Artivion, Inc. (AORT) Earnings Call Transcript & Summary

Greetings, and welcome to the CryoLife PROACT Xa Clinical Trial Conference Call. [Operator Instructions] Please note that this conference is being recorded. I will now turn the conference over to our host, Pat Mackin, President and Chief Executive Officer. Thank you. You may begin.

Thank you, and good afternoon, everyone. Welcome to the CryoLife investor call to discuss our PROACT Xa clinical trial. I'm just going to throw up our forward-looking statement briefly so you can look at that. So during this 1-hour call, I'm going to kick it off with a few opening slides to put the trial into context for our overall business. Then I'm going to turn the call -- the majority of this call is going to be Dr. John Alexander presenting, and then we'll have time at the end for Q&A. So just a few slides on CryoLife. So when I joined about 6 years ago, we had a couple main product lines, BioGlue and our homograft tissue valves and vascular tissue. These products were primarily focused on the treatment of the aorta. And really we set a vision at that time to become a recognized leader in providing technologies for patients with aortic disease. And you can see the kind of the picture there, if you're on the slide show. And what this set off is we formulated our strategy. And as you can see, if you're on the Webex, we've been executing against this for the past 5 years. We've completed 4 transactions, and as a result, have assembled one of the most advanced portfolios in the world to treat aortic disease. In addition, we've got a very robust pipeline, which will help fuel our growth into the future. So here are just a few high points on the strategy because this will feed into the discussion about the PROACT Xa trial. We are focused on delivering double-digit growth once this pandemic subsides. We're focused on cardiac and vascular surgeons. We focus on globe, the full globe and expanding internationally. And I think one of the points you'll see here is, at the end, we're focused on solving our customers' biggest clinical challenges by acquiring or developing simple and elegant solutions. And I think you'll hear in the call today from Dr. Alexander, this is a big clinical challenge for our heart surgeons and cardiologists out there, and that's managing the anticoagulation of patients with mechanical aortic valves. So this is the pipeline that we've assembled, and I don't think many people really fully comprehend the depth and breadth of this pipeline. We've got 13 products, all to treat the aorta. In addition, through these 4 acquisitions that we've completed over the past 5 years, we've actually increased our total addressable market from about $600 million in 2015 to over $6 billion in 2020. And again, as you can see, we've got everything from heart valves to sealants to stent grafts to cover from the aortic valve all the way down to the iliac artery. The second byproduct of those 4 acquisitions is a very robust product pipeline. We have 12 projects overall, 7 of them are PMAs. All of them, but ProAct Mitral, are focused on the aorta. I think another interesting point for investors, our anticipated gross margin for all of these products in our pipeline are in the 80% to 90% gross margin range. So today's call, we're going to be focusing on the PROACT Xa trial. And I thought I would just set this up briefly for Dr. Alexander. As you think about putting this trial into context, I thought it would be helpful to have some backdrop from the market. So this slide really seeks to give a comprehensive overview of the valve market. I picked the U.S., but you could apply this around the world. So 40% of the valve market is in patients under 70 years of age. And if you look at this slide, let me just quickly walk through the kind of the left side of the slide, those are the 4 different valve options that you could receive if you needed an aortic valve. Across the top, that first line is the percentage of people who get an aortic valve by age. So 13% of aortic valve patients are under 58, 17% are over 85. If you look at the circles on the lines in the graph or in the document, that's the mean age of a patient of when they see one of the 4 technologies. So for example, a pulmonary homograft for an aortic valve operation, the mean age is 38. The mean age for a mechanical On-X valve is 56. The mean age for a bioprosthetic valve is 68. And the mean age for a TAVR, and that was in our PARTNER III trial, is 73. And these arrows basically show you kind of the age ranges that each one of these technologies would be used in. So age is by the way a proxy of which valve you're going to get. So you can see that you typically don't see someone getting a Ross pulmonary valve operation if they're over 50. Similarly, you don't typically see people getting a mechanical valve over 75. And really what this trial is about is the trade-off between a mechanical valve, you see there, the On-X valve, and a bioprosthetic valve is -- in one sense, the mechanical On-X valve gives you the ultimate durability, one operation and great durability. But the side effect is you've got to take an anticoagulant for the rest of your life. Whereas a bioprosthetic valve, it's pitched to the patient that you don't have to take an anticoagulant with the valve, but the problem is the durability of the valve and how long it lasts. So therein lies the trade-off. And the circled area is really what we think is the sweet spot for PROACT Xa. We think that the average age of an On-X valve after -- if PROACT Xa were to get approved, would go from 56 up to 66. So there's about a 10-year window where we would eat into bioprosthetic valves because you could get a one operation and a very mild anticoagulant. And that's really the hypothesis around the PROACT Xa clinical trial. So now let me introduce Dr. John Alexander. As you're going to hear in his conversation, we've worked with him and his team for about 2 years to design, get approval and execute this trial. And he currently serves as one of the co-PIs. He's a cardiologist and Professor of Medicine at Duke in the Division of Cardiology. He's also the Director of Cardiovascular Research at the Duke Clinical Research Institute, where he oversees a broad portfolio of cardiovascular clinical trials. He's published extensively. He serves as a PI on numerous multi-center clinical trials and survey, including ARISTOTLE, which you'll hear about in this presentation, which investigated use of apixaban, also known as Eliquis, to prevent stroke in patients with atrial fibrillation. So with that brief introduction, it's now my pleasure to turn the call over to Dr. John Alexander.

Thank you, Pat, and it's really my pleasure to be here today and talk with you a little bit about the background behind PROACT Xa and the trial design. The PROACT Xa, as Pat alluded to, is really -- we spent about 2, 2.5 years on the design of the trial, getting it right. And that was done in collaboration with a steering committee, a broad group of clinical trialists and surgeon investigators and with a lot of input of the U.S. Food and Drug Administration, so both the devices and drugs groups at the FDA. So we're confident that this trial's going to answer an important question. These are my disclosures, and they're publicly available on the DCRI website as well. So I just wanted to talk a little bit about drug-device combinations, specifically related to aortic valve replacement and antithrombotic therapy. And as Pat alluded to already, there are really 3 options for aortic valve replacement. I left out the pulmonary autograft and Ross from this slide. And those are mechanical valves, which require an operation to put them in, but as Pat alluded to, really last forever. They require combination aspirin and a vitamin K antagonist, most commonly warfarin or Coumadin, lifelong or a bioprosthetic valve, which also requires an operation to put it in and require aspirin, a simple once-a-day pill, plus/minus a vitamin K antagonist for some period of time but not lifelong. And these valves don't last forever. And then finally, transcatheter aortic valves, which have -- TAVR, which has really come about in the last 8 to 10 years and has really revolutionized the aortic valve market and practice and can be put in without an operation with a catheter, and we're still sorting out the anticoagulation antithrombotic therapy but are typically treated with aspirin plus clopidogrel. So to get approval of a new device by CDRH and the FDA, you need to have a reasonable assurance of safety and effectiveness. And they have -- that's the criteria for getting a new valve approved. And prosthetic valves are developed with some antithrombotic regimen. And this antithrombotic regimen is arrived at either by guess or extrapolation from other clinical settings or based on historical precedent. And then the valve and antithrombotic regimen, in combination, have to have a reasonable assurance of safety and effectiveness. And it's really important to recognize that there's typically no systematic evaluation nor any regulatory requirement for an optimal antithrombotic regimen to optimize safety and effectiveness for device approval. And then also really important is to recognize that antithrombotic options are a major factor in surgeon and patient valve choice. And that's, as Pat alluded to, really most true with mechanical prosthetic valves, which, as I said, really last forever, last for life of the patient, but require warfarin anticoagulation with lots of the food and drug interactions and regular monitoring for the life of the valve. And so these factors have led to a shift in the distribution of valve types over time. And this is data that's somewhat dated now. It's only through 2015, and really importantly, it doesn't include TAVR. But you can see that for patients aged 50 to 65, there's been a gradual shift away from mechanical valves toward tissue valves, again, all surgically implanted, over the last decade. And this is mostly due to concerns about the requirements for oral anticoagulation with warfarin with mechanical valves. Now we all, for those of us who practice medicine, live in a world of practice guidelines, and these practice guidelines make recommendations for antithrombotic therapy based on levels of evidence. And these are the antithrombotic guideline recommendations for patients with bioprosthetic aortic heart valves. And you can see in that circled red column, the levels of evidence are not great. So we have A, B and C. And A is data from multiple well-designed, randomized, controlled trials. There's none of that. B is data from multiple observational studies, although -- but nonrandomized. And the level of evidence C is expert opinion. And so you can see for TAVR, clopidogrel is recommended for at least 6 months on top of lifelong aspirin and -- based on expert opinion and then anticoagulation with a vitamin K antagonist to achieve an INR between -- of 2.5 may be reasonable for at least 3 months. With a bioprosthetic surgical aortic valve, aspirin is recommended, level of evidence B, and anticoagulation for a period of 3 months with warfarin is maybe reasonable, again, based on nonrandomized data. And in TAVR, there are clearly still questions remaining about what the optimal antithrombotic regimen is. These are the results of the GALILEO study, which randomized patients undergoing TAVR aortic valve replacement to either a rivaroxaban, that's a factor X inhibitor like apixaban, regimen or dual-antiplatelet therapy. And while the rivaroxaban regimen did reduce subclinical leaflet thrombosis, there's thickening of the TAVR valves that sometimes causes acute problems and may lead to earlier valve degeneration. It didn't result in an improvement in clinical outcomes. And actually the rivaroxaban group had worse clinical outcomes than the antiplatelet group. So this is with TAVR. And we obviously still have work to do to determine what the optimal antithrombotic regimen is here. And then for mechanical aortic valves, the mainstay of antithrombotic therapy is anticoagulation lifelong with warfarin. So for people with a mechanical valve anticoagulation with a vitamin K antagonist and regular INR monitoring is recommended. For people with an older valve, a higher level of anticoagulation, up to 3.0, is recommended. For people with an On-X valve, in the aortic position and no thromboembolic risk factors, a lower INR of between 1.5 and 2 may be reasonable, and I'll show you some data from the PROACT trial that led to this recommendation. And then background aspirin is recommended. And then very importantly, specifically related to the new oral anticoagulants, anticoagulant therapy with an oral direct thrombin inhibitor or a factor X inhibitor should not be used in patients with a mechanical valve. And this is based on 1 trial with dabigatran, a direct thrombin inhibitor specifically, that I'll show you some more data on as well. Now all of these guidelines are making the assumption that these classes -- that these groups of valves, that mechanical valves are all similar, that bioprosthetic aortic valves are all similar and that TAVR valves are all similar and should be managed with similar anticoagulant regimens. And there's good reason to think that these valves may not all be the same. Specifically for the purposes of this discussion, the On-X mechanical aortic valve may be different. So the leaflets are made of pyrolytic carbon deposited on a graphite surface that may be less thrombogenic than other valves. The orifice inflow area has a flared inlet. The leaflets form a nominal 90-degree relative to the orifice plane, again, maybe leading to less turbulence. And then there are some studies suggesting that the On-X valve can be managed with lower levels of oral anticoagulation that I'm going to show you in just a second. So previously, CryoLife and others conducted the PROACT trial. This was really a multi-arm study looking at patients with low-risk AVR, another cohort of patients with high risk -- a higher-risk AVR and then a cohort of patients with a mitral valve replacement. For the low-risk AVR cohort, they randomized patients to either warfarin with an INR of 2 to 3, or clopidogrel plus aspirin as the antiplatelet regimen. For the higher-risk AVR, they randomized patients to warfarin with an INR of 2 to 3, plus aspirin or warfarin with a lower INR, 1.5 to 2.0, plus aspirin. And then -- and I'm going to show you the results of both of those studies -- or those arms. And then the mitral arm is still ongoing. They've enrolled 408 patients randomized to warfarin 2 to 2.5 or warfarin 2.5 to 3.5, both plus aspirin. Those results aren't yet available. So here are the results of the PROACT trial, looking on the left at the 2 different target INRs; and on the right, the INR -- warfarin with an INR of 2 to 3 versus clopidogrel. And on the left side of this slide, you can see that warfarin with an INR of 1.5 to 2 led to a substantial reduction in the primary composite outcome that included both thromboembolic events and bleeding compared to warfarin with an INR of 2 to 3. The thromboembolic events were statistically equal between the groups with just a few more events in the lower-INR group than the higher-INR group, but again not a statistically significant difference. And the major advantage of the lower INR was this dramatic reduction, more than a 50% reduction, in bleeding compared to patients managed with a higher INR. So these results created that guideline recommendation and led to the change in the labeling of the On-X valve to allow this lower target INR. The low-risk AVR cohort that randomized standard warfarin with an INR 2 to 3 or clopidogrel had worse outcomes with clopidogrel than with standard warfarin. And this was driven primarily by an increase in thromboembolic events with clopidogrel without warfarin. But there was also a numerically higher rate of bleeding. And so this just shows you that in patients with an On-X valve who aren't getting an oral anticoagulant they -- but are getting potent antiplatelet therapy, they do have thromboembolic events. So there's obviously some thromboembolic risk with the On-X valve that warfarin is effectively preventing. Now in the last 10 years, there have been 4 new oral anticoagulants that have entered the U.S. and global marketplace. And one of these, dabigatran, is a factor II or thrombin inhibitor. The other 3, rivaroxaban, apixaban and edoxaban, are direct selective factor X inhibitors. I'm going to be talking mostly in this and in my discussion of PROACT Xa about apixaban. But I just wanted to give you a sort of lay of the land. All these drugs have around a 12-hour half-life. There's 2 doses of dabigatran and edoxaban that were studied and really 1 dose of rivaroxaban and apixaban that have been studied with some dose adjustments. These drugs have been studied in very large numbers of patients, mostly with atrial fibrillation, ranging from 14,000 patients up to 21,000 patients. And in ARISTOTLE, which I led with apixaban, there were 18,201 patients with atrial fibrillation. They were -- all these trials were designed as noninferiority trials. They all included patients with atrial fibrillation with various risk of stroke, and they all compared these new drugs to warfarin with an INR between 2 and 3, just the same warfarin INR range that's used for mechanical prosthetic valves. And then on the bottom part of this slide, I have the results, and I won't go through all of them. All these drugs are noninferior to warfarin for stroke prevention, and 2 of them, high-dose dabigatran at 150 milligrams twice a day and apixaban, are superior to warfarin for stroke prevention. Apixaban resulted in a 21% relative risk reduction in stroke. And then their bleeding profiles are quite different and are the biggest driver of differences between these drugs. And apixaban resulted in a 31% reduction in bleeding compared to warfarin. And it's really important that it's not like apixaban doesn't cause bleeding. It causes bleeding, but it causes a lot less bleeding, about 1/3 less bleeding than warfarin does. So the reason that mechanical valves have a prohibition against using the new oral anticoagulants currently is because of the RE-ALIGN trial. And the RE-ALIGN trial took patients with mechanical prosthetic valves, any valve type, and actually they were mostly St. Jude valves, and they could be either in the mitral or aortic position, and they randomized them to either -- to one of 3 doses of dabigatran or warfarin. Now there were 2 cohorts of patients: cohort A was a population that was enrolled almost immediately after surgery; and cohort B was a population that was enrolled later, at least 3 months after surgery. This was a pharmacokinetic study. So it was a short-term, 12-week outcome drug -- dose-ranging study to determine the ideal dose to take forward into a larger Phase III trial. Now the RE-ALIGN trial was stopped early after only 252 patients were enrolled because there was an excess of both bleeding, mostly pericardial bleeding, in those early enrolled patients with dabigatran, and an excess of stroke in dabigatran patients, mostly related to thrombi forming on the valve annulus, again, mostly in the early implants after surgery. And so we learned some important lessons from RE-ALIGN. First is that patients with mechanical valves are different from those with atrial fibrillation and venous thromboembolism, mostly because the valve is different, the people aren't so different, but the valve exists and needs to be anticoagulated. Early enrollment post-op may have impacted bleeding, particularly pericardial bleeding, and possibly valve thrombosis and stroke. Mechanical valves may not all be the same. The -- for the purposes of this discussion, the On-X valve may be different from the St. Jude valve and AVRs maybe different from lower-flow, lower-pressure mitral valve replacements. Dabigatran may not be the best new oral anticoagulant for valve thromboembolism prophylaxis. It's a factor II inhibitor, and so [ leaves ] thrombin generation going and just prevents thrombin activity whereas the factor X inhibitors that prevent thrombin generation may be better, and it has relatively poor bioavailability. And then really importantly, preclinical studies were not predictive of drug safety and efficacy in RE-ALIGN. The animal model work they did looked favorable even though the clinical outcome did not. Now apixaban. So apixaban is a direct selective factor X inhibitor. And it has also been studied in animal models of mechanical prosthetic valves. And apixaban looks like it inhibits valve clot formation in this heterotopic valve model similar to how warfarin does. And so this preclinical data suggests that there's reason to think that apixaban may be effective as an anticoagulant for mechanical prosthetic valves. This is with a St. Jude valve. And dabigatran work also done in this model showed similar benefits, even though the clinical trial ended up not demonstrating that dabigatran was an adequate anticoagulant. So this is my one slide about the design of PROACT Xa. It's a simple, pragmatic trial that's answering a clinically important question. We're taking patients with the On-X aortic valve replacement at least 3 months prior. So one important lesson we learned from RE-ALIGN is that, that early period may have increased risk because of lack of endothelialization of the valve annulus. We're planning on enrolling 1,000 patients and randomizing them to apixaban 5 milligrams twice a day, which is the same dose that was used in ARISTOTLE, with a dose reduction in selected patients, that will be very rare in people with an On-X mechanical valve, or continued warfarin with a goal INR between 2 and 3. It's going to be open label because of the challenges of blinding warfarin management. And then we're going to have 2-year follow-up of all patients, resulting in 800 patient years of follow-up in each arm. Our primary endpoint is the composite of valve thrombosis or valve-related thromboembolism. And then we'll have a secondary endpoint that includes the components of the primary and major bleeding. And after a lot of discussion with the FDA, we're going to have 2 co-primary analyses that, one, is that apixaban will be noninferior to warfarin with an absolute noninferiority margin of 1.7% per patient year, that is, we'll exclude with certainty what we expect to be about a doubling of thromboembolic event; and that the apixaban primary outcome, 95% confidence at interval, will be below the objective performance criteria of 3.4% per year. And these 2 co-primary analyses are almost certainly going to be aligned. So we've really just started PROACT Xa, and it's a little bit of a -- all clinical trials are a little bit of a challenge to run in the COVID pandemic. PROACT Xa is actually well suited to run in a pandemic given that most of the trial can be done remotely without patients having to come to a participating center. We enrolled the first patient on May 7 at St. Vincent's Hospital in Little Rock, Arkansas. And this is a picture of Dr. Rao and the team there who enrolled our first patient. We currently have 39 U.S. centers that are open. 21 of those are enrolling. Most of the delta there is because the other centers were opened recently. We're planning on having about 60 centers. Fortunate for us, we know the centers that have put in a lot of On-X valves and so know the centers to go to that have and are following these patients. As of today, we have 81 participants enrolled. The number on the slide is a little bit dated. And on average, these patients are 22 months out from AVR. About 1/3 of them are 3 to 12 months post AVR; and 2/3, more than 12 months post AVR. So we're not enrolling patients who are having new AVRs put in. These patients are out there walking around, and we just have to contact them and convince them to participate in the trial. As of late November, we had 10.9 patient years of follow-up out of the 1,600 or 800 per arm that we're planning. And we had our first -- we have a great data safety monitoring board that's looking at open-label data during the trial. They met on November 20 just to review very early data, and we'll be meeting regularly during the trial reviewing the safety of the participants enrolled and letting us know if there are any safety concerns. So that's just the high-level overview of sort of the background behind the design and status of PROACT Xa. And now I'll turn things back over to Pat Mackin to talk a little bit about time lines, and then we'll look forward to taking some questions. Pat?

Yes. Thanks, Dr. Alexander. And this is our last slide on the presentation -- oh, went the wrong way -- yes, here's the last slide. So we just wanted to give you a quick update on kind of the timing of the projects. We get this question a lot, just because of how long it's going to take to enroll. So I would say that, clearly, the pandemic, as Dr. Alexander alluded to, hasn't helped really more with kind of just opening centers, and a lot of these hospitals have been preoccupied doing other things. The nice thing about PROACT Xa is it's certainly a trial that was designed for -- to be implemented during a pandemic because the first patient was enrolled online, and we can do this all kind of remotely. We also -- if you look at the enrollment, we're projecting that we'll -- this trial will be enrolled in by the second quarter of 2022. We probably took about a 6-month hit, both from just the pandemic logistically rolling out centers as well as on our earnings calls we talked about backing down some of the funding just because of the uncertainty around the pandemic and protecting our cash situation. So we're expecting to spend pretty significantly next year, but it has slowed us down a little bit. You could see if we end in Q2 -- end enrollment in Q2 of '22, as Dr. Alexander said, it's a 2-year follow-up, we'll then package up the data in the next quarter, submit it to the FDA. So we should have received an approval kind of at the end of '24, beginning of '25, depending on how fast we can enroll. So with that, let's open up the line for questions with the operator.

[Operator Instructions] Our first question comes from Mike Matson with Needham & Company.

Question for Dr. Alexander. So I guess, first, what do you think the likelihood of success of PROACT Xa is? And then how would it affect your practice if it is ultimately successful? How much would it increase your use of the On-X valve? And then do you think just the fact that the trial is being run in itself is enough to get doctors increase their use of the On-X valve just because they know that the patients may have an option to switch to Eliquis within a few years?

Yes. So good question. So regarding the likelihood of success, we don't know. I mean that's why we do these clinical trials. There's -- I obviously believe there's enough chance of success to invest a lot of my time and energy and investment in this. But ultimately we don't know for certain, which is why we do these experiments. The one important thing about PROACT Xa is that events are going to be accumulating over time. Our DSMB is looking at the data. We're trying to show that apixaban is at least as good as warfarin. And so the only way the DSMB would stop the trial is if apixaban is turning out to be worse than warfarin. And so we -- the longer the trial goes on without it being stopped, the higher the likelihood of success at the end of the day. Regarding the impact of the trial on my practice, I do think that having a mechanical valve that one can take with a drug that doesn't require any monitoring would be a substantial advantage. And it would move a fair number of people from bioprosthetic surgical valves to mechanical valves, specifically the On-X valve. TAVR is still a large issue and is expanding into lower-risk patients, but some substantial proportion of patients would move from a bioprosthetic -- especially the younger ones, the area that Pat alluded to, would move from a bioprosthetic surgical valve to a mechanical valve. And then the question about whether people are going to be putting more On-X valves than now. I do think there'll be some of that for 2 reasons. I mean, actually, we've talked to some site investigators who are excited about using the On-X valve in order to be able to enroll people in the trial. And so 3 months after -- if you get your valve put in today, 3 months from now, you have a 50% chance of being randomized to apixaban and 50% chance of needing to stay on warfarin. And that's been -- some of our site investigators have told us that, that's a reason why they have interest in implanting the On-X valve. I have not heard a lot of what you're describing, which is we'll put in an On-X valve now and because in 2025 we'll know the results of the trial. But there may be some of that as well.

Okay. And then just a question for Pat. So this isn't on PROACT Xa really specifically, just a higher-level question, but I noticed in the slide, at the beginning, you specifically said you're calling for double-digit growth from 2021 to 2023. I think this is the first time I've heard you give that kind of guidance for that time frame. I know you've talked about double-digit growth in 2021 specifically, but can you just comment on that? And is this something new? And what gives you the confidence and the ability to give those projections?

Yes. My caveat was post-pandemic. We're all dealing with the same kind of continued kind of headwind with the pandemic. And although our business has done quite well in the pandemic, we're pretty confident once this thing lifts that we can get to double-digit growth. And it goes back to the portfolio that we've assembled, right? So if you look -- I don't know if you're online, Mike, but if you look at the portfolio that we've assembled, we acquired the On-X franchise almost 5 years ago. And you just heard about the PROACT Xa trial. We're also selling the benefits of the PROACT valve today. We're the only company out there that has FDA approval for a low INR. We then acquired JOTEC. And we're launching some of the new products like you see here, the E-nside device, the E-vita OPEN NEO, both in Europe as well as in select markets in Asia and Latin America. We then acquired -- or did a distribution agreement with NEXUS through Endospan, and that product is out. And again, it's been challenged by the pandemic because of the training requirements. And then we most recently acquired AMDS. So we've got this very powerful portfolio while we're expanding internationally. And we think once this pandemic kind of lifts, we're locking down for double-digit growth.

Our next question comes from Suraj Kalia with Oppenheimer & Company.

Can you hear me all right?

We can hear you fine.

Perfect. So Pat, first question for you, then a bunch of questions for Dr. Alexander. I saw in one of the slides, Pat, it said greater than equal to 800 patients in each arm. Has the trial size changed? I was always under the impression it was around 1,000 or 1,200.

I think what you said, it's this slide here. It's 800 patient years.

Sorry, Pat, you just -- I missed out a little bit. I couldn't hear you.

Yes. So I don't know if you're on the Webex. So the end, there's 1,000 patients in the trial, 500 each arm. But the 2-year follow-up gives you greater than 800 patient years in each arm.

Okay. Never mind, my mistake.

That's an FDA kind of requirement kind of that greater than 800 patient years.

So that's what it is. Okay. Fair enough. Dr. Alexander, obviously there have been some trials in dabigatran and other NOACs. I guess when we look at apixaban, right, how do you think about risk severity of patients? And a subset of that is, how do you see the performance, especially when you consider hemodynamics and the valve size? And for the second part, really, the reason I ask is, if you look at the original PROACT trial, right, there were differences in the valve sizing, and they had -- I don't have it right in front of me, but there was differences in outcomes. How do you think through severity and the sizing of the valve?

So good questions. We have included in PROACT Xa all sizes of the On-X valve in the aortic position, and in addition, the aortic conduit with the valve implanted in it. So we're enrolling a broad population of patients in terms of valve size. And in terms of various other -- certainly in other risk characteristics and in terms of time from surgery, after 3 months. And so we're assuming that apixaban, if it works, will work across the spectrum of valve sizes and across the various risk profiles of patients. We've seen that in other populations that these NOACs are equally effective across various risk profiles, obviously not with mechanical valves. And so when I think about the risk-benefit ratio here, I think the risk is almost all related to the valve and whether a selective direct factor X inhibitor is going to be adequate to prevent clots on a valve, on the artificial surface of a valve. The bleeding, on the other hand, I think is, valves don't bleed, patients bleed. And it's going to be -- I would be shocked if the bleeding is not more favorable with apixaban than warfarin. But the big unanswered question has to do with thromboembolic risk.

Right, right. Well, to that effect, Dr. Alexander, how would -- how is compliance in each arm, for Coumadin versus Eliquis, because in the original PROACT trial you superimpose the bell curves for INRs. Between the control arm and the treatment arm, they were skewed a little bit. So as you think through this, because there isn't one -- the median isn't like, I shouldn't say median, but there isn't like one INR for the entire treatment. It's a curve, right? How do you think through that in terms of outcomes for primary and secondary endpoint?

Yes, yes. So warfarin, as you know, has a very long pharmacodynamic half-life. So missing a day or something of warfarin doesn't matter as much. Apixaban has a 12-hour half-life. So missing doses will be much more of an issue with apixaban than with warfarin in PROACT. The good news is that we're enrolling patients who have been on warfarin for some period, and we know what their level of compliance with warfarin has been. I mean these are all patients who've had a mechanical valve for at least 3 months. And so we're enrolling -- first of all, patients who get mechanical valves are a relatively compliant population to start with and then we're enrolling them 3 months out after we know what -- something about their INR management. And so are focusing on enrolling a more compliant patient population and then reinforcing with them the need to be compliant with both apixaban and warfarin and having frequent follow-up with patients via telephone and other mechanisms.

Got it. And finally, Dr. Alexander, one last question and I'll hop back in queue. So let's say Suraj comes in, right? He got an On-X valve for whatever basic condition, right? He got an AVR, fine. Let's forget the severity for now. X number of months later he gets an Eliquis. As you think through it, what does your experience say -- you know what, within, let's say the first 3 months of crossover, that's really when we should start seeing an issue. Based on my experience, there is going to be some thromboembolic event or should be. How do you think through -- if I cross this period, Suraj is through.

Yes. It's a good question. The upfront risk that happens, there's certainly a period of higher risk right after surgery. We're going to avoid that largely because we're -- there's -- we have the 3-month window until people are eligible. After that, with warfarin, there's a pretty linear rate of thromboembolic events. It's low. It's about 1.7% per year. But it's pretty constant over time. I don't have any reason to think it will be different with apixaban, but that's why we will be carefully monitoring people and our DSMB will be carefully monitoring our outcome data by treatment.

Our next question comes from Jeffrey Cohen with Ladenburg Thalmann.

So a couple of questions from us here. Firstly, are there [ set looks ] from the DSMB at a certain number of patient years or a certain number of months? And will that be disclosed to you and/or will be disclosed to the public?

So what's written into the protocol and our DSMB charter is that the DSMB will meet at least every -- at least annually, so -- but they're -- but we have built in a much more frequent monthly look that the DSMB chair is getting. We won't -- we will get reports from the what are likely to be Q6-month-or-so meetings. And those are sent to all investigators. They're -- I don't -- they're not generally made public, if they continue as planned. Obviously, if they were to adversely affect the trial in some way that affected CryoLife, they would likely make that public. So the DSMB is looking frequently at the data.

Okay. And would we expect then, if the curves diverge to the advantage of On-X and apixaban so dramatically, will we see that -- could we see that the trial would halt, Pat? Or would they continue either way it goes to get through 2 years and the patient years?

So it's a statistical question. So this is a noninferiority trial, which means it's really -- think of it as an area under the curve. So long as -- and it's a lot along the lines of what Dr. Alexander has been saying, so long as the Eliquis arm doesn't cross over kind of the Coumadin upper level and stays under that margin, then as long as the trial keeps enrolling, then it's a success until it finishes. There's not a -- it's not like you power an efficacy trial. And if you really beat it, you can stop early. So maybe Dr. Alexander, you can talk. I mean we're not going to have -- we're not going to get any updates about the trial along the way. There are no interim analysis. It's statistically powered to the 2-year -- the last patient gets to 2 years. And so long as you're under that margin, then we should be fine. Maybe Dr. Alexander, you can comment on that.

No, that's correct, Pat. I mean, I think there's -- basically, for largely mathematical reasons, there's almost no way the trial can get stopped early for a win for apixaban. And that's largely driven by these very low -- I mean, the event rate is 1.7% per year. And so we're talking about very small numbers of events. And you need to accumulate enough events to be confident that apixaban is not worse than warfarin. And so unless the event rate is much higher than we projected in both arms, there's really no way the trial will end early because apixaban is better. And so really the only way the trial will end early is if apixaban is worse.

Yes. Okay. Got it. And one more, if I may. Could you talk about -- you touched upon a little bit about just kind of general compliance amongst a large amount of patient population and the ramifications from 1 times or 2 times daily from medication, and if you're aware of any other NOACs working on longer delivery times -- times between delivery's half-lives?

Yes. So I'm -- all the NOACs have about a 12-hour half-life. If you go back to my -- there's a slide here about the various NOACs. They all have about a 12-hour half-life. And 2 of them are dosed BID and 2 of them are dosed Q-day. And all that means is that there's a bigger peak-to-trough ratio with rivaroxaban and -- with rivaroxaban, which is -- and edoxaban, which are once-a-day and a smaller peak to trough with dabigatran and apixaban. The -- obviously, as I mentioned, compliance is important. I mean if patients stop taking apixaban, the anticoagulant effect will go away. We don't know exactly how much that matters. I mean we don't know if being not anticoagulated for 12 hours, for example, is a problem or not. All of this is on a back -- all of this is based on data with warfarin, which has a very long half-life. There -- in some ways, the short half-lives of these drugs, the relatively short half-lives, are advantageous because when people are bleeding, they go away or when you need to stop them for procedures, they go away. So that's what we -- that's the extent of what we know.

There are no further questions at this time. I'll turn it back to Mr. Mackin for closing remarks.

Well, I appreciate everybody joining us, and I want to -- special thanks to Dr. Alexander. Obviously, this is an exciting trial and one that we're obviously investing significantly in. We spent a lot of time with a lot of smart people arriving at the trial. So I mean things are up and running. We're enrolling now. As you said, the patients are being added, the sites are being added. And as the pandemic starts to lift, I think we'll start to see really strong enrollment next year. So we're excited about it and obviously look forward to sharing other technologies in future investor calls. So thank you for joining today.

Thank you. This concludes today's conference. All parties may disconnect. Have a good day. Thank you.