Penumbra, Inc. Stock

Good afternoon, everyone. I'm Robbie Marcus, med tech analyst at JPMorgan. Very happy to introduce our next company, Penumbra. Adam Elsesser, the CEO, will give a presentation, and then we'll do some Q&A. Adam?

Thank you, Robbie, and of course, JPMorgan for hosting us. Robbie was reminding me, I think this is our 10th or 11th time since we went public. So it's great to be here and a pleasure to talk. Let's see. Here's our safe harbor.

Just a quick sort of overview. Our company was started in 2004. So over 20 years ago, we're headquartered here in the Bay Area in the town of Alameda, which is right across the Bay in the East Bay, a beautiful community.

We have over -- a little over 4,500 employees worldwide. All of our products today are manufactured in California. That's going to change. I'll get to that in a minute.

And we've been around a long time. We have -- our products are available in over 100 markets throughout the world.

On the manufacturing front, I wanted to just sort of touch on this because I think it's important. In Alameda, we have 9 buildings total and obviously, a lot more square footage than this, but 260,000 square feet of it is dedicated to manufacturing here in Alameda.

I think it's important to point out why that's what we do that and what the benefit of that is. We are known for iterating our products a lot. We're quite innovative. We launch many, many products every year.

And having the manufacturing right there with all of our other engineering and support staff dramatically increases the pace that we can bring products from R&D, move them into manufacturing and up and running. So that's a really, really important part of our success is the ability to really bring products from R&D into manufacturing and up and which by being altogether is part of the success.

We needed to expand as we grew, we opened up a facility about 2.5 hours from our Alameda facility up in Roseville, near Sacramento. That is about 300,000 square feet, and that has been obviously quite successful.

And as we look forward, as we continue to grow, we're going to need more space, and we have secured a facility, a site in Costa Rica that we own and are building -- building that facility that will be about 330,000 square feet. That's due to open sometime in mid- to later 2027.

And then we have an option next door to build another building of that same size, to continue to grow. We've learned a lot about being in Costa Rica. We're excited about it. I think it will augment our ability to make cost-effective products as we move forward.

I want to really talk about the 2 sides of our business. Over the last couple of years, we have focused mostly on our thrombectomy side, where the growth has happened. But for a number of different reasons that I'll go through in a bit, we now have our embolization side of the business, both in peripheral and in neuro also growing pretty significantly.

And as I said on the last earnings call, I think that has some sustainable growth behind it. So I'll walk through the point of that and how we can do that. I want to sort of level set, just sort of what is the thrombectomy markets we're in.

The total, if you add it all up, again, these are U.S. numbers. It's easier to get these numbers, and they're more accurate than if we try to look at this worldwide, so one can extrapolate, of course.

And again, even so -- there's some variability in people's estimates of these markets. We try to be on the conservative side. So if you look at other estimates of these markets, they tend to be higher.

If that were to turn out to be true, great, but we want to be sort of a little bit on the conservative side. So just sort of think about that. Arterial, that's legs, limbs, not all arteries because the stroke patients are different, but that has a little over 250,000 patients.

The majority of those patients now get either open surgery or dripping TPA or lytic. That's the majority. On the device side, the mechanical thrombectomy side, we have the vast majority of that share.

But the real effort for us there in that market is to continue to convert people who are doing these cases. So they're being treated.

They're not having to be found and brought to the hospital. They're being treated. They're just using more traditional older technology. I think that has a lot of room for growth as we move into the era of what we call CAVT or computer-assisted vacuum thrombectomy, where the outcomes and the time and the pace are pretty extraordinary and I think have shown a significant benefit.

So more work to be done there, but I think the opportunity is pretty obvious. The biggest of these numbers, you can see is DVT. What we've done with this number is try to limit it to the iliofemoral DVTs, the ones that have been shown in clinical trials to benefit from mechanical thrombectomy as opposed to the larger definition where is yet to be fully demonstrated.

I think there's room in the future to show that for all veins, but we haven't done that yet. So we're limiting it. Notwithstanding that, obviously, you can see this is the -- in terms of number of patients, the biggest opportunity.

Next is PE or pulmonary embolism, it's around a little over 150,000 of the type of patients that we've identified. Some numbers have that higher because it's a broader definition. That's not unfair, but we've just tried to limit it to a smaller number that is clearly appropriate and viable.

Of that, that might be the -- other than stroke, the more penetrated, it's somewhere around 15% or so that is currently penetrated in that market. Coronary patients is pretty significant.

Our product there is a mechanical product. It's not -- it's aspiration based, but it's not yet CAVT. That market, I think, is relatively stable. Work to be done. Is there a need to have the computer-assisted vacuum thrombectomy in that space.

There are certainly some patients we've heard that might benefit, but we're doing some analysis and trying to determine how many of those are and is that a viable opportunity to help those patients.

And then stroke, which has been our historic -- we started the company based on stroke and trying to do something about stroke. That is about 200,000 in the U.S. roughly.

There is a huge opportunity there to continue to grow that. It's the most penetrated. It's over 30-some-odd percent. There were about 10 years ago, landmark studies that showed you should treat these patients mechanically.

And recently, it sort of stalled out a touch. That's not atypical over the 10 years. We've watched the market sort of move in on a quarterly basis, it's never a straight line when you're talking about this growth, so I think there's optimism that as we move into '26, we'll see that revert back to some growth.

So that gives you sort of a starting point. I want to focus a little on the technology, thrombectomy technology. We used to be a catheter and a pump, a very sort of analog system. You would turn it on and try to suck out the clot.

That's what got us started for many, many years, and it's a viable technology. We moved as we try to continue to improve and improve and make ingestion of the clot all the way into the canister there that you can see on attached to that pump, the standard, like it just happens and it happens quickly.

We brought in CAVT technology. Again, now we're on many generations of that. And it really has fundamentally changed the way cases are done, both on the arterial side and of course, on DVT and PE, most recently in the STORM-PE trial, pretty remarkable results.

And also on Thunderbolt, which is, as everyone knows, under submission with the FDA. I know I'm going to get this question, so I'll at least answer it first, but I'm sure it will come up again.

As I -- we said on the last earnings call, we answered their collective questions in the later part of the fall. I said at another public event in December that we did get some follow-up questions.

We're addressing those and answering those. I -- there's nothing about this that -- I think this is in good shape. It's a thorough process. As you look at other products in the neuro space, there's some other ones, some liquid embolics that recently got approved. The time frame is 18 months from start to finish.

They got approved. They ran trials as well. We're at 9 months now. So we're in good shape, and we'll wait and see how that process goes. But I wanted to give that sort additional update. Moving on to STORM-PE real quick. This is a randomized trial, the first in really a generation in pulmonary embolism, randomized against anticoagulation alone.

Our arm is anticoagulation plus our device. The primary endpoint was a reduction in RV/LV ratio. That was extremely positive statistically, very significant. The safety data was really just stand out and has been commented over and over by people in the field for being amazing.

And then the other thing that people have commented on is the procedure time, both the low procedure time, but also the device time itself, which in this trial was 25 minutes. Keep that in mind because I'm going to show you some cases with our latest product, which we didn't even use in this trial, and you can calibrate that 25 minutes, which was best-in-class at the time, which was now 2 months ago and see where we're headed with our constant innovation.

The thing that really stood out in addition to those other endpoints, which were really, really significant was both the clot volume and the 6-minute walk test at 90 days, both statistically significant in the CAVT arm, and that wasn't powered to be significant, but it was still notwithstanding that, statistically significant.

And just to measure this in the 6-minute walk test, so this is 90 days later, you measure how long it takes -- how long you can walk in 6 minutes.

The people in the CAVT arm walked around a football field longer than the people in the anticoagulation arm alone in that 6 minutes.

So it wasn't like a few yards longer. It was a football field longer. So it's pretty significant, and that has definitely caught the attention of folks. I'm looking at the time, so I'm going to sort of speed through this a little bit faster. I apologize. I'm going to show quick cases.

These are with Flash 3.0, our latest product that we didn't have for the trial. The image on the left is before a pre-image. You can see the blockage where you see the tip of the catheter and there's sort of no darkness. And then the 2 pictures following where a lot of that clot is gone.

And then, of course, the image of the clot that was taken out. The device time of that was around 1 minute. I mean it's extraordinary. And that is a hardware and an algorithm change. This is another one, again, similar pre and post and all the clot that was taken out. That was a 3-minute case. This is both a DVT and a PE in the same patient.

Again, the ones on the left are pre, then the post, the same on the vein, which is the next and then the clot. This is worth noting, you can see the color differential on the clot.

The lighter clot is very -- is much older. It's more fibrous, usually harder to get out. And again, we did that the DVT took about 5 minutes and the PE less than a minute. Just in DVT alone, again, another image more clot under 2 minutes for that.

And this is an arterial case using our Bolt technology, again, pre-post and then they form the clot on the table in the image of what the artery, how it laid in the artery when it was sitting there.

Again, pretty quick 10 minutes for an arterial case. A couple of stroke cases as well. This is silver label, our current gold standard.

You can see where the circles are the pre -- where the blockage is and then in the images after that, how wide open it is in the brain. And this is a much distal stroke and M2 occlusion with the 43 and the picture of the clot.

So it gives you a little sense of where the technology is headed, and it's really just in a different league than some of the old analog technologies of a catheter in a syringe or a catheter in a pump or some kind of grabber.

I want to talk briefly about embolization as well. Two different parts of embolization. The peripheral side, we do -- there's about 175,000 procedures done.

We have a significant share, but not a majority. We, I think, have the largest share, but there's a lot of room for us to go. We saw that last quarter with both our new coil launch, but also the ability of our sales team to be able to focus more because the sales team didn't have to cover both thrombectomy and coils.

The next thing I want to talk briefly about is MMA and the number of patients here in the U.S. is about 140,000. These are chronic patients.

This is somewhat new. It's sort of come up really this year. We didn't plan on this. This is a unique procedure that is coiling the middle meningeal artery after someone's had a subdural hematoma and trying to make sure it wouldn't bleed again. It's an effective procedure. Our liquids have been used.

They have some topics that people sometimes potentially get anxious about. So they started using our unique coil called the Swift coil that really doesn't have a shape. It just sort of runs. And I'll share with you some images.

We have a large portfolio. These are brand names of the coils. So it's not one coil or one small family of coils. All of these coils have different sizes, functions and capacity.

So we're really now at a point where we can handle embolizing any part of -- for the most part, any part of the body, a lot of different options for doctors to use. Just real quick, just sort of where things are in the peripheral, the types of things.

The dark part that you can see there on the slides are where the coil mass is where you need it to embolize the area. So you have splenic aneurysms tend to be ground, big spaces that you're filling. If you go down and look at some of the other issues, it images, it's much more of a fine like you're embolizing a part of an artery or a little piece, it's a little different.

So just to give you a little sense. And then to end talking a little more, this is what an MMA embolization looks like. It's not filling a volume of space. It's really filling the artery itself. So the coil has to be really soft. It just has to sort of float out, make sure it's in the right place.

It's not blocking any of those little linkages between middle meningeal artery and ophthalmic artery. That's where the risk happens. And so you're not doing that and you're able to see that before you detach it, which obviously isn't true when you're looking at liquids. So that's just one example.

And the final example is when you have bilateral MMA. And again, you can see where both of those happened and just what that looks like. I want to end. I'm trying to time this pretty perfectly with 20 seconds. We're known for innovation. We've done a ton of innovation, all of it in-house.

We're super proud of our team and how we've structured that. And I just want to end by saying we're not done. We have great products, but they're going to get better and better and expand applications. And the future, I think, looks pretty good for us. So with that, thank you, and we'll take your questions. Robert?

Well, great. Maybe just a quick clarification because in some e-mails during the presentation. I want to make sure people don't misinterpret anything you said.

Just on Thunderbolt, you said it's been 9 months since submission. You got some questions in December. You've responded to them. It can take -- there are some that take 18 months. And I just want to make sure you're not saying it will take 18 months.

I do not know how long it will take. If I did, I would tell you. What I was simply saying is this is normal. The normal I -- look we've been working with the neuro division at the FDA for over 20 years.

We did this with our first company as well. I have nothing but utter admiration. They are doing a great job. The brain is different than the rest of the body. It requires a lot of being super careful. So we're addressing those questions. Nothing at all. I mean, am I anxious? Do I want the product out there? Of course. Do we want to do it the right way? We're in good shape.

I just wanted to get that out of the way. So maybe we could start. 2025, we'll get the full results in February, I imagine, with fourth quarter earnings. Guidance implies something like mid-teens with 30% plus growth on the bottom line.

That said, you have a lot of businesses that are doing really well, and you have just a few businesses and some outside U.S., difficult comps that are kind of limiting the reported growth this year when all is said and done. What are some of the businesses that are doing really well that have solid growth? And what are some of the headwinds that are going to go away as we move into 2026?

Well, yes. So I can talk conceptually, obviously, the last thing I'm doing is giving implied guidance right now.

I'm not asking for numbers.

I'm just talking structurally. So I want to -- we'll obviously cover all of that on our upcoming earnings call.

The biggest thing that is just easy to talk about because we talked it a bunch before. China had been a couple of point headwind. I think we won't see that in 2026, obviously, as it compared to 2025.

So I think we'll have just on a growth rate basis, that benefit. And then the difference is we have -- as you could just see, we're in a different place. For a long time, last 2, 3 years, we were talking only about thrombectomy and our coil business was just doing well, but not growing at the same rate.

And now that's changing. And I think we will see both provide benefit to the growth of this company. And so we'll talk about what those numbers on the next call.

Those are 2 important growth drivers. Maybe we could start with peripheral thrombectomy. You had the STORM-PE data at TCT. And I think the words you used were this is a landmark trial or maybe it was the PI of the trial, you said that?

It was not the PI, and it was not me. It was the Chair of that session who was not involved. Dr. Kenny Rosenfeld, who is independent and had that reaction. That was not a marketing story. That was a reaction from an outside physician. It's important.

Yes. So very positive 100-patient trial. There are 2 more trials coming up over the next 12 to 18 months. We have HI-PEITHO from Boston Scientific and PEERLESS II in 2027 from Stryker. How are you thinking about what STORM-PE does for the market and Penumbra?

And then assuming those other 2 trials are positive, I imagine that probably helps the market as well, just having more confirmatory evidence that is the best way to treat these patients.

So how do you think the market will benefit from STORM-PE and Penumbra and then the market overall once we have more complete data sets?

Yes. So I think it's -- the market is already benefiting. Lots of people are talking about this. The larger medical community has responded incredibly favorably because the -- it's not just for interventionalists, obviously, it's for all of the referring physicians, pulmonologists and emergency room docs and so on.

So those reactions now sure there's somebody one could find, but across the board, it's just been really, really positive. So people are starting the work within their systems to change their protocols inside.

That's how it goes. And we're seeing that. It takes some time. Hospitals can't just in one day, declare a change protocol. They go through a process. There's committees, there's back and forth to make sure what they're doing is accurate. So those have started, obviously, after the trial. No one is waiting for additional data. Whether those trials are positive or not, I don't think will matter to our success.

And partly, that's because of the things that we already know that are part of that, that I went through, which is the safety profile and the time of the case, both things that aren't -- that are related to its success, but sort of somewhat independent.

And I've never ever met a physician who says, I'm going to use both products were successful in a trial, but one was 3x faster. I'm going to use the slower one. I think that's the better decision. I don't think that's viable.

So I think those things are what -- and we actually called that out on our last earnings call where, again, it's the technology is at a different place.

It's no longer everyone is the same, what's happening. The technology has changed the procedure to be different and faster and with an amazing safety profile.

So I certainly welcome the additional data from the trials, but I do think this technology is setting up a standard in and of itself.

And it's been a really fast-growing market for a while now. There's still a stubbornly high number of docs, who don't use peripheral thrombectomy. Some of them will point to guidelines. Do you think STORM-PE is enough to change guidelines to maybe like [ it to be ] and how impactful will that be?

Protocols first get changed and then guidelines sort of follow. There are very few people that wait until everyone else has changed and then do it.

Some do, but most sort of react that way. So that's the position we're in. And I am happy to talk to somebody who's conservative or stubborn, but the number of people who want to get on with making sure these patients get treated is a fairly significant number, and we got a lot of work to do, a huge amount of work to do in the next year with people who want to do this. So we'll get to the others as we need to because there's an awful lot of energy right now and work to be done.

Maybe if I think on the arterial side, you have a majority share in this market. It's still growing...

Mechanical...

Mechanical thrombectomy in arterial. And it's still a really nice double-digit growth business for you. Where do you stand in terms of penetration in that market? And how do we think about what sustains that double-digit growth moving forward?

I think -- well, we stand in a great spot in that we're really the standard for mechanical, where the work we have to do, and I think we're positioned to do that in a more concerted effort in the next -- this year.

And that is really focused on making sure physicians who are otherwise using open surgery, doing open surgery or dripping lytic that this is a better alternative. And I'll be honest, our sales reps, they covered coils, they covered DT, they covered coronary and they covered arterial. That's a lot to do all at once.

And when you wake up and where do I focus, what do I do, you're going to focus on the things that are moving and have momentum and convincing someone who's done a certain procedure for 20 years to switch is maybe a slightly harder task and so I think we're now at a place where with the coil business going to a dedicated sales force, our thrombectomy reps have the ability to focus on both arterial and BT at the same time.

It's a great transition. Maybe we could talk about coils and this hasn't been a growth driver for Penumbra historically. And all of a sudden, it's become a very nice growth driver, over 100% growth the past several quarters.

MMA is getting a lot of buzz at medical meetings. And clearly, we're seeing the benefits in the Penumbra results. You've talked about a billion-dollar opportunity, I believe, in this market over time, the Penumbra...

I think somebody else said that.

Maybe somebody else.

I mentioned they said that. I did not...

Are you willing to put out a market?

I think it's early. I think everyone has gotten a sense from other people that it's not insignificant. I did talk about the number of patients that could be treated.

I think people can do the math. This is a different type of thing in that it sort of just came upon us. It wasn't -- we already have the coil. We made it for other arteries. It's obviously indicated for using in neuro arteries.

So that part is fine. But it was somewhat just appeared from -- and so just as investors were surprised and like what is this, we, as a company, had to also get caught up.

I think in our second quarter call, we mentioned it as this is something that people are now talking about. And then in that short period of time to the next call, it became like, oh, hold on, this is actually real.

And I think we're seeing that continue. So we're doing a lot of work to be able to be helpful in this space where reps are getting trained that aren't as trained, which is wonderful.

We're looking at retrospective data to make sure that people who are using our product that it's all going as planned. We are starting the conversation of do we want to run a prospective study.

So there's a ton of stuff that we're doing to sort of make sure these patients are getting the best care that they can. It's a great opportunity, a lot of excitement. And it makes a lot of sense. If you can have this treatment and not have to have more invasive surgery or boreholes drilled over time, that's great for patients. we're happy to be part of it.

Do you feel like -- obviously, the product just kind of fell into this. Are there any more product-specific enhancements you can make for using it in M&A and MMA? And also, are you out there now aggressively starting to detail this and say a word on education?

So I think you know me well enough after all these years that, one, we are constantly innovating. And you also know that I'm not going to tell you what that innovation looks like.

But yes, we always look at ways to make the products better and better. The interesting thing is our current product is really good in this space.

So it doesn't have obvious limitations, but there are some possible things. I'm not promising, but some possible things that could make it potentially even better.

So that's cool. But right now, as it is, it's it doesn't have a ton of limitations. And yes, we -- as I've said, we're reacting to everyone's interest. These are our customers, these are the people that we've been working with for many, many -- 2 decades to make sure that they have the support that they would expect from Penumbra.

It grew over 100%, arguably off a small base, still doing nicely. Is it greedy to think that given how small the revenue base is and big the opportunity without...

I'm not going to answer the number of question, but there is lots of opportunity here.

Great. If we think about stroke and put Thunderbolt aside for a second, the market decelerated in 2025.

In the second half of '25.

In the second half of '25, where do you think overall market growth is right now? We see there are some companies investing in stroke. There are a lot of companies exiting and pulling back investment in stroke. How do you think about this market over, let's call it, the medium run, 3 to 5 years?

Well, I think there's a lot of pros and cons like if just as a market. You know how I feel this is incredibly important work. These are patients that need to have this happen. There's so much randomized data that supports doing these interventions and them.

Probably more than any other [indiscernible] stroke is...

It's overwhelming. So we have to deal with the structural issues of making sure patients get to places where they can be treated.

There's enough physicians able to do that. All of those things have been the biggest challenge, not -- and that's more unique in stroke than any other things the nature of it being neurospecific.

We have been committed as a company. I personally have been to help drive that work, working with the societies and so on.

And I think we have to do that even better as we go coming out of COVID and lots of structural reasons that we saw sort of a different attitude.

But the products are profound. We also are a little bit suffering from -- we have a lot of companies all over the catheter, and it's a little bit overwhelming. So I think what we're seeing and even though the market did that, we did keep growing and gain share just because we're the standard.

And as you've tried a bunch of things, you come back to the one that's the best and I think we're seeing that happen. So I think that consolidation around the best product, which we obviously believe we have, I think will continue into the year, and we'll continue to put the energy into doing this. There are some great bright spots.

There's a young physician who is new in his practice a year or 2 is really passionate about bringing stroke care deeper into the community that wouldn't have this type of care available.

And with that effort has seen just a huge uptick in the number of patients that are getting treated. And sort of on a quarterly basis, they just share those numbers, and it's really heartening to see.

So there are models out there, that are bright spots. We just have to continue to put that energy into it. And I think we can continue to see a lot of people.

It's personal to me. I've had patients come up to me in various settings who knew who I was and had been treated and thanked me. And for a guy who's not a doctor to be thanked for somebody's well-being, I got to tell you, there's nothing better than that in the entire world. So we're committed to keep doing that.

With 1 minute left, doesn't do it justice, but margins in a bright spot, you've committed to exiting next year at 70% plus. How do you feel about the company's ability to continue to expand margins coming through?

Yes. I think we're in really, really good shape. I think if you look at a couple of things sort of put that in a worst spot. We expanded rapidly into the new facility in Roseville coming because of COVID, that press things. We launched a ton of products, which always takes a while to sort of get up to speed.

We were in markets, in international markets with some of our products where it really depressed those things. And as you know, we exited some of those. So I think we're in really good shape.

Great. Well, let's end it there. Thanks for a great discussion. Thanks, everybody, for joining.

Thank you.

All right. Good morning, everybody. Thanks so much for joining us. My name is Matt O'Brien. I'm one of the med-tech analysts here at Piper Sandler. Picking up day 2 here with a phenomenal company in Penumbra. From the company, we have Adam, who's the CEO; Shruthi, who is the President of the company. We're also joined by Arani, who is one of the co-founders, and then Cecilia and Jason are in the audience as well. Thank you so much for coming all the way across the country to be with us.

Happy to be here.

So can we start with what happened in Q3, especially in the embo and access business. As I was saying to you earlier, it was so good, I actually thought there was a mistake in that press release. But just talk about the strength that you saw there, splitting the sales force, and then just the durability of that growth, because I think that's being underappreciated by investors.

Yes. Well, Shruthi, why don't you start with the sales force and why we did it, and then some of the new products and all that?

Yes. So towards the end of last year, we saw this growing sort of interest in the CAVT portfolio, the franchise. And to get our team to be 100% focused on that, we said, now is a good time to actually expand and add an additional set of people that can focus on our growing embolization product line as well. We've had our embolization products for multiple years since 2013 in the periphery. And physicians have just loved these coils and how they perform. So it needed its due attention. And so with that in mind, we said let's bring on this really tenured group of people that are going to be spread around the country, a little over 50 people that came in and joined the team from really tenured areas like TAVR and aortic and other coil companies. They just came to be a part of this innovative culture that we have at Penumbra.

And we sort of got lucky with the timing of Ruby XL, which is the sort of latest addition to that coil portfolio. And so right when they came in and got trained up and running, they then got to go and launch Ruby XL to our customers. And as you saw and as you pointed out in Q3, that was the first full quarter of launch of Ruby XL, and we really saw it take off. There's a huge amount of interest in that coil.

What it essentially does is it fills spaces faster. It's a larger vessel coil. So any time you're trying to fill large spaces, you can just use the Ruby XL, and that's what physicians have been looking for. There's been very little innovation in that area for a while. And so they were excited to see Penumbra bring the softness, the volume advantage that we are known for with our coils, bring that to the large vessel sort of space.

And these reps have, like I mentioned earlier, all this tenure, and so they are continuing to elevate our overall sort of commercial team and how we operate. So we're seeing sort of the early innings of all that play out right now, and we're pretty excited about what's ahead.

Got it. Okay. So maybe I'll just stay there for a second. Hemorrhagic stroke, it's a smaller market, obviously, than ischemic. And then it's also thought of as maybe a little more commoditized. So when you talk about a soft coil, can you talk a little bit more about Ruby XL specifically, the uniqueness and how that's scaling?

Yes, let me -- so what Shruthi was talking about is on the peripheral vascular side.

Okay. I'm sorry.

No, no, it's totally fine. There's a lot of different parts of the body here that we focus on. So you do a lot of embolization in the peripheral vascular. The neuro, we'll get to in a second, but that's a very specific thing that also is having a new area sort of birth, if you will. The coil -- the platform that Shruthi is talking about has really been around for a long time. We have a new offering with the Ruby XL, but the whole platform fills spaces faster, softer, better.

And it's been around for a while. We've had lots of the companies we compete against in those spaces have tried to match our technology and not yet done that. So it's unlikely that they're going to keep trying. So I think we have a pretty likely run here of being able to continue to grow on the peripheral vascular side of our business, on the coil side.

On the neuro side, you're right, the hemorrhagic, which is sort of cerebral aneurysms, if you will, as another way of framing it, has been somewhat not only commoditized, but coils and flow diverters and various other tools. The growth that we're seeing on the neuro side, which is also significant, is coming from a new procedure called MMA embolization. And that really wasn't being done even a couple of years ago, and it just started being done. This is where you have a dural hematoma and you're trying to embolize it to stop that.

The early returns, if you will, have been really, really positive. We play in that world against sort of liquid embolics, which can have a safety challenge. Once a liquid leaves the catheter, you don't control where it goes. With a coil, obviously, you do. We happen to have a coil in that world, in that space. We did not design it for this. We designed it for other applications, but it happens to work. Well, it doesn't really have a shape, it just sort of runs and fills the length of the artery in a way that really is incredibly productive for these patients. So we're seeing a pretty significant growth.

There was a report came out recently that predicted -- that sort of tried to -- again, we haven't done this, but this report was saying that there's a lot of interest and the opportunity, if you will, and MMA embolization alone could be a $1 billion opportunity for a while. And that's wonderful. We're excited. We're seeing that, too. So you combine that with what's happening on the peripheral vascular side. And for the first time in a number of years, our coil franchise, our embolization franchise is going to be growing at the same pace as the rest of our business. So it's pretty exciting. And they're accretive to margin.

That was my next question. Well done. You read my mind, which it's not hard to do. But what about the sales force? Is it -- for the embo sales force, is it complete? Is it good to go? Or are you going to keep adding?

On the peripheral vascular, it is in a good spot now. Obviously, as the business keeps growing, there'll be a little bit of add here and there, just as we always do, but nothing -- like we don't have a plan to hire another 50 people.

Right. Got it. Okay. And then when you talk about the durability of growth there, I mean, similar kind of levels as just the overall business, or even faster than what you've been doing kind of the mid-teens or even upper teens for that business?

Yes. So we'll wait until the next quarter's call to give sort of numbers to that guidance. But needless to say, I think between where we are on our thrombectomy and what's happening in the STORM and everything else in our VTE product business and what's happening here, you combine it all, yes, I think we don't have the China headwinds anymore. International without China is now back in a good spot. We've done a lot of hard work to get the business there. I think we're feeling pretty confident that we're going to have some fun years here.

Good. Excellent. Okay. So you touched on STORM-PE a little bit. I would love to talk a little bit more about that. I know it's early, but data was good. It was really good. What kind of feedback have you gotten early days on that? And some of the pushback that I hear is like, oh, it's a small study. It's not really going to influence the overall industry, but we haven't had much clinical data like this. So why is it not a great perception?

So there is no -- we have not heard that other than through the investment community, to be very candid with you. The size of the study -- I think most people do understand basic clinical science that the size of the study is based on how little -- how many -- how few patients you can use to prove your point. In this case, we did it with 100 patients. It doesn't make it bad, that makes it even better. Most physicians understand that. That is not a -- obviously, if you're a competitor, you say that. But that's obviously not a valid scientific idea.

The fundamental thing here is the reception to this data, both on the primary endpoints, but the secondary endpoints, which were really extraordinary. The study was not powered to show any superiority on the secondary endpoints, and the 4 most important ones showed superiority. You add the safety element, which showed just how unbelievably safe and fast our procedure is compared to even other mechanical systems, has been really just extraordinary. I mean the reaction through both the interventional community, but the noninterventional community has been really kind of fun. So the work started with out there doing the work. The hospitals are changing protocols as we speak. It takes some time, but we have not heard -- I can't -- have we heard stories there? Like, we're not going to address this.

No. And in fact, the physician community is thankful that someone finally decided to run this trial that showed that mechanical thrombectomy with CAVT is superior to anticoagulation. Essentially, it's as safe as not intervening on these patients at all, which is a big statement. So not only do you have the safety profile that everyone, the interventional and noninterventional community, look for, but then you also have the RV/LV ratio reduction, that leading to better long-term outcomes as we've shown in the secondary endpoints in times that were a record 25-minute procedure times, device times, and 56-minute procedure times are unheard of.

So I think for the physician community to see that, to see that there was a very low requirement, each operator only had to have 2 prior cases with CAVT before they could enroll in the trial. So they talked about the device being really easy to use, and that was a big topic at the conferences. And that has then prompted physicians to say, well, not only is this safe, it's really easy to use. I don't have to do 100 cases before I get comfortable with the technology, it's time to switch. And that's what we're seeing.

Okay. And how big a deal was the RVRL (sic) [ RV/LV ] update? I mean, is that an issue that the clinicians are dealing with quite a bit and that it's top of mind where it's like, wait, we're doing -- this is superior here and this is a big deal and really going to influence my utilization?

Yes. Actually, RV/LV ratio is how the classifications are actually decided. So when you have a right heart strain or a high RV/LV ratio, you fall into the category of an intermediate/high-risk PE or a high-risk PE. And so that's the diagnostic measure that physicians use to then decide should this patient actually be treated or not. And the goal there is to then reduce the RV/LV ratio like we did in the study, and be able to show that you're relieving the pressure on the right heart and the patient is able to get back to normal function.

And I think the point that was made at Veeva and then beyond, and has been repeated hundreds and hundreds of times, is that the rest of the data in the trial, the 6-minute walk test as well as the other functional outcomes, all validated the use of RV/LV ratio as a primary endpoint. They all work in sync. Obviously, if those endpoints weren't as incredibly strong and statistically significant, one could say, okay, well, what did we actually do here? But when you have it all, just across the board, it validates that is obviously the right endpoint given the use of it in the diagnostic setting.

Got it.

So yes, we have not heard -- again, the 2 things that one hears from the Wall Street community is the size of the trial and that endpoint, that is not at all what we're hearing in the medical.

And what are you hearing from the referring physicians? That was interesting, too, that got their attention as well. Because I know that's always been a challenge of getting the patients in, getting mechanical thrombectomy done.

Yes. I mean, it's been just extraordinary.

Yes. I think the makeup of the steering committee, having 4 noninterventionalists as well as 3 interventionalists that were part of the steering committee, that in itself was a fact they were involved in the design of the study, they were involved in the execution of the study, they were involved in the results, the rollout, the publication.

And so now they've got what they need to go and talk to their sort of peers and counterparts. And that work is well underway, both through some of our education efforts, but also through the PERT consortium, which is the largest sort of group that's solely focused on PE. And their members are both interventionalists as well as noninterventionalists from all the main institutions around the country that are treating PEs. So they're getting the word out through their channels as well. And they also announced what was called the PRISM program, which is meant to go and get centers to update their protocols and digest this information and start to refer patients on for treatment that should be treated.

Got it. Okay. I guess the penetration rates of mechanical thrombectomy are pretty low, right? Is that because pre-STORM-PE, it was viewed as, as you mentioned, the RV/LV levels, like this is really just for the more challenging complex patients. And then STORM-PE can really open up kind of that intermediate group. Is that the way to think about it? Or is that too simplistic?

Yes. I mean that's, I think, a fair way to look at it. And you now have data, randomized controlled data that shows that mechanical thrombectomy is superior to anticoagulation. Up until now, that hasn't been available. So the standard of care up until now has been anticoagulation, unless the patient clinically deteriorates, and that's when they escalate to some sort of a treatment. That now is going to be what changes where they're not waiting for the patient to clinically deteriorate when they come into the ER. They say, okay, they have high RV/LV ratio. Let's go in and treat them with CAVT.

Got it. Okay.

Yes. I mean I just want to put a real fine point on that. That is sort of why some people were being treated, because they were being watched, they get a little anticoagulation, and they wait until they start to crash and get bad. And then they're like, okay, now we have no choice but to treat them. That's what started sort of building the whole market.

And if you think about it, okay, now we have really strong evidence, just don't wait. And there's just a common sense. Like if I were to ask you or anyone else, if you had a PE, would you have -- saying, no, no, no, I want to see if anticoagulation works. And I understand there's a risk that I could crash and die, but I'm going to take that risk, because I don't want something that's been proven really safe to take the clot out first. Like it just doesn't make any sense. And so now we have the evidence that supports the common sense approach here. So that's why it's being so well accepted. And the safety profile being now serious is a deal too, because it's like [indiscernible].

Okay.

If it was not as safe, then there's a risk to doing that. But there is no risk to doing that, because it is as safe. And numerically, in the trial, at least it wasn't statistical, but it was numerically even safer. There is just no reason not to do that. And people have wanted that data to support that behavior. No one wants to watch their patient crash and then hope that they can be rescued. It doesn't make any sense.

So you have a huge installed base already. How quickly do you think they can start to implement these protocols? Sorry, we're all numbers here. I mean is it second half of next year? Is it even faster than that?

Yes. We're, what, a month barely into it. So it's hard to give you the kind of numbers. But the way I would look at it is the first phase here is the acceptance, the knowledge of this data, the ability to sort of now engage with the teams in your hospitals to change protocols. All of that is well underway. It didn't happen in 10 seconds, but it is already happening. To get the sort of full benefit will take a little time. But again, that's why we at least want to stress here. The reaction to this has been, across the board, positive. People want to do this. There's no logical reason to not do this. There's no group or team saying it's just about education, just about doing the work to change hospital protocols. And as that happens, I think more and more people will have the benefit of waiting to crash.

And the protocols will get updated, but that doesn't mean that the referring physicians are going to wait for that when a patient like comes in the door today, for example, right? They just see, okay, the patient is...

Yes. And we've seen that already. We've seen that already.

Got it. Okay. That's fantastic. It sounds like the market is ready to accelerate. You have a new product, 3.0. I thought 2.0s are very quick. What are the features of 3.0 that are different from 2.0?

So the way I would describe it, when we launched 2.0, there was a lot of discussion, well, it's just incrementally better. Yes. But 2.0 was what got us here, because that incrementally better meant it was way faster and it took out less blood -- there's a lot less blood loss. So it just worked better than 1.0. 3.0 does the same thing. So it works faster than 2.0, and there's even less blood loss. So those are the 2 fundamental things.

How did that happen? And this, I think, is worth me explaining sort of what the product change was, because I think it's important. With 2.0, it was just a software change. And the algorithm got a little bit more fidelity, so it can open and close the valve of the aspiration source faster. With 3.0, there was both a software change, but also more significantly, a hardware change. And that is -- in Flash, that technology, the computer unit that controls the valve and the valve itself is at the end of the tubing that connects directly to the pump. So it's all the way up the source of the pump.

With the Bolt technology that we use for the arterial side, and also with now Bolt 12 on the venous side as well, now also Bolt 16, the unit is in the middle of the tubing, which is in the sterile field that has the valve in it, and that's where the computer is. So the valve for 3.0 -- so we're using that same hardware for 3.0. So the valve is halfway further toward the tip of the catheter. So when you open and close the valve and you're that much closer, the fidelity of how fast the valve can open and close increased because of the physical proximity as well as the software being updated to adjust to that.

So you now have -- the Bolt 16, which also got, cleared and 3.0 use the same hardware. They're separate products today, but one can obviously imagine them someday being the same thing that you could toggle between the two, which would be helpful for a clot that might be a little more organized or fibrous and you can just go back and forth. But 3.0 has that huge fidelity advantage. It's just that much faster.

So if you look at the trial, our device time were, what, 25 minutes, which was unheard of in the PE field. If that can cut down even by 10 minutes or so, blood loss is less, you're talking about a whole different thing. When you're asking people to treat 80% more patients, time does matter, time in the lab. You don't really have time for 1.5 hours plus case anymore. You want to get it done really quickly.

Got it. Okay. Makes sense. So I know everybody in the room and on the call are going to kill me if I don't ask, but I'm going to have to ask about Thunderbolt. Would you classify the questions you got back as pretty straightforward? Did they ask you for more data?

So the questions -- yes, so let me give you guys some -- the FDA, particularly the neuro division, is very thorough. They've always been thorough about new products. We have never had a situation where one has known how long we've been under review for a new product, because we haven't had a trial since we've been a public company. So there's never been this clock that everyone is measuring. There is nothing about this process. They ask a lot of questions. Some of them are, why are they asking that? Others are, that's a logical question. They're thorough. We want them to be thorough. That's a good thing. We then answer those questions. No conspiracy. I look at the data, it's extraordinary. This product works. We're going through the process. Since our submission, they've asked a few clarifying questions. We're going to answer those. It's the process. There is nothing that one can be anxious about.

Okay. And there's nothing external that was outside of your control as far as shutdown goes or anybody left because of shutdown or...

Not that we have been made aware of. I mean, we might not know that, obviously, but not that we have been made aware of at all.

Okay. And maybe just talk a little bit about ASP on that product. When it does get cleared? How do we think about that?

We haven't set that yet. It obviously will be higher than our current ASP. There's a sort of a total amount that you don't want to exceed. And so there's a little bit of trade-off. If someone is using our catheter and inner catheter and sort of add it up, the delta will be less than if they're just using a primary catheter of ours. There's room in there for an ASP uptick. That will take process. You got to go through VAC and all, which is a little more straightforward in neuro, just because there's less accounts. So we're ready to go. We've launched a lot of neuro products over the years. We know how to do it. I know it's gotten sort of an outsized view of the world. It is an exciting product. We can't wait. We have so many other things that are -- I mean, if you just look at MMA, and again, I'm not saying look over here, not here, but we have to pay attention to other things, too.

If the MMA -- if it's true that it's a $1 billion opportunity for us, and we have a unique product that no one else has that seems to work really well and is the safest alternative and potentially the most cost effective, it would be strange for us not to also want to do that at the same time as have Thunderbolt, right?

Yes. Makes sense.

But we're going to do it all. And that makes for a really fun view.

Absolutely. Okay. All right. Well, I think we've taken us over. I apologize for that. Adam, Shruthi, thanks so much for the feedback. Really do appreciate it.

Yes. Thank you. Appreciate it.

Thank you. Great questions.

Ladies and gentlemen, good afternoon. My name is Abby, and I'll be your conference operator today. At this time, I would like to welcome everyone to Penumbra's Third Quarter 2025 Conference Call. [Operator Instructions]. Thank you. And I would now like to introduce Ms. Cecilia Furlong, Business Development and Investor Relations for Penumbra. Ms. Furlong, you may begin your conference.

Thank you, operator, and thank you all for joining us on today's call to discuss Penumbra's earnings release for the third quarter of 2025. A copy of the press release and financial tables, which includes a GAAP to non-GAAP reconciliation, can be viewed under the Investors tab on our company website at www.penumbrainc.com.

With me on today's call are Adam Elsesser, Chairman and CEO; and Shruthi Narayan, President; and Maggie Yuen, Chief Financial Officer. Also joining us for the Q&A portion of the call is Jason Mills, EVP Strategy.

During the course of this conference call, the company will make forward-looking statements pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995, including statements regarding our financial performance, commercialization, clinical trials, regulatory status, quality, compliance and business trends. Actual results could differ materially from those stated or implied by our forward-looking statements due to certain risks and uncertainties, including those referenced in our 10-K for the year ended December 31, 2024, filed with the SEC. As a result, we caution you against placing undue reliance on these forward-looking statements, and we encourage you to review our periodic filings with the SEC including the 10-K previously mentioned, for a more complete discussion of these factors and other risks that may affect our future results or the market price of our stock.

Penumbra disclaims any duty to update or revise our forward-looking statements as a result of new information, future events, developments or otherwise. On this call, financial results for revenue and gross margin are presented on a GAAP basis, while operating expenses, operating income and adjusted EBITDA are presented on a non-GAAP basis. The corresponding GAAP measures and a reconciliation of GAAP to non-GAAP financial measures are provided in our posted press release. Non-GAAP operating expenses and operating income exclude expenses related to the wind down of our immersive health care business in the third quarter of 2024 of $5 million, and adjusted EBITDA excludes wind-down expenses stock compensation expense, depreciation and amortization, provision for income taxes and interest income expenses. And with that, I would like to turn the call over to Adam.

Thank you, Cecilia. Good afternoon. Thank you for joining Penumbra's Third Quarter 2025 Conference Call. In the third quarter, we generated total revenue of $4.7 million, representing underlying year-over-year growth of 17.8% on a reported basis and 16.9% on a constant currency basis. Our strong third quarter results reflect broad-based execution against our strategy and significant progress in further enhancing our competitive technological and market positioning across all franchises. Total U.S. revenue was $275 million in the third quarter, an increase of 21.5% compared to the third quarter of 2024.

Total U.S. thrombectomy sales increased 18.5% year-over-year to $192 million. Our VTE franchise once again led overall corporate growth. delivering 34% year-over-year growth. U.S. embolization across an access revenue of $83 million increased 29.2% year-over-year. Growth in our embolization and access business exceeded our expectations, reflecting the benefit of a dedicated peripheral embolization sales team alongside our investment in continuous innovation and new product introductions. Internationally, the work we previously undertook to enhance our long-term market positioning is beginning to play out. We are now operating from a position of portfolio and geographic strength with our commercial execution in the quarter, more than offsetting the China growth comp headwinds we continue to face.

Total international revenue of $79.7 million increased 6.6% year-over-year or 3% in constant currency. Below the top line, gross margin of 67.8% expanded 130 basis points over the prior year period, and we delivered operating income of $48.8 million or 13.8% of revenue, with our operating expense in the quarter, reflecting the full impact of our embolization sales force build-out. We remain on track and are well positioned to achieve a gross margin profile of over 70% by the end of 2026. And expect operating margin expansion to outpace gross margin expansion for the foreseeable future as we prioritize delivering profitable growth and an expanding profitability profile.

Turning to our U.S. peripheral business. Our third quarter thrombectomy performance reflected both the clinical benefit and competitive strength of our CABT technology alongside enhanced commercial focus from our now peripheral thrombectomy dedicated sales force. We delivered the highest sequential quarterly increase in VTE case volume growth to date in 2025. Our U.S. arterial business also delivered another strong performance with a combination of both 7 and Bolt 6x, supporting further physician conversion from open surgery or the Usolytics to CAT.

Penumbra received FDA clearance during the third quarter for both Lightning Bolt 16 and Lightning Flash 3.0. Lightning Bolt 16 brings proprietary modulated aspiration technology to our 16 French system and Flash 3.0 improves the fidelity of the algorithm by updating both the hardware and the software. These new products significantly add to the growing portfolio of advanced CEBT devices, opening up a clear path for us to reach the over 80,000 patients in the U.S. who suffer from VTE and arterial clot with our CIBT technology. I also want to remind everyone that the core technology in our CVT products is protected by a robust patent portfolio. We recently secured a victory from the Court of Appeals for the Federal Circuit which summarily and unanimously affirmed the Patent and Trademark Appeal Board's IPR decisions in validating several competitive patents challenged by Penumbra.

We will continue to vigorously defend all of our patents and IP for our CABT technology. At the TCT and Viva conferences, we presented the landmark results from Storm PE, the prospective multicenter randomized controlled trial evaluating CEBT plus anticoagulation versus anticoagulation alone for the treatment of acute intermediate high-risk PE. Shruthi will provide additional detail on the trial and study outcomes in her prepared remarks. But needless to say, taking into account the medical community's reaction to the data, we are highly optimistic that Storm PE will act as a major catalyst and positively impacting the treatment of PE while significantly increasing the number of patients receiving intervention with CAT.

In our U.S. peripheral embolization business, our new Embo dedicated 50-plus member sales team delivered strong 21.2% sequential growth and embolization revenue in the quarter. We believe the team's performance this quarter and enhanced focus across our comprehensive current coil portfolio represents a new phase of sustainable growth in our U.S. embolization franchise. Shruthi will discuss how the integration of this new team is unfolding. Shifting to our Neurovascular business. recent market trends in the macro U.S. stroke market showed a slight decline in the third quarter.

Notwithstanding that, our stroke thrombectomy portfolio delivered positive growth and share gain. Historically, new innovation has catalyzed the stroke market, and we are optimistic that with the anticipated introduction of Thunderbolt we will see a similar dynamic play out. In neuro embolization, the recent interest we've seen in utilizing our Swift coil for MMA embolization, continues to translate into extremely strong above-market procedure growth, further contributing to sustainable growth in the U.S. embolization franchise. Now I'd like to give you an update on Thunderbolt. In October, we submitted thorough responses to all the outstanding questions received from the FDA during the 510(k) review process. We are now in the stage where we address any final questions or provide any required clarifications to our answers.

For obvious reasons, we'll be prudent in our discussion today about Thunderbolt. However, we remain excited about the prospect of introducing CT to the neurovascular field. Overall, based on our third quarter performance and reflecting the trends we've seen in our business, and our target markets highlighted previously. We are raising our revenue guidance for the year, to $1.375 billion to $1.380 billion. We reiterate 20% to 21% year-over-year growth for 2025 for U.S. thrombectomy. Now I will turn the call over to Shruti, who will discuss a few details on the expansion of our sales team as well as the STORM PE trial.

Thank you, Adam. Good afternoon, everyone. Starting off with an update on our recent commercial expansion initiatives. Based on the significant ramp in physician interest in CVT, we experienced throughout 2023 and 2024, and Entering 2025, we made the strategic decision to build out an additional peripheral sales team to focus on our embolization business enabling our existing peripheral sales team to shift exclusive focus to CAET. Through the first half of 2025, we added over 50 new peripheral embolization sales reps under our existing peripheral sales leadership team. The build-out and integration went extremely well with the new team fitting seamlessly into our organization.

This allowed the team to execute the launch of Ruby XL while also maintaining our momentum in our established portfolio, achieving strong 21.2% sequential growth in embolization revenue in the third quarter. As we execute the shift in our commercial structure, our existing team helped integrate our new team and supported the transition of embolization case coverage across our accounts. This transition process will taper in the fourth quarter, well positioning our teams to enter 2026, fully focused on their respective sales objectives, CABT and embolization. Turning to Storm PE at the TCT conference early last week, Dr. Rob Loken presented the study's primary endpoints alongside key safety data. Additional secondary endpoint data, including key physiological and functional outcomes was presented by Dr. Rachel Bazosky at Veeva earlier this week. The trial proved CAVT superior in reducing right heart strain in intermediate high-risk PE patients with a comparable safety profile to the current standard of care.

In addition, while not powered to show statistical significance on secondary endpoints, the trial data demonstrated statistical significance in favor of CAT across multiple secondary endpoints. In addition, the initial trial results were published in circulation on Monday. Storm PE proved CAVT superior to the current standard of care across the primary as well as multiple secondary endpoints showing CEBT patients recovered earlier and have significant long-term improvement in functional outcomes. The trial also highlighted the ability of the field's most advanced technology to deliver faster procedure and device times, reestablishing the baseline and expectations around acceptable case times. With procedurals in the trial requiring minimal prior experience with flash ahead of enrolling patients in STORM PE, the strong results also highlight CABT's ease of use and exceptional safety profile.

The physician feedback across the interventional and non-interventional community has been enthusiastically positive. Since the presentations, we are seeing a shift to CAT from older mechanical thrombectomy options for PE. We expect to see this continue. We are also working with societies and organizations to disseminate the now published data from STORM PE to hospitals so they can update their current hospital protocols. Over the next year, this will be one of our top priorities. We are executing at a high level across our peripheral and neuro business units, and are well positioned to continue to build off our work to date. As we look towards 2026 and beyond, we remain highly confident in our long-term strategy grounded in meaningful innovation data generation and investment in our team, supporting durable, profitable growth and the ability to treat a significant number of patients with our technologies I'll now turn the call over to Maggie to go over our financial results for the third quarter of 2025.

Thank you, Shruthi. Good afternoon, everyone. Today, I will discuss the financial results for the third quarter of 2025. We financial results on this call for revenue and gross margin on a GAAP basis, while operating expenses, operating income and adjusted EBITDA are on a non-GAAP basis. The corresponding GAAP measures and a reconciliation of GAAP to non-GAAP financial measures are provided in our posted press release. For the third quarter ended September 30, 2025, our total revenues were $354.7 million, an increase of 17.8% reported and 16.9% in constant currency compared to the third quarter of 2024. Our geographic mix of sales for the third quarter of 2025 was 77.5% U.S. and 22.5% international. Our U.S. region reported growth of 21.5% and driven by 18.5% growth in our thrombectomy franchise and 29.2% growth in embolization and excess driven by our Ruby XL product compared to the same period last year.

As we previously shared, due to easing of China's headwind and double-digit growth for the remaining international regions, our international business has returned to growth. increasing by 6.6% reported and 3% in constant currency compared to the same period last year. Moving to revenue by product. Revenue from our global thrombectomy business grew to $236.4 million in the third quarter of 2025. We an increase of 15.8% reported and 15.1% in constant currency compared to the same period last year. The growth was primarily driven by an increased 18.5% increase in our U.S. thrombectomy business. As expected, our international thrombectomy revenue also increased by 5.6% when compared to the same period last year.

Revenue from our embolization and access business was $118.3 million in the third quarter of 2025, an increase of 22% reported and 20.8% in constant currency compared to the same period last year. primarily driven by an increase in U.S. peripheral embolization sales due to the momentum from our successful launch of Ruby XL combined with strategic investment we made in expanding our U.S. embolization team in the first half of the year. Gross margin for the third quarter of 2025 is 67.8% and compared to 66.5% for the third quarter of 2024. And consistent with our expectations, we delivered sequential gross margin growth of 180 basis points. driven primarily by favorable regional mix, product mix and productivity improvements. We are very pleased at how quickly the team stabilized our Ruby XL build of which our Ruby XL product has an accretive impact on our gross margin and the sales model supports a more efficient working capital dynamic.

In addition, we are on track to achieve our full year gross margin target and remain well positioned to deliver our long-term gross margin profile of 70% by the end of 2026. Now on to our non-GAAP operating expenses. Non-GAAP operating income and margin and adjusted EBITDA. Total operating expense for the quarter was $191.6 million or 54% of revenue compared to $160 million or 53.1% of revenue for the same quarter last year. Our research and development expenses for Q3 2025 were $22.7 million or 6.4% of revenue compared to $22.6 million or 7.5% of revenue for Q3 2024, which reflects savings of $3.6 million due to our immersive business way down, offset by continued investment in product development. G&A expenses for Q3 2025 were $168.9 million or 47.6% of revenue compared to $137.4 million or 45.6% of revenue for Q3 2024. As we have stated previously, we have made targeted hires in our commercial and market access teams which will support customer demand and allow us to capitalize on long-term growth drivers.

Sequentially, our SG&A expenses increased by $8.9 million, reflecting the full quarter presence of our embolization sales team investment along with other variable spend. With this buildout now complete, we are positioned to capture sales and operation leverage in future quarters. We recorded operating income of $48.8 million or 13.8% of revenue compared to an operating income of $40.3 million or 13.4% of revenue for the same period last year. We posted adjusted EBITDA of $66.7 million or 18.8% of total revenue compared to $56.7 million or 18.8% in the third quarter last year. Turning to cash flow and balance sheet. We ended the third quarter of 2025 with cash, cash equivalents and marketable security balance of $470.3 million and no debt. which is an increase of $45.7 million sequentially. This increase includes improving working capital ratios in both receivable and inventory turns and strong profitability. We continue to expect positive operating cash flow trends to continue in 2025 and beyond.

Turning to 2025 guidance. As Adam previously stated, we are raising our revenue guidance for the year to $1.375 billion to $1.380 billion. We reiterate 20% to 21% year-over-year growth for 2025 for U.S. thrombectomy. Finally, we maintain our previously stated 2025 growth and operating margin expansion guidance. This concludes our prepared remarks. Operator, we can now open the call to questions.

[Operator Instructions]. And our first question comes from the line of Travis Steed with Bank of America.

I'll start with Thunderbolt since it's topical. This sounds like you're a lot closer to approval than you were. I assume nothing surprising with the FDA back and forth. But just maybe how you're thinking about the product when it hits the market, how much share you can take and how much -- when investors will see the data from Thunderbolt? And anything else you can share on the FDA process, so it would be helpful if you could have the product approved by year-end or not?

That's great. Thanks, Travis. A lot of nuances of that question, so let me try to cover as much as possible. Let's maybe start with the process and give a little additional color, and then we can talk about at a point in time of what happens subsequently. I know there's been a lot of comments and thoughts around the timing of the process. I want to maybe level set the process and the timing because there's some -- I think, some maybe misunderstanding. Thunderbolt is a brand-new product. It doesn't -- it's not already approved. It doesn't -- it's not just seeking a new indications. So I know in the past, there's been comparables to products that have been already on the market that have a history that have safety information that obviously, the FDA sees and collects. And all they're looking for is a new indication.

That's a different time period usually, and those can be shorter. For products that are brand new, those processes at least in our experience with the neuro division sometimes take a while. For example, this is not fundamentally different timing than our original RED 72 silver label, which was an update on an existing product, but it was a new product with some new materials and so on. And it takes a while because the FDA and I applaud them for this are incredibly through. So there's nothing about this process and timing. But we're feeling at least hopeful and optimistic. Needless to say, without sharing competitive information about what does a launch look like. And obviously, we don't want to do that. We obviously are ready and to launch this product, and they're excited about it when that time comes that we can do that. So we'll stay optimistic. And again, got to have a little prudence until we have a clearance letter from giving you more sort of detail on that.

Okay. Makes a lot of sense. And maybe a follow-up on the storm PE secondaries now that those are out earlier this week and you've had a chance to talk to doctors curious how you're thinking about kind of the impact to the market and acceleration of the market and the potential to kind of take share given your safety profile on the device?

Yes. There's a couple of interesting and exciting things that have happened. One, the reaction to the data within the medical community has been really just incredibly positive. I think folks who were at either TCT or wear both saw that, they felt it. The safety -- the endpoints, the primary endpoint and of course, the secondary endpoints are really, really positive and strong and will have a big impact on noninterventional as well. But in addition, the safety profile of the product was really striking, and I think that's led to a number of interventionalists who have not traditionally used our product call and want to start doing cases and have done cases in the last really weeks since last Sunday with our product really for the first time in multiple cases. So that momentum is exciting as the field moves from the older technology that sort of got the field started to sort of the more modern CVT technology.

And as it relates to the noninterventional community and sort of what that reaction is which is really core to building this market and growing it. Shurith maybe can give you some additional comments. She's had a number of conversations with people and can share her perspective.

Yes. Thanks, Adam. So yes, on the non-interventional community, for example, specific conversations with like pulmonologists, hematologists, they're all now very happy that there's validation that right heart recovery then leads to -- leads to the sort of 6-minute walk test progress that we saw in the functional outcomes. And so they're now doing the work to start updating protocols within their institutions and start to see more and more patients get access to care. That was a comment, Dr. Rita Rezoski made on the podium right at the end of the presentation. and now different physicians in the noninterventional community are looking at their own hospital protocols to make those updates. Just since the release of Rachel's presentation on Monday, I've had conversations myself with physicians that have had the same reaction. So I think it's really positive for the field, and hopefully, it's going to open up access to care for a lot more patients.

And our next question comes from the line of Robbie Marcus with JPMorgan.

Great. Congratulations on a nice quarter. wanted to touch on margins here. I'll stay away from the revenue question. And reiterating gross margin and operating margin guide. Just wondering how you're thinking about the ability to drive margins? Is it is Thunderbolt to be a big component to get those moving higher versus where you were thinking originally? And how should we be thinking about the impact of tariffs and how that impacted full year '25 margins?

Let me start briefly and then Maggie can address most of those. All of our CABT products have very, very strong margins. So all of them, whether it's just Thunderbolt or any of the other areas that we've just spent a lot of time talking about have the ability as that mix changes. The other thing is with our newest coil that we launched, that's also accretive. So I think we're feeling pretty good. But Maggie can maybe go through some of the more specific.

Yes, thanks. With the remaining factors, this quarter, you see favorable, especially on a regional mix standpoint since we have seen a lot of U.S. growth. by continued product mix and also the recovery from the operation teams to stabilize our Movie Xcel yield improvement and productivity. And for the -- going into Q4, we'll continue to see this trend. I think we'll continue to see sequential improvement from product mix, regional mix and productivity improvement and also some volume leverage. I think what we have seen in this quarter already reflected some tariff impact. We do not have a material impact on tariff although still a little bit. But the number trend that we've seen has already reflected in absorbing those headwinds.

Great. Maybe just a quick follow-up. If I zoom in on thrombectomy, U.S. and OUS, but I guess more specifically U.S. here. It's been decelerating throughout the year. I imagine a lot of that has been pressure in the stroke market, offset by really strong venous growth, market growth and Penumbra growth. How do you think about if -- is it fair to assume that it will continue to downward trend? Or do you think there might be a stabilization in stroke as we move out to '26 with venous offsetting that and returning it more towards accelerating growth

Yes. Robbie, thank you. And thanks for -- in the body of your question sort of highlighting and pointing out the dynamic in the market, you're exactly accurate. Obviously, BTE has led the growth sort of quarter after quarter and some of that market growth, obviously, some significant share gain. I think arterial has also been very, very strong. Stroke over the last couple of quarters has definitely been the drag that notwithstanding, we've held our own and use that opportunity to continue to take share. So we're not negative there. we're not seeing a decline in our business. It's just not growing at the same rate, obviously.

Over the course of the many, many years we've been in the stroke market, we have seen this. This happens. This is not new, and you've known that from watching us for many, many years. It doesn't -- it comes in waves. It doesn't happen always linearly. So there's nothing at this moment to be particularly concerned about. I do think having new technology that's very, very novel, not sort of similar to others is usually been a pretty significant catalyst. And so that certainly could help, but even without that, the market typically sort of ebbs and flows a bit, and we would expect that to turn around. The fourth quarter is usually a growth quarter traditionally in terms of the volume that grows.

But overall, I'm optimistic about the future. It just comes in these waves.

And our next question comes from the line of Larry Biegelsen with Wells Fargo.

Congrats on the nice quarter here, Adam. Adam, I wanted to ask about Embo Access, which was in the U.S., which was extremely strong in Q3. And in fact, worldwide MBO access grew faster than worldwide thrombectomy. So was there anything you would characterize as onetime in Q3? And how much of the strength was Ruby XL versus Swift, which you called out? And going forward, should we expect Embo Access to now outpace thrombectomy? And I had one follow-up.

Yes. It's a great question. We're really -- to the answer the immediate question, there was nothing onetime about this quarter. The two things that we really stood out in the U.S., which was what was driving that. Well, two things around the peripheral part of the business. One, we have a dedicated team for the first time in a while. So their focus is on that. Obviously, over the last few years, our focus with our peripheral team had been split between our coils and our thrombectomy making it pretty hard to do both well. So I think that bodes well for our future that the integration, as Shruthi said, was really seamless.

And I'll be honest, I was incredibly impressed with every member of that team and the leadership of the sales team for getting to allow that to happen that way. It's not -- it's not a normal process to be that seamless and it really was. So I think we're in really good shape. I think what this shows is that there's a huge market and appetite for our very distinctive products that are different than others on the market, and I think that will continue. And I said that in my prepared remarks. As it relates to the neuro side, I'm really fascinating to watch that. The growth continues to happen there. the number of cases that people are doing in general for MMA embolization is increased notably.

And more and more people are moving to our product because of the safety profile of it compared to more traditional means of treatment. So I think we're pretty optimistic about this business. And I think, together with the thrombectomy, they both become equal drivers of growth. Obviously, you're going to see quarters the quarter that you launch a product have slightly bigger growth on a sequential basis. But overall, I think going forward, we're not going to be looking at that business is dragging down our growth. I think both Embo and thrombectomy will be an important part of our growth in '26 and beyond.

That's helpful. Adam, one quick one on the guidance. It does look like the guidance implies a deceleration in Q4. So my question is why? And the guidance for the year of 15% to 16% through 2025, it's about a 4% headwind from China. So I guess on a reported basis at 15% to 16%, is there any reason why growth would slow next year?

Yes. Look, I think we've been pretty clear about learning our lessons on guidance, not getting ahead of ourselves. I do not -- we're obviously not going to give sort of back to our guidance, if you will, for 2026. I'm not going to will give our guidance on the fourth quarter call. But obviously, if you listen carefully and you did to what we've just said about our business and where we stand, we feel particularly good about our where we stand, not just this quarter, but for a while to come well into next year and the years beyond.

Thank you. And our next question comes from the line of Joanne Wuensch with Citibank.

I wanted to dig a little bit into international sales, please. It seems based on your commentary that China headwinds are waning. At what stage are they done? And if we adjusted for said headwinds, what would OUS growth have been this quarter?

Yes, thanks. I think pretty much by next year, early next year, a lot of our headwind will be very minimal. We do have some China revenue a little bit this year, but I would look at it as a headwind for next year. In terms of our -- rest of the international region growth, we mentioned that other than outside of China, the rest of the international regions are growing in double digits. I think in the prior quarters, sometimes we have highlighted that throw back to me part of the OUS a lot of time growing in the mid-teens. So that has been kind of our trend throughout this year and still a lot of good momentum next year. Thank you.

And our next question comes from the line of Vijay Kumar with Evercore.

Congrats on the nice wins here. Adam, maybe my first one for you on the store. I think in the past, you alluded to strong periods having the potential to change clinical practice. I'm curious now that we're seeing the primary and secondary endpoints. Is this enough to change practice should we start seeing an acceleration in procedures adoption? Or are there any sort of hurdles, if you will, on the adoption of thrombectomine treating PE cases.

Yes. Look, I can only tell you what we have heard in the last week or so, lots and lots of conversations. I will tell you that a number of non interventionalists. And I think Shruthi alluded to 2 pulmonologists. She's had personal conversations with and a number of other noninterventional have indicated to us that they would be leading the effort in their particular hospitals to change their protocols based on this data. So I think the reaction the positive reaction to this data has been uniform just across interventionalists, non interventionalists.

So now the work starts to get protocols and the hospitals change. And usually, as we've said, you heard members of our steering committee say, guidelines are lagged behind practice sometimes. So I think you'll see the behavior in the protocols and hospitals change and guidelines will follow that effort. But it's been really heartening to see the excitement around this and not just with interventionalists but with non interventionalists knowing that they now have a very strong alternative for when they know their patients need something more significant. So I think it's a good day for the field and a good moment in time, and we're very optimistic about that.

Understood. Then maybe one P&L question. Gross margin is up 180 basis points Q-on-Q, with SG&A growth was quite striking up 25%. Was there any timing benefit on -- or timing impact, if you will, on SG&A, OpEx? Or is this -- is this the number being proactive as you look at the pipeline on products coming to fruition?

Yes. No, thanks for your question. On the operating expense side, most of the increases that you see is pretty much all from the investment of our embolization team and the commercial team structure. We have pretty much completed the investment or the build out by the end of this quarter or last quarter. So going forward, we'll start to see more leverage, and it will be -- the investment level will be lower than what you have seen in earlier in the year.

And our next question comes from the line of Brandon Vazquez with William Blair.

I wanted to stay on the MPE train of thought for a minute, but I'll ask this question slightly different, but it sounds like there could be kind of an initial benefit from the current interventionalists performing more mechanical thrombectomy. But one of the other tertiary benefits that might come is really that like smaller hospitals that aren't using any thrombectomy at all today. might start performing these procedures. We saw in the data that it was a very low learning curve, a lot of naive users in this trial. So the question that I'll kind of frame here is like, one, talk to us a little bit about the time lines and your expectations of kind of turning on some of these accounts, how difficult or easy that might be? And then two, once your foot is in the door with PE treatment. Are there other venous opportunities now that you've placed the lightning in these smaller hospitals that you could start to go after as well?

Yes. There's a lot to your question. So at a respectful time, I'll try to give you a fairly complete but brief answer. There's so many layers of opportunity here. There's practices that are not treating everyone, which is the most of them who now will hopefully start to change their internal protocols. There are hospitals that don't really do intervention for PE that, as you alluded to, that might now take it on, sometimes that could also have a benefit because once you're doing PE, you really are sort of related to its cousin of DVT treatments. Some of those places are using our technology for Arteria, for example, which is a little bit of a different dynamic.

So there's some overlap. All in all, what it does is really put a spotlight on the newest technology which is CBT, the speed of those cases, the safety profile of those cases. In the past, as you know, we've talked about and presented really robust data on the health economics of that, which justifies doing this. It's a procedure that is clearly validated not only clinically but also economically. So I think it just really opens the door for a lot of people. I personally had the experience at TCT of talking to a cardiologist who runs a very well-known national practice, but their group doesn't typically do PE. It's covered by another specialty in their hospital. And after hearing the data, they were particularly excited about getting in there helping their colleagues opening up the coverage capacity pretty dramatically. So -- and that was just one of many, many conversations. So it takes some time. It can't all happen in one quarter, but it puts us in a really good spot for 2026.

Okay. And a separate follow-up question here. Forgive me if you guys had kind of gone over these details already before, but this is the first I'm hearing a little bit more of Lightning Flash 3.0. Can you just spend a minute talking about timing for when that goes into a broader launch, how important it is? Where do you think we'll see the benefits of this 3.0 over 2.0?

Yes. So with 3.0, as Adam alluded to on the prepared remarks, it is really a hardware and a software update. And what it does is it really improves the fidelity. And so what that means is these cases, as you saw from Storm PE, 25-minute device times and really the fastest on the market right now. But Flash 3.0 is going to make it even faster, even better in terms of overall safety profile, blood mitigation, all the things that our physicians have now sort of come to expect. And so this really moves that even further forward by improving those case times even more.

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

Yes. Great. The first is going to be -- I understand on the guidance you want to be conservative, but the high end of that guidance is up 2% sequentially. And you did up 4.8% and 5.1% in the last year and the year before. And I'm just trying to -- is there anything specific that you're concerned about that you're only guiding up to and not the 5% that would be in line with the past 2 years?

Not something sort of I look at in all the various ways you look at it. I look at what we beat by and what we raised by. And I think that's a pretty solid guide up. We're obviously not going to want to get ahead of ourselves. And there's so much positive work to do over the next period of time. Let's give us a chance to get started. Let's get going. We're only a week into storm PE. And I think we'll feel pretty comfortable. We got couple of products to launch in peripheral and neuro strong bear with us, we'll be good.

All right. And then just to follow up again on the FDA. I truly appreciate the update -- a lot of detail today. My question is, is the FDA requiring any additional testing and for clinical data from you? Or is it a purely back in quarter at this point?

Yes. So again, I want to remain prudent. I was pretty clear about that. What I said in my prepared remarks is we have already answered very thoroughly all the questions that we had received at that point. So that has already been done. They are now in the phase where they can obviously ask for clarifications or additional questions. But -- so I can't tell you what that's going to be yet. But we've been through, obviously, for a period of time here, very, very thorough questions, like any new product, that they review in the neuro division. So totally appropriate and not totally unexpected.

And is the shutdown impacting their responses? Or are they continuing to work through.

I can't speak to that on a sort of daily basis. But as of now, we have not experienced any sense of delay because of the shutdown. But obviously, that's a process that's in real time and continues. I can't speak to that what might happen end of this week or next week or what have you.

And our next question comes from the line of Michael Sarcone with Jefferies.

Seen a lot of focus on PE and pulmonary embolism. Your U.S. VTE business continues to grow at a healthy clip. Maybe you can give us kind of the latest and greatest on trends on the DVT side of the business and how you're thinking about growth there?

Thanks. That's a great question. So on the DBT side, what we're again seeing is just a positive response to flash we have commented on the VTE segment as a whole, growing consistently the past several quarters, and that trend is expected to continue. And you're just going to see physicians as they hear about storm results, start adopting the technology on PE. And if they haven't yet used us on DVT, you're going to see that dynamic play out. So I think it just sets us up really well here for the future. Hopefully, that answers the question.

That does. That's really helpful. And just one quick follow-up there. In the past, you've talked about DBT being a little bit different in that you're focused more on the health economic side and working in collaboration with some hospital accounts. I guess do you still have a focus there? And are you going to continue to execute on that kind of health economic data on the DVT side.

Absolutely. We're starting to see those results get presented at different conferences, just some of the market access data sets, working with Vizient and the premier databases. So the physician community is starting to hear about that. We also have our hospital engagement team that is actively having concessions with different folks within the hospitals to communicate that information. And what you'll also see is, again, the benefit of storm PE applies to the overall sort of VTE segment. So you're just going to see more excitement within the hospital to treat these patients, the level 1 evidence that's now available will help on the DBT side as well.

And our next question comes from the line of Pito Chickering with Deutsche Bank.

A few follow-up, Peter, on U.S. to become. Can you talk about market share versus market growth in the quarter? And can you quantify how many new accounts you added this quarter and how you see that in the fourth quarter post the STORM trial.

Yes. Obviously, we have -- there's comparative information out there with other companies. So obviously a huge chunk of -- this is all before storm obviously came out, but a huge chunk of that was market share. And I think that's known. It's not just number of new accounts isn't really the complete way to look at it because, obviously, there number of different physicians within each account. Some use one product and other users another. So there's mix between that. So that's not a good measure. But when we look at the people who were using something is to us, that has continued all year long, obviously, at the end of last year and all the way through this year.

And really what was surprising was I guess not surprising, but pleasant to see is just in the last week because of the data and the safety profile as one of the cited reasons for their switch. We saw people who have moved from older technology to newer and all signs are that, that will continue. So it's a combination going forward of continued share shift to -- from older type of technology to the newer CABT and what will play out, I think, over the next year or so is the increase in the number of patients treated. And I think they're tied. DVT and PE play into each other that way. And I think you're going to see some growth over the next 2, 3 years.

Okay. I sit back and look at market growth, how many per teams do you think were created this year? And how is longform hospitals look post-storm trials. There seem to be a lot of excitement last TCT about more programs.

Yes. I think, look, there's two different groups of per teams. There's the formal per teams that are also members of the PERT consortium. And then there are lots and lots of sort of local teams that aren't -- haven't joined the Pert consortium, some of the time, smaller hospitals or just newer in the process. What I was very pleased to see at the beginning of TCT, per the consortium announced a pretty significant effort over the next couple of years to dramatically increase the number of members of their consortium as well as really help get sort of with the most current technology and the idea that these people can be helped in a different way.

So I think there's a lot of energy ready to go and coming out of the TCT meeting and also at Veeva, that was really clear that people are going to have either informal programs, which is a cross-functional group of people caring about these patients differently or formalize them and be part of the per consortium. Either way, I think we're seeing a huge change in energy and focus around SP-4 Pulmonary embolism patients.

And our final question comes from the line of Ryan Zimmerman with BTIG.

I appreciate that very much. So I'll ask the 2 questions upfront. We haven't talked much about the interim update on Strike out of TCT I'm wondering if you can kind of give us your high-level thoughts at on that. We look at the data, good kind of RBL changes. But I think certainly, there is some other aspects to it that may have stuck out or been on people's minds, I'm curious kind of get your high-level thoughts. And then the other question is just directed Truth, which is embolization and access. You talked about it. These markets typically don't grow at the kind of rates that you're seeing. And so is it just the products themselves? Is it higher use of MMA. I'm just curious if you could kind of talk specifically about what is going to drive this durably for the time being?

Yes. Great questions. So on the first one there with Stripe BE just like you commented on, I mean, there's obviously store is kind of the highlight right now, but StripePE certainly confirms the longer-term benefit of CABT. As you may recall, that study goes out to 1 year and looks at functional outcomes longer term, how do these patients do out to 1 year. And so it just adds to the overall body of evidence. At Veeva, there was a specific subset presented on just the use of the flash product in these patients and their follow-up out to 1 year at TCT, there was a data set around the high-risk PE patients and just showing that CIBT can have a real benefit in that patient group.

So overall, I think it's just a growing body of evidence in PE. And so I think the -- what you'll start to see is that these teams and PE teams across the board are just going to have more of a focus around treating more patients with CABT. On your second question, as it relates to the embolization products and that category as a whole, the team now is in place to focus 100% on our embolization products. And while we are market leading, we do not yet have 100% of the market. So I think the there's still opportunity to continue to grow that. Our physicians love the way the coils perform. And so the focus just allows our team to be able to go and make sure that the calls are available to everyone. On top of that, you see MMAs are sort of a newer procedure overall. And as Adam alluded to, there are some of the more traditional options like embolics that have been available, but the coils, the way they perform our coil specifically have really been -- our physicians have been really been responding well to that. So I think you're seeing here a combination of what we have with the peripheral embolization focus as well as the growing number of procedures in the MMA category.

And that concludes the question-and-answer portion of today's conference call. Ms. Furlong, I will turn the call back over to you.

Thank you, operator. On behalf of our management team, thank you all again for joining us today and for your interest in Penumbra. We look forward to updating you on our fourth quarter call.

And ladies and gentlemen, this concludes today's call, and we thank you for your participation. You may now disconnect.

All right. I want to welcome everyone. My name is Adam Elsesser. We're webcasting this event. But welcome to the follow-on discussion of STORM-PE. We have members and steering committee and as well as just Dr. Benenati, and I'm going to not talk much, but turn it over to Shruthi Narayan, our President, and she will make the proper instructions, and we'll dive into the conversation. Thank you.

Thank you, Adam. Thank you all for joining us this afternoon to talk more about STORM-PE, the data we just presented a couple of hours ago. Needless to say, it's a very exciting time for PE patients, for PE intervention as a whole and certainly for us here at Penumbra. As you heard from Dr. Lookstein as well as on the panel today is a landmark trial and foundational evidence here. And hopefully, we will all have the opportunity to be involved in treating more and more PE patients.

I do want to just introduce the members of the STORM-PE steering committee that are here with us today. We have Dr. Rob Lookstein, who is from Mount Sinai, New York, the international co-PI of the STORM-PE trial. We have Dr. Rachel Rosovsky, hematologist from Mass General, also international co-PI of the STORM-PE trial. And then we have Dr. Stavros Konstantinides from University Mainz who's also on the steering committee and provides global sort of representation. Along with that, other than Adam and I from Penumbra, we have Dr. Jim Beninati, our CMO, that will also moderate some of the discussion.

In terms of format, we're going to actually first have Dr. Lookstein go through just the highlights from the study that we just presented earlier, and then we'll open it up for Q&A with Adam and Dr. B moderating the Q&A session.

I do want to reiterate that this event is only going to be discussing STORM-PE results that were presented today. Any questions about the business, we will defer to our third quarter earnings call in early November. And with that, I will turn it over to Dr. Lookstein.

Thank you, Shruti. Thank you, Adam. It's a pleasure to be here in front of all of you. Again, a great day for advancing pulmonary embolism care. Can we go to the next slide, please. I'm advancing it?

She's advancing the screen. You're doing just [indiscernible] They're not connected.

Okay. Thank you. Sorry about that. So this idea has been brewing for 10 years. The endovascular era for acute pulmonary embolism started 10 years ago with the ULTIMA trial. And despite the fact that there have been 7 devices that have been approved by the U.S. FDA, there have been no trials since ULTIMA that have compared any form of endovascular therapy to anticoagulation alone, which is currently the gold standard of care and in the guidelines for intermediate high-risk pulmonary embolism. There have been no randomized trials comparing mechanical thrombectomy versus anticoagulation. And so this trial that we presented today is the first randomized trial evaluating any endovascular therapy versus anticoagulation in over 10 years. And it's the first randomized trial comparing mechanical thrombectomy versus anticoagulation. And obviously, the mechanical thrombectomy device that was used as part of this trial is the CAVT, the lightning flash system by Penumbra, who's here today.

We chose the RV to LV ratio as the primary endpoint because it's been adjudicated in countless studies in the cardiovascular literature to denote strain on the heart. The higher the RV ratio, the more strain there is from the heart from the acute pulmonary embolism. And so the idea here was very focused. We set out to demonstrate that the introduction of a CAVT procedure on top of the gold standard anticoagulation could reduce the strain on the heart faster and in a more uniform way than just treating the patient with anticoagulation alone.

We use multiple safety endpoints: A composite endpoint of clinical deterioration; PE-related mortality, symptomatic recurrent PE; and major bleeding. The primary efficacy endpoint was assessed by an independent blinded core lab; and the secondary outcomes here, the composite safety endpoint, was assessed by an independent clinical events committee.

The study schedule and data collection are listed here. There are numerous functional endpoints and quality of life endpoints that will be presented later at a subsequent meeting. We're very excited to share that with you, but that was not presented this afternoon. Suffice it to say, the patients, after being assessed, were randomized to either again, continue to receive anticoagulation alone or to undergo the CAVT procedure, and the patients were followed for a total of 90 days.

The baseline assessments of these patients in both arms matched almost perfectly. They were well matched grouped. There were no significant medical comorbidities between the 2 groups. When they presented and were assessed, they had similar vital signs, they had similar presenting symptoms, they had similar medical comorbidities. All the assessments that are been shown to have a poor prognosis for outcomes, we're well matched between the 2 groups. And the tools that we, as physicians, use to denote the degree of thrombotic burden or the clot burden in the lungs was also matched between the 2 groups. We set out to see if there was going to be a treatment effect by adding on the CAVT procedure.

The endpoint is shown here. It showed that there was a reduction in the RV to LV ratio in both groups but the reduction in the CAVT group was significantly superior to anticoagulation alone. This procedure took the strain off of the right heart in a more reproducible, more uniform and in a statistically superior way than anticoagulation alone.

The FDA has set out that a therapeutic effect is defined as reducing this ratio by 0.2. So we use that to demonstrate a therapeutic effect. We saw this in 78% of the patients treated with CAVT and only 52% of patients treated with anticoagulation. We also denote a normal RV to LV ratio showing that there is no strain on the heart as defined as the ratio being 1 or less, and this was observed in 39% of patients in the CAVT group and only 13% of patients in the anticoagulation group.

So again, just to paraphrase this, just by introducing the CAVT procedure, almost 40% of the patients returned to a normal level with no evidence of right heart strain on their follow-up CAT scan.

The safety endpoints, again, are listed here. There were 2 major adverse events in the CAVT group and 4 in the anticoagulation group. These were not statistically significant whatsoever. And so again, this is a foundational trial. It's a game changer. It's the first-of-its-kind trial that demonstrates the efficacy and superiority of mechanical thrombectomy, obviously, in this trial, the CAVT procedure over anticoagulation alone in the treatment of intermediate high-risk patients. The procedure -- the CAVT procedure dropped the pressure on the right heart uniformly across the cohort, and this led to a statistically superior reduction in the RV to LV ratio compared to anticoagulation alone. The safety profile was comparable between the 2 groups. This once and for all, establishes on a foundation level, the role of mechanical thrombectomy to treat and rescue patients with intermediate high-risk pulmonary embolism. And again, I'll say this again, our entire trial leadership are excited and looking forward to sharing our functional data and our quality-of-life data at future meetings in the next weeks and months to come.

Thank you, Rob. I appreciate that. We're going to open it up to questions in just a moment. We are -- we're going to have mics. So we have to wait only because it's webcast, and we've got to hear the question appropriately. So we're raising hands again. Oh, come on. You guys are going to make me pick. Let's see. Hold on. Why don't you pick? Okay. So first, to be fair, the first hand I saw go up was yours. Second row. Okay. And then...

Bill Plovanic, Canaccord. Thanks for sharing that. Congratulations on the study. I just want to understand or at least help me understand, how do you think the physicians will look at the fact that there were like 767 patients that were randomized and only 100 patients treated? And how do we think about that as just the opportunity? What's the true TAM? Like, I mean, if you're only treating 1 out of 7 patients or less, just any color would be helpful.

If I could correct that, 767 patients were screened to participate in a randomized interventional trial comparing a medical therapy with an interventional therapy, right? Of those 767, a very significant percentage were not eligible to participate because they were either too sick or too healthy.

The points I made before that we wanted very well matched -- very symptomatic, but well-matched patients in this trial was intentional. We had a high threshold in order to allow patients into this study. And the patients had to understand and agree to be randomized between a medical therapy and an interventional therapy.

So we screened 767 patients in order to randomize only 100, all right? 100 patients didn't get treated with CAVT, only 43 patients were treated with CAVT. This is actually a very, very good screening ratio for a large cardiovascular trial, especially a trial that randomizes medical therapy to interventional therapy. I think Dr. Beninati and I have both participated in interventional or endovascular trials where the ratio of screen patients to randomize is 10:1 or 20:1. And so we were very happy with these results that we only had to screen 767 patients to randomize 100. I don't know if you want to...

And I'd just tell you that from 30 years experience participating trials, 1:7 screening ratio is outstanding. We, as Rob said, a 1:15 ratio, we've seen in many trials. And part of that is because when you randomize someone, you have to look at a patient and say, you have to agree to a coin toss. And most -- in a single-arm study, you could say I have a new device, would you like to try that. That's an easy -- much easier question for a patient to answer then for them to have to say, I'm willing to.

There's another part of his question, if I can play the role of follow-up here. And that is what -- what is your perception? It's unrelated to the number of patient screening, which is where that got sort of on to a different topic. But what's your perception of the opportunity? How many more patients, given this trial, can we help in terms of like a percentage more?

So the prevalence of intermediate high-risk pulmonary embolism across the United States is estimated somewhere between 15% and 25% of all pulmonary embolism. The incidence of pulmonary embolism with screening with CT pulmonary and geography can be as high as 900,000 in the U.S. So I'm not going to do the arithmetic, but you're talking about tens of thousands, if not hundreds of thousands of patients just in the U.S. And then obviously, you have the opportunity to expand the reach of this disruptive therapeutic option to the entire globe. So it's a huge opportunity to really improve the care that we're providing just to this subset of acute pulmonary embolism.

I just wanted to add, up until today, if you look at the guidelines, the American Heart Association guidelines, the European guidelines, just guidelines, PERT guidelines, when you look at the intermediate high risk and how to treat those, it's all anticoagulate and then monitor and consider rescue and perfusion if the person deteriorates. So there's been no guidance up until today to help us think about, is there a better way to treat those patients. And this is the first trial, which is really a game changer because we now have evidence. We have evidence that you don't actually have to wait until these people deteriorate that you can actually do it upfront.

So I think this is the trial we've all been waiting for literally for a decade. So when you think about those patients, it's the intermediate high risk. And to your point, it's those people that had strain of their heart not just on their echo or their CAT scan, but also in the biomarkers. So it's those high risk.

And if you look at the mortality rate, we know patients that have high risk. Those are the people that are hypotensive, their mortality rate can be in the 40s, 50s of 3 months. But we also know the people that are in the intermediate risk, their mortality rate can be up to 15%. So this is a real important issue because when you think about the RV LV, that strain on the heart, when you look at those patients and you wait until they deteriorate, you wait until they become hypotensive, then if you look at that curve of who dies for mortality, it shoots way up. And so you want to prevent people from getting to that point where they're straining their heart so much they become hypotensive because the risk of mortality goes way up.

Stavros, any comments about the screen, the number of patients, the randomization process?

No, I only want to fully agree that a screen of 7:1 screening ratio is a good ratio. There are 2 things to that. And trials, as you all know, are highly regulated, right? We have very strict rules on whom to include, particularly on whom to exclude from trials. So sometimes, it could be that, first, the patients do not meet the criteria of the protocol of the trial or sometimes there are other issues, some other diseases that prevent us from randomizing these patients. So this is a good ratio.

And another part and another mark of the quality of this trial is the fact that the screening ratio was very well and consistently documented. This is not something that happens always. Sometimes some centers do not document all the patients that they screen. They say, we had a very high randomization ratio, but it has to be done in a diligent and honest manner, and this is what was done in this trial. And so we have a representative, and I would say, objective screening ratio of 7:1, and I also agree that this is a good one.

Next question.

Yes, Cecilia, please pass the mic to the next question.

Robbie Marcus from JPMorgan. Congrats on the nice trial. Two questions for me. First, how important will the secondary endpoints be? I believe mortality is one of them. Do you think this -- the RV to LV loading alone is enough to drive adoption? Do you think you'll need the secondary endpoints?

And then the second one, I'll ask up top. How do you think about guideline changes? You talked about the need, more data evidence. We have, I believe, 3 more trials coming, looking at this. The trial is great, but it's still fairly small in size. Do you think you'll need, along with the other 3, assuming they all read out positively, will that be enough to change guidelines?

So I'll answer the first part, and then I'll defer the guidelines question to my colleague down the row, Stavros, who I think has a little bit more insight to that. So we had to demonstrate the RV to LV ratio as the primary endpoint because they've never been done before. Again, just to reemphasize this issue, there's not been an endovascular trial in 10 years. There have never been one for mechanical thrombectomy, not the least of which CAVT. So that -- we had to demonstrate efficacy.

I'll say this again. We are very excited to share the functional out points with you in the near future, but we obviously can't go into details about that for obvious reasons. But we do think that the wealth of evidence that's going to be shared over the next weeks and months will be very, very exciting for the entire field. Stavros, do you want to talk about what's going to go into the guidelines and how the other trials will frame into that?

Yes. So what the guidelines need for recommending some sort of therapy is randomized controlled trials. And I must say it was thought for a long time, actually until now, that such trials are not possible for devices in PE. So people thought there will never be state-of-the-art trials, and this is why there was all this discussion because there were believers and nonbelievers, but there was no one to judge, I mean, can this now really be formally recommended. And the criteria for recommending some device or a drug are quite -- are the same amongst guidelines. So you need to run a state-of-the-art randomized control trial.

The guidelines, there is no number of patients that the guidance -- that the trial has to have in, right? So it's not the number of patients. It's the effect how relevant the effect is, and this is what Dr. Lookstein alluded to, I mean, RV to LV ratio is something that has been established for years, actually since the early '90s. And by coincidence, I didn't know it back then, I was quite young. It was first proposed in Mainz, actually, the university by my later supervisor. And then it was taken up also by the team of Professor Goldhaver here in Harvard Medical School.

So this is -- so one is that the guideline says there is superiority based on these criteria. And then if this is also a clinically relevant endpoint, and this is the case for RV LV, the guidelines will say, yes, we recommend this type of treatment versus the standard of care until now with a level of evidence.

Now when more trials come, what changes is not the recommendation, provided that the results are consistent, of course. If one trial says one and the other, then it's a different story. But if the results are consistent, then what changes is not the recommendation, it's the level of evidence. So if you have more than one randomized control trials, then you move up from level of evidence B to level of evidence A. But the recommendation can be based -- a strong recommendation can be based on a single state-of-the-art randomized control trial.

Okay. Let's grab another question.

Chris Pasquale, Nephron. I wanted to ask a long similar vein about translating these data into actual clinical adoption. Curious about the audience that needs to really be convinced for this to become -- to move that needle on what's considered standard of care. You're in front of a very interventionalist audience today. The additional data is going to be a different audience, but again, lots of folks that do procedures. With pulmonary embolism, how broad is the audience you really need to get to? Are there a lot of physicians who are not going to be reading this manuscript, not going to be seeing the slides that need to be convinced before patients can get consistent therapy?

Rachel, do you want to take that and comment about the publication? I think there's a lot of people who aren't interventionalists that this is targeted to.

Yes. So I'm not an interventionalist, I'm a hematologist. So the fact that on their steering committee and actually, the co-global PI was a noninterventionalist was intentional and very important. And I think the question you asked is also very important.

I think this is the first place that we've announced this. We will be having a publication come out once we get all of our data, the secondary endpoints as well. And then that will be -- our goal is to get that in a very high-end journal, which will then be able to reach not just interventionalists. There's also various other conferences over the next year or 2 that we will be targeting.

And then the other thing is this was in partnership with the pulmonary embolism response team consortium. And we've recently launched something with Penumbra called PRISM. And the goal of PRISM is actually to exactly what you're saying -- is to get this information out to all of the people. And as Rob alluded to, the majority of patients that were on -- put in this trial had put teams already. And if you think about a PERT teams, that's multidisciplinary, that's not just the interventionalist, that's the critical care, the pulmonologists, the emergency room physicians, the hematologist, the pharmacists, the nurses, all of those people. And so we need to think about educating all of those people about this trial so that it can be adopted.

We recently did also -- this was sponsored by a Penumbra called [indiscernible] clot, which kind of looked at once people's understanding right now of how these devices are used and we were shocked to find that actually the majority of people might not have the best understanding of where these should be used. And I think this is a huge opportunity. The fact that this trial has shown this. And when we get secondary outcomes, we'll be able to share those as well. I think that's going to launch this and allow us to really educate everybody that's taken care of. I think that's so important. And I think your question actually speaks to just raising awareness about PE in general, if you look at, well, thrombosis stated have a survey that looked at just participants who knew what a PE was and they looked at MI, about 90% of people knew what an MI was, 90% of people knew what a stroke was, only 50% of people that they surveyed knew what a PE was. So our work is cut out for us, and I think this is opening the door to allow us to really share this.

Wouldn't you say that in the PERT membership, the majority of people are noninterventionalist? This trial is done in conjunction with PERT. So as this gets through the PERT as this is to disseminate through PERT, the majority of people this is going to are noninterventionalists. So that was a specific target in the design of this trial. It's the reason we collaborated with a noninterventional group primarily. And your point, again, well taken, and it's something that we targeted and executed on completely. Next question?

Joanne Weunsch from Citibank. Thank you for presenting this data. It's impressive. I think the question I have is, to take off of everyone else's question, what has to be done. It's the additional data. We hit up on that. It's the publication. We hit upon that. Do you need changes in guidelines? Do you need changes in reimbursement? Do you need more than 100 patients? I'm trying to go from today's wonderful data presentation to adoption and what steps have to come from here to there.

I'll take a stab at that. I think that this trial -- to quote my dear colleague, opens the door. There are countless providers across the United States and around the globe that have a little bit of skepticism don't necessarily understand that this is a safe and effective therapy. And I think this trial allows us to embrace this technology as a safe and effective therapy to rescue patients that are presenting to our emergency rooms with the right heart strain.

We have to start somewhere, and I believe this is a foundational trial that allows the majority of the medical community that are treating patients with acute pulmonary embolism to recognize. Again, this is an effective therapeutic option in our armamentarium. Our hope is that more and more patients in the next weeks, months and years to come will be offered this therapy. This is the first step in the right direction to allow more and more patients to be rescued with this therapy.

Can I -- a couple of comments to address what's next. We, at Penumbra, we're obviously very, very involved over a decade ago in what happened after the trials in stroke. And it's a similar moment. What -- it was amazing to have that data. But to address your question, that's when the real work started in terms of what to do. There -- we're in a very different spot now than we were in stroke. And I say that just about structurally what the field looks like there, we had to deal with where the patient showed up, potentially moving them to other centers. We don't really have that issue here. And we're years ahead because of the work that PERT has already done, building cross-functional teams where there is an awareness and a desire to treat the group of patients with pulmonary embolism differently. So we're not looking at the same type of road blocks.

It doesn't mean we don't have to educate, get out and get that information. And we were thrilled to see the announcement that PERT made about their very ambitious program around education, training at centers, many more members coming in, the level of that engagement that PERT is going to do gives me hope that we're going to see a very different arc around the ability to treat more and more PE patients and get what -- the treatment they deserve. So I'm incredibly optimistic.

It is just groundwork. It is just the hard work of now the trial is done, how do we educate everyone, how do we get that out there. And I can tell you, we've already had -- we scheduled this at 4 because for the last 1.5 hours before 4 since the presentation, we were on many, many different calls laying out and now that we can publicly talk about it, that work that I've just alluded to.

So we haven't -- we didn't do this one first. We did that one first because it matters. No offense, please. And we're just going to do the work. We haven't stopped.

Okay. Let's see other questions. You've been waiting a long time.

I guess just a couple of follow-ups. How quickly do you think guidelines could evolve here? For this to be practice changing, it seems like awareness is the gating factor. So should we think about it being more gradual? Or are there a subset of people kind of waiting out there to get a data like this and you could switch them? And then lastly, how do you think about a class effect here?

Can I address guideline for a second and then open it up for you to disagree with what I'm going to say? Again, living through what we saw in stroke, there is no central body in the world that issues guidelines. There are lots and lots of different organizations that have guidelines suggestions, they call them different things. So there is no -- when you all asked a question about guidelines, that is not a thing that we think about in quite the same way because there's so many different organizations. Some do it every couple of years, you missed their cycle. You don't -- they're not going to do it. It doesn't mean people don't get treated. So guidelines matter. But the most important thing here in the short term for the next period of time is getting hospitals to be aware at the treatment level of what this trial says and what they could do differently. So there are different things, and we understand that difference. We're not waiting for the other. It doesn't mean we're not more than happy when organizations change their recommendation to a guideline. So again, I think I'm stating that right, but I wanted to say that up there.

Stavros has written guidelines and has been lead author on many guidelines. Could you comment?

Yes. That's how it is. I mean each guidelines is -- each society that issues guidelines had its own cycle. It could be 3, 4, 5 years. But there's always guidelines that are coming, right, that will be -- will appear, let's say, in a year from now or within a year from now. And this is also the case here. Actually, let's say, from the European point of view, this is a very good moment because the guidelines are being finalized now. And they would -- they should be final early next year and appear some a little bit -- a couple of months later. So this is a moment which is a good one, but also the American Heart Association is going on with guidelines. There are many, many societies all over the world. So you always can have the evidence needed to provide to the next guidelines that will come. And generally, although there are many guidelines in many countries or continents, generally, when the guidelines issue a recommendation, they are in agreement. So it is extremely rare that you will find, let's say, guidelines in North America, saying that guidelines, and Europe saying the opposite or disagree. So this is something when usually -- when the evidence is here, the process of evaluating this evidence of issuing a recommendation is quite similar. So we think this will not be an issue in this case.

If I may, about the class effect, okay? I think there will be a slight class effect but this is a very important concept, and I alluded to this in the presentation, the safety profile that we witnessed during this trial were completely naive operators, the 2 cases and had no device-related complications, no access site complications, no device-related transfusions, I would argue, potentially sets a new gold standard for safety. And I think that as more and more interventional operators, not only in the U.S. but around the world, think about how they're going to incorporate mechanical thrombectomy into their treatment algorithm, they're going to look at this data and look at previously published data, and they're going to question whether or not perhaps a device that they've used historically can offer them that degree of safety. And I think it's a testament to the operators that we had to participate in this trial, but it's also a testament to how uniformly they all said, this is an easy device to learn how to use. It's incredibly safe in my hands, and I think the data bears that out.

And Rob, may I mention that one of the things that distinguishes a device used in this trial from everything else, is this computer algorithm, the flash. And that algorithm, which helps regulate and modulate blood flow is only present in one device. And so whether that can be -- this data could be reproduced by others is something we'll have to see. But I think that when you look at the things that Rob mentioned and when you look at the technology that differentiates this device, I believe this device is its own class. I do not believe that we can generalize every device that's out there that work very differently. We don't function the same. So we'll have to see all that.

Can I add one thing because there was also this very interesting aspect of the question of -- on the proof of cost effectiveness. So I'm not an economist, but we have been also working on a little bit health economic evaluations. And it is very interesting to see, especially with any medical intervention, but in this case, with a catheter intervention, of course, the upfront costs. But the question is, first of all, how many life years you potentially save and how many quality adjusted life years you potentially save and how many indirect costs to reduce by having someone live better, live longer and not seek medical help again and again because they have a lot of thrombus and this causes them to go into the hospital again and again after the acute PE.

So there is -- the data that we already have is really very promising. And what we were waiting for was also the direct comparison of medical outcomes to strengthen this data. And I am very optimistic. I mean, this is being done by many specialized academic institutions, and we're also collaborating with some of those that the cost-effectiveness data that will follow after the medical data will also be very positive because there are many ways in which you can affect long-term costs also with -- by relieving the patient from the acute clot overload.

And I think that goes into your question about the functional outcomes. And I think that's why the functional outcomes are critically important because we want these people to get back to the life that they had, get back to their jobs, get back to their families. And so those will be critically important.

Yes, Cecilia, go ahead.

Philip Chickering, Deutsche Bank. So every center in this trial had a PERT team. Is that a limitation for adoption going forward? And how do you see expanding our CATV in the hospitals that don't have PERT teams?

Go ahead.

Do you want to go for it?

Well, first of all, not every center did. The majority of centers did have a PERT team. And I think that with the new PRISM program with PERT, we will be expanding the number of centers really around the world. So PERT started in the United States. we now have PERT international. So we already have 12 countries, and we have, I think, 50 more that are waiting to sign up. And when you think about PERT teams, even if somebody does not have a formalized PERT team, I think the way that people think about treating PE is very different because it's not -- most places is not one person making that decision, it's more than one. So right, it's the ER doctor that sees the patient and calls the interventionalists or the ER doctor sees the patient calls the IC person. So even though they might not call it a formal PERT, it's not one person. So I do think that this can be adopted. Whether centers are calling themselves PERT or not, I do think this opens the opportunity to allow people to think about adopting PERTs and actually formalizing those very easy to do. And again, the PRISM program is really going to help us with that.

And if I can add, having a PERT program was not a requirement for this trial. It wasn't really even a consideration. It just is data we collected. And the second thing I'd mention is that when we talk about the PERT consortium, that's a formal consortium and this formal recognition of programs. But the term PERT is a vernacular term that everybody that does PE uses now. So if there's just 2 doctors in a hospital working together, they call themselves PERT. That doesn't even mean that they're registered with the consortium. So we look at it, it's a vernacular term that's used widespread. So I don't -- just again, to emphasize, we did not seek anybody out that had PERT team is just the way they define themselves.

And it stands for -- I think everybody knows this, but pulmonary embolism response team. So as long as there's a team at the institution to take this on, they should be able to sort of triage patients and determine the appropriate course of treatment.

But again, just to reiterate, you didn't have to have a PERT. And if you think about it, when PEs are identified, they're not identified, no offense, by the interventionalists, right? They're identified in the ER, they're identified on the wards, the internal medicine, hospitalist, pulmonary. So automatically, it's more than one person. So even though it's technically called a PERT team, that is what they're doing.

Next question. Do you want to -- yes. But after this, we have to go to the back a little bit.

Rich Newitter, Truist Securities. And congrats on the data. Just one, were you surprised by anything positively or negatively in the data relative to what your expectations were? And I'm going to -- well, let me let you ask that, and then I have one that I wanted to ask about.

So I, for one, I think I mentioned this on the podium earlier today. I was really impressed with the safety profile. And I think that, that was something that we were obviously all optimistic, but the fact that this was a new technology, the fact that we all on the trial leadership believed in the technology and the fact that we were able to get it into the operator's hands with almost no experience, and it had such a high safety profile. That was the most impressive data. We were all optimistic that we were going to show that removing the clot out of the heart was going to offload or take the strain off of the right heart the safety profile was just so impressive and so uniform that was -- it was very pleasing to see with a novel technology.

I want to -- for those of you who weren't in the room watching the presentation, that was actually highlighted not even by people in the trial, but by Dr. Simski who wasn't in the trial, but was one of the moderators of this session. And he commented on that safety profile and the fact that with very little experience, this could -- the operators could have these results. He then asked the question, does that mean that this can democratize the procedure at a pace that's much faster? So somebody who's not associated with trial has a desire for it to be successful, if you will, had that observation in real time. That was great and a pleasant surprise because I think that's how we have felt about this technology. That was the point of developing this technology to make it not only faster but safer, and to see that obvious in the presented data, I think, meant a lot to us. So thank you.

Control arm, the anticoagulation arm did. Were you surprised at all by the amount of benefit and RV to LV ratio in that arm of the trial? And how does that stack up to kind of real-world experience?

Yes. So we know that anticoagulation works very well. That's why it's the mainstay of therapy. There was a trial that we did with a few years ago where we looked at the thrombus burden decrease from when they got the anticoagulation plus this TAF inhibitor. And we found that it was -- that was a negative trial, a different trial. But in that trial at 24 hours, there was a 40% decrease risk in thrombus burden in anticoagulation arm and the other arm. So we were not surprised to see that there was going to be a decrease. We were hoping that there would be. It's just a dramatic difference between the CAVT arm plus anticoagulation versus anticoagulation alone. We know anticoagulation -- anticoagulation doesn't cause the thrombus to decrease. It's your own body of resorbing that clot, Anticoagulation is preventing you from getting new clots. So over time, we do expect that to happen. And we often see anticoagulation making relief, but not to the degree that, that combination did. So I wasn't surprised to see that. In fact, I'd be shocked if there was no change with anticoagulation alone we were looking to see if there was a significant decrease and there was by over 2.3%.

Rich, can you comment because we've spoken offline about this about the quality of the anticoagulation delivered as part of this trial. And how it compares to other anticoagulation trials and then how it compares to your practice as a hematologist of whether or not this level of quality of anticoagulation is representative of real-world patient care. And then how you can impute the results.

If you were at the presentation, you saw that Rob put up a slide where 80% of patients had full dose anticoagulation at 48 hours. That is pretty remarkable because in general, the way this country -- world treats PE, a lot of people use unfractionated heparin versus logran fracturing heparins through the IV, you have to follow levels. And we've done studies in our own institution where we found that at 48 hours, less than 50% of people were fully anticoagulated. So if you can't even get these people with pretty high-risk PEs fully anticoagulated, we're in trouble. And we are not the only ones who have done that.

So the fact that this trial was able to get -- make sure that people will fully anticoagulated, we've moved away from unfractionated heparin, we're now doing heparin, so we're getting more and more people therapeutic. But I think this is one of the strengths of this trial is that there was no difference in that quality of getting people fully anticoagulated. Because if there was, that would have been a real confounder, we would not have been able to say this is what we found, and this is the reason. So we've taken that off the table.

And also, it was exactly a benefit of this study that for the first time, there was a direct comparison to optimal anticoagulation. Because until now, studies said with the catheter, the thrombus was effectively removed, and that was great. but no one knew how things would have happened if there was also an arm that was a control arm. And this shows that optimal anticoagulation also works, I mean, in some patients, but the difference was of a factor between 2.3 and 3:1 depending on which parameter you were looking at, which is quite remarkable. That means as many as 3x more patients can benefit from that in as many as 3x more patients, you can prevent the compensation or you prevent death. 48 hours later, you know, but in the beginning, you don't know who of those patients who will be treated with anticoagulation alone and again, we are talking about severe PE here, right? We're not talking about pulmonary embolism. Who of them will survive, who of them will decompensate. So the difference itself is something which is now shown for the first time directly. Until now, there were comparisons to historical controls, which are not correct or comparison to what many people thought that anticoagulation would achieve. And in real life, that's also the problem. As was highlighted, anticoagulation is bad. I mean, all those of us who are doing we've been in ICUs, and we know that achieving therapeutic anticoagulation within 6 hours approximately as was in this trial is totally utopic. I mean, in most cases, you achieve it within 24 to 48 hours. That's unfortunately realized. Anticoagulation is difficult even with low molecular weight heparins, it's getting a little better.

Next question

Brandon Vasquez from William Blair. I wanted to go back to a question from the start of the conversations on kind of the enrollment criteria. You made a comment that some patients were either too healthy or too sick. Can you spend a minute on rough proportions of how many were healthy, too healthy, too sick and then where those patients are going after? Are they getting maybe the sick patients are getting mechanical thrombectomy anyway. And then maybe we've talked a lot about this being a foundation. Outside of commercial execution, I assume there's more clinical data to build on this foundation. What are the clinical data sets to expect next? And what are kind of the time lines for those?

I'll take a stab at that. So patients who were more symptomatic or more ill than the patients that we enrolled in this trial would be widely considered to be high risk or the highest risk. Those mortality rates are as high as 25% in that cohort. There are guidelines that say those patients need to be reperfused immediately. And there's a multitude of different options available to that. So it's not necessarily ethical to randomize those patients to anticoagulation alone, all right? We were looking for a very uniform cohort because we were trying to really demonstrate this treatment effect. And as Rachel clearly stated, these patients got best-in-class anticoagulation strategy, all right? And so we showed that we were superior best-in-class anticoagulation strategy. To do that, we needed a very uniform cohort in terms of symptoms in order to demonstrate that. The more that we allowed a very heterogeneous patient population in, it would question the durability of the results or the applicability of these results here. So again, the fact that it took screening 7 patients to enroll 1, we were enrolling the right patients to demonstrate the treatment effect in a specific risk categorization against best medical therapy. I hope that answers your question.

What about the other end sectors [indiscernible]

So we did not study the minimally symptomatic patients. These patients were very sick as demonstrated not only by a very abnormal size of their right ventricle on the CAT scan, but also that their blood work showed that their heart was under significant strain. So these are the patients that are right under that high-risk category. And still to this day, in 2025, the guidelines say they should be treated first with anticoagulation alone. And so that's the cohort that we wanted to study here. Obviously, future trials can and will randomize patients that are much less symptomatic to try and demonstrate benefit for that. But we were studying a very specific cohort of patients in this trial.

Mostly because this has not been looked at. And again, if you look at the guidelines, it was a huge gap in care that we have now filled.

And just to clarify for everybody, that is the group of patients that we've been talking about, which is the acute intermediate high-risk PE patients that were what were matched up in both arms to then determine the treatment effect of both therapies.

Sure. Right.

Mike Kratky from Leerink. A lot has been asked about the different new physicians and centers using mechanical thrombectomy versus the existing ones. As we think about the near-term additional impact from a commercial standpoint, where do you expect to see that come from between new physicians versus existing ones that are now more comfortable using this in that intermediate high-risk bucket.

Can I -- I think there's -- we have to talk about what physician groups we're talking about -- we're not so much talking about intervention list at that level. And I want to make sure that I heard the question right. I think most interventionalists would treat any of these patients that are referred to them. I don't -- I'm not aware of interventionalists who would say, an ER doctor or pulmonologist refer a patient to be treated and they said, you know what, I'm not going to treat them. But I could be wrong. So what we're really talking about is the group of physicians who are seeing them on the front line. And I don't think one can judge sort of their reaction. I mean we just have to educate what the data is, but I could be wrong.

No, I would agree with that. I mean, I think when we -- when you think about who the first person has seen -- if -- are you asking in general providers? Or are you asking interventionists specifically and how we're going to get this message out to interventional or how we're getting it out to non-interventionalists?

[indiscernible]

Yes. Yes. Yes. Well, I think an earlier question was asked very similar to where it's about education, right? And if you go back to the stroke data and where we've come with stroke, if you look at the first, even MI and stroke, how many years it took to get the first person thinking about something to do interventional, it was like 20, 30 years. ULTIMA trial was...

30 is about right.

Okay. 30. ULTIMA was 10 years ago. So in 10 years, we have like closed that gap. So it's all about education. And I think one of the things that I mentioned is the PRISM program, which is being by [indiscernible], Penumbra is one of the sponsors of that program, but I really think it's about education and thinking of where are those front lines. In the front lines as the ER doctors, the front lines are the hospitals in internal medicine folks pulmonary critical care, and it's targeting those people and really educating them. And it's about educating PE in general, right, because there are so many -- PEs in the hospital a lot. So I think it all goes down to education and figuring out all these different venues through that's where that's going to happen. And we've already done that through PERT, and we've already expanded that. So I think it's we already have kind of a platform for that.

Yes. The only final point I want to add to that is that these non-intervention are not going to now start doing mechanical thrombectomy, right? What this will then do is have them now refer the patient to the interventionist at their institution to now treat the patient with CAVT. And that's sort of the education and the awareness that is being done through this initiative and -- but feel free to add any...

So I'll just [indiscernible] slightly as an interventionalists. I think that interventionalists that have not previously treated acute pulmonary embolism will look at this trial data and look at the efficacy and the safety and the ease of use and think maybe I can participate, maybe I can get involved, maybe I should reconsider whether I'm going to participate in the care of these patients. I'm already getting contacts from my peer group around the United States saying that data is pretty impressive. Let's talk when you get back to in New York. So I think that the safety and the efficacy are going to resonate with a lot of interventionalists that previously had not considered getting involved with this work.

Okay.

Thank you, Mike Sarcone, Jefferies. Just a follow-up here on the interplay between this kind of cross-functional collaboration and like, call it, an official PERT team. I think the example was used, you could just have 2 doctors collaborating. I wonder if you could unpack how meaningful can adoption be if it is just kind of a loose collaboration versus do you really need more protocolized kind of systemic from the top down implementation of this.

I'll try that. So again, the statistics, the epidemiologic studies show that somewhere between 15% to 25% of all PE patients in the United States fall into this category, all right? And so I think this trial from a foundational point of view demonstrates that there is a therapy that is better than anticoagulation alone. I think you're going to see more frontline providers start to consider this as a therapeutic option. Interventionalists are going to be increasingly warming up to this. And I think you're going to see organically more and more partnerships form between noninterventionalists and interventionalists.

This is going to be organic. The same thing happened with stroke. Again, I think Adam spoke to that very eloquently. And I think you're going to see increasing adoption, increasing growth in the months and years to come. Obviously, Stavros mentioned that there's 3 other randomized trials underway. This is one giant step in the right direction to allow this technology to be offered in more patients around the globe.

I think we have time for one more question.

Dave Rescott with Baird. I wanted to ask about the technology itself, right? We've heard some feedback from physicians around having potential to do thrombectomy in some more of the tertiary centers, but they're referring those patients to maybe some centers that have higher volumes that are out there. And what we've heard even just post the data is that the potential to have more, I guess, tertiary is the word I'm going to use, centers doing this does fall back on maybe the ease of use of the technology and the understanding that they have been using this type of technology in other vascular states, which potentially means that the Penumbra technology itself, the size of the catheter is going to be the catheter driving that category adoption. So can you help us think about maybe the benefit that Penumbra has in driving this -- so the Penumbra catheter has in driving this expansion out there versus the broader kind of thrombectomy category.

I can -- I'll just start and Rob and others formulate their opinion. But your point is correct. We have products that work obviously in stroke and coronary for DVT and for arterial. And so fundamentally, the CAVT process is identical. So the experience that you can gain from one vascular bed can translate to the other. And I think that, that's a huge factor. In fact, we do have -- we have a lot of doctors who start -- because pulmonary embolism therapy in most people's mind holds a higher risk, a lot of people like to start in other areas, in DVT, for example, and they get their experience and their confidence. And the good news is in our other trials, we see similar complication rates and adverse events rates which are very low. So we do think that this is a very positive factor for us to be able to help democratize this procedure. And the use in other vascular beds is definitely a learning -- a confidence-building level for people when they get into pulmonary embolism.

So I'll expand on that a little bit. I think that the most common pattern right now in the United States is a more centralized approach to refer the patient to a tertiary center. It's all predicated on a, I would say, a moderate safety profile with the technology, which objectively was pretty crude. And so now that we have this, what I would say, is a very sophisticated or advanced technology and you have the safety data that we've shown today, you have the opportunity to explore expanding the footprint for patient access. Again, I'll say this a dozen times, this is one giant step in the right direction, to allow more and more patients to have access to this rescue therapy that is clearly showing to be superior to anticoagulation alone, but we have to generate more data and educate more and more patients and more and more providers across the country about it.

All right. Well, we're out of time. Thank you all for coming. And of course, thank the amazing PIs and steering committee for the work of designing and really executing on this trial. It's a great day for patients as we move forward. So thank you very much.

Thank you.

Thank you.