LivaNova PLC (LIVN) Earnings Call Transcript & Summary

Good morning, and welcome to day 2 of the UBS Global Healthcare Conference. This is the medtech track, and I'm Matt Taylor, UBS' U.S. medical supplies and devices analyst. And for this session, we're going to be joined by management from LivaNova. Just a couple of words about the session. It's going to be about a 45-minute fireside chat, and I have a lot of questions ready and prepared. But if you want to send some in for us to weave into the conversation, you can e-mail me at [email protected] or use the webcast tool to submit a question, and we'll take that into the conversation. So first, I just want to welcome both Damien McDonald, the CEO of LivaNova; and Matt Dodds, who's Senior Vice President of Corporate Development. Thanks a lot, guys, for making the time and for being here today.

You're welcome, Matt. And thanks for hosting us and having us along via conference. Really appreciate the time, and good morning to everyone. I got up early in the East Coast, and greetings from a stormy England afternoon in -- just outside London.

Well, that's every day, isn't it? I mean...

It has been the last 3 weeks, that's for sure. So...

Got you. Well, good. So Damien, maybe you can kick us off, if you could just start off with some high-level comments about the business, how it's been performing and maybe just touch on some of the recent trends. And then we'll get more into some of the detail.

Yes. Thanks, Matt. Look, I think overall, we're very happy with how Q1 landed. Like a lot of people, I think both businesses are starting to show recovery. January, February were a little soft. And in March, we started to see a lot of good momentum, and we carried that into the second quarter. So I think we're quietly confident about how procedure volumes are evolving, and it's still early in the year. As everyone is aware, there's still a lot of pluses and minuses around the world with the COVID impact, but we're quietly confident about how we're seeing the procedure volumes evolve.

Well, that's good to hear. Yes, some of the business turnarounds in Q1 are really notable, and it sounds like things are continuing to perform well. So that's great. What about -- maybe you could take us through some of the recovery trends by business and geography. Talk a little bit about what you saw in the quarter. And if you can give any color on what you're continuing to see here, that will be really appreciated.

Yes, sure. So let's break it. We've got business first and then geography. So neuromodulation, again, our procedure volumes are recovering. Neuromodulation has been an interesting one because the in-office visits for epilepsy have lagged a lot of other procedures. And I think also given that we skew pediatric, getting a carer and a child into a clinic right now is not simple. So both NPI, so new product implants, and as well as end of service are both improving. But end of service, much quicker. And that's, we believe, because patients can't wait forever to get a replacement. So we've got a bit of a backlog of patients in the end of service that I think is giving us a tailwind. But both NPIs and end of service are improving well. The other thing for us, too, is we don't skew ACS, ambulatory surgery centers, quite so much. So also, we're reliant on the clinics being fully open for the implant. So for both those reasons, we're seeing slightly lagging compared to other procedure types. But overall for us, we're seeing a good response. Cardiovascular, again, also very much linked to procedure volumes. So where procedure volumes are coming back, and I'm sure all of you are tracking that, we're seeing cardiovascular bounce well. In other areas, we think that -- where you're still seeing hampered volumes like India, we're not seeing quite strong trends yet. Heart valves is a different beast. It's a little softer, and we think that's kind of related to the acquisition by Gyrus as we switched supply chains and moved distributors, and people are just waiting to see that transition. So heart valves is probably a little softer than historical, but overall, again, as I said, very positive in terms of momentum. Geographically, without a doubt, the U.S. is bouncing back much more rapidly. You talked just as we were opening about how masks are coming off, the guidelines are clearer. I think those sorts of things give people confidence, and also clinics able to open. Asia Pac, apart from India, is showing some really good growth, China, Australia. Where we're still seeing softness obviously is Europe and Latin America, and both of those are lagging. But we expect them -- as vaccination rates increase, as restrictions open up, we expect them to come back in the second half of the year.

Got it. Got it. Have you seen Europe start to improve at all? We had a couple of companies yesterday that saw -- so they saw some positive signs there. Just curious what the feedback from you was.

It's very jurisdictional. Some countries that have seemed to get this under control, that have had good vaccination rates relatively -- I mean everyone compared to the U.S., the vaccination rates in Europe have been lagging. But where we've seen stronger vaccination rates, it's starting to come back. But again, it's -- we'd almost have to go country by country. But the signs are positive but again lagging what we see in a few other geographies.

Got it. It's a long fireside chat but probably not that long to go through each one. Okay. So I wanted to ask about your guidance. I mean you had a pretty nice beat in Q1. Neuro is coming back, which is really key especially for margins, but you didn't raise yet. So maybe just talk about your philosophy behind guidance. Are you being conservative? And what are the assumptions that are baked into the current guidance?

Yes. I think you really hit that on the head there. We're still, as I said, seeing a lot of noise. And while I'm in a positive about the momentum in the U.S., the other markets that we just talked about make us be a little conservative. So we thought that the more prudent thing to do was to stick with what we guided with at the start of the year, and I think that's important. As we start to see and get more confidence globally, we'll try to give more clarity around that. But I think the way people should model it is that the first half will be lower than the second half. The second half, we're more positive about. But we see a strong momentum coming out of Q1 into Q2.

Got it. Okay, okay. And you talked about in neuromod, the backlog, I mean these can't be deferred forever, especially with the batteries and end of life. But you're also using a new strategy in epilepsy to really kind of shake the tree and raise awareness in some of these comprehensive centers. Can you talk about that and how it's starting to bear some fruit?

Sure. I think one of the really important things we learned from 2019, as we took a step back, was to really think about where the patients are aggregating. We've had a very community neurology-centric strategy with some people selling well and systematically in the CECs, but it wasn't really a focused approach. So even though 50% of our sales came from the CECs, it was one unit, one unit, one unit. We didn't have an engagement with them. And as I said, with the step back in 2019, we really understood that being able to service them and be partners with them, I think, importantly, was going to be a different part of the growth strategy. So in 2020, literally right into the headwind of COVID, we launched this market strategy. It's a high-access, high-touch model but nevertheless is really key, we think, to being able to help clinics bring patients through. What does it do? It has a key account manager that supports the account broadly; a therapy consultant who does the day-to-day activity, who partners with a medical science liaison, who's a PhD or MD in neurology typically; and a patient nurse educator who are all RNs with an education background. And that quartet have a part of 4 to 5 CEC accounts, very geographically located around a particular point so that we're not burning calories with a lot of windshield time in -- like our old strategy. So very concentrated effort, 4 to 5 accounts each. The idea is to partner with the clinicians, partner with the nurses and partner with the patients to move them through our whole value chain. It's early days, especially relatively early given the COVID impact. But what we're starting to see is, in terms of the lagging indicators of sales, we're seeing a separation in the performance of these accounts. And we have 9 pods set up now that have roughly 50 accounts, so these 9 pods are starting to show a separation from the baseline performance but also from the other nontarget CECs. So I think that's the first thing that we think is positive in terms of a lagging indicator. In terms of a leading indicator, we're seeing a lot more access, and we call them firsts. We monitor this weekly, and it's about access to the account. So a major Midwest account for the first time allowed us to train on VNS with their neurosurgery department, and we had a new implanter, 2 weeks later implant of VNS, the first time that, that patient -- that clinician had used a VNS. We've seen similar firsts in -- on the West Coast where for the first time ever, we've been able to train fellows and residents on VNS. That's important now but also as the fellows and residents spread out post-fellowship to other accounts. The fact that we've now got access to them is really key. So leading indicators are positive. Early signs on the lagging indicators of sales are starting to show and bear fruit. We're extending these pods. We'll put another 3 in this year. And then again depending on what we see, we have plans for adding more pods next year.

Got it. Sounds like that's -- the strategy is doing really well. Can you remind us what you said on the last call about how they're growing more than the baseline? And over time, do you expect the majority of your strategy to gravitate toward these kind of pods?

Yes. We again have a really important aspect of our business that's tied to the community neurologists, so we'll never sort of fully step away from that. But we are definitively putting more resources and rewiring our operating procedure towards the CEC thing. So we'll keep adding more pods. So far, what we've been doing is some resource allocation. We've stopped a lot of other activities. We removed a layer of the field force, which was roughly 40 people, and taken those 40 heads, and we're putting them towards this strategy. And we'll also continue to monitor how long it takes to ramp. Currently, it's about 6 months to get a pod to what we think is good productivity levels. As we learn more about the type of people we hire, we hope to improve that slightly as well. So the real anticipation is that more and more, this will become our core strategy. At the moment, it's roughly 10% of our sales, but we expect fully that this will become the majority of our sales in the next 2 to 3 years.

Got you. Got you. Maybe we could back up and talk about the broader epilepsy market. There's been a few new drugs and also device strategies that have gotten more attention lately. Have you been seeing any changes in the underlying dynamic? And are new drugs or other device options a big factor in terms of how you're projecting recovery for epilepsy?

The short version of the drug answer is no, we're really not seeing that much of an impact. I think we're in an extraordinary period of drug releases for epilepsy. Seeing 3, 4 or 5 in a 2- or 3-year period is a rapid release rate. So typically, we wouldn't see many drugs launched in a particular single period. We know that when a drug gets launched, we have a sort of a 4- to 6-month lag on patients moving through the supply chain. So that, for us, we've now taken into account, and that was baked into our guidance and our modeling for this year. But we haven't seen an ongoing impact, for example, from Epidiolex. We think whilst 2019, that hit us, and we've talked about that extensively in the past, we now have moved on for that, and I think we have a much better handle on, firstly, how to deal with it because of our awareness but, secondly, our messaging. We -- VNS is drugs plus VNS. You want -- no matter what drug cocktail you're on and you're going to be on a drug cocktail to have VNS, and our messaging has really evolved around that. And if you think about a lot of these drugs, there's been 17 drugs in 50 years, and they're really based on 3 mechanisms of action, and yet, we've still seen no change in the DRE numbers. And with some of the drugs talking about, for example, 20% seizure freedom, even if 20% of the population of DRE patients were seizure-free tomorrow, there'd still be 800,000 patients that are not controlled. And to put that in context, our business is built on roughly 4,000 new patients a year. So our opportunity, we believe, is leaning into this DRE population regardless of what drug cocktail people are on. In terms of devices, I think perhaps the one that's getting a lot of airplay at the moment is RNS from NeuroPace. Candidly, we see them as complementary inasmuch as we don't really overlap much in terms of indication. And I like the fact, candidly, that someone is speaking about nondrug therapeutic intervention in the space. We've been the largest, almost sole voice in this area. So I think a well-capitalized group that are not totally overlapping us is a good thing. And in fact, we see in accounts where we are both present that we grow typically faster than not. So I think this is a good thing to be raising and elevating nondrug intervention.

Understood.

Matt, would you like to say anything there? Matt, you have a particular passion about this, having engaged really deeply on that topic.

No, it's spot on for NeuroPace. I mean they're very focused on local focal lesions. We're more multifocal lesions, so minimal overlap. On the drug side, you're exactly right. Epidiolex sales have been relatively flat recently. For us, again, this uncharted period of remote telemedicine, you've probably seen some additional usage of drugs versus VNS. But again, like Damien said, they're not going to really treat the underlying condition and get you to seizure freedom or dramatic reduction in seizures. So we think of this more as a delay in new patients than a loss.

Got it. Thanks, Matt. So I think that's pretty good for our epilepsy discussion for now. Why don't we move on and talk about some other areas of the business? I wanted to ask about TandemLife, which I know is another area that Matt has looked at very deeply, but that's been a good acquisition, and it's growing really nicely. And you had a recent launch of LifeSPARC, which is helping that business to grow. So maybe just talk about the outlook for that business. What's been helping the growth thus far? And touch on the new product launch and how that's differentiated.

Yes. Look, I love this business, and the team that's there is desperately passionate and not for many more reasons but for the impact that it can have on a patient's life. I mean by the time you get to ECMO, these patients are critical and very near end of life, and the opportunity to potentially impact that is enormous. The company had a good product with the Escort system, but they were moving towards launching a new product. And what we found attractive was bringing that to market with this team. So the LifeSPARC launch has been growing very well. We're particularly excited because it's a significant step-up in the technology for us. The user interface is much more simple. The ability to set up and train the product is much more simple. The flow rate is better. The metrics and control of those is better. So all around, the device has an opportunity, I think, to fundamentally change how ECMO is conducted in clinics. We launched it in Q3 last year largely around focusing on account acquisition in the first instance, moving people from our legacy system to the -- to our LifeSPARC system; and then secondly, into ECMO accounts where we just weren't present. To put it in context, there used to be about 250 ECMO facilities. Now there are about 350. So ECMO is expanding as a therapeutic option, but we were only selling in about 130 of them and really 70 of them on a repeat basis. So we see being able to expand into the ECMO facilities as the first opportunity. Related to that, inside that clinic is utilization, and we have a real opportunity to use our device in many more procedures. And we've got a tailwind with COVID, unfortunately so, but nevertheless, we believe we're making a difference. And a large percentage of our disposable sales right now are coming from those COVID procedures. But more broadly, we expect to be able to sustain this by focusing on 4 things: cardiogenic shock, RV failure, respiratory failure and cardiac arrest. So ex COVID, we have 4 areas where we've got a large opportunity. And if you really think about the fact that ECMO is about 9% penetrated for all possible procedures, we think that there's a long tailwind for this area for us.

Got it. And just digging deeper there a little bit, what have you been able to do to convert accounts that are stuck in traditional methods like cutdown or other kinds of access? How are you getting them to convert?

Yes. I think again, coming back to the device itself, the readout is simple to analyze and interpret. The device is wickedly easy to train and to implant. And then the third thing is it has this ability to execute its task because the motor is so powerful, because the flow rates are so good that you can get good patient outcomes. So those 3 things have enabled us to tell a story inside the accounts about adoption. And I think coupled with that, we've done a very good job in terms of our training and facilitating of the application. So when accounts do come on board, we think we've done a good job in terms of training and education. The gating item for us is really people. As with the epilepsy thing, we want to train people in very defined geographies, so there's not a lot of windshield time being lost. These patients are critical, so we want to be able to have proximal support. So we're very focused around attracting the right talent and expanding the sales force to continue this growth. We're planning to put on 20 additional field people this year, where we've got about half of them in the field already, and they take about 6 to 9 months to ramp. So we're very focused on getting the right people in the right territories now so that we can continue this support. And we believe that at least 20% growth is possible this year and into the future.

On that additional 20 hires, can you just remind us how that sales force has expanded over time and what that incremental 20 represents as a percentage?

Yes. So when we bought the company in '18, they had 20-ish people. We expanded that to 50 in '19 and mostly through '20 given the issues with COVID and just launching the product. And so we're adding another 20 on that 50.

Got it. Okay. And on the HLM side, so I want to touch on cardiovascular as well. We talked a little bit about the trends there. I mean you're going to be launching a new product for the first time in a few years, a next-generation HLM.

Sure.

But let's talk about that and what kind of growth you might expect from it.

Yes. You're generous with the first time in a few years. First time in about 15 years. And look, I think this is what's exciting about this opportunity. I mean we literally had to reboot our product development capabilities in this area because it had just been so long since we've launched anything, and I'm really pleased with how the team has grown into that over the last few years. The opportunity we see is taking a market leadership position that we have with our S5 and launching a new technology in an area that is very reliant on an upgrade cycle. How do we know that an upgrade cycle exists? When I started working with this team in 2017, we identified that there was an S3 device, which was the 20-year old device that had not been upgraded. And there were thousands of them out in the field. So we set about a strategy about converting the S3s to the S5s. And I think if people look at that time period of '18, '19, where in some quarters, we were at mid- to high single-digit growth, it shows that there's a willingness of accounts to convert and upgrade. Importantly, through that period, we also taught the team a lot about funnel management and how to work capital devices through the system and track them and place them. And so is there an ability to convert and upgrade? I would again point people to that period. So we're going to be launching a significant improvement. We've kept base technology like the pumps, which the perfusionists love, and completely updated the software and the user interface. And all the ethnographic research we've done and the customer feedback that I've seen has been tremendously positive about getting it right. So to put it in perspective, in terms of opportunity, there are roughly 7,000 S5s around the world. About 40% of them are past what we would consider their useful life. Now if you care for these devices, they can last well. I mean it's almost like a vintage Mercedes. But nevertheless, it's time to start that upgrade cycle. So we're excited about bringing this to market in a rolling manner in 2022. We'll start with CE Marked countries first, so predominantly Europe, and then start into the U.S. later in '22 and then expand internationally as we get registrations. But very exciting for a market-leading product.

Understood. And I know you touched on this, the software, and I think there's some workflow benefits. But what are kind of the key differences between this and the older version? Could you take us into the hands of the perfusionist or into the procedure? What difference will it make for the hospital?

Yes. So I think firstly, if you look at the S5, each pump, each unit is controlled separately. The software is -- it's almost like running on a Pentium chip when we all should be running on the new ARM chips from Apple. It's that sort of difference in terms of the firmware, the hardware. The user interface is, again, very, very advanced. It looks like a large iPad. And for a perfusionist, what that means is they have much more precise control. They have a single point of control from this cockpit, and they're going to be able to modify their system and essentially have it set up for their preferences and on a repeatable basis. And we're very focused on this idea of goal-directed perfusion. If a perfusionist has a particular set of metrics that they are targeting, they're going to be able to do that in a much more precise manner.

I like the Pentium chip reference. I remember when that came out, that was a big deal. Oh, boy.

I was going to say 286, but I didn't think anyone would know what a 286 was apart from me. So...

All right. Well, cool. Well, we covered a lot of the base business, and I want to spend some time on the pipeline because that's gathering a lot more interest these days as some of these programs are starting to come towards big readouts. So maybe we could just touch on RECOVER first and talk a little bit about DTD and remind us what the next guideposts are for that program?

Yes. So DTD is exciting for us. The application of VNS for severe depression is increasingly well characterized. I point people back to the Aaronson paper late 2017 about what is possible in terms of response, in terms of remission and even before you get to the statistically meaningful changes there, the quality-of-life improvement. And that, I think, is the powerful aspect of what we're investing behind here, that we know what the real-world evidence looks like. And we've put together, I think, the single largest and longest well-constructed depression study. It's going to be first-class evidence that I think will support not only the U.S. reimbursement case that we're making but also support us well internationally. The milestone is getting either 250 unipolar or 150 bipolar patients implanted or both by the end of this year. And then the goalpost is to submit for assessment by the CMS at the end of '22 or early 2023 and at that point be able to move to a registry. We know the registry number. It's about 5,800. And again, to put it in context, the business for epilepsy is built on roughly 4,000 new patients a year. So we've got a lot of runway once we move to registry. And we believe once we're in the registry, even while we're negotiating, we'll be able to have that extended as a lot of other registry products have done over the last decade. And so that's what's exciting. We're very focused on site activation. We have more than 60 sites activated now out of a potential 80 to 100 where we've got a flexible site activation. That's been a little bit hampered by COVID, especially last year when we more or less turned off the whole study. But the activation of sites has continued well. We've continued to consent patients in, which is the last stage before they begin their run to implant. And the consents are sequentially improving as are now the implants. And we're really pleased with that run rate, too, and that -- seeing that improve now that the clinics are starting to open up again. So we are very excited about what this could mean for the patients and for the company.

Okay. So I have a few follow-ups there. So first one was, I know that COVID had been a little bit of an impediment to enrollment, but it sounds like getting things going again. How much of a drag is that now? And do you still expect to hit the enrollment targets by the end of the year?

Yes. So I mean it's still a drag in some geographies. I mean you guys know well that the U.S. is not unidimensional with its COVID response. So some clinics are implanting and recruiting faster. But again, we remain confident that we'll get to one or both of those milestones, the 250 or the 150 by the end of the year. Again, it's progressing well, and the consent rate is really the key focus area for us because once patients consent, it's just a matter of the clinic being able to pull them through baseline 1, baseline 2 and implant. And then we're very focused on making sure that the follow-ups occur well, the titration and the scoring. It's monthly scoring, so we're heavily focused on making sure that, that is followed well.

Got it. And you referenced the Aaronson paper, the registry and the difference that people saw there in their response, their remission and quality of life. Is that the right model to look at to think about what the difference could look like in these populations as well? Or maybe you could just describe what we might expect to see in this study.

Yes. Well, I think that's -- again, we're very focused on -- well, there's a whole bunch of metrics that we're focused on. There's a number of clinical questions that we're answering. But it's response -- time and response is the key aspect of the study. But remission, we believe, is important as is quality of life and suicidality. And in the paper, again steering the whole conversation with CMS opening up a noncoverage decision after a decade is the fact that we see all of those metrics improving. Response rate, 41% to 68%; remission, 26% to 43%; the quality of life, as I said, improving even before you get to a statistically significant change in response. All of those things are quite meaningful for these patient groups who candidly have no alternative. There aren't options after this. Often they've been treated with ECT. Often, they've tried TMS. They've had a cocktail of drugs for up to 10 or 12 years. This is really the last stop. Matt, do you want to talk about the time and response and how that scoring works?

Sure. So as Damien said, the Aaronson paper, the numbers quoted, those were the 5-year numbers. But you can see in the paper, there's -- he's got the actual trends at 6 months all the way through 5 years. So -- and this is in our dossier on RECOVER. What we're looking for in a year in response is a percentage for the turned-on implant in the low 30% range, about half that for the turned-off rate. That's what we get as statistical power. So -- and the way we're looking at it also is it's time and response. It's actually scored monthly. So we don't have to wait the full 12 months for each patient. If the patient hits that number enough, they would essentially hit that endpoint. So it's more Bayesian in design when we're looking at the total number of patients and the total months in response. And that's why there's not an exact time line for when we would hit the statistical significance. Well, the nice thing about this trial, since it's not an FDA trial, it's a CMS trial, is we can look in both arms every 25 patients implanted at the data, so a pretty regular cadence of being able to get an updated look. So there shouldn't be a long lag of when we hit the statistical significance in both arms. And we're then able to start analyzing the data in full to submit.

Got it. Thanks, Matt. Thanks for that detail. I keep getting nervous every time Damien goes to Matt because I'm thinking he's asking me these questions.

I'll say, Matt, could you just explain -- I'll say Dodds from now on. But look, it's tremendous how much excitement we have in-house with Matt. And our Head of Quality, Clinical and Regulatory is a PhD in Statistics. So when you start talking about Bayesian analysis, they get very excited.

[indiscernible] that component.

The investigators are equally excited. Again, this is probably the largest nondrug intervention in depression that's ever occurred. And the fact that we are going to be following these patients for quite a long time is a key aspect. This -- the registry -- setting up this registry is going to be a tremendous opportunity for these patients and for the clinicians who I think are seeking to make a substantial difference to the therapy options.

Right. Let's talk about that part of it. So I have 2 questions on the registry. One is when we get to the point late in '22, let's say, things are looking pretty good and you're ready to submit. How is that information going to come out to The Street? Are you going to put out a press release or an 8-K? Will we see a submission? What should we be looking for next year to understand how the program is going?

Yes. I will tell you. This is, again, as Matt said, different to an FDA trial where you'd probably release the study results in -- at a clinical event. This is different. This is a CMS study. We're going to have to work with CMS to figure out how best they would like the data released. And so I think for us, the milestone will be if we're submitting to CMS and which we declare, people should assume that we're seeing the sorts of results at that point that are valuable.

Very good, very good. Okay. And so this conversion to registry, let's assume you submit and things are moving that way. When could the registry begin? What kind of sales should we expect during the registry phase?

Yes. Well, the registry gives us, again, a lot of latitude, up to like 5,600 patients. So -- and as I said, it's typical in a registry that if the negotiations are moving beyond that period, that the registry is extended. And I would point to the WATCHMAN or even TAVR for that example, still living in a registry environment. To put it in context of where we think the market is, there was a period in 2005 when we were approved and reimbursed before the noncoverage decision happened. And in that period of about 18 months, we implanted 3,000 patients. We had about another 1,500 patients queued up before the noncoverage decision kicked in, and there had been 12,000 prescriptions written for the device. So again, I'd point it back to the math that is built around the epilepsy business with roughly 4,000 new patients a year. That -- we think there's a significant opportunity for us. You should model, in our opinion, roughly $25,000 a unit. I think people know our neuromodulation gross margins well enough. So hopefully, that math, again, this sort of thing we've been talking about for 6 or 12 months now is starting to help people model what this could possibly be if we're successful.

Understood. And back in the 2005 example, I had known about the implants in the queue, but I didn't know the prescription number. That's interesting that there is a much higher level of prescriptions there. What was the delta between the queue and the prescriptions? Are those just one's a good...

It's where you are in the value stream.

Right. Okay.

So you're moving through stages. And that's the delta, and it's moving from one step to the next.

Got it. Okay. Why don't we spend a few minutes on ANTHEM? And then I had a couple of margin questions. So let's talk about the heart failure program. Maybe just give us an update on where that is, and we can walk through some of the same kinds of questions that we did for RECOVER.

Right. Well, similar to depression, heart failure, nondrug therapeutic options, fairly limited. There are a couple of other devices that are approved. Again, we think we have a significant opportunity in this area. There are hundreds of thousands of patients if you take it at the narrowest point of the funnel, millions of patients if you go further up the funnel, who are Class II and Class III New York Heart Classification, which is our target area. We have 2 stages that's an embedded trial, so the primary endpoints of mortality and hospitalization are the end of the study goals. But the way it was constructed was to get at this embedded stage, which is to get the functional endpoints, things like 6-minute walk, left ventricular ejection fraction. And those points were predicated on reaching 300 implanted -- 300 patients followed for 9 months with at least 400 implanted. So we passed the 300 implanted last month. We're very pleased about the team's ability to do that again given that nearly all of the clinics were shut down last year. So we're continuing to see good recruitment and implanting. And so now we start following those patients for 9 months and look towards the secondary functional endpoints. At that point, if the data proves to be positive, we can submit for a PMA at that point and potentially release the product on the secondary endpoints, as was done with CRT-D a decade or so ago, or depending on our commercial strategy, continue to enroll and -- through to 800 patients for the primary endpoints. So that's sort of the flow. Dodds, is there anything there that you wanted to add on?

I'll just jump ahead, and probably Matt's next question is when he'd see that data. And there, what we're shooting for is most likely in the fall of 2022. It's either the European Society of Cardiology or the Heart Failure Society Meeting. Those are usually September. Outside chance, if we can get everything buttoned down really fast and we start analyzing the data, in January of next year at the ACC. But right now, I would target the fall of 2022.

Okay. Very good. I appreciate you not giving me heart failure there with the -- [ in relation ] to Matt. All right. So it's kind of unique for a company to have these 2 large programs with large unaddressed needs coming out around the same time. Could you offer any thoughts on your confidence in the probability of success in one versus the other? And is there any way to frame how you think the sales could evolve over the next several years? Could one of those be much bigger than the other? Or do you think they're both similarly sized opportunities?

Yes, that's a really interesting one. Look, I think both of them have support that signals a potential positive outcome. Again, the Aaronson study with that registry shows what the follow-up looks like with multiple year points. So we're encouraged by that because we almost sort of know the answer to the study. We're just going to have it in an extremely rigorous structure, and I like that. And the patient set, as I talked about, is pretty extensive. I think if you look at the potential therapeutic area, our TAM is 700,000-plus patients. I mean there's multiple ways. Again, I'm sure all your listeners have got different models. Is it 2 million to 3 million people? Is it the 100,000 ECT patients? Is it the 25,000 TMS patients? There are funnels of -- either model would point you to a large patient step, and that's what's exciting about that. And again, we more or less know what the pricing ought to be around that. Heart failure is going to be a bigger step because we don't know the outcome of the study to the same extent, but we've got early feasibility work in the -- that signals good, positive outcomes statistically and significant outcomes in these functional endpoints. We believe there's a large unmet need. Again, you can slice the populations in multiple ways, but that the -- let's call the patient set 300,000 patients in the U.S. alone. So I think again, we've got a tremendous area of opportunity to operate here in a patient group that, candidly, don't have many therapeutic options at that point.

Great, great. We're almost out of time. Let me make this our last question. I wanted to ask one on margins in the P&L. And so the question is, we know that you're spending kind of $50 million to $70 million on these strategic initiatives. Can you help us understand how long that level of spending is going to last? And maybe connect it back to -- you're more than $3 in '18 and '19. How should investors think about the underlying earnings power of the business post-normalization?

Yes. Look, I think these VNS opportunities are almost like bio farm, right? It's a single molecule being applied in multiple therapeutic areas requiring large clinical sets of data to support their release. So I ask people to sort of think about the modeling is a bit like a bio farm in terms of the length of time required, the study size. So again, I think we've been pretty clear about the length of the study. So that $50 million to $70 million is going to continue for several years. We're -- ultimately, it will start to come down. We'll then be looking to commercialize. And so we're going to have an SG&A expansion to commercialize them. But again, given the upside in the market, we think that's significant. In terms of normalizing, how do we do that? Things like resource allocation, we took out $65 million of costs last year in response to COVID. About $20 million to $30 million of that savings is going to stay out of the P&L. And so that's how we're trying to focus the effort internally to be able to fund this. We also, as everyone is aware, have a significant debt structure that's costing us a lot in terms of EPS. We'll look to optimize that appropriately. So it's possible to start heading back towards an operating margin that we believe should be north of 20% and earnings power that significantly heads towards the historical levels. Again, it's going to require some discipline, and it's going to require patience for people on the studies. But I think both of those are achievable.

Understood. Great. Well, thanks for all the color. It's been a great discussion, get an update on all these different aspects of the business. And glad to see the momentum is shifting in the positive direction, and it's going to be exciting to watch out for these guideposts on the pipeline. So thanks so much for your time.

Yes. Thanks for having us, Matt. Really appreciate it. And for everyone that joined in, we look forward to further discussions with you. Thanks for having us along.

Great. Take care, guys. Thank you.

Cheers.

Bye, Matt.

See you, Matt. Bye.

Yes.