CVRx, Inc. (CVRX) Earnings Call Transcript & Summary

Greetings, and welcome to CVRx shares preliminary results of the BeAT-HF post-market clinical trials. [Operator Instructions] As a reminder, this conference is being recorded. It is now my pleasure to introduce your host, Nadim Yared, President and Chief Executive Officer. Thank you. You may begin.

Good afternoon, and thank you for joining us today for this important update. I am excited to share with you the preliminary top line results of the post-market phase of our BeAT-HF trial, as presented at the Technology and Heart Failure Therapeutics Conference or THT in Boston today by Dr. Michael Zile, Chair of the Executive Steering Committee of the trial. Before we go further, I need to state that the remarks today will contain forward-looking statements including statements about expected product developments, regulatory matters and business impacts. The statements are based on plans and expectations as of today, which may change over time. In addition, actual results could differ materially due to a number of risks and uncertainties, including those identified in the press release issued prior to this call and in the company's SEC filings. The next section of slides that I will walk through are those presented by Dr. Michael Zile on behalf of the Executive Steering Committee of the BeAT-HF trial at THT earlier today. As part of the presentation, Dr. Zile included a disclosure statement of his financial interests. As a reminder, the Baroreflex Activation Therapy device or Barostim or BAT, consists of 2 surgically implanted components. The carotid sinus lead is placed under the skin, but outside of the vasculature on the carotid artery, and the pulse generator is placed under the skin in a pocket created just below the clavicle. Electrical stimulation of the carotid baroreceptors results in 2 balanced efferent signals from the brain. The first increases the parasympathetic activity also known as the resin digest mechanism. And the second decreases the sympathetic activity, also known as the fight or flight mechanism. This type of autonomic modulation produces beneficial cardiovascular responses, such as those shown on this slide. BeAT-HF was a prospective multicenter randomized controlled trial. Patients were randomized into 2 groups: those with BAT labeled in red in this presentation, received the Barostim device on top of optimal guideline-directed medical therapy, while the control group in blue in this presentation only received optimal guideline-directed medical therapy. The main eligibility criteria are shown on the slide. Barostim received the Breakthrough Device designation from FDA. To qualify for the Breakthrough Device designation, a device must address an unmet need and show that it has the potential to provide for more effective treatment of life-threatening diseases or irreversibly debilitating conditions. The BeAT-HF trial was performed in 2 phases: the premarket phase, in which 4 primary endpoints were examined at 6 months after randomization led to the FDA approval of Barostim for the improvement of symptoms in 2019. In the post market phase, the primary endpoint was cardiovascular mortality and heart failure morbidity. Additional prespecified endpoints were also examined. An intention-to-treat analysis was performed on the 323 randomized patients who experienced 332 primary events over 1,036 patient years of follow-up with a median follow-up of 3.6 years per patient. The baseline characteristics were well balanced between the BAT and control groups. Patients in the trial has characteristics typical of patients suffering from heart failure with a reduced ejection fraction that are classified as NYHA Class III or Class II with the recent history of Class III. The patients in this trial also had the expected level of comorbid diseases. The heart failure treatment at baseline for the patients in the trial were well balanced between the groups and followed the guidelines in place at the time patients were randomized, especially noteworthy. Approximately 80% of patients have previously received an implantable cardiodefibrillator. For the post-market phase of the BeAT-HF study, the primary endpoint was a composite of cardiovascular mortality and heart failure morbidity assessed as a negative binomial model. CV mortality was defined as cardiovascular death or receiving LVAD or heartens plant. HF morbidity was defined as a nonelective heart failure hospitalization or a heart failure emergency room visits. Additional prespecified endpoints included a hierarchical composite analysis using Win Ratio and all-cause mortality endpoint and the durability of the safety and patient-centered symptomatic endpoints. The top line summary of key evidence from this run is shown here. This is an outline of the data that I will explain in more detail in the next few slides. For each endpoint, the colored dot represents the observation in the trial and the horizontal black line represents the confidence interval. Dots on the right of the vertical line favor Barostim, for both the primary endpoint and the components of the primary endpoint. There were no statistically significant differences between Barostim and Control. However, both all-cause mortality and the hierarchical Win Ratio were significantly improved in the BAT group. In addition, the long-term measures of safety and symptomatic improvement significantly favored Barostim. Next, I will review each of these end points in detail. In this chart, the cumulative number of primary endpoint events per patient are plotted in blue for the control arm and red for BAT. The overall rate ratio was calculated using a negative binomial model. There was no statistically significant difference between BAT and control. On this slide, the 2 components of the primary endpoint are plotted separately. The table summarizes the data, which shows the number of events per patient per 100 years in each group. The results were not statistically significant. However, there are important detailed analysis related to these components that we will discuss later in this presentation. To gain a better understanding of the totality of the data in this study, the Executive Steering Committee performed a prespecified Win Ratio analysis. This Win Ratio method has now been used as a primary endpoint in many cardiovascular randomized controlled trials, powering a trial for a mortality and morbidity endpoint has 2 major limitations. The first limitation is that it only uses information from patients who have had an event, and that may represent only a fraction of all of the patients who are enrolled in a trial. The second is the lack of a hierarchical aspect that treats all events in a similar fashion. Death and heart failure hospitalization may be equal in what it means to the number of events in a trial, but they are not equal to what they mean to a patient and to their physician. The hierarchical composite analysis using a win ratio method captures the totality of the experience of all of the patients comparing the outcome for each patient in the BAT arm against each patient in the control arm. The prespecified Win Ratio was assessed using the hierarchy described on this slide with the order of importance being cardiovascular death, LVAD or transplant, heart failure hospitalization and then quality of life. Approximately 26,000 possible pairs of patients were analyzed and the results will be described as a ratio of the total wins for the Barostim arm divided by the total wins for the control. And a result greater than 1 means the patients in the Barostim arm were more likely to have a better outcome than control. In BeAT-HF, the prespecified hierarchical composite endpoint favored as a BAT group with a Win Ratio of 1.26 and a nominal p-value of 0.04. This means that BAT patients had a 26% greater likelihood of having a better outcome than control. Given the increased prevalence and better understanding of Win Ratio analysis, I believe this information will be considered very important to patients and their physicians when considering if this therapy is right for them. As shown in this pie chart, each of the 3 main hierarchical components was balanced and contributed equally to the total Win Ratio. As was shown previously, the cardiovascular mortality as a component of the primary endpoint was not statistically significant. However, the all-cause mortality which tracked all trial patients who stayed alive and did not receive an LVAD or a heart transplant, showed that patients who received Barostim had a relative risk reduction of 34% and with a nominal p value of 0.054. I need to remind everyone that approximately 80% of the patients in the trial had a life-saving ICD already in their chests and all of the patients who are on guideline-directed medical therapy that are also known to reduce mortality, yet, despite all of these background therapies, we saw a significant reduction in mortality. I believe this information will be considered very important to patients and their physicians when considering if Barostim is right for them. And now we turn our attention to the durability of the safety and patient-centered symptomatic data. As you can see on this slide, the trial demonstrated a MANCE-free rate of 97%. This is further evidence that the device and the rated procedure are safe. The change in quality of life score from baseline as assessed with a standardized tool Minnesota Living with Heart Failure Questionnaire, is plotted here at 6, 12 and 24 months. The change for QOL for BAT patients from baseline is shown in red, control is in blue and the difference between the 2 groups is shown in green. Now focus your attention on the green. At each time point, there is a significant large improvement in BAT versus control. The difference was shown to be durable in time and sustainable in extent. The change in exercise capacity as measured by the standardized test of a 6-minute hall walk is shown here at 6 and 12 months. The study protocol did not seek an assessment at 24 months. At each time point, there is a significant large improvement in BAT versus control. Also here, the difference was durable in time and sustainable in extent. Finally, this slide shows that the percent of patients that improved by at least 1 NYHA class was significantly greater in BAT versus control at 6, 12 and 24 months. In summary, and while the primary endpoint and its components were not statistically significant in BAT versus control, both all-cause mortality and the hierarchical Win Ratio were significantly improved in the BAT arm. And both safety and measures of symptomatic improvements were durable over the assessment period at 6, 12 and 24 months. These favorable data led the executive seating committee of the trial to conclude that the totality of evidence indicates that BAT is a safe, effective and durable treatment for patients with heart failure with reduced ejection fraction. There were additional analysis shown during the CVRx-sponsored lunch symposium at THT, where Dr. Abraham, Lindenfeld, McCann and Zile shared with the audience their interpretation of the data. In the next set of slides, I will share with you some of the key additional findings that they disclosed with the audience. A clinical stability analysis examines whether patients' clinical status improved, worsened or stayed the same. Using the analytics defined on the left, similar to those used in the emperor-reduced HF trial published by Dr. Packer. The Executive Steering Committee performed a prespecified clinical stability analysis in BeAT-HF. BAT resulted in significantly more patients who improved and fewer patients that worsened. The proportional odds of the analysis was 1.9 and the nominal p-value 0.009. This is a reminder of the Win Ratio shown earlier with CV mortality, which was the one pre-specified. Note here that CV mortality contributed 34% to the final ratio as shown on the pie chart. Many physicians prefer to consider the hierarchical Win Ratio with all-cause mortality instead of the cardiovascular mortality as they consider it to be more representative. This Win Ratio with all-cause mortality was performed as a sensitivity analysis. The results remain consistent and significant and as shown in the pie chart, the hierarchical components were still balanced but importantly, all-cause mortality contributed to 41% of the Win Ratio. Back to the primary endpoint and the graph I've shown you earlier. If you look closely, you can see during the time period between 1 and 2.5 years post randomization that appears to be at least some separation between BAT and control. We will look more closely at this time period, the next slide. At 1 in 2 years, that decreased the primary endpoint event rate by 24% and 18%, respectively, neither reaching statistical significance. The Executive Steering Committee explored potential confounding factors which may have resulted in this pattern. Since COVID-19 has been shown to have a significant impact on previously reported randomized clinical trials such as GUIDE-HF the Executive Steering Committee examined first the effect of COVID-19. In the insights, which we have shown earlier, heart failure morbidity data is plotted with all patients starting at time 0 when they were randomized in the trial. In the main graph on this slide, the same exact data is plotted according to a calendar date. This explains the visual differences between the graphs and importantly, shows the clear and significant effect that COVID has on the event rate. In the control group, the heart failure morbidity event rate fell significantly during COVID. While in the BAT group, there were little to no changes in the event rates comparing before and during COVID. The table clearly demonstrates that the primary effect of COVID was in the control group, where the heart failure morbidity event rate fell from 29% to 7% during that period. In all other years before or after COVID, the average event rate was 26% in the BAT group compared to 29% in the control group. This data led the executive steering committee to believe that COVID definitely impacted the hospitalization rate of the study. The COVID impact was differentially expressed more in the control group than in the BAT group. Why COVID had these differential effect has not been thoroughly investigated yet. Whether and to what extent COVID acted to limit the ability of BeAT-HF to identify an effect of BAT on the heart failure hospitalization awaits further analysis. In summary, despite the potential confounding effect of COVID and other potential confounding factors on the primary end point both all-cause mortality and the hierarchical Win Ratio were significantly improved in the BAT group in both safety and measures of symptomatic improvements, we are durable over the assessment periods at 6, 12 and 24 months. These favorable data are leading the committee to include that the totality of evidence indicates that BAT is a safe, effective and durable treatment for patients with heart failure with reduced ejection fraction. And now to our own takeaways. Barostim is currently FDA-approved for the improvement of heart failure symptoms based on the pre-market phase of BeAT-HF trial at 6 months. The post-market phase of BeAT-HF confirmed the long-term durability of safety and symptomatic improvement and the sustainability of the extent of the improvements. The survival free from LVAD and transplant is meaningful by showing a 34% reduction with a nominal p-value of 0.054. The prespecified hierarchical composite endpoint was while balanced and demonstrated meaningful benefit by showing a Win Ratio of 1.26 with a nominal p-value of 0.04, which was stable over multiple sensitivity analysis despite the presence of multiple confounders, including COVID. One or more manuscripts will be written by the Executive Steering Committee for submission to peer-reviewed journals. The PMA clinical report is being prepared by CVRx to be submitted to FDA, to seek an expansion of the labeling, commensurate with the recommendation of the Executive Steering Committee of BeAT-HF. We agree with the committee that the totality of evidence supports the use of Barostim as a treatment for heart failure. I appreciate all of you taking the time to join us today to hear the more detailed explanations of the results of BeAT-HF. I hope that we are able to communicate with you our enthusiasm for the results. And with that, I would like to open the line for questions. Operator?

[Operator Instructions] Our first question comes from the line of Robbie Marcus with JPMorgan.

This is Allen on for Robbie. I guess just to start off, I think the main value proposition of Barostim and really why it's been adopted up until now has been improvements in QOL. So it's definitely encouraging to see that maintained out to 24 months. And while it is disappointing to see that the primary endpoint didn't read clinical significance. As you said, there's some confounding factors and definitely some areas where it does look like Barostim has a meaningful benefit to mortality and morbidity. So how should we really think about the physician response that you've seen so far at the conference to the data? And how do you see it really affecting your adoption given the lack of a clinical and meaningful primary endpoint?

Thanks, Allen, for joining us. Great question. So you are correct. We've been commercializing Barostim so far based on the 6-month symptomatic data, and we've seen a very nice traction so far. Now we're showing that this 6-month data is sustainable to 12 and 24 and safety is sustainable, too. And we're answering the question on whether there is a cost to the improvement of symptoms. Let me explain that in a second. The current medical treatment for those patients is the guideline-directed medical therapy for classes of drug, all for improved mortality, none of them improve symptoms. Yet when patients knock on the door of a physician, they first want a better life, not a longer life, a better life. So some physicians discuss with the patients, the possibility for the patients to take inotropes drugs. Those are not part of the guideline-directed medical therapy, and they are known to shorten the life. So patients are willing to trade a shorter life for a better quality of life by taking those positive inotrope drug. Now we come forward with the proposition and says, "Hey, wait, you don't have to do this. You can get Barostim and it improves much better, much more the quality of life. It doesn't require compliance. It doesn't require taking even more drugs, more pills. And now we can say, and this doesn't come at a cost of shortening your life and maybe it could prolong your life. So that alone, Allen, is a very interesting value proposition that most of the physicians here at THT that listen to the presentation today and came back and interacted with me, they were repeating this over and over again. Now there'll be a lot written about this. What to do with data like this, do you go the dogmatic way and you say, no, you missed the endpoint, you don't look at anything else? Or do you go the pragmatic way where you say, no, it's the totality of evidence. We cannot ignore the rest. And often what happens is if a trial wins the primary endpoint, but there is a negative signal somewhere else. Let's take the opposite example for a second. If we had a 34% increase in mortality with a p-value of 0.1, do you think physicians would have ignored that? Do you think FDA would have ignored that? Absolutely not. We'll be digging right now, trying to understand is that really a mortal risk to patients. So why would we, in one case assume that it's important. And when it's positive, assume that is not important, right? So that asymmetry and the interpretation of the data works to our advantage in this way at this time because we do have a good data. It's a 34%. And answering one of the questions today during the symposium, I think it was Dr. Abraham asked Dr. Zile about the number needed to treat with this therapy at the median follow-up around 4 years. And the answer from Dr. Zile was 10.4. 10.4 is an amazingly low number of patients needed to be treated to save a life if this was a primary powered end point. So I think here is we cannot ignore the data. I think physicians will not ignore the data, and patients definitely will not ignore the data. We don't know yet about regulators and payers, right? So that would be a different question. To the question of adoption, it's a little bit early right now to speculate. We've had a good run so far. We plan to continue the run. And you may want to ask me this question maybe in 6 months or 9 months when we've had more clarity from FDA and some more time with this data with physicians and see how -- whether it's impacting the number of activation of new centers or the number of implants in activated centers.

Got it. That was really comprehensive. So I'll just leave another -- just a final quick one. Obviously, the effect of COVID seems pretty drastic in 2020. But to my eye, it's a little confusing why it would have had such a dramatic effect on the control arm and not on the BAT arm. So you're looking into it yourself already, but do you have any preliminary theories for why that might have led to such a market difference in HF morbidity for control but not for the Barostim arm.

Great question, Allen. So second, a very good question. And we don't believe we have the full picture yet, neither did our steering committee, to answer this question. However, we do have some hypothesis and some speculations at this stage. One of the elements in here is the asymmetry of treatment of the patients in the trial. When COVID hit, the HF collaboratory that gathers academicians, FDA and industry, discuss this issue, particularly based on the data that was observed at Vanderbilt with the reduction of hospitalizations. And we added all of the safeguards in the trial to ensure that we're collecting the COVID data, but also we offer the opportunity of doing remote follow-up for patients. And while remote for what could be done in patients without a device, it is almost impossible to do it with a device when you need to reprogram the device, particularly in the titration phase. So that's one difference between the 2 arms. The second idea or hypothesis came to us from a very prominent statistician who was blinded to the data. So this statistician did not see the data until today. This statistician, when we ask him what should we consider in regard to the COVID analysis, his answer was to look at the adjudication differences. The stricter the rules about what to consider being a heart failure hospitalization primary event and whatnot. The more difference you might see during COVID, and these were his words, and let me try to explain this in layman's terms. How do we collect heart failure hospitalization? The site report any event. This event then is sent to an independent adjudication committee that looks at all of the documents, and there is a very developed charter that they have to go through it and verify if this qualifies as a primary event or not. So that's -- what the site declare there's being a heart failure hospitalization might not be one. And that ratio, we will be analyzing it and looking carefully because one of the hypothesis from the statistician is that during COVID, hospitals were overloaded, the flow of information collected and reported back would have been inadequate at best. And that would have made the adjudication almost impossible. And that would be differential between the 2 arms because the Barostim patients. So patients who have received Barostim would have gone to the sites that implanted the Barostim, while patients in the control arm if they had a congestive event, they would go to the nearest hospital that is not on lockdown. So those are 2 examples where we could have a differential impact. And it's we don't have proof yet. It will require a few more weeks of analysis here to understand it. And I tell you here, everyone is asking the same question, why and what are the explanations. And oh wow, it looks eerily similar to what CardioMEMS, Abbott have seen during GUIDE-HF, but not a single physician told us that they don't believe this or where are we looking at this? It's -- as soon as they look at this chart, the answer is, wow, there is an impact. So thank you for the questions.

Our next question comes from the line of Matt O'Brien with Piper Sandler.

This is Sam on for Matt. Congrats on the data today. I guess, first of all, kind of a follow-up to that prior question as well. We were hoping you could give us just a little bit more detail on kind of potential label expansions and what to expect there and maybe when we could expect it? And then also maybe as it pertains to the -- what all data would be included and maybe some of this COVID and adjudication analysis maybe could be in there as well? Any thoughts there?

Great questions. The first one, I have to be careful in here to make sure that preface -- I will preface everything I say, saying this is our position as a company. And we take this position. We don't know how FDA would react to our position -- our interpretation of the data and the arguments we make versus their own interpretation and the analysis they make. That said, based on the recommendation from the steering committee, we agree with them that we should upgrade our labeling from being considered a device that improves symptoms of heart failure to a device that treats heart failure. That word treatment is not used lightly in this context. It took a lot of discussions internally within the steering committee between the steering committee and CVRx and inside CVRx to make sure that this is what we want to go with. And it is an important claim. If we get disclaimed that this device treats heart failure and not just the symptoms of heart failure, I believe that's a significant upgrade to where we are today. That said, we could elect to go with a lesser one, which will be about the durability of symptoms, longer term, et cetera, and/or we could go for a more aggressive stance saying this reduces or improves outcome in patients, right, based on the Win Ratio. I would say right now, the position is we will go for treatment and everything else will be discussed with FDA, and we might end up combining the treatment labeling with durable symptoms as well. You had a second question, Sam, and I think you mentioned about whether the COVID analysis will be presented.

Yes, yes, that's right.

Yes. So the -- I don't know yet if the COVID analysis will make its way in the first, the main manuscript, or whether it justifies having a separate manuscript just to focus on the COVID impact. So that has not been decided. But clearly, our PMA submission to FDA will include all of the analysis that we would have concluded to date. It will be the list and the exhaustive list of all of the analysis that we as a company and the steering committee have done. So all of this will be submitted to FDA. Now the submission to FDA is confidential. And FDA has 180 days or 6 months to respond to it if everything goes to plan. They could also decide to go faster, we don't know. But the baseline here is after we submit, it will be at least -- I'm sorry not at least, it will be up to 180 days for an FDA response.

Our next question comes from the line of Margaret Kaczor with William Blair.

I wanted to maybe start with -- there's been a few discussions this week on the pros and cons of the Win Ratio. And you spoke a lot about that, but there's also been this discussion over when odds and taking that into account ties, if you will. So if you look at the 1.26 hierarchical composite Win Ratio, but you didn't disclose the win odds ratio. I guess -- have you looked at that? And how maybe your Win Ratio compares to other heart failure trials [indiscernible], of course, coming into play and coapt kind of being discussed as well. And are they [indiscernible].

Yes. Great question, Margaret. And very nice hearing from you. Listen, the [indiscernible] more familiar with is the tricuspid valve to ILLUMINATE that was presented at ACC 2 weeks ago. And you are correct. Some people received it very favorably that they win the primary endpoint. Some other says, well, we're a second -- the reason they want it is about 50% of the effect was driven by the symptomatic endpoint, which was a KCCQ of 15 points or more, and there were a lot of ties, lot of inconclusive comparisons between pairs of patients. In our situation, and although this number was not presented today, it's not a material number, but I can go and tell you about 5% of the comparison ended up in a tie. And the competitor we have for the quality of life is at least 5-point improvement in the Minnesota Living with Heart Failure Questionnaire. So 95% of the comparisons led to a winner, one way or the other. So in my opinion, it's much harder to criticize our analysis the way the data played out. And the reason we showed the sensitivity analysis was to strengthen that point because one of the criticisms will be to say, well, where a second you're showing cardiovascular mortality, but that's not the way to do Win Ratio and so forth. We went during the symposium, and we've shown all mortality version of the Win Ratio, even though this was not the prespecified analysis. So we're showing it as a sensitivity analysis. But actually, it doesn't weaken the argument, it strengthened the argument by showing now that most of the effect is driven by the mortality or a higher effect is driven by mortality than previously. Also footnote in the Dr. Zile's presentation that we did not mention, there was a 24-month competitor as well. So the way the Win Ratio analysis works. All of the data is compared. The quality of life that was used was the 12 months. This was prespecified. But as you know, we also collected quality of life at 24 months. So we ran this as well as a sensitivity analysis. And I think you can see it in the footnote of the slide. It shows 1.34 with a p-value of 0.01. So no matter which way we are cutting or slicing this win ratio data...

Our next question comes from the line of Bill Plovanic with Canaccord.

Most of the questions have been asked. I just have a couple of follow-ups here. One is, one of the comments by the physicians was something regarding the FDA TAP program for HFpEF and HFmrEF. Just any thoughts on next steps for the company in terms of clinical trials to run and data together. Any thoughts on priorities in terms of do you think you need to go get more data in the HFrEF market? Or you're going to look for more label expansion into other areas?

Bill, nice hearing you. Can you hear me okay?

I can, yes.

Oh, perfect. Yes, I thought I may have lost the connection. I'm becoming paranoid with this. So yes. So what is the TAP advisory program, the TAP. This is part of the MDUFA 5 program that FDA started. They're accepting 15 programs this year, all of them in cardiovascular. And we were, I think, the first to submit the request required is a product to be a breakthrough designation to be considered in the TAP program. So think about it like a breakthrough designation on steroid. It doesn't mean that we have decided to start the program now. And to answer your second question, and I can come back to this first one. I don't think we need to conduct more trial in HFrEF. We will be conducting more health economic data collection in the future. We may be requested to do other data collection, registries or so forth for coverage purposes with private payers or CMS, that's different. We don't intend to run a double blinded or a blinded, randomized trial in HFrEF. So as we think about our future clinical trial investment, they'll most likely be in adjacent therapies. And you know that we have 2 breakthrough designations, one in HFpEF and the second in hypertension -- resistant hypertension. And the one that we submitted that request to FDA to upgrade the breakthrough designation to a TAP program is the one with HFpEF, which actually, when you look at the definition we're using for the ejection fraction, it's more HFmrEF plus HFpEF. I'm sorry about this. HFmrEF, which is heart failure with midrange EF and the HFpEF, which is heart failure with preserved ejection fraction, those are the terminologies based on different ranges of EF. So that's the status here, Bill.

Okay. And then I think this was asked, but I just want to clarify. Just in the past, I mean, can you give us examples of FDA really looking at the Win Ratios in terms of approvals and labeling and what their kind of position is, at least in your opinion?

Yes. Actually, the latest one is the triluminal. It was accepted by FDA to be their primary endpoints, and they merit and they're submitting now that request to be approved for the tricuspid valve repair. You want an example of a product that has been approved with the Win Ratio. I don't have one right now. I thought the coapt was one. But one of the things to consider is the Win Ratio methodology did not exist for a long time, meaning it's pretty recent. So for this to have happened, you have had to have a program that selected Win Ratio as a primary endpoint, finished the trial, submitted to FDA and got approval. And I don't have an example right now, Bill, in this regard. That said, let me disclose one interesting tidbit here. We didn't -- we're not aware of the Win Ratio analysis as a methodology when we designed this trial. But you know very well that we worked very interactively and very collaboratively with FDA when we're designing this trial, and the audience yesterday -- in the FDA session here at THT heard so many examples, talking about the BeAT-HF trial as plowing the way in here for the new way of developing clinical trials. So part of those conversations and interactions we had with FDA, they actually hinted to us that we should add in the statistical analysis plan, a Win Ratio analysis. And when we heard this, it was like what is that, right? So we went and did the investigation and we figured out that a statistician, a famous statistician by the name of Dr. Stuart Pocock from the U.K. has taken an existing approach and applying it to cardiovascular trials and so forth. So we worked with him to verify that the approach that we're using is valid and so forth. That's the story, right? So it's not a foreign concept to FDA, but I don't have an example where FDA approved the product yet based on it. And I would associate and get back to you with it when I find one.

And then my last question, Nadim, is you and I have talked about this, but as we sit here at THT, how do you feel that the physician added to is relative to end points in these trials today versus maybe 3 or 5 years ago, as it pertains to mortality, hospitalization, quality of life? And I'll end it on that.

Yes. It's a long answer to your question, Bill here. Let me try to select a road where I don't go down deep into the weeds. Listen, the -- I don't know, in my presentation, I mentioned that this trial, the analysis were done on an intent-to-treat basis, there was a trial for AF ablation recently, where the data on an intent-to-treat basis did not show any effect because a lot of the patients crossed between one arm and the others. And when they did it on a per protocol basis, the data strengthens and was statistically significant. And yet you see where AF ablation is these days in terms of penetration. I think it depends on the physician. You have some that are very theoretical, academic, dogmatic in their approach, and you've got some others who are very clinical, down to earth, pragmatic. And depending on where they sit on the spectrum, they could have a different appreciation of the data. Now in regard to the quality of life, versus duration of life, I think that language has changed dramatically over the past 5 years. When we started this program 5 years ago, cardiologists will basically criticize us because we're focusing on quality of life instead of focusing on the double quote and hear the real outcome, which is mortality. 5 years later now, they are realizing the community at large that actually what patients want when they knock on the door of a doctor, they want a better living. It's more about the living experience rather than the duration of life that they need, that's what patients are asking for and physicians are starting to respond to this. And what has helped in this dialogue is a study that was conducted under the signs of patient input initiative, that was driven by FDA and the Medical Device Innovation Consortium, MDIC. And heart failure was the first project elected, 6 companies, cardiovascular, all 5 large and one small, CVRx was the smallest, plus FDA contributed financially to this project. And the study was done around [indiscernible], the statistics were conducted by MIT and the results were published, where it was clearly shown that when you ask patients using a conjoint analysis, about comparing quality of life versus duration of life, almost 3:1 ratio, they prefer quality over quantity. I think that is helping to change a little bit the attitude here of many physicians toward that debate, quality of life versus outcome. Now we don't have to ask this question anymore at CVRx. I think we have a relatively good outcome data. when you look at the totality of evidence. So I think we've got both of both words here, the best of both words.

Our next question comes from the line of Alex Nowak with Craig-Hallum.

Okay. On the cardiovascular mortality and also on the all-cause mortality, there was this huge split that really occurred at the years for the patient groups. So I'm just curious, as you dig into the data there, what's really the biggest factor with that major split occurring at that time point?

Alex, great question. We don't know the answer, and it could be that we're trying to read too much into it because those curves, you have to think about it like they are surrounded by a cloud of uncertainty. Those are the 95% confidence interval around each of these lines. So the line has just happened to be a line and you're trying to figure out where it is within that cloud. So it could split earlier or could split later. So I don't think there is a physiological reasons. Now some might speculate that, hey, you know what, 50% of the patients with heart failure end up dying after 5 years based on the AHA statistics, et cetera. And that's why you're seeing that type of curve in the control arm, while as in the BAT arm, it's pretty steady. I think that's a speculation. I think that's reading way too much into the data. And maybe I'm not enough of a salesperson here too much of a scientist, but that's my interpretation of the data.

Okay. That makes sense. And then in the prepared remarks, you mentioned beyond COVID that you hinted that there was potentially other confounders that would have impacted the morbidity side of the study. So could you maybe expand on that, what are you seeing in the data when you look at a patient by patient that would have maybe hurt the morbidity results not necessary the mortality piece?

Yes. There are 2 other confounders that we can talk about. The first one is the competing risk of death. So we know from the all-cause mortality data that we had more deaths and LVAD and heart transplant, so more sensoring terminal events in the control arm than the BAT arm. Now let me ask a question and I'll answer it. Who are the patients who would die or get an LVAD or a heart transplant. They are the sickest patients. So you're taking a control arm, and you're taking the sickest patients and more of those and you're exiting the trial because they get these events. So who is left in the control arm, lesser sick patients, more so or lesser sick and the control arm than in the device arm. So when you think about those recurring heart failure hospitalization events, now you have that risk of shortening the population in the control arm and focusing on the healthiest patients because the sickest patients left the trial in the control arm and trying to compare it with the device arm. You have also -- the other elements, which is the difference between ER visit and heart failure hospitalization. This one will have to wait for the manuscript that will be very interesting data in there. There is another one which is the interventions during the trials. So the drug uptake and device uptake during the trial. So LVAD being one, but there are others, CardioMEMS, other devices, other drugs added Entresto, SGLT2i which was made available. And the question is, was the uptake equivalent between the arms. And if it is not, it could be a confounder. And because I'm mentioning here that there was a confounder, so I'm saying that this uptake was not the same in both arms. And again, we'll have to wait here for the manuscript for all of this data to get out in the public domain.

Okay. Makes sense. Just one quick clarification, timing for the FDA submission. And then would you expect the panel meeting based on the data you have in hand?

The timing of the submission, I would say, not immediately, is going to take us some while. I estimated the document that we submit to FDA to have about 600 pages. That's my own estimate. It is going to take a while to verify every single number, every single table, the text surrounding it and routed with our advisers, our steering committee and so forth. So we're going to be at a few weeks down the road here. And then I don't have a specific time to say. And then it will be up to 6 months for a response from FDA. Do I expect a panel? I don't know. I really don't know. FDA had, had a panel to discuss heart failure, both for CardioMEMS and for impulse dynamics. So they already know the feedback from physicians about what is important in terms of efficacy, symptomatic and mortality, morbidity and the safety. If FDA doesn't feel that the data needs to have a vote from an independent panel that the data speaks for itself, like what happened with our data back in April of 2019, then they would not convene a panel. They don't need to. Convening a panel is an expensive avenue for FDA. It requires a lot of time of preparation. It's expensive for the company. So FDA does not take that decision lightly. They will convene a panel if they believe they need one. If they don't need one, they will not convene a panel.

Next question is a follow-up question from the line of Margaret Kaczor with William Blair.

All right. I seem to have been knocked off earlier, so I'm going to follow up with one more. Just a commercial question and I'm just trying to get a sense based on the data today, now that we've seen it, are you guys going to, I guess, invest more aggressively or cautiously as you kind of push or educate on Barostim? And then would you, I guess, wait in some ways until you get the indication expansion or commercial reimbursement approval to change that behavior set?

Apologies. I thought that the problem was on my side when you were cut off, every time I try to answer the question, last time it happened the same thing. I really apologize for this. So listen, the -- it's the latter. I will wait until we know what we have in hand from FDA before we make a decision to change our investment trajectory. So right now, we gave a guidance for the year, and we will keep investing towards that same guidance.

Okay. And I guess a follow-up to my follow-up and no problem it happens. But how do we kind of consider -- what do you consider, I guess, the success, whether it's 6 months, 12 months, 18 months from now, if you kind of look at it in chunks, what would you consider both kind of a commercial and regulatory success?

Yes. I mean, from my perspective, I'm modest. I would love to continue with the same trajectory that we've been on now the past couple of years in the United States. I should hope that this data would help us to change the trajectory in Europe. It's a bit early to know that in Germany, right, to get a similar trajectory. That's a hope. But if we stay on that same trajectory that has been very aggressive. And as we grow, it become harder over time, right, to stay at that growth rate. And if we stay at that, I would consider this a success. So I don't know if I answered your question here.

No, that helps. And so you brought up Europe, but I guess 6 months from now, you're pretty certain that you're going to have filed for approval at that point, right? Waiting to hear back from the FDA, right? And so 12 months from now, you should get a good sense of whether it's an indication and expansion, et cetera. So that's, I guess, what I'm looking for.

Yes. I'm sorry, I misunderstood your question. So Margaret, from that regard, it's just one word. If we can get that treatment worth in our labeling that this Barostim device treat heart failure. That for me is success. I don't know if I'm the only one who sees this as a big deal, but it is a big deal. Now do we need this? No. we've been doing very well without it, the long-term symptomatic endpoint and the fact that now we have long-term symptomatic improvement in a safe way that does not shorten the life of patients, and we have the proof for that, will be helpful and will help us keep on this growth trajectory, but having that treatment will be really nice.

There are no further questions in the queue. I'd like to hand the call back to management for closing remarks.

Yes. It seems that Jared got it easy this time. I had to answer all of the questions. Well, with this, thank you, operator. And I'd like to thank everybody here for joining us today for our update call. Also I'd like to extend my sincere thanks and gratitude here to all of the patients, physicians, investigators, nurses, regulators who help us with this trial. Now that chapter is the first part of it is closed. We still have a lot of work to do to get the paper published and everything else. And as soon as those will be available, we would be very glad to share this with you. And now with this, we look forward to updating you on our progress during our next quarterly earnings call. Thanks, everyone.

Ladies and gentlemen, this does conclude today's teleconference. Thank you for your participation. You may disconnect your lines at this time, and have a wonderful day.