Takeda Pharmaceutical Company Limited (4502) Earnings Call Transcript & Summary

Good afternoon, everyone. Welcome once again to TD Cowen's 46th Annual Healthcare Conference, and welcome to the Takeda session. We're very pleased to welcome Dr. Andy Plump, who is the President of R&D at Takeda, joined by his IR team. And let me say at the top, if anybody has a question in the room, please feel free to raise your hand and we can call on you. But I, of course, will take the liberty of the first question, and it is a big picture question. setting the stage for Takeda. 2025 was really a phenomenal year for the company. You had 3 NMEs, rusfertide, oveporexton and zasocitinib, all of which had positive Phase III readouts. We're now headed into launches of those drugs. We're also looking at a maturing in-line portfolio, and you have a new CEO coming in June. So maybe you can help us understand as investors how to think about the next chapter for Takeda.

Thanks, Mike, and hello, everybody. So I was in a town hall a couple -- a week ago with my team, and I said I turned 60 last year. I don't know if I'm allowed to give that information confidential. And so I spent a lot of years in this business. And I have to say that the last year was by far the most exciting and successful of any that I've been in for the reasons that you just suggested. So amazing time for us with Julie Kim coming in as CEO. Julie, I've been working with for 6 years now. I know her very well. She's just an amazing individual. We're all really excited about her transition. We'll miss Christophe immensely. I started with Christophe. He came 12 years ago, I came 11 years ago. So this is a huge transition for me and for the whole organization. But I can't imagine someone better poised to take over. And of course, an internal candidate tells you how enthusiastic the whole organization is up to the Board to carry forward the momentum that we've generated. So just from a big picture, the past couple of years, a lot of focus on VYVANSE, loss of expiry, '24, '25 took the big hits, '26, that will start to level off for us. While at the same time, we have a group of products that we call our launch and growth products. I think in Q4, we'll come up with a new name for these because we're going to have another set of products that are going to be our launch products. And what we're seeing with our launch and growth products is that they're starting to mature. And so their growth starts to become much more moderate over the coming years. And this is then where the excitement happens with these new launches. So 3 great readouts. I know we're going to talk a lot about them. Last year, we're really looking forward to approvals and launches starting this year, and we're already starting to lay the foundation for what those launches will look like. And of course, each of these 3 products from oveporexton to rusfertide to zaso have a very different market and a very different structure around the launch.

Great. Well, maybe we could start off talking about oveporexton in that market. PDUFA date in August. How should we think about the launch? And what's the sort of ideal candidate right from the get-go for a drug like oveporexton?

By candidate, you mean?

Patient candidate.

I see. So if you have type 1 narcolepsy, you are a candidate for this drug. This is really one of the most exciting drugs that I've ever been a part of. The data that we've seen in our Phase III study has just been phenomenal. Essentially, not quite, but essentially all patients benefit from the drug and the vast majority normalize on whatever efficacy endpoint we studied, whether it's daytime sleepiness, whether it's cataplexy, whether it's cognition, whether it's nighttime symptoms, this drug oveporexton has benefits across this whole spectrum. And for the majority of patients, it's normalization. So what we see in this market is that if you're a type 1 narcolepsy patient, you're likely to need one medicine to fulfill all of your needs, and that's oveporexton. We're quite excited about the -- getting it approved and launched. Of course, we have the PDUFA date set in August, but we're doing everything we can that's in our power to see how that can move up and get accelerated.

What should we anticipate in terms of the label? There's a pretty broad set of endpoints that was measured in the pivotal trials. What among those might make it on to the product labeling...

Yes. So first, I'll say that the amount of data that we have in this program is extraordinary. It's the endpoints that we can measure are just so quantitative. So we've talked about the fact that we have in our -- each of our Phase III studies, 14 different endpoints. We've hit with very high significance and clinical meaningfulness on every single endpoint that we've studied. And in addition, we have a number of exploratory endpoints that are quite interesting. I'll say that for those of you who follow this field closely, we've disclosed a fraction of the data that we will be disclosing, so less than 50%. So expect over the course of this year to see a large amount of additional data coming out. We are keen to fully own this space since that means better understanding the journey of a type 1 narcolepsy patient. And so we're working very closely trying to unravel biomarkers, whether they're digital biomarkers or other biomarkers that can help us further understand patient needs. So you've probably seen we've announced a partnership with Beacon using their Dreem device to help measure sleep. That's both to help us with oveporexton uptake, but also our broader interest in the sleep wake cycle. And then I'll mention, you asked about the label, and you're asking what we can expect from the label. We have that same question because, of course, that's still part of the process in the review. But I think our base case is that it's going to be a very attractive label for physicians and for patients and include as many endpoints as possible.

Great. Before we move off the orexin franchise, maybe we could talk about TAK-360. Phase II NT2 NIH data are coming later this year. Are you more optimistic about prospects for this molecule given competitor data readouts? And maybe you could put the entire competitive landscape into some context for us.

Sure. So I wouldn't say I'm more optimistic based on what we've seen from the competitors. I think we've been optimistic all along. We've had a lot of experience, not just in type 1 narcolepsy, but in type 2 NIH. You'll remember, Mike, and many will remember that we've done with TAK-925, the first orexin agonist to go into the clinic, our IV molecule with TAK-994. We've explored quite broadly a number of indications beyond NT1. So a, we know -- we have a good sense, at least with short durations of therapy, what one can see with this mechanism across a range of different indications. And then secondly, we have deep expertise internally and especially that translational expertise that help us understand what kind of molecule one needs in the various diseases as much as one can and how to develop those molecules across these diseases. So we feel quite excited about our franchise of orexin agonist, starting with oveporexton launch in NT1 likely that will be the single indication for oveporexton. TAK-360, which, as you mentioned, is in Phase II studies for IH and NT2. We hope to have data rolling out this year. And then the third molecule, TAK-495 that's just gone into the clinic. And all 3 of these molecules has very different pharmacological profiles that allow us to tackle these different disease states. I will -- if you don't mind, I just wanted to go back quickly. We are running one additional Phase III study for oveporexton. It's a study that will enable registration in Europe, but it will also provide additional information that will help us to further position it in the U.S. So it's a randomized withdrawal study. So we'll know how patients who are taking a medicine prior to washout do when they transition to oveporexton. And I did mention that the vast majority of patients normalize with the doses of oveporexton that we studied in our existing Phase III program. There are some patients that may need a little bit more agonism. And so we have the ability in this now ongoing Phase III study to test slightly higher doses for a small percentage of patients.

And you said that trial is required for registration in the EU.

That trial will be -- that's right. That's right. It's going to be a relatively small trial. The studies that we've run have been about 150 patients each. This one will be slightly smaller. And given our experience with sites and with investigators and the excitement around this mechanism, as soon as we open this trial up for enrollment, it's going to -- it will enroll like that. So that will happen fast.

And the rationale for higher doses is potentially better efficacy, I imagine.

Right. As I said, I think that we've hit it for the vast majority of patients with the 2 doses that we've tested. So right now, just so everybody knows, we have 2 dosing regimens. We have 1 milligram and 1 milligram given about 3 hours apart in the morning. We have 2 milligrams and 2 milligrams given about 3 hours apart in the morning. And again, the vast majority of patients are fine, many at 1 and 1 the majority of 2 and 2, there might be a small percentage of patients that just need a little bit more. So we have a chance in that trial to go a little higher. You can also imagine -- I mean, everybody's physiology around sleep wake is slightly different. So having these 2 doses, having them administered 3 hours apart, at least by way of our trial, you can imagine flexibility, one dose slightly higher, 1 dose at a slightly later time point. So we think that with that flexibility in the administration schedule, we're going to be able to achieve the needs of essentially every patient with NT1.

I actually failed to ask you about other indications that you might be exploring for oveporexton outside of NT1? Is there anything of note that we should keep on our?

So right now, our focus for OV will be NT1. For 360, you mentioned we're looking at other rare sleep wake cycle disorders like IH and NT2. And then as 495 matures and as additional molecules come in the clinic, you think of it as a pyramid with NT1 at the top, then these other rare sleep wake cycle disorders like NT2 and IH. And then it starts to open up to conditions that have larger patient sizes have different market considerations. And our goal is to focus on any indication that would lend itself to orexin 2 receptor agonism.

Is there an example or two you can give?

Sure. So there are a number of different conditions. So let's -- I'll give you buckets, let's say. So there are other rare disorders that we don't talk about like, for example, Prader-Willi syndrome. There are disorders that this mechanism has unique biology and pharmacology and that are driven by respiratory issues. So the prototype there would be obstructive sleep apnea, but there are many others that fall into this realm. Of course, the challenge there is that you need to maintain orexin 3 probably beyond the daytime and into the night, perhaps more so than we do in narcolepsy. And so how you thread that needle requires a lot of precision around the pharmacokinetic profile, but also the pharmacology of the molecule. And then you could imagine many neurodegenerative conditions that are characterized by either cognitive impairment because we're seeing just unprecedented results on cognition by sleepiness. My mom actually has Parkinson's disease and her #1 complaint, and this is common for many Parkinson's patients is that is the fatigue and sleepiness that she has through the day. So you can imagine a broad range of potential indications. I don't think you're going to find a single molecule that's going to work across all of these indications, which is why our strategy has been start with NT1, that's OV, move into other rare sleep wake cycle disorders and then start to expand and start to understand the pharmacology that you're going to need.

Any questions from the room?

Are you highly confident that the FDA will not want to see the data from the Phase III trial?

So it's a great question. So the trial hasn't started. And so we were cognizant of that question as we went into this. And I think that based on the package that we submitted because we had discussions originally with the FDA that one trial would be sufficient, and we knocked it out of the park with the first trial and the second trial then finished 5 days later. So we, of course, included both. So we feel very confident based on the quality of the package we've sent in, the discussions that we had with FDA before we started Phase III and then the interactions that we've had since. And I think that, that would be highly, highly, highly unlikely.

Yes. Eric?

There are so many different sources of idiopathic hypersomnia sometimes they can vary to medication prescriptions... Secondary to medication and other things. Where does that figure in your list of potential indications? And could you size that opportunity relative to others?

So maybe we should repeat the question because I don't think it was...

Yes.

So the question was about idiopathic hypersomnia and thinking about the heterogeneity of hypersomnia conditions that don't have a clear etiology and then the size of that disorder. So it's -- I mean, I dissect that question in a few different ways, Eric. One is that there is a condition that in and of itself is called idiopathic hypersomnia that has a code, a reimbursement code that's associated with it. And then there are other forms of hypersomnia that exist as part of other indications. So you have a disease and then you have symptoms, let's say. So the size of that disease is small relative to NT2 and NT1, but on the larger side relative to rare diseases. So maybe we're talking as opposed to 100,000 plus like you'd see in NT1 or maybe 200,000 plus that you'd see in NT2 to something in the order of 50,000 patients with IH. I'm ballparking that number for you. But as you can imagine, it's a difficult disease to diagnose and it's extremely heterogeneous. We were quite surprised and actually excited by the fact that the first molecule we brought in the clinic, which you'll remember, TAK-925, we did a proof-of-concept study in IH. And the levels of efficacy that we saw were essentially equivalent to what we're seeing in type 1 narcolepsy. I think that just speaks to the fact that regardless of why you're hypersomnolent, this mechanism will wake you up. In terms of the size of the population of patients who have hypersomnolence as part of some other disorder, there, you're probably talking millions, right? You're talking millions. If you look at diseases like Parkinson's or MS or even Alzheimer's, there have been estimates that suggest that 20% to 30% of these populations could have hypersomnia. Now how you target those populations, how you access that market, not simple because they're so heterogeneous and there you're not talking about the disease itself, you're talking about a symptom associated with the disease.

You talked about the data you've always been optimistic throughout. Can you just maybe discuss profile and...

Sure. So the tolerability and safety profile of oveporexton has been outstanding. I would say that the issues that we've seen are known mechanism-based issues. And the 2 most significant are obviously insomnia and then the second are urinary symptoms. Interestingly, we now have data that extends out well past a year, looking at patients, and you see efficacy that continues from day 1 to 1 year plus. You do see sensitization to many of the tolerability issues, particularly the insomnia. So patients that have insomnia, the vast majority, 80% to 90% will have it within the first week or 10 days of dosing. And then for the majority of those patients, it will go away. You do have some patients that have persistent insomnia. It's hard to distinguish what those patients are experiencing than what any of us might experience. But you do still carry forward that AE, that tolerability issue of insomnia. And it's pretty much the same with the urinary issues. You see sensitization early, perhaps not 80%, 90%, but the majority of those patients then sensitize and no longer have those issues. We don't see visual disturbances. We haven't seen them in any of our trials. The first time we really saw them were in our Phase III studies after there was just an immense amount of public awareness of visual disturbance that have come from our competitor molecules. And so of course, when physicians are asking continually and patients are primed, you're going to see this. And interestingly, we saw a fairly similar incidence of visual disturbance in our Phase III trials between the placebo arm and the active arm, which is suggestive that it's a suggestive phenomenon, not a real phenomenon. Whether there's something there because you do see -- you've seen reports of higher incidents with some other molecules, particularly at higher doses, I can't tell you, but we've had a hard time explaining mechanistically why orexin 2 receptor agonism would lead to visual disturbances.

Maybe we can move on zasocitinib. So you presented or at least disclosed the first Phase III psoriasis data. The full data, I think, will be presented very soon. Can you recap what you've disclosed so far and how it fits into the competitive landscape in your mind?

Yes, absolutely. So really excited. We had a range of possibilities for what our Phase III trial would look like. And I would say that this was at the kind of higher end. The results were at the higher end in terms of what that range would look like. They're very consistent with what we saw in Phase IIb, almost identical, maybe even a little bit better, which I think is quite unusual when you look at the work that we do. You typically see some weaning of activity between Phase IIb and Phase III for a number of different reasons. So zasocitinib, I think as everyone knows, is a highly potent, highly selective TYK2 inhibitor. It's clearly once-a-day dosing with a beautiful 24-hour coverage with the once-a-day doses that we are administering for psoriatic arthritis, which is our 30-milligram dose, we have 90-plus percent inhibition of TYK2 over a 24-hour period. There's no food effect, which I think is going to be quite important for this class. So a tolerability and safety profile that was very consistent with what we had seen in Phase IIb. And then an efficacy, as I mentioned, it was really just phenomenal. Over half of the patients had complete or nearly complete clearance of their psoriatic plaque on skin, slightly over 30% had 100% clearance. It is really just phenomenal. And as I said, it was as good as we could have expected. We looked at almost 45 different endpoints across our 2 trials, prespecified endpoints. And again, similar to oveporexton, all of them were significant and many with the number of zeros before the one beyond anything that I've ever seen before. So it's a really highly effective drug and really quite excited about it.

Great. So slightly over 30% PASI 100. When we get the full data at, I believe, 80, what data points would you urge us to focus on.

Yes. So we are presenting at AAD. So we just found out that we have a late breaker. So we're excited to have an oral presentation. I would look at -- so it's a lot. I would focus on the efficacy and the safety data, of course. And on the efficacy data, I would look at a couple of different elements. One would be the rapidity of activity, right? So many of the biologics can take months before you start to see effect and then it can take even more months before you see plateauing of that effect. We see benefits as early as 4 weeks, which is quite unusual. It's hard to do cross-study comparisons, but I think our conclusion, and you've come to your own conclusion is that we're as good or better than any other oral option that exists on the market today or that is coming on to the market in the near future. Yes. So we'll have an investor call as part of that. So it will be the presentation from the principal investigator and then we'll -- I'll be out there as well. We'll do an investor call and share our perspective and be able to answer questions as well.

Any question in the room?

[indiscernible]

So the question was how we feel our Zaso psoriatic arthritis data compares with the [ Ilumis ] data. It's -- I mean, it's hard for me to say because all we've seen from [ Ilumis ] is a press release and you see you get snippets. I think we'll have to wait and see what happens in their -- with their presentation. Just the one obvious differentiator that's just obvious is we're once a day and they're twice a day. And I think in a market like psoriasis, having a once-a-day option is going to be a considerable differentiation.

So zasocitinib is also in a Phase II IBD trial, ulcerative colitis and Crohn's, which is due to read out later this year. What are the data that support the idea that other TYK2 inhibitors have failed in IBD because they didn't achieve sufficient inhibition of the trial?

So interestingly, the reason we brought zasocitinib in was because of our strength and history and strategic interest in IBD. And then we became interested in the immunology conditions, psoriasis and psoriatic arthritis, of course. So there are 3 lines of evidence that suggests that TYK2 inhibition should work in IBD. The first is pharmacological evidence. You see biologics like IL-23 inhibitors that work upstream of TYK2 are highly effective. Now they have other pathways that don't necessarily require TYK2, but it's a strong, I think, a very strong rationale. The second is we look at animal models of colitis with all the caveats of an animal model, they're highly effective. And the third that has been, by far, the most compelling is the genetics. So we know that there are loss of function TYK2 variants that exist in a fairly high percentage of the population, 1 in 25 individuals of European descent. And so you have many patients who are heterozygous and many patients who are homozygous. And for both heterozygotes and homozygotes, they are protected against Crohn's disease and homozygotes, but not heterozygotes are protected from ulcerative colitis. So yes, the existing TYK2s that have been tested in IBD have not been effective. We strongly believe what you said, Mike, was that none of those have been tested at the level of inhibition that you need to see efficacy. So interestingly, the animal models, the exposures that you need to see efficacy in colitis are much greater than the exposures that you need to see efficacy in models of psoriatic arthritis, right? The genetics, the fact that you see protection and you see in the homozygous and the heterozygous suggests again that you need more exposure. And the doses that we're testing in our Phase IIb studies give us exposures that are probably 3-plus fold more than what's been studied previously. So we'll see what happens over the course of this year as those data read out, but we continue to feel quite optimistic about the potential in IBD.

Questions from the room? So when we get those Phase II data, the first thing that people like me will do is put them in a chart next to other data. Should I include in that chart JAK inhibitors? Or are they so different as to be an inappropriate comparator, mainly because of safety issues?

Well, I mean, if I were in your shoes, I would include everything because why not? Because if you have a drug that's available, you want to make sure that you're including it in your thinking. It's not how we think about it. Firstly, I don't know whether JAK inhibitors are quite effective in IBD, and they come with the baggage that we know about, but the efficacy is quite substantial. In fact, as great as anything that's really that's out there. Other oral agents that are used in IBD are significantly less effective. There are very few oral agents that have the efficacy, tolerability, dosing ease and safety that we're expecting with zaso. So I think we'll have to see what happens. The other thing that we're looking into, we think that zaso in and of itself has potential as a monotherapy in IBD, and that's how we've designed our Phase IIb studies. I think if you look at where IBD is headed in the future, given the efficacy ceiling that we're seeing with every agent that's out there, if you really want to think beyond and as we're starting to think towards kind of having generic penetration, you have to start thinking about combination therapies. And we think that zaso, given its mechanism, given its tolerability and safety profile, lends itself nicely to combination therapies as well.

And a good combination partner presumably would be...

[ IBD1. ] That would be one, of course, yes, that mechanistically makes a lot of sense given the synergies between the pathways.

Okay. Maybe we could talk about -- a question from here.

When it comes to IBD, I'm just curious how you think when we see limited efficacy, sometimes what we see is right? I'm curious if you think that adding that second immunological agent is what will get us to mucosal healing? Or do we need to do something else to get the regenerative capacity of epithelium going?

Yes. So I'll have to get someone who knows much more about this space to come and speak with you. Yes. Well, I think that the answer to your question is I think that we know that there's still -- even patients who respond to then relapse, they still have an inflammatory disease. Right? They're still primarily an inflammatory disease. So I feel with high confidence that having coming at the immune system through different angles will give you added benefit and maybe the potential for more durability, right? Whether you need additional mechanisms to overcome other aspects of this disease like the stenosis that we see or other aspects of the disease, I think absolutely for sure. But that doesn't preclude the potential for combined immunological mechanisms.

Yes. And my question was about mucosal healing and prospects there.

Yes.

I don't want to neglect rusfertide in our last couple of minutes at launch in polycythemia vera is on the horizon. I think phlebotomy is the procedure that you hope to essentially replace. Can you just tell us a little bit about the competitive landscape there for rusfertide?

Sure. Well, essentially, there is none. But there isn't a very effective, very inexpensive standard of care, which is phlebotomy with its limitations. But maybe I'm glad we save time for rusfertide. I'll rank order what I see as the R&D head as the challenge. The 3 programs that we have and the rank order of the challenges of our launch, the easiest, although nothing is easy, is ovaprextin, right, then rusfertide then zaso. And that plays to really the market -- how the markets exist today. So rusfertide, the profile is just incredible in terms of what we're doing for patients. We're seeing -- we now have 4 years data, and we're seeing that with rusfertide, patients with polycythemia vera 90-plus percent are still within normal hematic ranges at 4-plus years, which is just extraordinary. The challenge is that -- and this is, by the way, rusfertide on top of standard of care. So standard of care would be hydroxyurea or other cytoreductive agents. About 50% of patients end up taking hydroxyurea. I don't know that patients will need to continue on both. It's just that physicians like as they start a patient on one, they add another, they tend to maintain that paradigm. The problem is that phlebotomy is about 75% of patients don't achieve hematic levels of below 45%, which is target with phlebotomy alone. A lot of that is compliance, right? But why are they noncompliant? Because it's an awful procedure to go through at the frequency that you have to. Secondly, there are issues with phlebotomy, so iron deficiency ironically, and there are a number of different sequelae of iron deficiency that range from organ defects to just constitutional symptoms. And the third is that patients who have phlebotomy just don't feel better. And one of the benefits that we've seen in the rusfertide Phase III program is the benefits on patient-reported outcomes that patients are reporting.

Great. Well, in our final moments, I just want to ask a forward-looking question. If we fast forward 10 years from now, what do you think will be most different or surprising at Takeda relative to how we see the company.

10 years is a long time. I can't even think. But if I were to think even 5 years, I'll take your question 5 years I think it will be -- everybody is talking about this and you and I have the conversation. I think it will be how much technology is supplanting our ways of working today. I'll give you one just example in R&D, which is we're about to move into a new building from our Kendall Square -- I'm sorry, from our Central Square location in Cambridge into Kendall. So it will be completed in the next 9 or 10 months. And 80% of our labs in the area will be in that building. 100% of those labs will be digitally enabled. So the whole workflow of how you take a program from target discovery to candidate, not that every part of it will be digital, but every part of it will have a digital enablement, including everything that we do at the beginning. It will look completely unlike any lab that we see today.

Great. Well with that, thank you so much.