Roivant Sciences Ltd. (ROIV) Earnings Call Transcript & Summary

Great. Well, thanks for coming, Matt, and thanks for everyone who's joining us both in-person and on the web stream. So maybe, Matt, just to start, could you please just provide a brief overview of what you view as the key programs that are going to shape the Roivant story as we think about 2023?

Sure. And thanks, Corinne, for having me and thanks to the conference for having us. It's great to be here. So I appreciate it. Yes. Look, 2023 is a pretty incredible year for us actually. So I'd say major catalysts across, let's say, 4 programs at Roivant. We have this anti-TL1A antibody RVT-3101 that we in-licensed from Pfizer just last month on which we put out top-line data from our induction study yesterday. So there's additional data coming from that program midyear, first half of the year from our [ maintenance ] study. We have atopic dermatitis data coming from our commercial tapinarof VTAMA cream for psoriasis, that data is in atopic dermatitis. We have data at Immunovant from our next-generation [ ADF ] antibody in the middle of the year. And then we have data coming toward the end of this year from brepocitinib, our TYK2, JAK1 that study in SLE. So a whole bunch of really exciting clinical data coming in 2023.

Great. And we'll go into each of those programs, I think, in a bit more detail. But to start, I just wanted to touch on capital deployment. And as a company, you just mentioned a couple of them, but Roivant has primarily built its portfolio through the acquisition of various assets with 3 deals just last year. So how do you think about the criteria you use as you assess potential acquisitions?

Yes, it's a great question. And we have built a lot of our portfolio around acquired in-licensed partnered programs. We've always been therapeutic area agnostic. As it turns out, we've built a pretty good concentration recently in I&I. But that's sort of not on purpose, as it were, just kind of worked out that way. I'd say we are in general obsessed around clinical data, obsessed around individuals or patient-level clinical data and trying to understand what's going on biologically with an agent, and that's been always core to our valuation strategy. And then other than that, it's all about drugs that can matter, things where we think we can do something different in development or where the class is going to matter in a way that we think we appreciate differentially. In an environment like the current one, talk about capital deployment, where the capital markets have been a little bit dicier, the bar for deploying capital is really high. But it's sort of an interesting moment because we're very fortunate. We have quite a significant cash balance. I think last reported number was just under $2 billion. And so we're fortunate from a capital perspective. And in our 8 years of existence, there has never been a more fruitful sort of opportunity environment. Just so many interesting programs are looking for homes. And so it's been -- it's been something that we had to really carefully balance and I've been happy with what we've been able to bring in.

Yes. That brings me to my next question, which was, as you think about the landscape for asset acquisition today, could you provide a little more detail on what you're seeing with respect to the availability of these deals, the quality and the value.

It's really never -- it's never been better for us. And some of that is specific. We have a multiyear track record. We have 6 products approved by FDA coming out of our platform. We have many Phase III studies successfully run, many others underway. Our partners now understand what we're capable of, and that's been idiosyncratically helpful, and we are a repeat partner of almost everyone we worked with. So this deal with Pfizer was our third major deal with Pfizer, for example. So that's sort of idiosyncratic to us. But it's also just been a totally unique environment in terms of biotech struggling and trying to figure out sort of where to go for homes for assets, pharma companies focused on prioritization, focused on patent cliffs, focused on sort of portfolio management. And then also once upon a time, not that long ago, when we saw a program, maybe like the TL1A program with Pfizer, for example, look, unavoidably, there would have been half a dozen other like -- other people mostly sort of company creation [ we see ] kind of people, standing in Pfizer's lobby trying to get the same program. And with that model, having at least for the moment dried up, we get to be alone in most of those rooms, and that's been a real privilege also.

That's helpful. And in the past, the company has also spun assets out and sold them off or gone through IPO processes. How do you think about the criteria for making those decisions? And how much do you think that will be a feature of the strategy moving forward?

Yes, thanks. That was the thing we did a lot early in our life. I think -- so taking a step back, we have the sort of unique, very modular business model, sort of multi-manager platform for drug development, where we put these programs down as [ Vants ], as you well know. And that gives us a lot of flexibility from an M&A perspective, from a capital planning perspective, we can do lots of things, and we've taken advantage of many of them. We've sold Vants, we financed Vants privately. We financed Vants publicly. We've partnered programs and I like that flexibility. And so for that reason, I'm definitely in like a never-say-never situation here. But our model has evolved over time. As we've become a public company, having public subsidiaries is not as convenient as when you're a private company. And so I'd say that's more been the exception than the rule for the last several years, and I think that will continue to be true. It's really not a primary piece of our model to be -- to taking Vants public and when we in-license program like the Pfizer program, we really love that program. I hope we can own just about all of it or as much as we possibly can for as long as possible. So I think in general, sort of IPO spinout route, less core to the model. But again, we're built in a way that maintains the flexibility to do whatever the sort of best opportunity is for each program and probably some of our programs lend themselves well the partnership for M&A and many of our programs lend themselves well to being internally developed and commercialized, and we'll follow the science and the opportunity in every case.

Sure. Maybe jumping into some of these specific programs, and I wanted to touch on VTAMA first. Could you just provide -- so we all get up to speed, a quick overview on the product including its current label and development data?

Yes, sure. So those who aren't familiar with this, so VTAMA is a topical treatment for psoriasis. It was approved in June of -- I'm sorry, May of last year. It was the -- this is a wild thing even to say. It was the first novel mechanism therapy approved for psoriasis as a topical in 25 years and the 1 that was approved 5 years ago was effectively vitamin D. So it's been a very long time since there was novel science applied to topicals in psoriasis. The mechanism of the drug is an aryl hydrocarbon modulator, which is an inflammatory target that's upstream of like IL-17 and Th17. It has an impact on memory T cell formation. So it's been a really cool program to work on. The science is really interesting. I feel like people think of topicals as having sort of old science, but actually it's a very cool program technically. The data have been great. So in psoriasis, it's got efficacy that is, I'd say, comfortably as good or better than anything that a topical has ever shown before. It's certainly safer and more effective on top of corticosteroids, which I guess this is the other point to make, especially for an investor audience. People think of psoriasis as a market dominated by systemic therapy, but that's just not true. The vast, vast majority of psoriasis treatment is done via topicals. So there's 90,000 topical prescriptions a week in psoriasis and the vast majority of those are for topical corticosteroids. And so yet to be able to put out new science that really matters that is safer and more effective than the standard of care is a great position to be in. So product works well. Label was incredibly clean. From a safety perspective, there was no warnings in precautions section. There was no limitation on disease severity. There is no limitation on where on the body, you can use it. So it's very different from the profile of like a corticosteroid.

Great. And reimbursement has been a key focus for the VTAMA launch given kind of the history in the space. And so could you give us an update also on the progress with respect to establishing access here?

Yes. Again, it's a good question. It was obviously -- look, in May of last year when the product got approved, I think a reasonable question was what's access going to look like for novel topicals. Earlier today, Herve was here from Incyte talking about Opzelura. They've gotten pretty good coverage. We've now signed our first major PBM and payer contracts. The PBM contract that we signed that we've talked about is a great contract. It's perfect. It's got really only 1 step through steroid. That step has a couple of different possible ways of achieving it, including an automatic look back that requires no physician action or no action at all. So look, really good coverage. The other thing I think is like generally interesting. So I get questions about gross to net, and we have not sort of guided to a steady-state gross to net. But it sort of feels looking -- forget like topicals or dermatology for a second, like across the field like a new normal has been established for pharmacy-dispensed products, and that's like a sort of 50% steady-state GTN yield if you're a biotech company. And I guess I feel like that's going to apply to us too, that we're going to be in the sort of normal category for these products. It's the dynamics of the [ payer ] landscape we're continuing to change all the time.

Yes. And you mentioned the look-back process. Part of the question is like what's the paperwork and the demand on the office. How do you think about the applicability of this deal as you expand into additional indications and then we can talk about [ AD ] in more detail?

How do I think about the applicability?

Do you think you'll see these deals like these kinds of contracts supply beyond psoriasis into the AD market as well?

Yes. Look, it's hard to speculate until we know what our AD label looks like and what our AD data looks like that's coming in the first half of next year. That said, I think all of the factors that have led VTAMA to be really popular and exciting to docs and patients in psoriasis could certainly equally be true in AD depending on the quality of the data. And so I'm pretty optimistic. Look, I think the world hasn't totally sort of processed the fact that topical corticosteroids are not great drugs. They're somewhat efficacious, but if you want to give yourself bad dreams, you can go like Google or go on TikTok and look for people who talk about topical corticosteroid [ withdrawal ] injuries. It's terrifying. And it's just been the standard for -- I mean, I think like my uncle's -- my uncle's grandfather or something, it was like 1 of the like patent inventors on the triamcinolone patent in 1955. Like these drugs have been around for a very long time. And we've just gotten used to this idea that we slather relatively toxic [indiscernible] on us. And look, I think there's a new world now where we've got these safer, more efficacious topicals, and I'm pretty excited to be part of it.

Great. And as you think about the AD landscape and the coming Phase III results from ADORING, what do you think you need to show to be competitive in the space? And where do you think the drug fits within the current treatment paradigm?

Yes. So look, again, I think the hope and the expectation is safer and more efficacious than topical corticosteroid. So I mentioned 90,000 [ scripts ] a week in psoriasis. There are 320,000 topical prescriptions written every week in atopic dermatitis. Again, the vast majority of them are topical corticosteroid. So that's a low but important [ bar to clear ], safer and more effective than topical corticosteroids. On the 1 end of the spectrum for topicals in AD, you've got Opzelura which is a very efficacious drug. It looks very good. It has some JAK class sort of labeling and safety liabilities, but it's a good drug. It's labeled currently, I think, for [ 12 ] and opted for noncontinuous therapy and things like that. But for patients who are [ so that end ] in the spectrum, it's important. And then on the other end of the spectrum, frankly, you have [ as already ] of our competitors' product [indiscernible] to the topical [ PDE4 ] that just put out their 2 studies of AD data. I think that data was fine. I think it's probably as or more efficacious than a steroid or at least close and certainly well tolerated, but not particularly sort of jump-up-and-down exciting. I think 1 sell-side analyst called it "[indiscernible] like." Look, I think that sets a pretty wide range in terms of where we might fit in. Our Phase IIb data was definitely much closer to the top of that range than the bottom. I think our gross efficacy on whatever IgA response rate was like, 49% versus 13% placebo at the 8-week time point. I think if we had anything with a 4 handle, it would be an absolute grand slam in terms of where we would fit. I think anything with a 3 handle is perfectly adequate and it's good enough and puts us in the position of a strong drug. So that's kind of what I think the sort of efficacy bar is. And my hope is that we can go after the mainstay of therapy there that we can be a real viable alternative for topical corticosteroids.

Okay. Great. Maybe switching gears now to RVT-3101, which has been super topical of late. Maybe just provide a quick history on the deal that you guys did there, including kind of the background story.

Yes. So look, Pfizer has been a very close partner of Roivant for a long time. They are a partner of Myovant on relugolix. We did the brepocitinib deal, which we may talk about brepocitinib later today as well to stick to JAK1. We meet with them frequently. We know them well. And I think that's sort of the first part of the story as we love working with them. We've talked to them for a long time. Our RVT-3101 for those who are not familiar with the drug is an antibody to TL1A, it's a very hot topic right now. Thanks to Prometheus having put out [ quite a data ] last month and us having put out, I would say, equivalently good data or better data yesterday. And it's a class that we think is going to matter in inflammatory and fibrotic disease. Pfizer has been working on the class for a while. We've been talking to them about it for a while. And mostly, we're getting told that they were not interested. But that's sort of the nature of the business, and we stayed on them. I won't speculate here about the exact reasons why Pfizer's opinion on that changed. But they clearly think the program is great. And so once they decided it didn't fit in their immediate R&D strategy, they have a little bit of a puzzle on their hands, which is they needed to, on the 1 hand, find a home for it that allowed them to benefit from its long-term success. I think that was important to them. On the other hand, find somebody who was going to execute on the program well. Look, looking at our data, I think there's no question if they had wanted to, they could have just sold it. But instead of doing that, they chose to partner with us and the deal construct involves them keeping 25% of the JV basically, and then keeping Europe and rest of world, and we have U.S. and Japan. And once they had decided that, that was the construct they wanted, yes, they need to find a partner who is going to accept that sort of deal and be able to execute well. And I think we were in a relatively small, potentially in of 1 of companies that fit that bill. And so yes, we pounced on the opportunity, and we're, again, really excited to be working with them on the drug.

Yes, helpful. And you shared data yesterday, as we've talked about, which demonstrated 21% placebo-corrected responses in all comers, compelling efficacy in a biomarker-enrich population some of the other data sets. Can you just contextualize how these results fit within ulcerative colitis today? And also kind of relative to the competitive landscape in TL1A?

Yes. Look, for starters, look, I think placebo-adjusted deltas for other classes of drugs, certainly for standard of care are like single digits, 10%, 12%, 14%, like much lower numbers. And I think if you sort of listen to sell-side expectations for Prometheus' data before they put it out, I think everyone was like, yes, it's got to be in like that 10% to 15% context to matter. And clearly, TL1A class, both we and Prometheus comfortably in the 20s on a placebo-adjusted basis, has blown that expectation out of the water. It's interesting, I think, like, for example, you mentioned our biomarker data. I think like before we sort of had data from Prometheus or from us, there was a version of the world in which these drugs did like 12% gross or 12% net placebo-adjusted efficacy in the all-comers population and you were looking for like a 10 or 12-point bump from the biomarker. So you got into the 20s where you could be an interesting therapy. And instead, we all just start there, which I think is a pretty remarkable place to be. So we feel really good about where this could fit in the treatment landscape. There's obviously a lot of interesting therapies in development for UC and Crohn's, but our data is sort of right up there at the top end of what people have shown competitively within the TL1A category. So there's really sort of -- 2 sort of directly comparable programs, ours and Prometheus'. There's a couple of other anti-TL1A antibodies out there, but none of them have put out data in these indications of substance. In short, I like our data. Our gross efficacy, not placebo-corrected, was in the low 30s in the relevant treatment arms for all comers, with a sort of 12-ish percent placebo response rate compared with, I think, Prometheus was like 26.5% with a 1.5% placebo response rate or something in that general environment. I think the arm of data that I am most sort of comfortable with is our sort of drug arm of our study had about 200 patients in it in total across doses. So that's a pretty robust data set that sort of gives a lot of heft to the gross efficacy number. And then our placebo arm was a bit smaller and a bit less sick. And Prometheus' placebo arm was also a bit smaller and a bit more sick. So I think like the [ placebo ] will bounce around. But I feel like the gross efficacy here is pretty well supported by the data we've got. So I feel really strongly about where we are.

Great. And we mentioned the biomarker data and TL1A development has been in part characterized by that biomarker drive approach. So as you think about the approach from a biomarker perspective, what can you share about what you're pursuing with RVT-3101?

Yes. So again, I think it's sort of interesting to contextualize biomarkers in these therapies. Biomarker therapy is like a relatively newer thing in immunology compared to like precision oncology, where obviously, it's very well established. These are not like binary biomarkers. These are sort of computational algorithms of sort of genetic markers and other things. There's sort of representation of who the patients are. And as a consequence, there's a lot of flexibility in what those algorithms are. And so you could imagine sort of clinical biomarkers that like roughly every patient has. And they obviously would provide no benefit in separating out responders from nonresponders, because everyone's got them. And you could imagine biomarker algorithms that cover a very small percentage of patients, which ideally will provide a quite big separation in terms of the biomarker response rate versus the non-biomarker response rate. And there's some curve in between those things. And we and Prometheus are on pretty different parts of that curve. I think they've said their biomarker covers about 30% of the addressable patient population, ours covers about 60% of addressable patient population. And our biomarker data was about 10 points better than our all-comers data on a placebo-adjusted basis. So about a 30% placebo-adjusted response versus that sort of 20-ish-percent placebo-adjusted response. And then maybe even most exciting to me, in the sort of treatment -- in the treatment refractory like the biologics experienced patient population, we showed a 41% gross efficacy with no placebo response. It's very, very hard to get those patients to respond. And so you generally don't see much placebo response in those subsets. And most therapies erode significantly in that patient population, but our TL1A appears to sort of hold its efficacy, which is great and which sets us up for a pretty interesting sort of second-line therapy approach as well.

Yes. And you touched on some of this already, but as you think about maybe zooming out in I&I and potential for precision approaches in that therapeutic area. What do you think are the necessary components for a successful precision approach in I&I and what do you think are the indications where this is kind of the most right for disruption?

That's a hard question because it's not something that's like so well established at this point. I think there's like a few different versions of it. So 1 is the kind of thing that you're seeing now in TL1A, where you've got these like sort of computational or algorithmic approaches to stratifying patients based on various [ static ] markers. Look, I actually think there's -- the thing to me that's like interesting and exciting about this is there's actually just like a lot of different commercial models that you could imagine flowing from this. Again, it's not like precision oncology, where it's like here's a drug that affects a certain mutation. We're going to be for patients who have that mutation and that mutation only. This is about trying to understand across relatively, in many cases heterogenous immune disease like which patients fall where and what spectrum and how much of their disease is driven by 1 factor versus another. And so you can imagine companies taking very different approaches, companies taking very precision approaches, trying to get like very high response rates in very carefully sort of specifically selectively chosen patients more like a precision oncology sort of situation. And you can imagine situations where we're just sub-segmenting the patient population in bigger chunks. And so I think like there's lots of different sort of versions of it. Look, I think in terms of where it's going to work best? Obviously, there are genetically defined autoimmune diseases where you would imagine precision oncology -- precision immunology to work very similarly to precision oncology. So that's "obvious" place to go. There are also -- I'd say in any big immunological indication, there's like interesting possible sort of heterogeneity among the patient population. And the more you can do to understand that heterogeneity, the more likely it is that some kind of biomarker companion diagnostic can help. So I think about things like lupus, which is a very heterogenous disease and whether you could do something interesting there.

Yes. That makes sense. Maybe that's a great segue as well for brepocitinib. So maybe we'll just start with a quick history on the drug and Roivant's ownership of the asset, which was about last year.

And notably, even though we are setting brepocitinib in SLE we are not pursuing a biomarker guidance strategy.

Yes, good clarification.

So brepocitinib is an interesting drug. It's a dual inhibitor of TYK2 and JAK1. So TYK2 has obviously been a pretty popular target to talk about for the last year or so. That's in part driven by, obviously, the approval of TYK2 with BMS with a pretty clean label, we should know JAK-class labeling. And it's a pretty compelling clinical data. This drug also came from Pfizer, part of a broad portfolio. Pfizer was unquestionably 1 of the sort of leaders in JAK-class biology. So a really interesting drug as part of an interesting portfolio. Pfizer had studied it in a wide range of pretty large autoimmune indications. We have data with -- great data, best in category data, in many cases, in psoriasis, psoriatic arthritis, alopecia, ulcerative colitis funnily enough, and NHS and really good data across those indications. Unfortunately, for the JAK class, with the new sort of labeling, some of those indications are pretty hard competitively. And so the drug became available to us as Pfizer was sort of thinking through their JAK portfolio and what they wanted to do with which program. And we stopped and we said, okay, TYK2 biology is clearly interesting, it matters. JAK1 biology is clearly interesting, it matters. This is a big gun. It has worked in a lot of indications. What diseases do we think we can differentially [ win ] at with this drug, where we think the JAK class labeling is less likely to be an issue. And so we're focused on sort of rare and orphan autoimmune indications with high unmet need with not a lot of other approved therapies. And maybe this is an important point, JAK cytokine -- JAK kinases tend to pairwise signal cytokines. And so there are some cytokines that are specifically paralyzed signaled by TYK2 and JAK1. And so we look for diseases that were mediated by those cytokines. So far, we've settled on 2 of them. Dermatomyositis, where we're running a pivotal study where we got data in 2025, and it's a single pivotal kind of design. And then SLE where we're running a potentially pivotal Phase IIb. That program will read out later this year. And obviously, SLE is an interesting indication with very few approved therapies and a relatively mediocre performing standard of care. It's been a tough indication and many clinical programs have failed there, but we're looking forward to what we can deliver.