Roivant Sciences Ltd. ($ROIV)

All right. Welcome, and good morning. Thank you all for joining us. My name is Will Pickering. I cover US Biotech at Bernstein. I have the privilege of sharing the stage with Matt Gline, CEO of Roivant. The company has had a remarkable run over the past year, certainly with brepo is the standout, but far from the only value driver. We'll dig into that and the other assets over the next 50 minutes conference would also like to spend some time on bigger picture questions, the Roivant business model and Matt, your view as an asset hunter about the overall health of the biotech ecosystem today. For those in the audience, please submit your questions through the Pigeonhole app that we can make this as relevant for you as possible.

So with that as the preamble, Matt, how would you describe the evolution of Roivant over the past a few years into the company that it is today.

Yes, thanks. And thanks for having me. It's nice to be here. I appreciate the invitation. Thanks for all the work you guys have done on us. It's been a fun early run here. So look, it's always funny we add more generalist vendors, like asked to talk about Regent because I feel like there's so much complexity in our history -- we got our start as sort of whatever, these words are all overused, sort of a Maverick outsider biotech company, doing things our own way, walking our own path. We're built in kind of a funny way that I'm sure we'll talk a little bit about with the van model. And I think we do bring a pretty different philosophical lens to the industry and that we are a mixture of experienced industry insider drug developers, but also a lot of outsiders like myself. I was a physicist. I was an investment banker. A lot of my leadership team had been public markets, biotech investors before joining Roivant, private market back tech investors before joining Roivant. And so you take all that history and you hold it and it sort of creates opinions attitudes, a feeling of not fitting in as it were. And then the moment that we're at in the business actually is -- in some ways, it's the most normal moment in biotech. It's the moment where finally, we've invested for years. We've wandered through the desert. We've reached the other side. We have a portfolio of programs that I think can convert us into RECONNECT for lack of a better phrase, a real business, right? We're at this precipice of launching one and then multiple drugs and on and then multiple indications that I think stand us a real chance of building one of the next large cap biopharma companies, and that's just tremendously exciting moment to be in.

That's great. And BD has been a huge part of your approach historically, but the current pipeline is very full. So I think 10 indications across brepo, mostly Immunovant, if I'm counting roughly right. .

I trust that you have.

How much of a focus is further BD for the company today?

Yes. So I'll say, for those who don't know the company, look, I think if you're thinking about biotech from the outside, I think people take a relatively reductionist view on what sort of "science" -- and I've used like mental image of science as like a thing that happens in the lab at the bench for the white lab code on. And in fact, there's like many different kinds of science that happen within biotech. And bluntly, I make that maybe defensive sounding case because we've never been that good at the white lab code version of the activity. But we have been, I think, increasingly, and I'm super proud of this, very good at the kind of science that takes place in the clinic in doctors' offices around the world, in clinical trial design and indication selection and understanding patients and diseases and figuring out the right way to develop a drug ideally in a disease where the patients have high unmet need and where not a lot of novel science has been done. And if you're good at that latter kind of science, but don't fancy yourself that good at the former kind of science, you're sort of stuck, right? Because it turns out every drug must be discovered in some kind of lab before it is studied in the clinic. And so we have, as Will has alluded to, built basically our entire pipeline via collaboration, via BD, via acquiring programs. And in fact, another thing that I'm proud of that I think we're pretty good at in addition to sort of asset hunting, spotting things in the world that are attractive is most of our partnerships have come from big pharma companies. And I think we have built a pretty good understanding of and some very good relationships with some of the largest pharma companies in the world. And they all have a problem, which is that they do all kinds of science. And sometimes the first kind of science that happens in the lab and the second kind of science that happens in the clinic and third kind of science that happens when you're trying to build a commercial business don't all align in a nice train that makes you want to start on the same projects that you want to end on. And some big pharma companies often have really great things that fall by the wayside, not because they're not great because they don't fit with the industry's strategic commitments the companies have made. And I think what we've built a lot of our portfolio around is bringing in those programs. It is profoundly a part of our DNA to be on the hunt. And frankly, biopharma is a treading water industry. It's like whatever the -- I'm not sure it's actually scientifically accurate, but they say they have sharp stop swimming, they die. If pharma companies stop replenishing their pipeline, they die because our drugs are all wasting assets. From the minute you invent a drug, it has fixed life before the patent runs out. And so we are always on the hunt. And I think that still serves us well in developing the drugs we have now. But absolutely, we continue to look for programs, and we'll continue to bring more in. That's it, I completely agree with the comment that you made. Our pipeline is rich and exciting and there's plenty to do just within the context of the late-stage clinical programs that we're bringing to market.

And as you look at the biotech landscape today, both from an innovation perspective, but also looking at valuation. How attractive is that for an asset hunter?

I mean very -- look, I think the last really 25 years in biotech have been pretty remarkable and that we went from an industry in the early 2000s, where 99.5% of drugs were small molecule pills of the same general sort that has been around a long time. And the big story from the previous 25 years have been the invention of modern synthetic medicinal chemistry that lets us do some version of designing those drugs intentionally. To now, we have mRNA vaccines and monoclonal antibodies or use left right in center and people are injecting themselves with unapproved peptides, manufactured by small Chinese CDMOs. Like the world is crazy in terms of like the kinds of things that people are working on. And while our understanding of human biology remains what I'll call nascent across the industry, it's gotten better. And I think that has made for an explosion of ideas. And look, I think anyone living in the world knows that we are still far behind in terms of our fight versus disease illness and aging. And so with the explosion of ideas and a lot of territory left to cover, it's a good time to be an asset hunter. There's a lot out there that could be valuable. And then there's all kinds of more localized shifts, right? There's geopolitical things going on. There's the incredible pace and quality of antibody discovery, especially in China, there's the pace with which you can run early clinical trials in places like Australia, New Zealand and China. And there's a series of global and U.S. political factors, buffering big pharma that creates a need for shift in their portfolios. And I think all of that also benefits us.

Excellent. Well, why don't we dig into parts of the pipeline, start with brepo, you're approaching your first commercial launch in dermatomyositis. Could you start by framing the opportunity and then talk through the key priorities for your commercial team to drive that launch?

Yes, perfect. I get ever so slightly defensive well, maybe it's not serving me well when people call it our first commercial launch because -- in fact, we have had now 6 or 7 FDA approvals come effectively through our business. Many of them have been commercialized externally, a number of them through Sumitomo a company. We partnered with a few years ago on the first generation of our pipeline. And then we did, in fact, launch a drug called VTAMA in psoriasis. I'm not sure it benefits me to bring it up in a setting like this because it didn't go spectacularly well, but it went fine. And then we sold it to Organon a couple of years after we launched it because it just didn't fit with the quality of the pipeline that we had in-house. But we learned an enormous amount about what it takes to launch a drug and what we think will make us successful, including enormous amount that makes us more and more excited about launching brepocitinib. So I think that's kind of where we're at now.

And what would you say are like the top priorities for your commercial team to really through that launch?

One of the cool things about the current moment is if we were at this stage, if we were about to launch a drug like brepocitinib and it were 2019, I think there'd be a lot of healthy skepticism would be a sort of short launch story. But you look out over the past 5 or 6 years and probably first horizon with TEPEZZA, but Horizon, Argenx, obviously, Insmed Now, Madrigal, Verona, BridgeBio, there have been a number of very successful commercial launches from and out of biotech. And in fact, I suspect that at least some of those drugs have been launched better in the hands of companies like Argenx than they might have been in the hands of big pharma companies that Argenx's creativity in crafting a modern commercial strategy has been really transformative. And so the first thing that we're trying to do is, look, I hope eventually people describe us in the same terms that they describe companies like Argenx as a commercial innovator. But before that, I hope they simply say they learned the lessons that Argenx taught us. And so we are trying to do everything that Argenx and Horizon and others have done. And that includes -- look, the world has had some structural shifts on access, for example. We're rebating at least an orphan disease is less a part of the landscape and where patients support has been an enormously important part of access. And so for example, we hired this woman, Le liberator who built TEPEZZA's patient support organization, and she is building our patient support organization. I think one of the things about launching an orphan disease that's become increasingly true is these patients, and this is true in DM, are treated at a more and more concentrated set of referral centers and, a, we've built excellent relationships with those referral centers. They were the people who run our clinical trial in many cases, and they are aware. We are spending a lot of our commercial time and medical education time now. And then, b, I think you build a field force there for that is the right people not necessarily to call in a community dermatologist, but to call on an expert in the field who spent their entire career treating dermatomyositis patients. And that means we have recruited some of the -- one of our field force personnel, one of our sales reps is a career-long myositis KOL who left clinical practice because she was so excited about what dermatomyositis might do that you thought you'd take a hand of like helping make it a success. And I think that is -- she's not really a sales rep there. before she served a medical fair professional. She's jack-of-all-trades but I think that's the sort of thing that is a really high priority for us and making sure we're making the launch of success.

The point that you made about short the launch, not being as successful of an investment strategy today as it was maybe a few years ago. I mean, part of that is these companies have done a good job. Part of it, I think, is also that I think expectations have been a little bit more well grounded. How are you thinking about what that means for communication with investors about the early launch about KPIs, what you're planning to share for brepo?

This is a subject of healthy debate at Roivant and other places historically. And only, I think we've learned a few things from recent launches and from the investor trajectory of recently launched companies. The first is I have a great deal of confidence in the ultimate commercial potential of brepocitinib as a drug, specifically in dermatomyositis and across indications. These are -- there's a lot of these patients. We can talk more about it. They have high unmet need. It's not a very competitively intense field right now in terms of people who are offering them options. And look, at some level, this isn't rocket science. The only other approved therapy in dermatomyositis requires on label that you spend 8 hours a day, 5 days a month consecutively in an infusion center, receiving an IV infusion, and we are a once daily oral that provides probably better clinical benefit. This is not the most uphill battle in terms of convincing people that it's an attractive alternative. So I think all of that pulls together to a high set of expectations. That said, no one has launched a novel targeted therapy in dermatomyositis ever. And so the pace of the launch, how quickly we'll be able to get doctors changing clinical practice, how quickly we'll be able to work with payers in this specific population and do the education work. I just think it's basically impossible to know. And we've certainly seen launches rocket out of the gate, and we've seen more slow and steady ramps. My general view is, I expect, I think everyone should expect slow and steady and also there is 0 benefit to providing guidance. So standing up there and giving people a sense of what I think is we're going to have to wait and see, and we're going to do our level best to make it a long-term success. And along the way. I'm sure we'll have fits and starts, but I'm excited for where it heads. And frankly, I think the companies that have attempted to provide guidance have not been rewarded for providing that guidance anyway. And so I'm not sure -- that's also a lesson learned from the Darwinian process.

Yes. other indicators like new start forms have you thought about whether you going to be sharing that information.

Yes. So the [indiscernible] most of our decisions on that basis, they're going to be rooted in commercial and competitive dynamics. We will be, as all orphan launches do now using tailored quite narrow focused limited distribution. And in general, for a variety of commercial reasons when you do that, you pretty rarely wind up sharing scripts. And so our scripts will probably not be widely available on a regular basis just because of our commercial distribution plans. And so that's just how that's going to play out.

Got it. Got it. What are you hearing from docs on the likely mix of patients in terms of prior therapy? And one of the questions that we get a lot is the pace of JAK switching.

Yes. Look, I think -- so taking a tiny step back. In claims data sets today, there are about 40,000 patients actively treated for dermatomyositis, that is a portion of the total dermatomyositis population. I think epidemiology suggests maybe that number is 70 plus. It's also a relatively difficult to diagnose condition. And so there's probably more patients out there who have either not received the diagnosis at all or have received a lupus diagnosis or whatever, who like may very well have dermatomyositis and show up later once we arrive anyway. . Of those 40,000 patients, about 75% of them are on steroids and immunosuppressants, prednisone, methotrexate. And many of those patients are poorly controlled. And the way we know they're poorly controlled is they're on quite high doses of steroids. In many cases, more than 6 months a year on greater than 10 or even greater than 20 milligrams prednisone, if you've ever had, like an allergic reaction, you've spent time on 20-milligrams of prednisone. It's a miserable existence for 3 days. I cannot imagine doing it for 6 months. And so you know those patients are poorly controlled because they're making that choice. That is absolutely one of the early groups of patients that we are most enthusiastic about. Those patients are not on other therapies for a variety of reasons. They can't spend 5 days a month in an infusion center. They don't want to take an unapproved B-cell depleting drug like rituximab or whatever that has failed clinical trials in dermatomyositis. And so they're sort of stuck with steroids and DMARDs and dealing with the unsuccessful treatment that comes with it. You can imagine an IVIG patient spending all this time in an IVIG infusion center, thinking that maybe a one-pill regimen sounds good. So that's like -- that's another place where I think we will get early patients. There are -- so about 25% of the patient population is on something other than steroids and DMARDs. That's like about half of those patients are on IVIG and the other half were on a collection of off-label stuff. The vast majority of the off-label stuff are literally drugs that have failed dermatomyositis trials, but they probably provide some benefit in inflammatory diseases or docs are trying it because they don't have anything else to do. Some small low to mid-single-digit percentage of those patients are on off-label JAK inhibitors. The truth of the off-label JAK inhibitor population, when you actually do the math, low single-digit percentage of a 40,000 patient number. It's hundreds of patients basically, maybe a couple of thousand patients in total. It's actually a pretty concentrated group. For example, I think at another bank's KOL called Julie Pake at John Hopkins said she had 70 patients in off-label tofacitinib. What she said she that's like a relatively significant percentage of the total patients off JAK inhibitors. What she said she would do is switch those patients as soon as she could basically. So I think like what will wind up happening is prescriber-by-prescriber. Some docs will be eager to switch, some docs will say, look, at least the off-label JAK patients are like kind of on something that works for them, I'm going to go with my high dose steroid patients. I think some of that will come down to access and how easy we make it to get patients on drug. I think some of it will come down to how the docs that were on our trial and have a lot of familiarity with brepocitinib, maybe faster to use it, the docs that were not in our trial are so coming that may do a little more experimentation. So I think it will be a mixture.

Yes. In terms of competition, VYVGART, they've got the Phase III IM later this year. Assuming that works, like how much of a swing factor are the actual results really in terms of brepo's future market share? Do you think

Yes. I mean, look, they're 18 months behind us probably. And so I'm not that -- right now, we do have to outrun the bear as it were. That is it's not about the competitors. It's just about the disease. Later, we only have to run them or something. Look, I -- first of all, I think DM is one of these diseases where the unmet need is so large and there are so many patients that more share of voice by the industry, more new therapies, more options for doctors is mostly going to be a rising tide. And I think VYVGART in MG has benefited from the existence of the complement pathway programs I think we will benefit from the existence of VYVGART. And I think I would much rather be VYVGART and MG than a complement company. I would much rather be brepocitinib and DM than VYVGART because we're first because I think -- look, we're getting a lot of inbound calls now from investors and the path they're taking to get to us is they are Argenx shareholders, and they're focused on DM. And so for the first time, they're doing KOL calls in DM, and they pick up the phone and they call it DM docs like what do you think of VYVGART and the docs, like, if you heard of brepocitinib, and so they're calling us because they're sort of coming at this from the side, and that's an enormously rewarding thing in the sense that it helps bring people to our story. I think it just underscores the lead that we've got. Practically speaking, I think there are reasons to believe that dermatomyositis is not going to be the strongest setting for VYVGART among the myocytes. The trial that VYVGART running is across multiple different myositis. It's running in IMM and in DM and polymyositis, I think IMNM is more biologically on point for an FcRn. And I think their efficacy will probably be better in IMNM than it is in DM. I think that based on biological rationale, I think that because Argenx' public statements to me suggest that they generally believe that too. And then I think if you compare our Phase III data to the Phase II data generated across the [indiscernible] they didn't break it out by subtype in their Phase II study, we were faster to achieve a moderate TIS response and achieved our responses against the backdrop of an aggressive steroid taper and still did comparably or better than VYVGART did. So I think we will have a competitive profile. The truth is, ultimately, you're only as good as your label -- and if VYVGART gets lucky and blows out of the park in the study, it will be more of a competitive factor than if they don't, and we'll just have to see what their data looks like when it comes.

NIU, next most mature indication, you've got the Phase III later this year. Phase II highly compelling small cohort. But I mean, frankly, I've not heard a real strong bear case for this trial. I'm not going to sit here and ask you to articulate?

You can if you want.

Maybe what were some of the risks that you sought to mitigate when you design the Phase III?

Yes. Look, I think -- so NIU, noninfectious uveitis is a basket diagnosis for inflammation of the eye that are not caused by an infection. It's a heterogeneous patient population. It has not been an area of very active clinical development. It has stymied a lot of others who have tried to develop there. It's a bad disease. And ophthalmologists tolerance for eye inflammation is very low because it's actually -- NIU is the third leading cause of blindness in the United States right now. It's like a bad disease and patients really don't want to go blind. And so they're willing to get treatment. . Against that backdrop, the only other sort of approved modern "therapy" for NIU is Humira, which bluntly doesn't work that well, somewhere between 15% and 80% of patients or 15%, 75% of patients fail HUMIRA depending on how you count it. et cetera. And they fail pretty quickly, right? The time treatment failure [indiscernible] was like 6 months median. So it's a disease with a lot of unmet need. We ran a study, it was a Phase II study. It was blinded in dose ranging, but it did not have a placebo. And I said HUMIRA's time filling their Phase III was a little under 6 months. We went over 12 months as medium-time treatment failure. So with a lot of cushion versus the field. But no placebo, the blunt level truth in immunology is the placebo response rates have crept up in every indication in history. And we didn't have a placebo on our Phase II. And you asked me what the risks in the trial are the biggest risk in the trial is whatever the bearer in this clinical trial is placebo, and we have to outrun it. And so I think that's a challenge that we're cognizant of. How do you do that? First of all, the heterogeneity of the patient population is always an obstacle. I think we've done a very good job with some creative and aggressive strategies to make sure we have the patients we need in the trial that they are sick in the right way, that they are sick NIU patients. We have specific -- what sort of look for adjudication criteria for making sure the right patients get into the trial. We have a very aggressive steroid taper. One of the things we did in the Phase II. The way these trials all work because tolerance for eye inflammation is poor, is you can't just bring a patient in and put them on your drug. You have to bring them in and put them on a very high dose of systemic steroids so you get the information under control and then taper the steroids and see if you can maintain a response on your drug. And the HUMIRA studies used a 12- to 16-week paper, I think, we used a much more aggressive taper than that. And we used it in the Phase II in [indiscernible], there was no placebo and we wanted to give ourselves a hard test, but it worked. And so we're using a similarly aggressive taper in the Phase III. I think that will help control placebo response as well.

How much of a headwind do you think the availability of biosimilar HUMIRA is in this market?

Look, 2 things. One is -- so there's about 40,000 patients on TNFs with NIU. And as I said, the treatment failure rate is 50-plus percent, and they tend to fail within 6 months in the clinical trials, even if at the price points that we have in mind for repicitinib, you live entirely in a immunorefractory population, it's a blockbuster indication for us. And because ophthalmologists tolerance for eye inflammation is low, if our data are good, I think there will be a strong desire to use us aggressively in early line settings. And so that will be a question of label. It will be a question of payer dynamics. But mostly, I think there will be a lot of people fighting for early access if our data supports it. And I think we will work with those people to get these patients on drug, so they don't go blind.

I had intended to spend a little bit more time on CS and LPP, but maybe let me just ask a broader question of what is your indication selection strategy for brepo and maybe just hit some of the highlights on CS and LPP?

Yes. I could launch into an, I think, interesting thematic history of JAK inhibitors here, but in the interest of parsimony. I won't. Look, I think in 2019, if you had said, by 2026, there will be a large successful important JAK inhibitor franchise targeting orphan disease, everyone around you would have said, "it's obvious, right? In 2019, JAK inhibitors were everywhere. They were going to be the future. And then the black box warning thing happened, and everyone kind of backed away. If in 2019, you said, "Oh, who is going to own that large franchise of JAK inhibitors in orphan disease, everyone would have just looked at you and said, obviously, it's going to be EILilly, it's going to be Sanofi. It's going to be AbbVie. It's going to be any of the companies that like had a place in JAK inhibitors and a place in orphan disease and a right to win there. The fact that a approximately random mid-cap biotech company owns that franchise is a combination of cleverness on our part that I'm proud of and random accidents of history that I'm glad to have benefited from. But as a consequence, we own right now the category of JAK inhibitors and orphan inflammation, and there are many orphan inflammatory disease. I mean, literally, almost any inflammatory disease, we call it, 30,000 to 130,000 patients is a good swim lane for us. And we are looking -- obviously, we're focusing on the diseases where we know the physicians, we're focusing on Th1-mediated disease, the places where JAK1 and TYK2 benefit for a variety of reasons. But we're looking across that space, and there are quite a lot of indications to go after. And I think even more than DM being an exciting indication, though it is or NIU being exciting indication though it is, the breadth of what brepocitinib could do was actually pretty staggering in terms of the number of diseases we could help.

Switching over to Mosley. Maybe would you like to start with just an overview of the drug and why you're excited about PHILD and then we can shift over into trial?

And I talked at the beginning of this year asked about PD around how we bring our drugs in. I didn't mention, we paid $14 million upfront to Pfizer for brepocitinib. We also paid $14 million upfront to Bayer from Mosley Sagawa. Mosley is a drug -- it's an inhaled vasodilator, mechanism is this thing called SGC activation. We got it from Bayer a few years ago. And basically, what happened here was Bayer had a history in this chemistry. They in fact sort of originated the development of drugs in SGC. They were partnered with Merck on respiratory disease around a drug called Adempas, that was a commercially successful drug, $2 billion or $3 billion in peak sales, a big drug. And then Bayer and Merck had kind of a messy divorce at the end of that process or at least a big foot ways, and each went on down their separate path of developing a successor drug to Adempas. Both of them developed inhaled SGC drugs. Bayer's in our view, was the better drug. And then Bayer went and did some M&A in the agrochemical space that was complicated. And they had to make difficult choices around their pharma portfolio because it's round up allegedly because it's cancer. And so they got out of respiratory diseases. And this was a while ago, and they no longer had a real footprint there, and they weren't sort of doing active research and respiratory disease. And meanwhile, United Therapeutics, one of the forefathers of pulmonary hypertension development and a phenomenal company, sort of whatever, struck gold for a second time. They found pulmonary hypertension from lung disease, PH-ILD as an indication that got Tevas approved there. They launched it commercially and it has been an enormous commercial success that has engendered an entire field of literal follow-on molecules of other treprostinil similar to Tyvaso, from Liquidia, from Insmed that I think are also really exciting drugs. We went to Bayer at that moment in time, realizing that mostly, which have been developed principally in the more competitive group 1 pulmonary arterial hypertension should also, in theory, work as an inhaled vasodilator, in PH-ILD and Bayer not at all really paying attention to the field and certainly wasn't going to run a novel development strategy in lung disease. And so we enlicensed it from them and set on retraining the program towards PH-ILD and now we are reading out a Phase IIb study later this year. The first, basically in a nontreprostinil mechanism of a late-stage study for PH-PLD.

What is a successful Phase II look like for you?

Yes. So in PH-ILD as with all pulmonary hypertension, the approval Phase III endpoints are things like 6-minute walk, which again, because it's a generalist conference. 6-minute walk, first of all, is a widely hated indication -- widely hated endpoint in biotech. And the reason is because roughly speaking, what happens is you're standing in a doctor's office, picture your doctor's office, and the doctor says, how far can you walk in 6 minutes. I'm going to set a stop watching and get it a tape measure and you're going to go. It turns out conditions very widely. Are you walking in a hallway, are you walking in a waiting room, are you walking in circles on a tile floor? Are you walking on a carpeted floor? This leads also like did you eat your weeds, when you woke up in the morning, did you drink a cup of coffee. There's just like a lot of things that make this a variable endpoint. . It turns out that it is a relatively well behaved endpoint in pulmonary hypertension, but it's quite variable. And so in Phase II studies in pulmonary hypertension, including in PH-ILD, the primary end point tends to be something called PVR, which is right hearted blood pressure. It is a literal measure of disease activity in that you are measuring the flow of blood through the right ventricle. The way you measure it is under general anesthesia on an operating table with a catheter, and so it's not something that you can do regularly in very large studies, but it is the primary endpoint of our study. And again, a measure of how sick these patients are, they subject themselves to a clinical trial that requires regular general anesthesia and catheterization. So that's the primary endpoint. There has not been across pulmonary hypertension, a mechanism that doesn't -- that delivers 20-plus percent reductions in PVR and has not gone on, not just to be clinically successful on 6-minute walk, but to be a multibillion-dollar class. And so our view is if we can deliver 20-plus benefit on PVR, nothing else really matters. That is almost certainly a good enough indicator that we will be able to deliver clinical benefit in a subsequent Phase III study that we will run the subsequent Phase III study. That said, we are measuring 6-minute walk in the trial. We are measuring other clinical endpoints. And while the study is not powered to deliver p-value [indiscernible] significance on those endpoints, it would certainly be helpful to understand the magnitude of effect in those endpoints as well.

With Tyvaso, I think that they had a pretty big gap between peak and trough 6-minute walk which I think kind of underscores that your daily dosing could be a clinical advantage and not just a convenience advantage. Are you collecting 6-minute walk at different time points post inhalation either in this or in a future Phase III? .

Yes. We are measuring a bunch of different time points. One of the great things about our drug is it is quite stable in the lung, and in fact, in Phase I studies, we have like elevated cardiac output, CGMP production out 48 hours after a single dose. And so I think we do actually get quite a lot of benefit from the time course because it's 48 hours, we will get meaningful stacking over multiple days over multiple dosing. And I think all of those things should contribute to the better clinical benefit. And I think we will measure at different time points so that we can begin to sort of dimension out that effect.

How do you envision the commercial opportunity for Moseley either as monotherapy or combination? And what kind of evidence would you need to generate to support that?

So there's another thing where we can learn from history. So in PAH, pulmonary arterial hypertension, which is a, I think, $10-plus billion category now, one of the -- whatever the sort of first to the pulmonary hypertensions as it were to be treated. What happened was actually [ treprostinil ] were the first modern drugs approved at prostacyclins and then a series of successive classes, including systemic SGC stimulators were approved. And each time a new class entered, two things happened. One is actually life expectancy for these patients increased by, in some cases, as much as 2 to 3 years. And the other is patients just went on multiple categories of drugs. This is a polypharmacy market. Pulmonary hypertension ultimately often kills you. It's a very bad disease. These patients go on everything they can get. They cycle drugs, they add them on top of each other. And so I think that is exactly what will happen in PH-ILD as well. It will be a polypharmacy market, and we will be used before Tyvaso, after Tyvaso, on top of Tyvaso in every combination and with other mechanisms, hopefully as well. Practically, that means from a clinical benefit perspective, worse than Tyvaso, better than Tyvaso, similar Tyvaso, it doesn't matter that much in the end because most patients will wind up on most drugs. Obviously, the better we are than Tyvaso the earlier will potentially be used. We have some other advantages, one inhalation once a day versus more for the others, one puff from a dry powder inhaler once a day. In fact, we don't likely cause cough as an on-target effect, what's prostacyclins right reasons why we could be using an earlier line setting, but mostly the inter is this is going to be a polypharmacy market, and everyone is going to be on top of everything. To answer resonant evidence generation, in our Phase III study -- sorry, in a Phase IIb study, we do not allow concurrent Tyvaso only approved in the U.S. It's a global study. It's like slightly complicated. In the Phase III study, we will allow some patients on concurrent Tyvaso precisely because it's important that the label will allow for concurrent use. However, it's important, therefore, to know a little bit about what we do together with Tyvaso, especially from a safety perspective, but also to get some sense of efficacy. So we are currently in addition to the main Phase IIb running an open-label combo study with Tyvaso that has just started, so we don't have any data from it right now, but that will help inform things like stratification in the Phase III.

Question on the iPad. What other indications in PH could you look at? And like when would you consider starting those?

Yes. Perfect. So look, I think the history of this is helpful, too. Most -- look, locally administered vasodilators are effective in pulmonary hypertension. We probably work in PAH. We have good Phase I data there. It's competitive, but we could go there eventually. I think PH-COPD is an interesting indication, although local vasodilation in lung disease patients with emphysema is a more complicated proposition. And there's more emphysema in the PH-COPD population. So that, for that reason, requires some more careful thought. Recently though, we've also seen the prostacyclin, [indiscernible] be successful, for example in IPF, and that is absolutely high on our list of things that we are excited to think about. Ultimately, we're going to learn a fair amount about all of that from this Phase IIb later this year. We're measuring FVC and IPF patients. We'll have a sense for how we did there, albeit in a small subset. I think all of that will inform next steps. PHL is a huge indication and we wanted to make sure we nailed it. But I think we will be initiating new indications if the Phase III data supports it pretty quickly after we get this data.

Switching to Immunovant. Could you start with how 1402 is differentiated from the other FcRn in terms of either product profile or the indications that you're...

So again, I'm sure many people in this room are familiar with FcRn as a mechanism because Argenx has been so enormously successful in creating that category. This is sort of a whatever, it's like a big moment in immunology in that this is sort of the Humira moment for a new kind of immunology. It's the first time we've had a drug approved that can treat what I'll call B-cell disease, autoantibody-driven disease, much of which isn't even inflammatory. Graves' disease isn't really an inflammatory disease at some level. And so it opens up an entire set of new indications and Argenx has done a phenomenal job in MG and CIDP in other places, creating those markets and showing real benefit for those patients. . VYVGART has some limits, nipocalimab, the J&J drug has some limits. Among them as each of them is currently studied practically. And in VYVGART's case, probably just like due to biological limitations, they don't suppress IgG more than, call it, 60% or 70%, whereas I believe we will suppress IgG by 80-plus percent with 1402. So I think we can get better efficacy. We are as a simple standard DUPIXENT style auto injector, which is something that VYVGART is not. VYVGART is a hydro low long push with dire injection site reactions and things like that or -- , I'm sorry, or an IV drug. So I think those are all advantages that we have. And then I think although not as important as being a better molecule almost as important as being a better molecule, I think we just carved out some really great white space indications for ourselves. We have pioneered modern drug development in Graves' disease where we have an ongoing payer of registrational studies that we hope will be -- we hope will be successful and will create the first approved novel therapy in Graves' disease since the 1950s. So that's an indication with millions of total patients, hundreds of thousands of poorly controlled patients. Imitation is the finance for flattery. We have created a coded industry of other companies now developing Graves' disease, but they're all years behind us. And I think we have really built out some expertise and positioning there that's exciting. Just last week, we showed some data in a sub-setting of rheumatoid arthritis that was frankly, more compelling data than we expected it to be and showed a potential role for FcRns in the late-line multi-mechanism failure rheumatoid arthritis that I'm excited to explore from here. So those are both indications where we're way out in front and kind of a lone right now, and there's many others that we're pursuing. Some of them like MG more competitive and some of them like CLE less competitive.

That's where I was going to go next was RA, if you want to share some of the highlights of that data and why you think you're able to achieve such higher ACR responses versus EPO?

Yes. So RA, obviously not an orphan disease, many, many patients with RA. It is a disease that is more complicated than Graves' disease and that it is not just an auto antibody disease. It's an inflammatory disease, it's an autoantibody disease in some patients. It's a -- there's different etiologies. It comes from different places. It's a complicated disease. The idea that FcRns could work in RA is not an idea of our own invention. And we know that some of the disease autoantibody driven. In fact, one of the exciting things about FcRns there is unlike all of the other drugs basically in RA, it's not an anti-inflammatory. And so it might work differently on a different axis, and therefore, work in combo or in later line settings where anti-inflammatories failed. This was the idea with which they had been studied historically. J&J has run 2 studies in RA. One was effectively a monotherapy study across a variety of lines of therapy, across a variety of different patients. And they showed fine responses, nothing -- they showed clear benefit, but nothing that looks like getting people out of bad. The 2 exciting sort of indicators in the data that were interesting. One, almost all of their responses came from the subset of patients that tested positive for autoantibodies. And so that was like a clear way to enrich on a biomarker. And then two, their efficacy appeared to be preserved irrespective of line of therapy. And so you look at that and you're like, okay, maybe there's something here in late line. And in fact, even though their efficacy was kind of middling in late-line multimechanism failure patients, it was potentially good enough to have a commercial plan. So then we and J&J after seeing that monotherapy data, each then plowed on with different strategies. J&J said, okay, we're going to run a combo study in early line patients combined with a TNF and that was a study they ran. And we were like, we're going to focus on the late line multimechanism failure population and run the study that we ran, which was in patients who effectively failed everything they can. They failed at least 2 of TNF, IL-6 and JAKs. A majority of them has recently said, have failed TNF and JAKs. These are like late-line patients without other options. The confusing thing is J&J's combo therapy pretty spectacularly didn't work. They basically didn't separate from the TNF. That was always a hard study to run. You're putting patients on a TNF for the first time. They're going to benefit significantly from the TNF. And so you have to kind of demonstrate benefit on top of something that's already helping them. But still, the level of separation was disconcertingly poor. And so things like that confused us when we saw it. Then our study read out, and it showed something sort of surprising in the opposite direction, meaningfully better responses than either of the other 2 studies. The data that we've produced so far is data from an open-label run-in period to a randomized withdrawal trial it is nuanced data across multiple axes. And certainly, one of the reasons it looks as good as it did. We showed ACR70s in the high 30s, which is as good as anyone ever shows basically. It's close to JAK-like and its efficacy, which in a JAK failure population is extradite especially for mechanisms is pretty safe like an FcRn. Part of the quality of that data has to be due to the fact that it's an open-label setting. There's some rising tide effect here, but you just don't spontaneously have an ACR70 response. Those are pretty big improvements in sick patients who have failed a lot of other drugs. So I think there's like our belief is there's clearly something real happening. The other 2 things, I think, that are relevant. One is we did a pretty good job selecting for an autoantibody positive population. The J&J combo study with CIMZIA was auto antibody selected, but not they allowed rheumatoid factor patients, for example, was not as rigorously specifically focused on [indiscernible]. I think we generated some benefit there. And then I sort of have to believe, although I don't yet have the evidence to back this up, but the other thing that happened for us is, it turns out that when you filter out patients who have failed, remember, failed is an important word your JAK inhibitors and TNFs, that those patients have tried anti-inflammatory mechanisms that are very effective for treating their RA and have not succeeded, that it must at some level be that we are enriching for a population whose disease is causally driven by autoantibodies in a way that even we didn't totally anticipate. But ultimately, we'll find out period 2 of this study is going to be harder now because we have to generate meaningful reversion in patients that have benefited a lot clinically from being on drug. And so we met not even in a p-value period, too. We're going to really study the patient level data here and trying to understand what's happening and then work with FDA on a trial design for a subsequent study that I hope will be tractable and lead to a lot of benefit for these patients.

Getting some more questions on the iPad, and this is more of a big picture thing, but the cash balance, you've got $4 billion, you'll be getting more from Moderna. What are your plans for that considering that you have not spent that -- you've not spent a lot of money on transactions historically?

Yes. Look, I think One of the things about Roivant is we are sort of culturally built to be good stewards of capital. It's just sort of part of who we work to begin with. And it's led to some puzzling decisions over time. Look, I think like biotech companies historically haven't bought back stock. That's changed a little bit, but like we bought back $1.5 billion of stock at $10 a share because we had too much cash. That turned out to be a good investment over time. We continue to buy back stock now as we bring in more money from Moderna. . What we said, lost in history at this point is that we bought a drug from Pfizer a while ago and then sold it to Roche for a lot of money. And so we had a lot of cash on our balance sheet. That's -- that plus the Sumitomo deal is how we wound up sort of highly capitalized and I think what we said all along is if we can't do it with $4 billion, we probably can't do it with $6 billion. That is like there's an amount of money beyond which it just like isn't required to build the kind of business we're building. And frankly, the kind of BD that we've done historically has been pretty capital efficient on the upfront, expensive clinical development, pretty capital efficient on the upfront. So look, I think as cash comes in above and beyond current levels, we will continue to be pretty aggressive about returning it to shareholders just because I think balance sheet leanness is something we believe in. But we have still a lot of cash even absent that, and we'll spend a fair amount of it on breadth of development for the existing pipeline. We will add indications for brepo, we will abdication for FcRn. We will add indications for motel. And we'll be active on the BD side. And I really hope we bring programs in with extensive Phase III trials. Those tend to be big indications. They tend to be exciting. They tend to be the kind of things we can make a difference at. So I hope we spend a fair amount of our money on that.

AI has been a big theme for this conference. What are your expectations for how that impacts drug development and over time? And what is Roivant doing today?

At the current moment at a generalist conference, I think what I'm supposed to say here is -- we're an AI-native AI-first company, building data centers and space. We're not building data centers in the space. I don't know how. But I think that's what we're supposed to say. Look, I there is a lot of talk about AI in biopharma. And a lot of that talk centers on novel uses of AI in drug discovery on medicinal chemistry on biology. And I think that stuff is super interesting. We have invested in it. We have built a company that is successfully using AI to model protein [indiscernible] interactions that we have a big stake in that I think could be. Look, I think it could be a company that invents designed molecular glues in a way that wasn't possible for AI. There's like cool things are going to happen here. That said, [Audio Gap] if quickly, that will be great for us. There will be many more drugs to put into clinical development, and we are very good clinical developers. If slowly, that's okay, too. That said, in clinical development, what is clinical development really in addition to a science problem. It is a massive logistical coordination exercise where you're reading hundreds or thousands of patients, medical charts coming through them, literally as human beings like trying to figure out which of these patients are good for your trial, you're trying to turn around legal documents like site agreements and IRBs and IBR agreements and protocols quickly there. all similar to each other, but different in little ways. These are all things, but like the AI of today, the Claude that is on all of our desks, it's really good at. And so like our ability to prosecute large trials at scale is dramatically better than it was a year ago, and we are coming to terms with that. And I'll say like, look, I think we made a concerted effort to get these tools out and available to everybody, and we are building tools every day to make ourselves better at this than again, if I were to try and defend ourselves as like an AI-first company, those are things I would talk about. But also, it's just been this incredible Darwinian process. You put cloud on the desk of a smart person running a clinical trial, and they're like, "Oh, my life is easier now. Here are the 7 ways my life is easier now. So I think a lot of what we're doing is just putting the right people in the right seats with the right tools, and it turns out that's making us better.

One more question to close this out. If we fast forward 2 to 3 years, what has to be true for Roivant to be a $50 billion market cap company?

Look, I think some of our launches have to go well. I think that's true. I don't think DM has to be an $8 billion indication. I think DM has to be a $1.5 billion indication in people's minds. And I -- look, if DM is a $1 billion or $1.5 billion indication, and I use a $1.5 billion indication in CS and LVP are $1 billion or $1.5 billion indications and FcRn in total is $3 billion or $4 billion, we're there easily, right? We're on our path to being -- I don't mean that literally in 3 or 4 years. People have to believe those things in 3 or 4 years or 2 or 3 years for us to achieve that kind of scale. So I think we have to do reasonably well commercially somewhere, and we have to keep stacking indications. And to be honest, I think that's about it there. I think if you asked me how we become a $150 billion company. There's a lot we have to do between here and there. But I think the path to $40 billion or $50 billion or $60 billion is good execution on what we've got now.

Great. Let's leave it there. Thank you, Matt.

If you take nothing away from this conference, just remember, we are an AI-first company building data centers in space.