Roivant Sciences Ltd. ($ROIV)

Great. Thank you guys for joining us, and good morning. I guess we're still morning at the Goldman Sachs Global Healthcare Conference. I was thrilled to be joined on stage today with the Chief Executive Officer for Roivant, Matthew Gline. And so maybe first, Matt, I just wanted to ask you maybe a little bit of a higher level question, which is, the company was originally developed on the basis of this Vant model. You had distributed and uncorrelated development programs across subsidiaries. But how are you thinking now about kind of the core competencies of the Roivant business model across programs? And how has that model evolved over time?

Yes, thanks, and thanks for having us at the conference. It's great to be here. I can see that you think we're a worse and worse client by the fact that our room keeps getting smaller and smaller. It's all good. No, it's -- look, it's been a fun year. I think there are things we've always been good at. And then there are things I think we've like more recently gotten better at, which I think is sort of worth the model. The truth of the Vant model, which for those of you who don't know, so we are a $20-ish billion biotech company principally focused on developing and soon to be commercializing a portfolio of late-stage drugs. We're really excited about. In that sense, we look like basically every other company. But if you sort of look beneath the surface, we're organized in this weird portfolio model as opposed to a single sort of R&D structure is what you're referring to the Vant model, where each sort of program lives within its own company which I think makes us a real devil to model. The truth of the model -- the truth of why we're set up that way is because every time we deviate from that model, we get worse from an execution perspective. And so it's just having tested everything else that is how we run our programs best. And I don't think we have any sort of near-term plans to change that. It's just been what's working for us. In terms of what I think we're good at, I think we've always been good at asset selection, going out to the world and finding programs we're excited to work on. I think we've generally been pretty good at like indication selection. I think we've gotten better and better at that over time as we've learned more and more actually how we want to do it. And then I think recently, we've built some real capabilities in actual -- just like clinical development and execution that have gotten. Again, it's something we've gotten much better at over the last 5 or 6 years that served us well. And those things, I think, together form the core of what Roivant is, is being able to like choose programs, choose smart places hope, smart places to develop those programs, and then hopefully run those programs while we're developing them. And I hope is over the next 12 to 18 months, we'll show that we can also now launch a product in one of those indications as well.

That is an excellent segue into my first question, which was on the commercialization front. Brepocitinib is your most advanced asset on that front, and you've got a PDUFA set now for September 2026. Maybe you could just refresh us on the highlights of the data in that indication and how you think that could translate into a label?

Sure. So brepocitinib -- again, I think everyone is familiar with this, but it's a dual inhibitor of JAK1 and TYK2. It's a drug we've had since about 2021, where we in-licensed it at a time where JAK inhibitors were at sort of meter of their own development. And our view was we know exactly what's going to happen in the class. It turns out the class has done just fine. But what we do know is there's an opportunity to develop these drugs in orphan inflammatory disease that nobody else is taking. And so that's how we built our franchise. The first indication that we chose, the one that we're now sharing our initial PDUFA date for is dermatomyositis, which is an orphan inflammatory disease. It affects probably 70-ish thousand people in the U.S., of whom 40,000 show open claims database right now is actively treated terrible disease. There's really very little for these patients. Most of them are sort of living on high-dose steroids and immunosuppressants -- some of them go on IVIG, where the treatment paradigm is 5 days a month in an infusion center, 8 hours a day or on a variety of off-label things, most of which has failed studies in dermatomyositis. So that's kind of what the setup is. Now our data Look, our data is great. In a field where very few people have succeeded. Great is a complicated bar, but we've shown really nice separation on the endpoint in dermatomyositis which is thing called total improvement score. It's one of these composite scores in immunology not only that, we got responses quickly, and we were able to maintain those responses against the backdrop of a pretty significant steroid taper. So these patients were both getting a lot better and also reducing their background steroid dose, which is important because these patients come in often on quite high doses of oral prednisone .

And as you think about what will kind of show up on the label or what you'd like to see show up on the label, what are you framing as kind of like a base or best case scenario on that front?

Yes. The truth is we haven't had labeling discussions with FDA yet. So I don't really know what's going to be on the label. I don't know that I'm that worried about how the label reads. We're really the first targeted therapy that hopefully will be approved in dermatomyositis. I think there's a lot of flexibility. I think the FDA is excited about the drug in the sense they're having good constructive conversations with us, but we'll see what they think in terms of labeling when we get there, that's really a discussion to have with them. I think mostly what we need is -- look, I think ideally, we get a broad dermatomyositis label, lets us approach all of the patients. Even if there were some restrictions on that, I think it would be fine. It would be nice to get some of the data on label around steroid use in the trial. Certainly, one of the secondaries involved benefit well on low steroid doses, but also some of the clinical conduct stuff around steroids. It would be nice to have in there. I think docs are mostly at this point getting familiar with the data and the way the trial was run, the way that makes the label a little bit less important. It's pretty academic field of physicians. But regardless, I think that's all coming. We will have, in all likelihood, JAK class labeling, there will be a black box on a label for the things that exist on JAK class black boxes mortality, et cetera, but nothing that we're particularly concerned about being unusual or mattering much in this patient population.

Okay. You just mentioned that it's a pretty academic kind of group of physician's. I guess, how is that -- how are you thinking about the size of the commercial sales force you need to build? And where are you in terms of hiring that out?

Yes. So we're making great progress. So there's -- about half of the U.S. patient population or a little more than half is treated at 200 referral centers, most of which are academic. It's things like Johns Hopkins and Mayo Rochester and stuff like that at Cleveland Clinic. And they -- about half the population is treated at those 200 centers. We will cover all of those centers and then into the tail of community derms and rooms. We will have a field force numbered in the 10s, exactly where in the 10s, we haven't said, and we've made tremendous progress at hiring those in general, a pretty experienced capable sort of medical personnel with a myositis KOL herself before leaving clinical practice to work with us. It's a field force that knows in many cases, this community. A lot of the people who are taking into the field from a medical perspective, worked on the clinical trial and have now moved over into the medical field force for this sort of launch. And so they already had the relationships from clinical trial sites coming in. I think it's been a great process for us.

And how are you thinking about pricing strategy, particularly given you have a sort of other indications coming behind DM?

Yes. Look, I think all of the indications that we were studying with brepo have in common that they are between, call it, 20,000 and 150,000 patient orphan inflammatory diseases without a lot of other options, they should all sustain or support pretty similar pricing bands. So I'm not -- I think whatever we do here is going to work within the envelope of the flexibility we have going forward, it's going to work just fine. All we said publicly about price is that IVIG on the low end is a $200,000 a year-ish therapy. And if efgartigimod is successful in myositis, it will be a call it $500,000 or $600,000 net price therapy maybe. And so those are reasonable bookends. I think anywhere within that range is still on the table. To be honest, given the size of the patient population, if everyone just models the bottom, then everyone can be pleasantly surprised.

Under promised, over delivered. So in terms of the you have just mentioned -- sorry, I lost my train of thought. But in terms of the patient population, how are you thinking about like what the right peak penetration is? And how do you think about like the path to peak, given this is a rare disease? And what would you point to as analog in that market?

Yes. I don't -- the truth is with 40,000 patients at the price points I just articulated, you don't need very deep penetration to be a great foundational indication. Remember brepo is in development in three other indications, a lot of other places to go. We will certainly add indications beyond these four. So I don't think we need very high penetration [Audio Gap] incredibly enthusiastic [Audio Gap] 30%, to much higher than 30% of their patients. any of those numbers would be just fine from a penetration perspective. I think I don't have analogs on launch trajectory. I think every rare disease is its own thing, and I think slow and steady is the right way to think about the launch in any new indication until you have a better sense of how the payer and physician dynamics are going to play out. I'll say in terms of like market size, first of all, DM does feel like a market where the total patient number could grow once there's novel therapies in the market, it's somewhat -- despite the already relatively high number of diagnosed patients, it's somewhat difficult to diagnose and you get patients with lupus diagnosis or you get patients with other myositis. So I think there is an opportunity for the patient population to grow. And without intending to give any guidance on trajectory, I'll just say in terms of like analogs to the patient population. There are not a whole lot of things you would have said about the myasthenia gravis market in 2019 that you wouldn't say about DM today, right? Before the introduction of efgartigimod, there were about 40,000 MG patients in claims data sets. About 20% of them were on sort of novel therapies, much of which was IVIG like a relatively similar composition. And obviously, that market over time has shown itself enthusiastic for a new option. I hope and expect we'll see something similar in DM.

Great. You mentioned there's three other indications in development. Maybe could you speak to what each of those are and how they match the indication selection strategy you've already outlined?

The other three indications are cutaneous sarcoidosis -- let me go in order. The first next indication is noninfectious uveitis, which is a noninfectious inflammation of the eye. We're focused on back of the eye inflammation that will read out a registrational data set in the second half of this year, which would then be a second approved indication if all that is successful. That is 70,000 to 180,000 patients, depending on how you count it. It's a very messy claims landscape or a very messy diagnostic landscape. So a little bit hard to know the precise number. But a severe disease third leading cause of blindness in the United States, low tolerance for optical inflammation. So if our drug works, and we have a really good Phase II study there. I think that will be an exciting market and pretty similar in size or maybe even a little bit bigger than DM. The next indication is cutaneous sarcoidosis, where we are beginning a pivotal study this year. We had a Phase II data set readout earlier this year. That's a slightly smaller indication, maybe 20,000 to 40,000 patients, but still quite severe, no approved options, really sick patients with a really bad set of inflammatory and potentially permanently disfiguring skin symptoms. So that pivotal program will be ongoing shortly. And then the most recently announced indication is Lichen Planopilaris, which is sort of form of lichen planus that involves scale scarring of the scalp and can involve permanent hair loss, very painful, high concurrent use of opioid pain medications. It's like really tough disease for the people that have it and basically nothing approved so far. We're, I think, the leading and potentially only mechanism or only program in late-stage development at this point. So an exciting addition and that's in sort of a continuous Phase II/III study that will start -- has started is enrolling nicely already.

Could you speak to the conviction you have in the Phase III NIU study that's coming based on the data you've presented to date and if there's any sort of like commercially relevant, secondary endpoints or bar that you think matter here?

I feel pretty good about the Phase III study. I lose sleep over it because it's biotech and you have to lose sleep over everything all the time. It's just the rule. But we had quite good Phase II data, which left quite a lot of margin. So the challenge of the Phase II data is a relatively small end study and didn't have a placebo. And placebo in all these conditions are variable and have been trending up over time. HUMIRA in their Phase III study in NIU, the primary endpoint is something called time-to-treatment failure, which is what it sounds like it's how long these patients take to fail treatment. Placebo patients failed in 3 months and change in the HUMIRA study and the drug delivered just under 6 months of sort of total benefit, so 2 or 3 extra months over placebo. In our Phase II study, it was greater than 12 months. That was as far out as the study went -- and so I think a lot of cushion. Frankly, I don't even think we have to be better than HUMIRA to be commercially successful, HUMIRA fails in a lot of patients. And the HUMIRA there's about 40,000 TNF patients with NIU now. So even in the half of those patients who eventually fail, that's a pretty big market, but I expect it would grow over time. But I think based on the Phase II data, there's certainly a possibility for better than HUMIRA across time treatment failure across treatment failure rate itself. Docs look a lot at indocyanine fluoroscopy and some of these other measures. So I think those things will matter commercially as well.

So recognizing this is somewhat early and you haven't provided guidance, how do you think about like the total sales opportunity for this product across the indications that you guys have outlined?

Across all of the indications. Look, I think it goes beyond those indications, too. Again, I think if it is inflammatory and has between [ 20,000 and 150,000 ] patients, it ought to be an eligible indication for us. I think over time, we could be in quite a large number of diseases. We have IP as it stands out to 2039. If each indication we're in is a low blockbuster indication, you can just start adding things up and finding a $5 billion to $10 billion drug pretty easily. And so I think as we continue to add indications and continue to find success indication by indication, the opportunity here is very large.

All right. Let's switch gears to Immunovant program, IMVT-1402. And again, let's start at a high level. As you think about the FcRn cost and your development strategy there, like where do you see unmet need? And how did you map the development strategy to that?

Yes. It's interesting. So we had a couple of ideas in mind with our FcRn franchise. One was just that we thought we could build a best-in-class agent, right? We have a better administration than the other drugs. We have a simple subcu injection, low-volume, sort of normal similar to DUPIXENT and other very successful subcu medications. And One of the key things in FcRn is the way these drugs work is by suppressing levels of IgG and there for autoantibodies, our view has been consistently and our clinical data has shown consistently the deeper you suppress IgG, the better you treat these patients. So the world sort of split into a few different categories. There are indications where others in the FcRn space, especially Argenx sort of the leader in the space, have carved a path forward, myasthenia gravis CIDP. I think in those indications, our goal is optimistically to deliver better clinical data and to take share that way to win share in route of administration. Those are really big markets, and the studies should work. And so it's like an opportunity to have sort of a base level of just even relatively modest penetration of those indications would matter. Then there are two other categories of indications, The indications I am most excited about are places where we have carved out a path either where we were first or where we are now first by end of the way that we've developed the drug. This includes things like Graves' disease, which is probably what I would call our lead indication, which is a truly massive opportunity, hundreds of thousands of poorly controlled patients where I would say there had been -- despite the fact that there are hundreds of thousands of patients who have failed every therapeutic option available to them. No modern clinical development had been done in Graves' disease prior to the introduction of our studies. And it feels like we have just a great opportunity to provide a new option to a ton of patients. We're now mutations of finest form of flattery. We're now the leader of a large number of companies doing different things in Graves' disease. And I think that will also help the market and help us in the long term. But now we've also added to that this sort of late-line multi-mechanism failure, D2T RA, rheumatoid arthritis, where these patients have failed multiple lines of therapy. We are not the first FcRn to be studied there, but I think we have found in our view, the right patient population is borne out with our clinical data for a real opportunity in FcRn. I think in both of those, we are now kind of the pole position FcRn. And then there's a third category of indications like Sjogren's disease, where we're not first, but we're much closer to first than an MG and where we think between the time line gap being shorter and the quality of our agent, we have a chance of being sort of a class leader or at least close to the front of the pack.

Great. You just mentioned the RA data from period 1, which came in earlier than we expected, and it was definitely pretty impressive. Maybe you could walk through the highlights of that data with an emphasis on contextualizing those results in the context of this patient population.

Yes. So this is a study that we ran. It is bluntly a weird study in the design of it. We had watched J&J begin development of nipocalimab in RA. And the thesis behind an FcRn and RA is RA is principally inflammatory disease, and most of the approved drugs are anti-inflammatory drugs, but at least in some subset of RA patients, there is an autoantibody component that is it seems like some of these patients have elevated levels of certain autoantibodies that are causal. That was the idea behind development of FcRns in general. And J&J showed something. They showed that if you suppress IgG, you can deliver at least modest benefit. And what they also showed is that benefit was significantly higher in patients who were positive on autoantibody titers than patients who weren't. And because it is orthogonal to the anti-inflammatory mechanisms, early evidence from the J&J data suggested that the benefit might be preserved even in patients that have failed anti-inflammatories. So we designed a study on that basis that looked at truly refractory patients who have failed at least 2 of the 3 late-line classes, JAKs, TNFs and IL-6 is about 60% of the patients we ran up setting had failed both TNF and JAKs, which is that's like basically the sickest of the patients or at least the most refractory of the patients. And what we showed in those patients and contextually, this is in the open-label lead-in portion to a randomized withdrawal study, we showed ACR20 response rates very high in the 70s. We showed ACR50 response rates above 50%, and we showed ACR70 response rates in the mid-30s, which even in an open-label study, you just don't see a lot of placebo response in ACR70. And so it's clear that there is real activity there. We selected for a high baseline antibody -- autoantibody titer of ACPA, the specific autoantibody that we think is most likely to be causal in the disease. I think in practice, in a multi-mechanism failure population, the data could be much worse than this and still see use because these patients don't have a lot of options. And this data is just pretty exciting data.

Okay. Broad brush strokes, knowing that you're still working on it. But what does the things I kind of have to look like here with respect to trial size? And how quickly can you start moving towards that?

That is mostly a question for FDA to answer. And I think there's a pretty wide range of possible answers. We would like to see, given that we are happy to accept restrictions in terms of being late line, et cetera. Smaller study, the typical RA study is 1,500 patients or 2,000 patients and FDA has historically been concerned that everyone in RA is trying to move into earlier treatment and so they want to make sure for the size of the RA population early line therapy that the drugs are sort of safe and efficacious. Our view is, if you're looking at that later line market, you should be comfortable from a scientific perspective with smaller studies, hopefully hundreds of patients. And that's certainly the conversation we intend to have with FDA. I think the most likely thing is that we would run a placebo-controlled study of a normal sort of head-to-head versus placebo design, but there's a lot of possibilities, including attempting to replicate the existing Phase IIb randomized withdrawal trial. That's just all a conversation for us to have with FDA. It's a conversation that we will have with FDA in the second half of this year and come back hopefully with everything kind of packaged in the bow in terms of what our plan is. One thing I'll say is I think we will be among the first companies to have a discussion with FDA about late-line RA studies, but it's coming, right? There's CAR-T studies, there's NK-targeted studies. There's T cell engagers there's people doing work. And by dint FcRn being quite a safe mechanism, I think like we're in an interesting position where all of those other mechanisms that I mentioned are not going to run 1,500 patient studies. You couldn't possibly. And so I think we get to approach FDA as kind of the first in this category, with a willingness to do more than I suspect others in the sort of general indication space are going to be able to do, and we can use that flexibility to hopefully have a constructive discussion.

You mentioned it already but Graves' disease is definitely the lead indication for the development program for 1402. Maybe you can refresh us on the clinical data to date and how it answers for the unmet need you see in that patient population?

Yes. So there are -- so Graves' disease is a disease where you have [Audio Gap] a way that causes the thyroid to become overactive. There are [Audio Gap] most of them are effectively treatable with the course of an antithyroid drug of some kind. And in fact, many of them eventually, if they're treated on relatively low doses of methimazole can get under control and then get off antithyroid drugs. For about 350,000 of these patients, 330,000 of these patients, their journey just can't end there. That is either they can never get off methimazole or methimazole is insufficient to control their Graves' disease and they continue to relapse and have issues. And those patients they're sort of stuck at that point. About 20,000 of them a year wind up having their thyroid removed, either surgically via radiation or radioactive ablation. And other than that, they just live sick. They live either on high-dose methimazole that's miserable or they live with symptoms of Graves' disease or both. Our clinical data suggests that for that patient population, we can get a very high proportion of them controlled. In our Phase II study at our high dose, over 70% ultimately got controlled and somewhere between 30% and 50% depending on how you count it got off ATDs and controlled. And in fact, of the people who were controlled and off ATDs, not that, but we were able to drive remitted benefit where even after some period of time, they could get them off our drug and remain off ATDs and control. Remember, these are patients who weren't controllable on ATDs before. So this is a pretty remarkable outcome for the data. Again, paves the way for a real clinical benefit, let's call it, 350,000 patients who otherwise didn't have a path.

Great. And you have Phase III data next year. I guess, what do you need to see there for clinical and also commercial success? And could you talk about the design of the two Phase IIIs you have relative to that?

So we're running two studies. One is a relatively straightforward 6-month study. The other is a 12-month study, where the primary endpoint is at approximately 6 months, 26 weeks. And the study is designed to allow us to show a remitted benefit in the second period. So responders in the first period are then taken off drug on a randomized basis and looked at for remission. So that study will read out effectively after 52 weeks. The other study will read out after 26 weeks. In terms of what we need clinically, again, first modern therapy ever in Graves' disease, a p-value we will do it. We think that will lead to an approvable drug, knock wood. That's really all we need, I think, also for commercial success, to be honest. Again, the size of that commercial success and how the competitive field plays out will depend on what we show. If we show anything that looks even remotely like our Phase II data, it will be transformative for tens of thousands or more of these patients. I think it's a huge commercial opportunity. But even with more modest data than that, I think we can have a big effect on the second of these patients.

You mentioned the size of the number of patients you have need here. I guess, how are you thinking strategically about pricing for this lead indication relative to a broader development path across indications?

We haven't given an answer to that question. But I think in general, our mind is on pricing comparably to in the same range as other FcRns and this is a refractory population that has no other therapeutic options. And so there are definitely large subsets of that population for which I think that price point could be sustained.

How do you think about the role of remission in the patient population? And what is your expectation in terms of how long patients -- like that the impact on duration of therapy?

Yes. Look, I think on the second question, based on what we know, there will be some patients who -- probably some patients who can't even get controlled on our drug. So they're better controlled on our drug, but not fully control in our drug. I suspect there are some patients who can get fully controlled in our drug potentially in perpetuity, but will need to be on our drug in perpetuity. And some of those patients will continue to require methimazole and our drug to be controlled and some will not require methimazole at all, but will require our drug to be controlled. And then I think there are patients who will go into remission. I think that whole spectrum will exist. And so from a duration of therapy perspective, obviously, the patients who can get into remission will be shorter duration therapy, but there will be some patients for whom this is unfortunately just going to be a chronic condition, whatever. Overly simplistically, I think 1402 suppresses IgG by about 80%. If you think the equivalent is roughly -- impact is roughly equivalent on TRAb, so if you're coming in at 5x the upper limit of normal antithyroid antibodies, then we will get you to within the normal range. If you're coming in at 20x the upper limit, we won't get you to it within the normal range. And I think that's sort of a rough heuristic for like how you might think this will evolve over time.

Okay. And how are you thinking about the emerging competitive clinical landscape? Obviously, there's other FcRns coming. There's also degraders that have entered the Graves' disease market? And how do you think about those agents?

Yes. As I said earlier, imitation is the finest form of flattery. I think this is a very much -- you don't have to outrun the bay indication. And what I mean by that is first of all, we're first, but even more importantly than that, it will be years before novel agents have saturated in Graves' disease. There are hundreds of thousands of patients treated in a variety of settings. I think the main question for everybody approaching Graves' disease is just like how do we build awareness of novel therapies? How do we get out there? How do we get patients out of the uneasy existing paradigm and into a space where there are actual treatment options. Practically speaking -- so that's the first answer. I think a rising tide lift all boats. I think this field is about to be transformed, and I don't think it's about sort of jockeying for position versus other competitors and won't be for many, many, many years. Of the stuff in development, I think general IgG approaches, other FcRn, the broad IgG degraders, et cetera, are going to look pretty similar conceptually to one another. And what we can do, they can do and probably vice versa. The degraders, I'd say like the one asterisk there is obviously like they haven't been tested in large-scale clinical trials at this point. So we'll learn more about how that actually plays out in Graves' disease and generally after we see that data, and there could be some issues. But mostly, if I had to guess, they'll all be sort of fine and pretty similar. Then there are two other approaches that people are taking. One is not autoantibody driven at all. There's just people working on therapies that either directly affect the TSH receptor in one or another. There's either small molecules or antibodies. Those approaches may very well treat patients even sicker than the ones we can treat, but they will send patients from hyperthyroid to hypothyroid and so they'll either need to be carefully titrated or you'll need to concurrently dose synthetic thyroid hormone. So my guess is they will be reserved in many cases for the sicker patients who can't be treated with a more elegant autoantibody-driven mechanism. And then there are a couple of companies now that are working on autoantibody specific degradation, meaning like TRAb degraders or whatever, I think that's a really interesting idea. It is the kind of mechanism that in the fullness of time could deliver even better data than we could, but it's just early to know how possible that's going to be. And how varied the antibodies are in Graves' disease is not something we have a very clear answer to scientifically yet.

I want to switch gears again to mosliciguat. And maybe you could just remind us the Phase II results that we should expect in the second half of this year and where you get conviction in, potentially a positive outcome?

Sure. So mosliciguat is our third late-stage drug. This is an inhaled sGC activator, which we're developing in pulmonary hypertension for lung disease patients. This is a mechanism that is a cousin to sGC stimulation, which was the basis of the drug called the mechanism of the drug called Adempas, that was a collaboration between Merck and Bayer that was quite commercially successful in Group 1 pulmonary hypertension. We are developing drug in PH-ILD as an inhaled formulation. Our Phase IIb data comes later this year. PH-ILD is an indication that has risen in prominence and relevance with the commercial success of Tyvaso and with Liquidia now coming into that market as well and Insmed eventually, all with different formulations of treprostinil. It's worth noting that Adempas, the drug I mentioned before, which was a systemic sGC stimulator failed in PH-ILD and had some pretty significant safety issues. Our expectations -- first of all, what do we know? We know that mostly in Group 1 pulmonary arterial hypertension patients, healthy volunteers is a very good local vasodilator that improves cardiac output and lung function, saw very deep reductions in PVR. We know that other systemic vasodilators in PH-ILD, the treprostinil, for example, don't work. We know that other locally administered inhaled vasodilator, treprostinil for example, do work. The end of those mechanisms is small, but that is where we get our conviction. Our drug is good as an inhaled as a dilator in Group 1 patients. The other drugs that have been good as inhaled vasodilator, Group 1 patients have worked in Group 3, and that other drugs have mechanisms that have failed systemically in Group 3 when administered from a local perspective, administered on an inhaled basis have worked. That's basically what gives us comfort, there's a lot of uncertainty there, and we won't know if it works until it works. I'll say we have the benefit and the challenge that if the drug works in hindsight, everyone will say, obviously, it worked, it was just as you described, inhaled vasodilators work in Group 3 and local and systemic vasodilators don't. [Audio Gap] everyone will say, obviously, it failed. Adempas failed, why did you think an sGC [Audio Gap]. Either way, we will look like with obvious in hindsight. But in fact on an end basis, who knows.

There's the reason we play the game, right?

That's right.

Yes. So in terms of investing further behind this program, what do you need to see from the Phase IIb coming to kind of move into pivotal?

The reason, I guess, you're implying is fear or gambling addiction or something, I'm with the hope of helping patients. Sorry, what do we need to see here? Look, I think if you can improve cardiac output in these patients in a properly sized Phase III study, it will benefit them clinically on measures like 6-minute walk. This study is not powered for 6-minute walk. I don't expect to see a p-value on 6-minute walk. I don't even know what kind of delta we're going to be able to put up in 6 minute walk. It would be nice to see some kind of trend. But the truth is if we see significant benefits in PVR and cardiac output, the rest will follow. And so as long as we see that and a safe agent in Phase IIb, we will run a Phase III study. It would be nice to see something a 6-minute walk. We're not looking for p-value. I'm not expecting a p-value to say that 1 million times in public settings so that the [ goofy Bloomberg chats ] about this will at least have me to contend with. So mostly, I think we're looking for good safety, the kind of thing that would support a registrational program and a strong benefit on PVR.

Okay. Great. And how are you thinking about positioning of this agent relative to the other things that are in development? You kind of mentioned a bunch of them in PH-ILD?

PH-ILD -- pulmonary PH, Group 1 pulmonary hypertension is a polypharmacy market. Basically, every patient in the fullness of time ends up going on every mechanism or at least many, many patients do. Right now, only treprostinil approved in PH-ILD. I think PH-ILD will evolve the same way. These are really sick patients if multiple mechanisms are approved, patient will try all of them. So this isn't really an us versus them. This is -- if we present a good compelling option, people will use it. I think given the PVR reductions we saw, given we are a once-daily puff of a DPI, which makes us nice from a form factor perspective relative to the treprostinil that are currently on the market or even coming. And given that treprostinil causes cough as it's an irritant. It causes cough as an on-target side effect, and we shouldn't. I think we have a chance of early first-line use in PH-ILD, but that's not really important. What's really important alongside these other agents.

Significant capital on your balance sheet and you have an upcoming launch. Can you walk us through the path to profitability from here?

Well, I think if our launches are successful, they will be high-margin launches and will become profitable. We have not given specific guidance on timing, but I think we've got plenty of cash to get there and shouldn't need any more money to do it. We've been buying back stock pretty consistently, and we'll continue to do so.

Okay. Speaking of that could you speak to the capital allocation priorities you have between returning cash to shareholders, business development and investing in the existing pipeline?

Yes. So last year, ending the middle of last year, we bought back about $1.5 billion of stock at about $10 a share, which has proven a good investment in hindsight. We are now doing further buybacks, we've sort of ramped them up after we settled with Moderna back in March and continue to buy back stock here. That's sort of on the premise that we've articulated for years that if we can't make it with $4 billion, we probably can't make it with $6 billion. And so we feel like we've got some flexibility. And then the $4 billion, we've said consistently is more than enough to fund both all of our existing programs to profitability as well as new stuff. The new stuff is new indications as you even announced for FcRn and brepo and mostly new stuff is business development, which we're continue to be focused on. And one of these days, we'll get a deal done and even we'll be excited about it. I hope.

In terms of the business development, we have 1 minute left. So what kind of criteria are you guys using as? Have you consider these potential opportunities?

In general, we like what we like, which is late-stage programs with relatively open competitive fields with a clinical development strategy that we think is interesting and differentiated. So that's most of what we're looking for. It's most of what we brought in historically. And we continue to see lots of opportunity approximately meeting that description.

All right. I think that does it for us today. Thank you so much, Matt, for joining us. And thanks to everyone, who joined us here and online.

Thank you. Appreciate it. Thank you.