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

Terrific. Well, delighted to introduce our next session with Roivant. I have the CEO with me, Matt Gline. I have Richard Pulik, the CFO and Investor Relations in the audience today. So Roivant is a very different type of pharma company with a fairly unique business model, but one that so far has enjoyed some considerable success, both with your marketed agents, but also with 3 agents you have, and I'm just going to say, late-stage development, but that's a bit generous for the mid-late, early late stage...

You can call them late stage.

Yes, we can call them. So maybe just starting with the TL1A antagonist and obviously, starting with the Prometheus data, the Merck acquisition and then your data in UC, there's an extraordinary amount of interest in it. And the transaction from your perspective, you did with Pfizer looks highly value-creating for you in that. How are you thinking about the clinical development plan because there's lots of complexity here. Number one, there's regulatory risk as Roche amply demonstrated with etrolizumab in running a head-to-head trial, which didn't pan out as they hoped. Number two, there's commercial risk in the sense that new drugs take a long time to get established among GI docs. And in addition, there's soon going to be very cheap alternatives, which could be combined safely and used instead of unknown new expensive drug. So for all those things, start where you want regulatory, all the way through to commercial, maybe [ meteorological ]. Talk about how you're thinking about the development opportunity and challenges.

Yes, perfect. And thanks for laying out the long road ahead. Look, I think the first thing is on the regulatory side, and I'll start with UC and then maybe we can talk about Crohn's a little bit later and some of the other indications. So the good news is, I think, the Roche experience notwithstanding. I think the regulatory path in UC is like relatively well established at this point that I think the market has coalesced around what I'll call the etrasimod design, basically, the sort of 2-study designs, one induction study, one 52-week study, and I think we like that design. And frankly, I think our view is the data from our Phase II from Prometheus' Phase II are compelling enough the best development plan, at least in sort of the broad scaffold of it, is just down the fairways to do what others have done. I think there is no need for our development plan to have an active comparator or anything like that. I think it will be a relatively straightforward plan. Now I think there's one node of complexity in this process that is sort of specific to TL1A and gets to your commercial question, which is one of the things that people have gotten excited about with anti-TL1A antibodies has been the existence of these predictive biomarker algorithms that help to narrow at least 1 set of eligible patients who may have a higher rate of response. And we have such an algorithm, and Merck has such an algorithm. And I think there was a version of the world a year ago today where the entire sort of data package for all of the ages in the class could have been called 10 points worse on every endpoint. And I think the people would have thought then, well, the biomarker is the way to go, this is a biomarker drug, and it gets to be sort of jacked competitive, sort of 20s, receivable adjusted delta clinical remission rates in the biomarker-positive setting. I think in that world, we would have developed the drug pretty differently. We would have developed it really in Phase III probably in the biomarker setting. The all-comers data here was so compelling for us and for Merck and for Prometheus. But I think at this point, it makes sense for us and for them to develop the drug in an all-comers population and to see nothing on the label, no line of therapy limitation, no biomarker limitation, et cetera. However, Mark Cuban selling, as you pointed out, Humira for $500 a month, at some point, we've got to realize that the commercial landscape by the time we hit the market is going to be competitive. Not only will there be multiple anti-TL1A agents on the market shortly after we're there, but also you'll have biosimilars and generic [indiscernible] wide variety of classes, maybe you'll have TYK2, depending on what the data looks like. There's a bunch of things that are coming in the pipeline...

As well as combinations...

And by the way, one of the things that I think is interesting to the combination point because I think the combination study that most people point to was the J&J VEGA study that combined the TNF with Stelara. And I think if you look at that study, and you look at our data in an equivalent population in sort of a first-line population, and it's notable that our data in that population looks [indiscernible] our comparison all the appropriate caveats actually looks better than combinations of those agents, which I think is an important point because those agents maybe biosimilar at some point while we're on the market. So that's all true. I think we will develop the drug with biomarker as a sort of key secondary endpoint. So it was not to restrict the label, but to set us up so that maybe we won't initially get great payer coverage in first line therapy for all-comers. Maybe the first place we'll go is second-line therapy for biomarker-positive patients. You carve out a niche there and then you build from there, first-line biomarker positive, second-line all-comers. And I think the data is good enough that eventually, we should get covered in quite broad settings. But I think the way the biomarker sets things up, it gives us an opportunity for a faster path to a subset of the population, and we can build from there.

Yes. And there some parallels with Astra's adventures with anifrolumab in lupus.

Yes, perfect. Exactly.

It went exactly the same path.

Exactly right. Yes, I totally agree with that.

So -- okay. So -- that's the development pathway. Then you've got the commercial landscape and dealing with clinical [ inertia ] among GI docs as well as the competitive commercial element because of price biosimilars. This is going to be -- this isn't going to ramp up with a hockey stick. It's going to take concerted effort, particularly if you have Entyvio being combined with Stelara, and there's all kinds of interesting investigator-sponsored trials that you can measure people Xeljanz with whatever. I mean how does induction with inexpensive, cheap Xeljanz, they're moving to a maintenance? I mean there's all kinds of things, which can add complexities to the...

It's hard to disagree with anything you just said. There will be a lot of flexibility. Docs will have a lots of flexibility. There will be a lot of agents in a lot of different places. I think the first thing is, the data are really differentiated for these programs, right? This -- I get that like the Vega combination data looked very good. And like you can cherry pick studies in like S1Ps or whatever that look good. And the JAK data looks good, right? There's no question JAK inhibitors work very well in IBD. But I think most of this other stuff, Entyvio, the TNFs, I think like there is a step function change in efficacy between those agents. I suspect even in some sense, those agents in combination versus what I think like we will do in a biomarker positive setting, for example. And so I do think it's just going to start with the fact that these are really sort of differentiated agents, and they're quite safe, right? Like even relative to a TNF, we really don't see infection risk materialize. The cool thing biologically about TL1A as it really concentrates in inflamed tissue. So you don't see a lot of TL1A building up in sort of systemic immunosuppressive context. And so you don't see a lot of infections. You don't see a lot of the things that are dose-limiting for some of these other agents. So I think we have a differentiated profile, which I think is going to be important in that competitive landscape. I also think the other benefit that we have going for us is this is a tremendously exciting class to the doc community right now. So like enrolling these trials is actually going to be very easy. The docs are really excited. I actually -- this is not a perfect point, but I'll make it anyway. I think like the existence of the biomarker and tend to be like commercially helpful. Actually, there is a sense among some IBD docs that biomarkers are potentially fundamentally useful and also cool and that they have sort of been outside the biomarker club for a while. And with the introduction of TL1A, like the idea that they get to use that to guide sort of a more precision approach for teaching their patients, I think it's actually something that we have found enthusiasm for in the doc community and has its own sort of branding differentiator for the class.

So could you -- I mean we're talking about IBD as a homogeneous category, but of course is [ not homogenous ]. There's 2 very different diseases with different patient journeys and different outcomes with -- I'd argue Crohn's is generally having -- because of the fibrotic elements are far more worrying. So when you think about where the acceptance is likely to be greatest, I would imagine Crohn's represents a low-hanging fruit, which is tricky because you haven't got the data in that setting yet assuming that you follow what we've seen previously. Anyway, just I'm interested in your observation and whether that concurs with what you're thinking?

It's a great question. Look, I think the first thing is -- there's a little bit of speculation involved here. The field has relatively limited data on TL1A and Crohn's right now. We have the Prometheus open-label study and really not much else. I think biologically, the mechanistic rationale is very strong. And actually, one of the things that people get excited about with TL1A is because of the [indiscernible] activity. You basically -- you down-regulate fibroblast formation, you should have an antifibrotic effect, which actually -- it should be particularly exciting in Crohn’s, exactly. So look, I think the answer is yes. I think we should have a very good uptake in Crohn's. We are currently running a Phase II study in Crohn's. That study was calculated to be the most straightforward Phase II study we could run, that could be sort of completed on a time line that let us remain basically first-in-class in Crohn's given the competitive dynamics. And so what that -- it's a dose-ranging study without a placebo because we want to be able to enroll it very, very quickly. But we're hoping it lets us get much of the dose ranging work out of the way before Phase III, so that we can go into a Phase III study armed with a clear view on dose and armed with a clear view that the biomarker that works in UC is the same as the biomarker that will work for us in Crohn's, which is something that we strongly believe should be true biologically, but Prometheus seems to have gotten a little bit tripped up on Crohn's translation of their biomarker strategy. And so we just need our own data to fully understand that. And then once we're there, I completely agree. I think if the fibrotic effect of this agent presents, I think in our own 52-week data, and you see you start to get a sense for it in the -- like the endoscopic remission rates and things like that. I think if that presents in Crohn's, this will be a very compelling agent in Crohn's and especially for long duration therapy. And I do look at -- we are studying this agent as induction maintenance combined. We will study a single-dose in both our UC program and in our Crohn's Phase II, a single dose for the entire 52-week period. Do I think patients will go on high doses of biosimilars or JAKs or something for a period and then look to TL1A as a maintenance therapy? I'm sure that there are some docs that will use the class that way.

And the average anticipated treatment duration, how are you thinking about that? Obviously, that may well change because this is more -- maybe Entyvio is a good land log because obviously, the tolerability is great. And so maybe the question should be what's the average treatment duration for Entyvio in different indications?

It's a good question. I don't actually know off the top of my head what's the average duration is for Entyvio in different indications. I would expect -- look, I think -- it's a hard question to answer because the data that we have right now is largely a single 56-week data set in a clinical setting. I think, the 2 things that I think are compelling in terms of people wanting to be on the agent for a longer -- comparatively longer time, what is it the safety generally looks very good, including out to 52. It's not like things presented in the maintenance date on a safety perspective that gave us sort of cause for concern. So I think that's helpful. One of the reasons that people like titrate down on doses with JAKs and things like that, you start to worry about the long-term profile. The other thing that I think is compelling, and this gets to the possibility of antifibrotic activity is, you look at the different endpoints that we studied and in particular, if you look at -- I think I mentioned endoscopic remission, which is what it sounds like. It's an endoscopy score of 0. It's clean endoscopic tissue. I think like the fact that we had I think in the biomarker positive setting, it was maybe 36%. It was like 1/3 of patients, 32%, somewhere in that range, 1/3 of patients were at an endoscopy score of 0. And even in like the all-comers population, it was over 20 at 56 weeks. I think like it's a therapy that appears to work better the longer you stay on it. And so based on the data that we have available to us right now, my expectation is that it's just an agent that people want to stay on. The other thing bluntly is I think a lot of the data about Entyvio, about the TNFs, about some of the other classes, is going to be less useful for predicting long-term adherence in some of the more modern classes, including S1Ps in addition to TL1As, because I think when you're talking about placebo just the clinical remission rates in the single digits or the low teens, there's just like it's going to be a high need for switching because a lot of patients are not seeing the benefit of the drug, whereas like when you're talking about placebo, I mean in our biomarker positive setting, we're getting into the 40s in some of these cohorts. Like there's going to be a lot more patients who are controlled on therapy with the TL1A.

So how much do you think the field is going to change in the next 5 years before you and Prometheus come to market, 3 years or whatever the number is. The reason I argue is that right now, I suspect if I did a sample of gastroenterologists and asked them what they think of orals, particularly the [ S1Ps ], they'll say, "God, their monitoring is pain in the a***, and I just can't be bothered with this." It's quite different from neurologists and other indications. And they're supposedly not doing very well. I didn't have desperately high expectations for [indiscernible], maybe a little bit better. But it's possible with time in the market and physician education and so on that the comfort level may change, and therefore, relatively the hurdles may go up. So I'm just curious, as you think about the transitioning of physician acceptance now, that's just the future...

I think, manifestly, the treatment landscape for IBD is going to be a lot different by the time we hit the markets now. I think uptake of orals is going to be a piece of that. The S1Ps are clearly going to be a piece of the uptake of orals. I think like the thing that I am most interested in is how the balance of the JAKs plays out over that period of time. The RINVOQ data is very, very good. And I think it's very, very good. And I think if there's a setting where a drug like RINVOQ is going to overcome the black box and become an important agent, I think I believe UC could very well be such a setting, especially with data of that quality. And then I think like the class that I am most "TYK2s" have not done well in UC and the data that we have already from them, but that data is fairly limited, and we're going to get a lot more data for TYK2s and IBD in the next couple of years. And I think if TYK2s start to look at higher doses or with sort of the diversity of agents better, I think the TYK2s and the JAKs could start to like shift people's perception on orals. I think the S1Ps could as well. And the S1Ps, especially long duration durable therapy, looked quite good, actually...

It just seems to be that the monitoring requirements, which is putting off the docs, but that can change.

I also think -- the honest answer is I hate needles. And up until very recently, my view was there's always going to be a large patient population who just like ultimately, when really they understand there is a valuable oral alternative, there's people who would prefer to take a pill. Maybe too much information in this room, but I recently went on [ Mounjaro ], and my opinion of injections completely changed. I would switch my statin to a weekly injectable in a second, if I could. It is so easy. And so I do think that the other area where there's maybe a slight up till battle here is like once-monthly injection versus a daily pill, I think that's a pretty different patient burden.

Yes. The clinical -- sorry, the pharmacological differentiation between your compound and Prometheus. And as you know, I don't cover Prometheus. So you have Roivant. So you'll have to give occasional lapses in my information base. But there are some differences, I think, in monomer versus trimers, and whether this has any clinical significance or not, perhaps you could indulge me and just give me the Roivant take?

Yes. So look, there are some differences in these drugs. I think the most notable one is the one that you articulated. They bind both the monomer and the trimer. We only bind the trimer. As I'm sure TL1A is only active like any of the TNF superfamily targets in a trimeric form. And Prometheus was competing with Pfizer at the time and needed a narrative to articulate as to why they were going to beat Pfizer. And so they focused on monomeric binding as a thing that was going to differentiate them. The truth, I think, is if we had bound the monomer or Pfizer had bound the monomer and Prometheus had only bound the trimer, they would have described the monomer as a likely antibody sync. And they would have said, well, you only want to bind the trimer. The next level of truth here is I think if you just look at the actual clinical data, our data is, if anything, strikingly similar to Prometheus' data. I think it's like we can point to gross efficacy differences and they had a very low placebo rate and a pretty sick placebo arm and stuff like that. But like I think if you take a step back and you look at the data sets in aggregate, first of all, I think the efficacy is very similar between these agents across a variety of parameters in a way that is surprising. And second of all, like, as validation for the class, you couldn't have wanted better. Like I think like there was a version of the world in which our data looked the way it looked and Prometheus looked worse. And you'd say, well, okay, so like Pfizer got lucky in their study and the range of sort of performance for these agents goes down from here more towards the like Entyvio's of the world. But in fact, I think these agents are like pretty much right on top of each other, both in the all-comers and to some degree, the biomarkers setting, all the biomarkers are different. And I think that gives me a great deal of confidence. I don't think -- barring something very surprising in Phase III, I don't think there's going to be some massive clinical differentiation in the all-comers clinical data. I think there may be some differentiation in the way this works in the biomarker setting. And I think there may be -- there are some risks to Merck involving clinical development that we don't have because they don't have subcu data and they don't have dose-ranging data and so on that could get to differentiation depending on how their study goes.

And they're getting to subcu. Are they using how to run those? How are they...

So I don't know the exact answer to that question. They have a subcu formulation. Prometheus had developed a subcu formulation. It was not used in their Phase II study. Their Phase II study was entirely IV. They, I think, have already done some amount of bridging work they've said. We have not seen that bridging work, but they've never dosed an IBD patient or a UC patient. And the one thing I think is worth noting and no one knows how this will play out yet, other than us through close analysis of our data, because TL1A concentrates only an inflamed tissue, there was a possibility that it actually behaves differently in patients to some degree, pharmacokinetically than it does in healthies. And so Merck has some real and important decisions to make on how they think the translation will work from IV to subcu and how to think about dose ranging in the context of first dosing a patient subcu right in Phase III.

So this is one of several compounds that were tied up in the Pfizer deal. And 1 of 2, maybe more, but 2 that I know of in relation to TL1A. So could you just talk to -- because I was just speaking to Michael next door about the P40 -- bispecific, and that makes an awful lot of sense in terms of potentially raising the bar with a likely a well-tolerated [indiscernible]. Could you just remind everyone in the room who may not be familiar with the -- why this is interesting to pursue? And secondly, just the economics on that compound separate from this compound and the terms and details of the option that you have?

Perfect. Yes, absolutely. So for those who are not familiar with it, so Pfizer has a, call it, a next-generation anti-TL1A antibody that is actually -- it is a bispecific antibody to TL1A and P40. P40 is basically, think of it as the same target as IL-12 or 23. And it makes sense for a lot of reasons. But among them, you can sort of think of TL1A as like a supercharged participant in the TNF access, right? TNF is one of many things that TL1A down regulates, but sort of that part of the inflammatory spectrum, and then P40 gets the sort of IL-12/23 part of the spectrum. And we know from the VEGA study that we talked about earlier that combining a TNF alone without 12 or 23 looks good. And so it makes a lot of sense actually. They're targeting both TL1A and P40, could produce complementary activity. And obviously, the P40 targets are, to your point, all pretty well tolerated, looks good, et cetera. So it's a nice pairing. I guess like one note of caution that I'm sure Michael would have given to, I mean Pfizer is among the best in the business of bispecific and trispecific antibodies. They're really, really good at this. There's not a lot of data for bispecific antibodies and immunology indications yet. So it's early days across all of these agents. And I think that's just a thing that we're going to have to learn more about through studying it. Anyway, the economic construct on it is Pfizer is currently running a Phase I study. They say that study will read out in 2025. They have indicated an openness to including some patients in that study, and I am personally hopeful, and I've asked Michael directly for some of those patients to be outside of IBD, so that we can learn a little bit about what TL1A does as a mechanism outside of IBD. But I think the inclusion of those patients and the sort of identity of those patients is still [ TBD ] and Pfizer completely controls the conduct of the Phase I study. After the Phase I study, we have an option to opt-in to a different construct than the 3101 construct. This is a 50-50 global co-development, co-promote arrangement, where we would each be responsible for half the costs, we would each receive half of the opportunity. It's a global deal, whereas I think you know, in 3101, we only have U.S. and Japan. Pfizer has retained Europe and rest of world. So it's a different construct both in the level of capital that Pfizer puts in as well as the geography of our exposure. And the construct is a full collaboration, and there's all the appropriate rules and controls and everything. I think the answer is like, in effect, if you get to the bottom of it, we both have equal say and veto and indication selection and things like that. And my personal hope, look, I think it's an interesting question in IBD. That said, the data from 3101 and the other TL1A antibodies are good enough. It's not obvious to me that this is necessarily an IBD drug versus an opportunity for other indications, rare diseases and other tissue types, expansion beyond IBD. I think that's all speculation at this point. We'll have to see what the data show in patients in IBD and outside of IBD. And I think it will be fun to work on it with Pfizer, honestly.

Yes. And when you're thinking about IBD, you're thinking of what psoriatic arthritis or you...

Yes. So for TL1A generally, and this includes the bispecific as well as other things, I think my view is like there's 2 pretty good plausible explanations for why the data are as good as they are. One of those explanations gets to the inflammatory activity, right? TL1A down regulates IL-17 and TNF alpha and a whole bunch of other things. And then the other is this antifibrotic activity. I think it is unclear today, which of those is primarily responsible for the quality of the data. So what I'd like to do is some set of small studies to just map out that playing field to figure out in some rare or fibrotic disease, maybe something with a biomarker what the antifibrotic effect really looks like and to figure out in some immune complex disease, what that looks like...

[indiscernible] or something like that.

Sure, something like that, exactly. And then I think once you take a step back and you look at all of that, then I think there's like a question of where you would go. I guess like mechanically on the inflammatory side, I think like psoriasis, as an example, you would think this drug would work well on psoriasis, but you probably wouldn't take it there because the bar is so high at this point. But you're like, I'm just scared to say this because people are thinking I'm going to run a 1500-patient study or something, but like RA. I'm not sure the set up in RA is so different than the setup in IBD, right? A bunch of agents, many of the same ones, but relative -- like they all work okay, but nothing works spectacularly well. And like there's room for improvement. So I think like if TL1A winds up working in broad inflammatory disorders, I think there's like -- you could imagine going quite broadly on the anti-inflammatory side. On the fibrotic side, NASH is a bad word these days, but like there's a wide variety of liver fibrotic indications, lung fibrotic indications that are sort of interesting and...

I understand. TYK2. So the data that Bristol showed and the high dose and my friends, who were involved in that trial, they say that they looked and there's just nothing there for deucravacitinib, the high-dose in lupus indication. So what's your real belief that this is something specific to deucravacitinib, rather than it's just not amenable to this particular target?

Well, so first of all just a point of important reference for us, our drug brepocitinib is not an allosteric TYK2. It is both the JAK1 and TYK2. And we know obviously well from the brepocitinib...

I was thinking -- sorry, I was -- okay, carry on and I'll explain why I was asking.

I guess the question is like better than -- why are we going to be better than brepocitinib. But I guess, like, first of all, I think the deucra data in aggregate looked pretty decent in SLE. The high dose was bad. And the low dose was strikingly good, right? Like a 20-something percent as...

Sorry, I said lupus and that's why your answer [indiscernible]. I didn't mean that. I meant in the IBD indication.

Got it. Sorry. Right.

Sorry for confusing you.

Yes, yes, yes, fine. Okay. So, now I should recenter myself here. So you're asking about -- so we are not studying our TYK2, JAK1...

No, but I'm talking from a competitive [indiscernible] as you try and [indiscernible] question about how the field...

Perfect question. Yes, exactly. Okay. So I think the base case expectation to your point has to be that the TYK2s are not going to work very well in UC. The deucra data do not look good. Deucra, you said high dose. My understanding is that most of the deucra data we've seen is at a lower -- relatively lower dose in UC, although I think they've started to produce some data in the higher doses...

I'm not hearing good stuff.

Yes. Look, I -- I guess for the field, you always hope things work competitively. I would rather not compete with allosteric TYK2 because it's got some advantages that are a little bit frightening. I think the answer is you would not -- mechanistically, there's lots of reasons to believe that TYK2 could be useful in these indications. I think the clinical data that we have, at least in deucra is not encouraging.

Shall we segue away from -- should we segue away from TL1A to the FcRn. And you obviously have 2 compounds, and I think this is an important month, and I see [indiscernible] covers in advance sitting there in the audience. You'll be far more familiar than me that with time lines. But why don't you just remind everyone when you expect both a single dose and the multiple ascending dose for your second-generation compound? And then we can talk about how through dose modification or dose reduction post induction, you can make the albumin lowering manageable for the first compound.

Yes, perfect. So in terms of the next in terms of 1402, what [indiscernible] said a single-ascending dose data this month and multiple-ascending dose data in October or November. That's the expectation for the time line. And I hope as I think everyone knows, is that we will not show a meaningful impact in albumin and LDL. And we've pointed out the assay variability in albumins around 5% and assay variability in LDLs around 10.

And the mapping of the primary data to humans, you would imagine would be pretty strong.

It has been excellent everywhere else in the FcRn class. I'm not a superstitious person, but the translation has been good elsewhere.

And then following that data, you're then in a position to initiate a clinical trial.

In fact, I think we could go -- so one of the great things with FcRn is that IgG has been a phenomenal biomarker for clinical activity. And that basically across all of the agents for a given level of IgG suppression, you see a similar amount of clinical activity. So we believe we will be able -- I mean we will be able to take 1402 direct into pivotal studies on the back of this data.

And so then that begs the obvious question, what about your first compound?

Yes. So I think the FcRn landscape has sort of evolved in an interesting way. And we have some very strong competition there in terms of quality of execution and in terms of access to capital and all kinds of things. I think the truth is, as I see it, is that efgartigimod first of all is largely a rare disease drug right now, that Argenx has developed it that way, that it is priced that way. The indications they're going after are kind of MG and rarer as indications. So I think the a, question about the field is, let's presume for a moment that 1402 looks good in this Phase I study. We've got like J&J, this was not a biological hypothesis that we were jumping up and down about, but J&J has run an RA study with nipocalimab. I don't really know what to expect from that, but they've said they're going to put it out this fall. If nipocalimab looks good in RA, I think 1402 will be a phenomenon, the sky is the limit, because then I think it's hard for me to imagine efgartigimod, Argenx, sort of really -- it's the price point is wrong. The sort of commercial model of Argenx is not right for that purpose. And nipocalimab, we believe -- we know it has an albumin liability, and we believe it has an LDL liability based on everything they've said, and I think they've been pretty clear about that point. And so we will be able to go in with a similarly IgG suppressive agent that is subcu from the start, which nipocalimab, I believe, is still not in clinical trials in the subcu format without an albumin LDL issue. And so I don't mean RA specifically necessarily, although I think obviously, if it looks good there, why not, like RA, lupus nephritis, Graves', like the sort of more prevalent indications. Then you get to the rare side and the Argenx competition. And I think the answer is like, there are going to be 2 sets of indications on the rare side. There will be indications where the only thing that matters is clinical efficacy and the LDL liability will be looked as a largely irrelevant fact. And the only thing that's going to matter is does deeper IgG suppression yield better clinical benefit. And I think of indications, like thyroid eye disease. And I think like it already is clear from our Phase IIb study in bato and thyroid eye disease, that the difference between mid-60s percent IgG suppression and 80-plus percent IgG suppression. And our view is the difference between a highly compelling drug that is especially in a world with their new label adjustments competitive with Tepezza, maybe not like literally an efficacy, but it's a balance of factors versus a drug that's probably underwhelming. And I suspect that's going to be true in a variety of rare diseases. And that to me is where there's still room for an agent like bato. How big that set is, depends on how good 1402 looks where nipo and efgart get developed and things like that. But I think bato to us becomes an exciting option in sort of the rare setting, and the opportunity is a function of what Argenx does and frankly, how much of the dose response we see to deeper IgG suppression across a variety of indications. And we're going to learn a lot about that next year. We're going to learn in the first half of next year, for example, a little bit about what our dose response looks like in the induction phase of a CIDP study. And I think if we can demonstrate better efficacy on a cross-trial comparison versus Argenx and CIDP deeper IgG suppression with the market attributed a lot of value organic to CIDP data.

So when we think about the monetization of these assets. And obviously, 3101 is an example of how you partnered with Pfizer, but that has a difference of providence. This came from a Chinese company. It's one of the two molecules, right. So how do you think when is the right stage to begin partnering, particularly if you're going up to the more prevalent patient populations? Or do you think that you can do this yourself because maybe you won't go in the high prevalent population. So I'm assuming you didn't do anything right now because you obviously have more data to derisk it. How are you thinking about the sort of journey here?

Yes. Look, I think on the one hand, TL1A and FcRn both, these are lightning in a bottle targets at some level, like it is rare for a biotech company to have an opportunity. We know FcRn can support a $30 billion biotech company because it's doing it right now. And we know that a TL1A with a single Phase II study can get acquired for $10 billion. So we know that these are like interesting targets. And there's a part of me that says we've got both of these programs is, in my opinion, quite rare, that a biotech company has 2 programs of that scale and reach. And it's incumbent on us to find the capital and find the word with all and develop these programs as much ourselves as we can, I think there's a real opportunity there. Now the flip side to that is these are both competitive races. In the case of FcRn, we are competing with J&J on the one side and Argenx on the other, which both appear to be impossibly infinitely deep pocketed. And then on TL1A, we're competing in a horse race for first-in-class with Merck, who also you can imagine is going to spend heavily behind the program. And so I think there's a concern that to maximize the value of each of these programs, you really need to make sure you're going as broad as you can, as fast as you can. And I think that's where it's incumbent on us to like at least ask the question around partnership. Now when you say partnership, I guess, like, look, the proxy statement for Prometheus' acquisition indicated broad interest in TL1A as a target. And I think you can imagine how that translates to a company like us. That's certainly an option that's available to us. But I think it's just a question of value and how we see the opportunity other than that. So I think we will evaluate those options sort of on a continuous basis. I think you can imagine on FcRn, it wouldn't make sense to think very much about it until after this data comes in. But once this data comes in, I'm sure pharma will take notice. And then it will just be a discussion about value and it feels like a high-class problem.

So just looking at a little bit even deeper into the future, maybe too early for this conversation, but I was speaking with the CEO of Alexion yesterday, and they obviously killed their FcRn. They clearly have a pre-IND, FcRn anti-APRIL conjugate. So the field is evolving. So I'm just curious as your thoughts and whether -- I mean, the extent to which Roivant is involved in discovery versus development?

Yes, perfect. So look, first of all, I think that success in these fields begets, whatever invitation is the highest form of flattery. We will see a lot of people trying various approaches to these targets. I think it is a testament to just how difficult it is to build an anti-FcRn antibody. But there are so few of them that are sort of in late-stage development now, right? We've known for a while, these are good programs. There are -- I don't know exactly how many, but many private biotech companies pitching various approaches to TL1A already. And that's like months after that data came out. But it's been genuinely difficult to target FcRn. But there are good approaches out there. There's the one you mentioned that I'm aware of. There's also Biohaven has an IgG degrader that is sort of going after a similar access of biology. There's obviously the UCB. So there's all the complement franchise kind of programs. So there's a bunch of different sort of approaches to this biology. I guess my fundamental view is, I think it's important biology and I think all of these approaches are merited and it's worth interrogating. We don't do a lot of basic research, but we do some and by and large, the basic research that we have going on is trying to take advantage of the [ execution ] we have in some of these later-stage clinical biology areas. So for example, we have a small molecule anti-TL1A program that's looking for interesting ways to modulate TL1A as a target using a small molecule, and we have a small molecule program using covalency to target FcRn, which again, like I think if that worked, it would be interesting. I think there is room for those approaches. They're all very far behind the field in terms of -- and they may matter in different ways. They will clearly have different tox and tolerability profiles and so on. So I think we're just going to have to wait and see on like each of these individual approaches. The thing that I think is cool about FcRn biology is you can sort of look at TL1A and say it fits within a genus of anti-inflammatory targets like IL-17 and like TNF and so on. FcRn is sort of out there on its own right now, like the IgG autoantibody. There's a bunch different approaches, but like relatively few late-stage clinical programs live in that construct, other than like IVIG. And I think it's going to be a big field. So I think there will be a lot of people working on stuff that heads that way.

So just pivoting to an organizational question. Merck has now completed, I think, the acquisition of Prometheus. Have you hired anyone from Prometheus? You talking to them?

I don't know if we've hired anybody. One of the things that's great about our business model is we managed to recruit very talented people because -- so as I think you know, we put each of our programs into what we call a Vant, effectively a biotech company. And so the company that is developing the TL1A program is called Telavant, and it has a management team. And one of the great things about working at Telavant, for example, is we could go to somebody who worked at Prometheus and say, "Hey, you were one step down from the Head of Clinical Development, you were a VP of Clinical Operations, like here's your chance, come be our Head of Clinical Development." And my guess is that like the very top tier of folks from Prometheus...

They are...

That's right are doing something else. But that there's a bunch of people who learned a lot from that exercise. And look, I think Merck has got a lot of incentive to keep those people. Bluntly Merck did not have a very deep...

Yes, they have exited [indiscernible] many years ago.

And so my guess is that Merck has a lot of incentive to keep those people. But you can imagine we're trying to get talent from anywhere that we can. The other thing I'd want to be careful about is just not tripping on noncompetes and other things as we go to develop the program.

So I think we've reached a lot of time. I'm just checking around the room if there are any hands, I didn't see any. Matt, many thanks for the conversation. I look forward to more conversations in the future.

Thank you. This was great. I really appreciate it. Thank you.