Okay, welcome to this fireside chat with Insmed. I'm Joe Schwartz from the Biopharma Equity Research Team at Leerink. It's my pleasure to host Roger Adsett, Chief Operating Officer, for an update from the company at a super exciting time. Thanks for joining us.
Thanks for having us.
Excellent. So I guess maybe we can start by having you just give us a quick overview of the company's portfolio and recent performance lately and the catalysts that many of us are paying attention to over the next few months.
Yeah, perfect. Thanks, Joe. It's an exciting time for us, for sure. We are expecting some data readouts over the first half of the year. Starting with TPIP and our PH-ILD study. We're expecting that in the second quarter of this year. That's a safety study, as a reminder. The primary endpoints there will be adverse events. It will be how many patients we're able to get to those higher doses. There are some exploratory secondary efficacy endpoints. Those will not be ready for top-line results. We'll probably release those later on at a medical conference this year. But we expect in the second quarter for ILD to have those top-line results pre-ASPEN. We also expect to release at that time about 40 patients we expect to have in the PH trial. The blended blinded data for PH.
We expect to give an update at that time as well. And then, of course, I think we're all excited about Aspen data. We continue to track for Aspen to be released in the second half of the second quarter. And we're on track there. We put our last patient in March 31st of last year. It's a 12-month trial. So we're looking for that last patient last visit to come up sometime in that first quarter. We continue to clean data as we go. And we're very confident we'll be able to meet those timelines for release.
Okay, great. And so let's start with brensocatib. No surprise. I'll start there. But can you talk a little bit about what you learned in the phase 2 WILLOW study? How does that de-risk of ASPEN? What learnings did you take from that? And then we'll go from there.
Yeah, we think that so the WILLOW study, we were very purposeful when we ran this study. Compared to some of the other folks who have tried in bronchiectasis, we've made this a pretty robust study. So about 250-some patients in WILLOW. And the primary endpoint for WILLOW was time to first exacerbation. Secondary endpoint was rate of exacerbations. And we powered that study at an 80% power to show a 40% reduction in exacerbations. What we took forward, what we learned based on that trial was to for ASPEN, we looked at a 90% powering for a 30% reduction in exacerbations. So we increased the power for ASPEN compared to WILLOW. One of the most important things we've done, I believe, is to make sure that we're enrolling enough patients to get enough events. So enough exacerbations for the trial.
In WILLOW, we use the same definition of exacerbation as we did for ASPEN. So three of six symptoms have to worsen. So things like coughing up blood, breathlessness, fatigue, et cetera. Plus there has to be a medical intervention. So the doctor has to look at this patient and say, yes, this patient's getting worse. I need to do something. Either put them into hospitalization, increase medication, something along those lines. And so based on that definition of exacerbation, we did. To be enrolled in ASPEN, we did a 12-month lookback. And you had to have at least two exacerbations in the previous 12 months to be enrolled in ASPEN. And those were things that we required medical documentation. So it couldn't be an attestation from the patient, yep, I felt worse for twice over the last 12 months.
That had to be documented according to that criteria going in. So we feel that what we need to do is give brenzo the best chance to show that they can impact on exacerbations. We feel very comfortable that we're doing that.
Okay. And then patients enrolled in ASPEN had to have an annual exacerbation rate above two per year before entering the study. And you announced some time ago that the blended blinded rate of exacerbations in ASPEN was running around 1.12-1.15. How do you think about how much of that exacerbation reduction that's implied is being driven by brensocatib's benefit versus a potential placebo effect? We've seen placebo effects in a lot of other bronchiectasis studies. And we even saw that the placebo arm in WILLOW went down from somewhere north of two. I don't think you've disclosed what that baseline was on the lookback before people entered WILLOW. But north of two. And then they went down to like 1.37.
So how do you think about how much of the treatment effect that you're seeing on a blended blinded basis might be driven by drug versus placebo or a combination of the two?
Yeah, so the honest answer to that is you can't tell based on that. Based on the 1.12-1.15. So we released that data because we thought maybe that would be comforting to folks. Because it's what we saw in Willow. And so we were right on top. Aspen's right on top of Willow for that rate. We said that we would update if that materially changed. We have not offered an update on that exacerbation rate. I will say too, as you pointed out, as you well know, that the placebo rate for exacerbation in Willow was 1.37. Aspen, we assumed a 1.2. So we were much more conservative on assuming the placebo rate within Aspen for exacerbations. There's a lot of variability in some of the other studies that have gone. You've talked about this before and the placebo rate.
I think some of the other studies we're looking in their secondary in their phase 2 trials, they were looking at primary endpoints that were not exacerbations. They were looking at things like bacterial density. They weren't measuring exacerbations. They took that into phase 3 then to measure exacerbations. And that's where they really ran into trouble. There are also shorter trials. One of the things that we're doing for ASPEN, it's a global trial, as you well know. It's extended for 12 months. In WILLOW, it was 6 months. And what we've done is to limit the exposure to any one region or country around the world. So no more than 15% from any country. One of the things, for example, in RESPIRE 2 , about 30% of the patients were Eastern Europe or Russian. We don't have any Russian patients at all.
We had, I think it was 13% of Eastern European patients in WILLOW. We're less than that. I think we're around 10%, something like that in ASPEN. So it's very consistent. So we've been very proactive in trying to control for those variables.
Okay. And I think you said on the recent or Will said on the recent earnings call that the definition of exacerbation is pretty strict. And so that should help with the placebo effect as well. I'm wondering, besides the definition of exacerbation, what do you think are the biggest drivers of placebo effect in these bronchiectasis trials? And how much can you control for these types of things?
Yeah, I think I probably alluded to it. I think the biggest factor is you want to have folks who have enough exacerbations, who are reliable exacerbators, to give the product a chance to show that it can actually work and reduce the exacerbations. And that's where that strict definition, but also the lookback period to make sure you've got documented two or more exacerbations over the past 12 months. The other thing I think is, and again, this is why the regional distribution is so important, because you got different standard of care. And this is why Eastern Europe can be particularly controversial. Just putting those patients into a trial and they getting care can actually reduce the number of exacerbations and increase that placebo rate.
So we think that by controlling, as we just described for that, also will make a difference for us. I will say that we think we're seeing the exacerbations we need in Aspen. In order to see a result, we're very comfortable about how that is trending.
Right. In terms of the powering?
Yeah.
Yeah. I think it was announced in your recent earnings call that your payer and KOL checks suggest that achieving an exacerbation reduction of at least 15% would make brensocatib attractive for bronchiectasis patients. Can you talk a little bit about that work that you've done?
Yeah, of course. And I want to be really clear. We're not expecting a 15% exacerbation rate, right? So folks already ask us, what's clinically meaningful for physicians if you're thinking about brensocatib and bronchiectasis? And the first thing they say back to us is, number one, you've got a very safe product, right? So if you look at WILLOW, the dropout rate in the placebo was actually higher than we had in the active arms. There's nothing approved for bronchiectasis. And so that takes away a lot of the barriers to prescribing. So they've told us that 15% is what they would see as clinically meaningful product. So a clinically meaningful effect for it to be prescribed for patients. And I think that we would see rapid uptake, even at 15%. Again, we don't expect to be there. But 15% would be enough.
has described this often as a heart attack for the lung. So preventing clinical exacerbations is a meaningful endpoint. You particularly want to stop. This is a progressive disease. You want to stop those subsequent exacerbations.
Okay. And how much is the market opportunity for brensocatib dependent on or influenced by the exacerbation reduction, the degree of exacerbation reduction that it can provide for patients in the health care system?
Yeah, I think the exacerbations and this is the reason why it's the FDA wants this to be a primary endpoint. The rate of exacerbation is probably the most important clinical driver for prescribing. So the better we are at that, I think the more accepted and more widely used we would be. I think that that has an impact on pricing as well. So we've previously said that a good floor price is sort of a Fasenra-like pricing, $40,000-$45,000 a year. We've done some pricing work with a couple of different individual agencies. And they've come back with about a year apart, they come back with remarkably consistent results. The higher that we were able to reduce exacerbations, the more room we have from a pricing perspective.
Yeah. Okay. And not to get too far into the weeds, but it's a question we get a lot from investors who wonder when they look into the Willow data on a more granular basis, because you've provided some detailed data and forest plots and such. And it seems like there's a fair amount of data beyond Willow supporting the merit of neutrophil elastase as a drug target in bronchiectasis. But the response exposure data doesn't paint a totally consistent picture. Since it seems like the patients who did best had the lowest levels of neutrophil elastase at baseline. And some of the quartile analyses and other things aren't completely consistent. But then I know that there's an art and a science to collecting and analyzing sputum. And so how does the company think about the merit of neutrophil elastase as a target?
How much should we think about these types of things ahead of Aspen, because it's really all we have to hang our hat on before we actually see the Aspen data? Are these types of things to tell us that the effect is truly reliable in the next study?
Yeah, it's a great question. So as you said, this is, I think, you're referring to sort of the sputum samples that were taken and the testing for any. A couple of things I would say about that. First of all, as you alluded to, sputum is really tricky to work with. So the samples you would get from patients vary significantly. So how well they're able to produce the sputum. And then talking to the physicians, sputum is tricky to work with, depending on how you take that sample. The other thing I would say is that the assay we use to measure NE is a research-grade assay. It is not. It's not commercially a great assay. It's not like you can use for eosinophilic asthma or something along those lines. It's still under development. So you get wide variation there.
The other thing I would say is that DPP1 inhibition, of course, NE is what we think is driving this. But it also inhibits other NSPs. And so the contribution of some of those other NSPs to the exacerbation is not really clear. So I think the better way to select patients is not by that NE level or the other NSPs. It's really by exacerbation. What's clinically driving these patients? And that's what I think we've done.
Right. And I know that you've been stratifying enrollment in ASPEN to try to balance out the proportion of hyper-exacerbators that are in each of the different arms, because there was an imbalance in WILLOW. How did you settle on three as the stratification threshold? And what was the range of exacerbations that you saw in WILLOW? Because some investors wonder whether you could still have an imbalance in the number of hyper-exacerbators if there's more patients with four or more exacerbations. What is your understanding of what this population looks like? And how did you settle on that stratification strategy?
Yeah, so great question. And you're absolutely right. What we saw between the placebo, the 10, which had a very consistent rate of hyper-exacerbators, and the 25, which had an outlier, I think we were 41%, something like that, with the hyper-exacerbators in the 25 arm. The way to adjust for that is to stratify. And we picked a stratification at three or more. Not totally arbitrary, because we looked at the data. And about 30% of our patients had three or more exacerbations. So if we go up more, if we say four or five, then you're reducing those number of patients. And so it's much, I think, is much more effective to distribute those three or more across those arms to make sure we got that balance.
If we get four or five or even six or something across a 1,600 patient trial, we don't think that that's going to drive the result in this time. It really is stratifying above those three or more.
Okay. Very helpful. Then lastly, on this topic, what is the powering and stats plan for ASPEN relative to the p-value thresholds that you've shared? I think the company has said that a p-value less than 0.01 is a clear win. Something between 0.01 and 0.05 could be a review issue. How is the trial designed to produce p-values in those ranges based on different exacerbation rate reductions?
Yeah. So to reiterate the powering assumptions, in WILLOW, we were 80% powered for a 40% reduction in exacerbation. In ASPEN, we were 90% powered for a 30% reduction in exacerbation. We used something and I'm not a statistician, Joe. So this is sort of reaching back into business school for me. And it wasn't Biostats. So it's an enhanced mixture model, which uses a truncated Hochberg model, statistical approach. And as it's been explained to me, is that that allows us to recycle alpha across our primary endpoints and also our secondary endpoints. And so the FDA has signed off on that approach. And it's been used previously. And so we're very comfortable that this is a very powerful statistical approach for this kind of a situation. To reiterate what you just said, we're shooting for the 0.01, because we're doing one very large trial in our phase 3.
If either one of those doses, either 10 or 25, hits the 0.01 statistical significance, we have a product. The FDA has told us if we're at 0.05, it'll be a review issue. Because keep in mind, we actually had a pretty large, well-controlled study in Willow. And so they would look at both of those trials.
Okay. Super helpful. And maybe one more question. We have to talk about this market opportunity, because it's untapped. And it overlaps with some other very big markets. So can you spend a bit talking about how you are planning to address such a big market opportunity? And how does that overlap work? Where, in particular, which patients and how many patients with things like COPD and asthma are impacted significantly enough by bronchiectasis symptoms that they would be candidates for the drug? And how are they diagnosed? And who would be the most likely patients to reach for the product first?
Yeah, it's a great question. So as we've talked about, there's about 1 million patients already diagnosed, confirmed diagnosis of bronchiectasis around the world, including 450,000 here in the U.S. So that's a large patient opportunity.
CT scans.
CT scans diagnose bronchiectasis. There is a code, there's a diagnostic code. So we can go and we can check the databases and confirm that they've been diagnosed with bronchiectasis. I think what you're alluding to is there's other respiratory conditions where bronchiectasis may be a comorbid condition and diagnosed. So COPD, for example, the literature says anywhere from 4%-54% of patients with COPD also have bronchiectasis. Let's just take the lower end of that, right? That's an enormous opportunity if you think about the proportion of patients who have COPD in this country. And assuming that bronchiectasis would play a role in driving that, I think that that's an untapped opportunity, primarily patients who are uncontrolled, making their way to pulmonologists with COPD.
If there's now a treatment approved for bronchiectasis and you're able to show an impact on exacerbations, I think it would make a lot of sense for a pulmonologist to say, let's get an image and make sure it's not bronchiectasis driving this.
Okay. Yeah, that makes sense. And so maybe we can shift gears to the treprostinil inhalation program, TPIP. What will you report when you report the next look at the phase 2 data in ILD? And I think you'll also be sharing some PAH data on a blind basis too. So can you walk us through what that will entail?
Of course. Yes. So in the Q2 , and again, we're expecting this to be before ASPEN. So in the second quarter, we expect to read out top-line results from a PH-ILD study. So this was a safety study, 39 patients, so very small. But what we expect to see here is how well tolerated the product is. And how many patients are able to get up to that 640-microgram dose. And so that's the data we will be releasing on ILD. There are some exploratory secondary endpoints that will take more time for that data to come in. We're not going to hold up those top-line results for that data. We'll just release them later at a medical conference, just to set expectations. A study this small, we don't expect to see anything sort of on lung function, six-minute walk test. There's more noise around that.
But safety is what we're going to be reading out. Then for PAH, we're anticipating around 40 or so patients that we'll be able to report on for PAH in a blended, blinded fashion. Again, we'll have that PVR data, how many patients are able to get to those higher doses. It'll still be blended and blinded, but it'll at least give us progress as to where we're headed with that program.
Okay. I know that one of the long-standing goals in PAH and ILD have been to deliver higher concentrations of treprostinil less frequently. I'm wondering, what do we know about if that is achievable, how that should translate into improvements in important endpoints like PVR and six-minute walk?
Yeah, I think that the physicians, and you're absolutely right, one of the struggles has been how do I get my patients to take enough of this treprostinil in order to make a difference? And as they're struggling, they treat to clinical effect. Am I getting enough in there in order for the patients to improve? We think that, obviously, with Tyvaso and Liquidia, that's a four times a day product. And so that's a burden for the patient. It's certainly been improved with dry powder, at least from a convenience perspective. You don't have to have the burden of the nebulizer there. But I think that the ability to get to the higher doses, what KOLs have told us, is this may change the way that I use treprostinil, right?
So if I've got a once-daily product that's very safe, very well tolerated, and I can dose it up to the levels that it appears that we could be able to dose, that may change the way as how do I think about treating the patients? For example, instead of going to an IV treprostinil where you're giving this continuously on a pump for these patients, maybe I can get to those dose levels with an inhaled treprostinil that will allow me to avoid that and to treat them more successfully.
Is the thought that there's a certain threshold that patients need to get to in order to have an adequate benefit? Or is the thought that the objective is to just dose the highest that could be achieved? And does that creep up over time? Or how do you foresee the adoption evolving in the real world?
Yeah, I think that the dosing to as high as you can get is really an evolution of how do I get my patient to feel better? How do I get enough drug in them to make them feel better? So it's not actually getting to that level of PVR. They don't measure that. That's not how they judge the therapeutic success. They get to are they able to walk longer? Are they feeling better? And so being able to dose to that effect, I think, with TPIP is going to be a real boom in practice.
Okay. And then what about getting to market? If all goes well and these blinded looks, that's encouraging. But then what are you hoping to see in the phase two and three trials in order to build a compelling case for the FDA and clinicians?
Yeah, so the primary endpoint that the FDA is looking for is six-minute walk test. So we'll certainly be measuring PVR as we go forward. And that's, we think, is a good indicator for translating into clinical benefit. But six-minute walk test will be the primary endpoint we expect for phase 3. We, of course, need to talk to the FDA and engage with them on that. But that's been the gold standard to show that you can have an impact on how a patient feels, functions, survives.
Right. Okay. Great. Well, I don't want to ignore ARIKAYCE , especially in the frontline setting, which I know you're working on trying to progress that program forward. So can you tell us what you saw in ENCORE and how you're working with the FDA in order to have them entertain the idea that data might suffice to form the basis of an application?
Yeah, absolutely. We were really excited about the results we saw in Arise. Just to remind everybody, the FDA requires us to develop a PRO, a patient-reported outcome tool, in order to get full approval of ARIKAYCE in frontline setting. What we saw in Arise was, we believe, we have a validated PRO. We submitted that down to the FDA and have had some very positive interactions with them. They gave us positive written feedback in the fourth quarter. We expect probably in the second quarter of this year to finalize with them the statistical plan, the endpoint that we're looking at, and move it forward. We're expecting with ENCORE to have top-line results in 2025. We're still on track for that. We think that based on the feedback and what we're seeing from PRO, we're very confident around ENCORE.
Okay. How does ENCORE differ from Arise?
It's just a longer trial. It's a larger trial. It's a longer trial. A 12-month readout. It's still measuring the PRO as we've measured it in Arise, just over a longer period of time. Just a reminder that the FDA is unique in that they're the only agency that's looking for a PRO for approval. We'll also be collecting culture conversion data in ENCORE. That's what the Japanese market is looking for. They're looking for culture conversion at that 12-month, but then also a durable culture conversion.
Okay. And then speaking of Japan, it looks like sales in Japan are coming through really well now. So how much is Japan driving the growth that you expect to see in 2024? Can you talk about your expectations in the different territories for ARIKAYCE and refractory MAC?
Yeah, sure, of course. So last year, we put up $305.2 million in revenue. This year, we guided $340-$360. So that's, again, double-digit growth for this year. You're absolutely right. Japan had a great fourth quarter, double-digit growth there. We saw the second half accelerate as we had expected. So in the first half of 2023, in June, we had a price decrease in Japan that was planned. So it was a 9% price decrease. And we had some FX currency headwinds as the yen was fairly weak last year. And still, Japan managed to put up the double-digit fourth quarter, so really strong. I would say that we expect that to continue in Japan. We expect them to continue to grow. We've got a very strong, capable management team there. They're continuing to do some things, some innovative things.
One of the things we're piloting is, as in the U.S., we had the ARIKAYCE trainers to go to train patients on the device. Japan is much harder to do that. But we're piloting a program with an external company at the doctor's request to go and train the patients to supplement the training that physicians and nurses might otherwise give. And we'll be monitoring that. Hopefully, that will have a positive impact. But in the Q4 , we saw a contribution from Europe. That's a much smaller part of our revenue. But still, a good contribution. It's good to see the growth there. And now we're entering our sixth year on the market for ARIKAYCE in the U.S. And we continue to see growth in the U.S. I think our commercial team is executing and our medical team executing at very high levels. A lot of excitement.
And so we expect we're comfortable with that $340-$360. I do want to just remind everybody that Q1, historically, is a light quarter for us. So we've got some seasonality associated with that business. So we have the copay reset in the Q1 for the U.S. And also the fiscal budgets in Japan, the hospital budgets, they run on a fiscal year. So they've got some that ends in the Q1 . So there's some little tightening around that. So historically, the seasonality we see the first quarter represents about 20% of the overall year. But we feel very bullish about the growth opportunities in front of us.
Great. And then just to quantify the front-line opportunity, how do you feel about that now?
I feel great about it. I think the ENCORE data really outperformed my expectations, at least. I feel that the size of the population, I think we've talked about 15-20,000 refractory patients in the U.S., probably somewhere 3x-4 x that in the front-line population. I think the clear takeaway for me when I look at the ENCORE data is you shouldn't wait to treat these patients. If you just look at culture conversion, I know the FDA is interested in PRO. But the KOLs still talk about culture conversion and clearing the infection. At six months, we were able to clear 80% of culture conversion for these patients. We know that from our CONVERT trial, 29%, right? So if you wait.
The clear message here, I think, is going to be for physicians, for patients, and for payers, give the most effective therapy you can upfront. That gives your patient the best chance of success.
Yep, that makes sense. Great. Well, thank you so much for the update, Roger.
My pleasure.
Best wishes heading into all these exciting data readouts.