is likely to enter guidelines first before the de-escalation component is widely accepted.

Okay. Got it. That's helpful. And then in terms of indication expansion, I know you mentioned generating new data for pancreatic and anal. Are these also the indications that you are prioritizing for as a next step for generating clinical utility evidence like you did for CRC? I know you mentioned that for these 2 indications, surveillance guidelines are already in place. So is that a reasonable way to think about it?

Yes, so the great thing about Signatera is it's really a tool, right? And we don't have to go develop a new test every time we want to go into a new indication. So the test that we're running in pancreatic is the exact same test that we're running in colorectal and it's the exact same test that we're running in anal, in breast and bladder, et cetera. So what we do is we basically apply this same tool across all the different pan-cancer indications. We have over 100 oncology clinical trials that are happening right now or data sets that we're analyzing, and we're going to be reading out data constantly. And every time we get a significant data set like this, it gives us an opportunity to take this and go try to get coverage. So when you look at pancreatic, this was over 100 patients and 450 time points. That's a very big data set when you compare to other studies that have been done in the solid tumor MRD space. Even in anal, there are 25 patients at 140 time points. I mean, that's a decent-sized set given the size of the anal-cancer opportunity. So what we plan to do is get these papers published. We don't think that would take any more than 6 months, probably at the upper limit. I think we're going to submit the papers to get coverage from the MolDx program. And prior to going in for a formal submission, we'll have a pre-meeting kind of feel out where we stand, but we absolutely plan on bringing these forward to MolDx and treating this as an expansion opportunity. We think the clinical utility here is very high. And I think there's an opportunity to help a lot of patients. And we think the data looks fantastic. As we described recently, I think at big health care conferences in previous earnings calls, we've got a lot of near-term and long-term opportunities to extend the total available market for ourselves. Just based on the data that we've presented on, I think, presented and published, I think we're in the range of up to 4 million tests per year that we'd be eligible to get coverage for. And then when you look long term, it's sort of in that range of sort of 10 million to 20 million tests per year based on the long-term studies that we have in place. So there's a very, very thick pipeline of data that's coming out where we get the opportunity to just use the tool in a different tumor type and then take it and get coverage.

Our next question comes from Tejas Savant with Morgan Stanley.

Just one quick follow-up to kick things off. Steve, as you think about NCCN guideline inclusion, can you just help us contextualize what underpins your confidence that this data set will be enough to get you over the finish line there? Perhaps some historical precedent or perhaps early conversations you've had with KOLs? Help us think through that a little bit.

Yes. So look, I think the -- when you look at the size of the study right now, we have 3,000 patients enrolled. We have over 1,000 patients that have been analyzed with a mean of 12 months follow-up, prospective real-world data and the performance of the data looks incredible. So when -- after we presented, we talked to a lot of KOLs, we got a lot of feedback. I mean everybody is feeling very, very good about the data. And when you look at recently with the ASCO guideline committee is saying, I mean, they're sort of saying as recent as December that, hey, like we're almost there. We need to see this predictive data. But now we think we generated the data and what is the largest MRD study that's ever been done. We've now read out data on a bigger cohort of patients than has ever been read out before in this field. And so we feel really strong about it. As far as sort of precedent opportunities, I'm not sure if Alex or Solomon, if you want to list any off, feel free to.

I think -- I'll just comment briefly that current guidelines, less use of prognostic features like lymphovascular invasion, T4 versus a T3 tumor. Many of these features are associated with hazard ratios of kind of 1.5 to 1.8 to help guide decisions which are rooted on a patient's prognosis. I think with this data set, we really unequivocally show that Signatera is not only the most prognostic feature that can be used to determine some of these risk of recurrence in early-stage disease, but also has predictive components. And we think that is a very strong indicator, at least a component of the guidelines are going to be changed to reflect this new information.

Got it. That's helpful context. And maybe just another sort of data-related follow-up for you guys. Do you see a time point where continued MRD test -- like ctDNA clearance is futile, i.e., there is no change in ctDNA positivity or negatively beyond a certain time point? And where I'm going with that is I just wanted to get your thoughts on how long MRD monitoring should continue based upon what you found in this study?

Yes. Alex, do you want to comment on that?

Yes. I mean we've definitely thought about this quite a bit. And we still stick with the point that up to 4 time points in the kind of perioperative adjuvant period. So up to 6 months after surgery is going to be important. Now we know that serial testing can improve clinical sensitivity of the assay, especially for patients who may be forgoing adjuvant therapy, either today or in the future, kind of the serial testing components would be important. Now I think that said, for certain patients, based on this data, there may be an opportunity to limit the number of tests done in this period. How much so? I think that's still an evolving question. But even with all that said, we still believe the assay is going to be very valuable in the surveillance setting to be done every 3 months for the first 2 years and then every 6 months thereafter.

Got it. That's helpful. And then just a quick point of clarification really on the stage II patient cohort. How do you envision the hazard ratio will look in the overall stage II population with and without adjuvant chemo? And do you anticipate physicians limiting the use of MRD to only the high-risk stage II population?

Yes. Alex?

Yes. So at this point, we presented the hazard ratio just for the high-risk stage II patients. I think the analysis that include all of stage II as well as stage I will be updated and presented when that is available, hopefully, when the results are published. But no, we actually envision this being used across all of stage II, especially for the low-risk stage II patients where currently chemotherapy is not being recommended. And what we've seen at least from the data presented, is that in the 3 patients who were either stage I or low-risk stage II, who were CTA-positive and did not receive adjuvant therapy, already 2/3 of them have recurred. So it really kind of shows that using Signatera, we can identify a patient of population, which has historically been neglected and hopefully, will be able to receive adjuvant therapy to improve outcomes.

Our next question comes from Catherine Schulte with Baird.

I guess, first, can you just detail what additional information the VEGA and ALTAIR trials will provide to further support guiding adjuvant chemo and how we should think about the time line for the readouts for those components of the CIRCULATE-Japan project?

Yes. Thanks, Catherine. Alex, do you want to take that?

Yes. We believe that the Vega and ALTAIR studies are going to be kind of the definitive studies that are looking at a randomized both escalation and de-escalation question. I think that the way we view these are -- really, these studies will confirm what we showed in the GALAXY cohort. And really, the timing for those we anticipate probably at the soonest, maybe in 1.5 years to hear back from ALTAIR and for VEGA anywhere from kind of 2 to 3 years in the future to have an update.

Okay. But you think that just the data that you have now, if you can go ahead and get that published from the GALAXY study that, that will be enough for guidelines and they won't be waiting for these randomized trials?

Yes, absolutely. We think that the GALAXY data stands on its own. I think if you look at the data, there is almost a component of natural randomization that's built in, where some patients are receiving and not receiving adjuvant therapy across all stages and even when adjusting for pretty much every known confounding factor that may influence outcomes, the hazard ratios are largely unchanged. The adjusted hazard ratios really do show a predictive component of the test for identified patients who will benefit from adjuvant if they're CTA positive as well as patients who have exceedingly good outcomes if they're CTA negative regardless of adjuvant therapy use.

Got it. And then can you just detail which additional indications you've already submitted doses for Medicare? And as we think about the path forward, I know you want to talk to Medicare about whether you should submit those one at a time or whether you could do multiples. Any clarity on that from your conversations with Medicare?

Yes. So yes, we mentioned we've already submitted additional indications. We've I think now published something like 15 peer-reviewed papers and presented a lot of data. So obviously, as papers come out, we're going to prioritize those and get those submitted. I think there's a requirement that you have a peer-reviewed paper to submit. So obviously, that could kind of point you in the right direction of which ones have already been submitted. I think what we believe is we can have a minimum of 5 additional coverages in 2022 and 2023. And then also within that time period, as we're talking about today, get NCCN guideline changed in colorectal cancer. And then beyond that, I think you start to get into this realm of, well, does it make sense to sort of cover indication by indication or does it just become a pan-cancer indication. And I think that remains to be seen. But the most important thing by far is that you have the peer-reviewed data in each indication. And that's an area where Natera has extensive, extensive leadership in this field because we started collecting samples in biobanks, speaking to PIs and running studies as early as 2015. And we have over 100 different clinical trials or biobanks that are underway are being analyzed that we're going to be reporting out on in the future. So this is an area of extreme strength for Natera, as you've seen, and this is only going to further separate us, I think, from competitors in the space.

And your next question comes from Alex Nowak with Craig-Hallum Capital.

Great. Awesome data here. Regarding those that are ctDNA positive treated and they stay positive, what additional studies could you employ here some of the use cases where you found the nonresponders and were able to change therapy? Just would that come from this study in CIRCULATE-Japan? Would it come from the U.S. study in a different arm, just thoughts there?

Yes. Alex, do you want to take that?

Yes. So we definitely think this is an opportunity for additional clinical research kind of beyond just the clinical utility of having this information today. For example, one study in this setting is a study we're doing with [ nFADP ] consortium, which is opening later this year in the next few months where patients who are persistently positive after completing adjuvant therapy will be eligible to receive an extended adjuvant course of treatment to see if some of these patients who were otherwise destined to recur can actually be salvaged. And we expect that this is going to be the first of many studies in this space.

Got it. Makes sense. And then just to go back to the private payers, are they specifically asking for NCCN guideline inclusion? Is there any other guidelines that would be included here, such as the American Gastro Association? And then I guess what else are they asking for, for coverage with CRC?

Yes. I think from our experience in the field, and we've gotten a lot of commercial coverage. And obviously, we're in network with all the major payers. We have routine conversations with these groups. I mean we have the sense that in order to get deep commercial coverage, you need a guideline change. Now it's possible to get partial coverage, some payers covering based on the strength of a study, particularly in an area like IO monitoring, for example, where there's potentially a cost savings opportunity for the payer. But generally, you need guidelines. And that's why we've invested so much in these market-defining studies that we're really, really far ahead on because we think these are going to be largely necessary like, for example, this readout that we're sharing with you today, to get guidelines change.

That makes sense. And then just last question. Just could you provide some more detail on that single time point sensitivity? I feel like on the competitive front, that's going to be the main point of argument. So just how is that measured? How does 67% compare it to standard of care and competition? And then where does that sensitivity go to when you're performing multiple collections?

Yes. So we think that performance is excellent. I think as Solomon mentioned, in the Reinert study, we had shown a sort of roughly 50% single time point sensitivity. And that trial was run, I think, back in maybe late 2017, early '18. And I think the reality is, we just made significant product improvements since then. And now this is a real-world prospective study that is 5x bigger than the Reinert study. So when you look at the number of recurrences, there's 5x more recurrences in this cohort that we're showing you today than what were in Reinert. So we were super pleased to see I think 67.6% single time point sensitivity. We think that's really incredible. We're proud of it. As far as the competitive front goes, I think you guys have to do your own investigations. We're not going to comment on that. But we're, I think, really pleased with our own performance in this particular study. And then I think as far as longitudinal performance, we've reported on that previously. And I think that will be reported on again as we submit the peer-reviewed paper.

Our next question comes from Mark Massaro with BTIG.

Congrats on the data. Maybe just to build off of that last question. We've seen a lot of your data for Signatera and other of your tests that have sensitivity between 90% and 100%. Can you just help us understand the 67.6% in context? What exactly does that mean? And why do you think it's significantly lower than the 90% to 100% we've seen before?

Yes, it's actually the opposite. So the 67% is actually significantly, significantly higher than anything that has been shown previously. And so when you look at I think previous sensitivities that we've shown where we're looking at longitudinal sensitivity, that's multiple time points over a period of time, leading up to a cancer recurrence. So what we're talking about, in this case, is a single time point, 4 weeks post-surgery, being able to identify recurrences that might be occurring 12 months-plus out, for example. So this is really exceptional, exceptional sensitivity, significantly higher than anything that Natera has ever reported before in colorectal in this particular single time point, 4 weeks post-surgery. We do see and we've continued to see longitudinal tracking, ongoing blood draws, significantly increases the sensitivity over time. And that's why we think longitudinal tracking over time is important to detect recurrence. So there's a little bit of a different time point. The first time point you're trying to make the decision, do you give adjuvant chemotherapy or not, right? Longitudinal tracking where you're doing multiple blood draws, you're saying, can we detect recurrence early and potentially intervene on that patient? So it's sort of a 2 different clinical scenarios that we're looking at.

Yes. That makes perfect sense. And then, Steve, one for you. You -- I think I heard you say that you've received a lot of commercial payer coverage. Obviously, that's very well understood in the women's health business. But maybe it's less well understood on the Signatera business. So can you just give us a sense for how many lives, if any, are covered for CRC for Signatera?

Yes. Well, right now, we have not been pursuing private payer coverage aggressively because we know from our experience that we need to have some sort of guidelines in place before we're going to be considered -- taken seriously. And I think our focus has been on getting Medicare coverage and then generating the data that's going to be enough to change guidelines. And so if you look back at women's health, I mean, we spent a decent amount of time working to get NIPT covered. We got half the payers, and then there was really a slog to get the other half until the guidelines changed. And so we get the sense and we know who the payers are that are willing to move sooner rather than later. And I think this data set that we're reading out right now, I think, is really probably the first time that we're in a position to go potentially influence private payers' pre guidelines. And I do think that this data here potentially could influence some of the payers, but ultimately, you're never going to get full coverage until you have guidelines. So we're spending our energy on getting guidelines changed, which I think is going to get to where we really need to be long term.

Okay. And then just a final clarification question. Is it right that NCCN typically requires 2 publications for prospective outcomes data. So if that is correct, and please tell me if I'm not correct. If that's correct, do you expect to submit this data through a peer-reviewed journal in addition to another data set?

Yes. Alex, do you want to talk about the specific publication requirements?

Yes, definitely. So we agree that the NCCN committee typically requires more than one study to actually make a change in the guidelines. They typically don't need to be prospective. They need to be prospectively done, but they can be retrospectively analyzed. And I think a good example there actually in CRC are the studies that led to inclusion of KRAS and then expanded RAS testing for determining who should or should not receive anti-EGFR therapy. So those were large, prospectively collected within retrospectively analyzed studies. And because of that, we believe the combination of the Reinert data as well as the publication of the CIRCULATE-Japan data for the GALAXY cohort will be sufficient for guideline change.

Our next question comes from Max Masucci with Cowen.

Just to keep a few quick ones here. There's a few Signatera SKUs an oncologist can choose from. So for Signatera CRC tests that are being used to guide treatment decisions in the adjuvant setting, can you just give us a sense for how many blood tests are being used on average in the, say, 6 and 12 months post-surgery for CRC patients?

Yes. So what we've said is we recommend, I think, up to 4 blood draws to be done in that adjuvant window roughly. And what we've seen is actually very strong compliance. Not everybody is following through with that. But I think we're kind of in that sort of 3 range, roughly or something in there. But we are seeing even in the recurrence monitoring setting, I think, which is a bigger opportunity when you look at the total market, that compliance is very high. I mean once people start using Signatera, they really feel comforted by receiving the result and they like to get it and they're proactively seeking it out from their physician in contact with us about when their next opportunity is. So the ongoing monitoring rates are high, in line with what we had expected. And I think that's part of what makes this a big opportunity.

Yes. Maybe just a quick follow-up on that. If you sort of separate Signatera tests used to guide decisions in the adjuvant setting versus those that are used for cancer recurrence monitoring for patients that are in remission, it sounds like the adjuvant setting volumes should dominate that sort of test volume mix in 2022. But I would just be curious to hear if there's any way we should frame that split between adjuvant test volumes and cancer recurrence monitoring volumes in 2022 and how that might play out over the next, call it, 2 to 3 years?

Yes. Well, I guess if you look at the sort of incident population, those would be the folks that are moving into the adjuvant window. And that, long term, is going to be a lot less patients than are in the recurrent setting, which would be the sort of prevalent pool of patients that are out there. So I think when you think long term, you just take a step back, there's, I think, maybe a small portion of patients are getting diagnosed every year compared to -- maybe not a small portion, but a portion compared to the total number of patients that are out there with the disease. I think we're seeing growth in both new patients coming in and volume growth from ongoing blood draws that are coming in from existing patients. So I think we're doing well in both the adjuvant setting and the recurrence monitoring setting. And then certainly, this data, I think, is going to be very well received in increasing utilization in the adjuvant setting. And then once somebody is using the test, generally, we're seeing that they transition from the adjuvant setting into the recurrence monitoring setting very smoothly. Solomon, do you want to make any further comments or maybe Mike, about the split that we expect in '22?

Yes. Quite simply, we're already seeing a very clear split where you're many more follow-up tests than new patients. Both are growing very well, but we've already hit that point where we're seeing many more tests that are subsequent follow-up tests for a given patient. Vast, vast majority of patients who start out with Signatera will order multiple tests, and we're working constantly to make sure that it's as easy as possible for patients to get those tests, whether it's through mobile phlebotomy or going back into their clinic.

Our next question comes from Dan Leonard with Wells Fargo.

Thanks for all the time and effort in helping us understand this data. Steve, a 2-parter for you. First off, you said you've made product improvements in Signatera since the Reinert paper. Can you speak specifically to what you've improved, how the methodology has changed? And should we expect that 67% sensitivity number could improve even further? That's question one. And then secondarily, what's the specificity associated with that single time point sensitivity, the 67%?

Yes, it's a great question. So look, we're always making improvements in the product whether it's in the operational workflow, whether it's in the calling strategy, whether it's in the bioinformatics, whether it's in the molecular biology. So there's a lot of things that we've improved over time. And I think we're feeling great to now have, I think, this excellent data out there showing the real-world current performance of the assay at a single time point. I think maybe Solomon, do you want to comment on the specificity?

Sure. On the specificity side, so this was not presented in the ASCO GI presentation last weekend. Because there are so many patients who get treatment and there's so many time points to check, specificity is just a more complicated analysis than seeing how many of the recurrences you've detected successfully. So we expect that, that information is going to be submitted for the peer-reviewed publication. In the past, we've published many thousands of time points now where pretty consistently specificity has been in on a per-test basis, north of 99.7% on a per-patient basis, anywhere from 97% to 100% generally. So we see no reason in this study or in this cohort to believe that there's a significant or material difference from that, but we look forward to submitting that in the near future. And yes, Steve, back to you.

Just one clarification. So Steve, you mentioned improvements in molecular biology. Is Signatera still a 16 marker test? Or has that evolved in any way to add more markers to the set?

Yes. So the results that you're seeing here are based on the personalized assay using 16 markers. We announced previously that we've already done, and we currently offer in an RUO setting, expanded panels of hundreds of different markers at a time. And that's something that we can take to CLIA in the event that we choose to do that. But we think the 16 offers excellent performance as we're seeing here in this colorectal study. And this is the 16-marker test that's currently on the market as these samples came in the run through the kind of standard pipeline.

I'll just add, yes, there's interest in the product improvement. So we we've made significant improvements in how we design the assays, so which variants we select, how we design the probes and also how we call variant coming out of the multiplex PCR assay as well. So there's been multiple opportunities for optimization. We're glad to see that playing out in this data set.

Our next question comes from Andrew Cooper with Raymond James.

A lot has been covered. So I'll maybe just ask one. Steve, you jumped back in earlier and provided some comment on sort of the fact that this could help you out in some other indications as well and really be sort of the start of an inflection. So maybe just a little more detail on your thoughts there and how you think seeing some of this more prognostic data or predictive and product prognostic data presented in a formal manner can really help you across more than just CRC when you start having these conversations and additional tumor types?

Yes. Thanks. So look, the vast majority of doctors out there or community oncologists that are seeing multiple different types of patients. And what we found is that when doctors become familiar with Signatera, they understand the workflow, they see the results, they use it on patients. They want to think about other ways that they can use it. And so certainly, this type of predictive data now available in bladder cancer, as Solomon mentioned earlier, now in colorectal cancer and being so strong, along with the incredible sensitivity that we saw from that single time point of the 67.6%. I mean, certainly, this will get additional penetration into new offices that are not using us. And as they become familiar with the test, they're going to look at the body of data in other cancer types and think about patients that might be eligible for testing. So yes, we think it's -- that's exactly how it's going to work.

Our next question comes from Kyle Mikson with Canaccord Genuity.

Congrats on the data. I just want to go to pharma quickly. So just obviously, guidelines and everything was really help on the kind of clinical side, but how could this type of data or maybe would the effect -- in the future could help maybe the companion diagnostic partnerships and the trials and so forth? Maybe how could those be accelerated? And just overall, I mean, how could the set of data really move the needle in terms of your kind of current and potential relationships with those biopharma companies?

Yes. Solomon, do you want to take that?

Sure. So we are very excited about what this data means for our research partnerships in biopharma and in academia. So we do already have some Phase II trials ongoing that use Signatera as an end point. In some cases, a co-primary endpoint and in other case is a secondary end point to evaluate treatment efficacy. I think we're seeing once again from this data why that makes so much sense instead of waiting for potentially years for disease for your overall survival outcomes, you get results back that are robust within less than 12 weeks of starting treatment. So this is not the first time we've seen this. I expect that this will become more widely adopted in early phase trials certainly. And then as Alex mentioned earlier, as time goes on and more evidence builds that can be presented to the FDA, we'll start to play a stronger role in registrational trials to help those get treatments that are working out to patients faster. So I think that's a big deal, and we look forward to discussing this data further with our collaborators.

Okay. Helpful. Just one quick follow-up. Could you guys share any time lines for the U.S. arm of the CIRCULATE study in terms of pacing of enrollment and how that's going to progress throughout the year, when we could see some initial data? I think I heard enrollment can start soon, but just wanted to get some sense for time lines.

Yes. Solomon, do you want to take that?

Sure. Given that the trial has not yet recruited its first patient, I think it might be a little early to talk about speed of enrollment. But this type of trial does get participation from many dozens of sites potentially over 100 sites around the country that we expect enrollment once it starts to be strong. And we also hope that there will be an opportunity for the Japanese and American teams to be able to combine data certainly on the MRD-negative arm into a combined analysis, which could further strengthen and accelerate the findings.

Yes. And I'll just add, getting selected for CIRCULATE U.S. is a very big deal. I mean, there's going to be many, many sites across the U.S. that are top KOLs, top academics that are setting up shop. They're enrolling patients and using Signatera in their practice to enroll in the study. It's also going to provide another very strong data set. So it's an amazing opportunity that we've been selected for. And I think we look forward to starting the enrollment very soon, and we'll give updates as it's available.

This concludes today's conference call. Thank you for participating. You may now disconnect.