Document Details

# GSK plc GSK plc presents at TD Cowen 45th Annual Healthcare Conference

Monday, March 3, 2025 9:50 AM

## Event Participants

Analysts 1 Steve Scala

Executives 1 Luke Miels

#### Steve Scala Analyst

Well, good morning once again. We're delighted to have GSK at the TD Cowen conference once again this year. Representing the company is Luke Miels, who is the Chief Commercial O!cer. So Luke, thanks so much for making the e"ort to be with us today.

#### Luke Miels Executive

Thank you, Steve.

#### Steve Scala Analyst

There's a ton going on in your business today, and we want to dig into all of it. But maybe we could start out by you identifying the 2 or 3 most important dynamics in the commercial operations that will transpire in 2025.

#### Luke Miels Executive

Sure. So I think number one is BLENREP, the launch of that. We have the PDUFA in July. Second would be the introduction of Nucala into COPD. And I think the third one would be just extracting more from Shingrix overall.

But...

# GSK plc GSK plc presents at TD Cowen 45th Annual Healthcare Conference

Monday, March 3, 2025 9:50 AM

## Event Participants

Analysts 1 Steve Scala

Executives 1 Luke Miels

#### Steve Scala Analyst

Well, good morning once again. We're delighted to have GSK at the TD Cowen conference once again this year. Representing the company is Luke Miels, who is the Chief Commercial O!cer. So Luke, thanks so much for making the e"ort to be with us today.

#### Luke Miels Executive

Thank you, Steve.

#### Steve Scala Analyst

There's a ton going on in your business today, and we want to dig into all of it. But maybe we could start out by you identifying the 2 or 3 most important dynamics in the commercial operations that will transpire in 2025.

#### Luke Miels Executive

Sure. So I think number one is BLENREP, the launch of that. We have the PDUFA in July. Second would be the introduction of Nucala into COPD. And I think the third one would be just extracting more from Shingrix overall.

But I can add to that list, you've asked for 3.

#### Steve Scala Analyst

Okay. We're actually going to get to Shingrix eventually. But let's start out with number one, and that is BLENREP. So I assume the review is going well and that we should not be overly apprehensive about the PDUFA. But therefore, describe for us the launch curve.

## What's the launch curve going to look like?

## Luke Miels Executive

Yes. I mean I think it's an unusual case because it's a relatively rare event that you relaunch a product. And the data that we've generated through essentially the relaunch program for DREAMM-7 and DREAMM-8 is very compelling. But there is an art to using BLENREP. It's relatively simple to use versus options such as CAR-T, and I would argue bispeci#cs, but we want people to have those #rst 5 patients a really good experience, understand how to manage the dosing, when to dose hold, when to infuse based on the patients, the status of their vision.

And so it will be a tiered introduction, will initially go to HCPs that have used hem/oncs, that have used the product in the past, and then we'll expand out. But ultimately, the destination, of course, is to use this very broadly.

Our argument as to why this can be used very broadly, you really need to look at the options that someone treating multiple myeloma patients has in second line today. The assumption is in the U.S., that you're going to be using a CD38 in #rst line; and so your choice in second line is a CAR-T, which is a minority of patients in a minority of centers; or a bispeci#c, which they don't have a label there, of course, they don't have survival data there, but they're really your primary choices if you're looking at chasing e!cacy; or you have BLENREP. And with BLENREP, with DREAMM-7, we're projected to have a 33-month incremental survival bene#t over and above daratumumab, which is obviously very compelling.

But there's an element of BLENREP's pro#le, which is blurred vision. When we #rst introduced the product, there was a lot of questions around it. We've now dosed over 7,000 patients. Every patient that's had any ocular issues, any blurred vision, it's been completely reversible to date. And so what we want to ensure is that physicians understand how to dose the product and when to dose hold and when to start.

And that's why that process there is going to be a little bit slower.

## Steve Scala Analyst

Okay. I should have mentioned at the outset, should you have a question anywhere along the line, just raise your hand, and we'll call on you.

So Luke, you mentioned that the #rst targeted physicians will be experienced or advanced hem/oncs and so forth. But what's the nature of the #rst types of patients that will be put on this drug?

## Luke Miels Executive

Yes. I mean the aim, of course, is second line. That's where the label is. These are patients who are largely in the community. This, ultimately, we think is going to be the choice for community oncologists because, again, if you look at the options -- and that's 70% of the patients in the U.S., by the way.

So CAR-T is an absolute minority. I think most CAR-T clinics have a capacity for about 250

patients a year. If you want to use a bispeci#c, you've got to admit the patient to hospital. You've got a complex dosing titration regimen and you have a pretty severe and unpredictable infection risk.

Whereas with BLENREP, which, again, it's likely to come down with bispeci#cs, but it was still 20% in their poster with Grade 5 fatal infections at ASH this year or last year, so with BLENREP, you're looking at managing eye-related side e"ects. If you treat 100 patients in the community or in an academic setting, 2/3 of them are going to have no visual disturbance. And that's the key thing that is often lost in these discussions: 2 out of 3 have no impact on their vision. The 1/3 that have an impact on the vision, the bulk of them, it's from some disturbance of, say, 20 to 50. But if you look at the time that they have that disturbance, it's around 11% of the time that they're on treatment.

So it's for periods around dosing. And what we've also learned is that, if you dose hold, you can get that patient to normalize relatively quickly and you're not sacri#cing e!cacy. We've shown that through DREAMM-7, and that protocol is now being used in our #rst-line approach.

And if you look at people that have severe visual disturbance, say, 20 to 100, 20 to 200, it's about 2% of patients. So it's dramatically less and a number of these patients can get to a steady-state infusion of 1 infusion every 10 to 12 weeks. You don't have to bring them in the hospital. You don't have to infuse immunoglobulin. So it's a much simpler regimen once you've got the patient up and going.

Hence, we ultimately aim to be the choice for community-based hem/oncs.

# Steve Scala Analyst

Okay. Now you mentioned that it will be a tiered introduction. But tell us, after this tiered introduction, what it's going to look like. So will there be in\$ections? And what will be those in\$ections?

Will it be guidelines, reimbursement, more data? What will drive these in\$ections?

# Luke Miels Executive

Yes. I mean guidelines are key, of course, and the sentiment around there. We're somewhat limited at this point to what we can do with guidelines because we don't have an approved product. But that is an important component just in terms, particularly with community oncologists and hematologists, hem/oncs. But I think the #rst 6 months of this year will be more steady state, carefully accumulating experience.

We don't want people just to just to go rapid-#re with the product. We want to support them so that their #rst few patients, they have a good experience. They understand that it's a very predictable product, and it can be managed very well in their clinic. And once they accumulate experience, then I think we can open it up. We'll start initially with people that have experienced with BLENREP in the past, then we'll look at community-based physicians and practices that have a lot of experience or have participated in our programs, go to them #rst and then ultimately, the broader community, so a stepwise approach.

## Steve Scala Analyst

Okay. And when would the product be #rst eligible for consideration for inclusion in guidelines?

## Luke Miels Executive

I mean that's really the gift of the guidelines committee. Sometimes they move very quickly. Sometimes they wait a number of months. So it's di!cult to predict. But all the members are well aware of the data.

DREAMM-7 is really quite striking. You've got a hazard ratio, which is approaching that of what you achieved with the CAR-T. And as I said earlier, you're adding 33 months of survival on top of what was standard of care in the form of Darzalex, daratumumab at that point. People have not missed that. We've got updates to DREAMM-7 and DREAMM-8 at ASCO this year.

So that timing is helpful and then further data at ASH at the end of '25.

#### Steve Scala Analyst

Would you push back on the assertion that it should be rapidly considered by guideline committees? Would that be too strong?

#### Luke Miels Executive

I'm going to hope so. But again, we have limited in\$uence as you do with good reason, right? These individuals need to be independent and seem to be independent. But I would expect it's something that's considered -- I don't think we'll be waiting 6 months. I think they'll be looking at it relatively quickly because of this gap that's opened up in second line.

And people do need guidance in terms of what's appropriate to employ in second line, if you're electing to use daratumumab in #rst line.

#### Steve Scala Analyst

Okay. GSK has provided a peak sales estimate of GBP 3 billion plus. Can you just dissect that a little bit for us? What's embedded in that in treatment duration, market share? And I assume you're not going to speak to price, but if you will...

#### Luke Miels Executive

The pricing, we are talking to payers about pricing. We'll be intelligent with pricing. But we have a good sense of the price that will provide good access but reward the pro#le of the product. If you look at the GBP 3 billion plus, that is second line plus, so it doesn't include any potential label that we -- currently, we have experiments running for #rst line. But yes, the bulk of those patients are in second line.

You can model the duration roughly of what we've seen so far with DREAMM-7, as I mentioned before. So 84 months would be an evidence-based time frame to model. The complicated element is that the dose initially drops relatively quickly in their frequency. So the initial dose is every 3 weeks, relatively high. And as I said, many of these patients are

going to get out to about half or 1/3 of that every 9 to 12 weeks.

So again, we've been quite thoughtful around how we price that and looking at an annualized or through treatment costs with payers.

## Steve Scala Analyst

Okay. One more question on BLENREP before we move to other topics, and that is that, in your mind, you have a best case outcome for the REMS. How does that best case outcome compare to the REMS that was in the original label?

## Luke Miels Executive

Yes. I mean the REMS that was in the original label was onerous. And for those of you that remember, in the AdCom, there's a lot of discussion at the time around what is the pro#le of this product in terms of keratopathy of the cornea. We only had a couple of hundred patients experience then. We're in a dramatically di"erent point now.

So ultimately, it's in the arms of the regulator. But our expectation is there will be some form of eye exam initially, but this is a simple slip lab examination. And the intent is that it can be conducted by an optometrist.

If you look at a typical multiple myeloma patient, 90% of them are within 30 minutes of an eye care professional optometrist because many of them, of course, have glasses. So the capacity to support those people and do that is there. And we also have a huge infrastructure that we're putting in place to help physicians navigate and their patients navigate that as you would with any other REMS program. And REMS #nally are -- I mean, most products in multiple myeloma actually have a REMS program. So it's not unusual for that to be there.

# Steve Scala Analyst

Okay. Questions from the audience? Okay. Let's move to vaccines and spend a few minutes here. First on Shingrix, do you expect U.S.

Shingrix to return to growth, plateau or decline through the end of the decade in the U.S.?

# Luke Miels Executive

I mean if you look at the total population that is 50-plus and presents for a regular adult vaccine, it's about 70 million people in the U.S., 72 million, 73 million. 4 million are added, obviously, some people pass away, but steady state is around 70 million. We have 50 million of them already vaccinated with at least one shot of Shingrix. So there's another 22 million steady state that we have a good chance of picking up. In the past, the penetration rate for Shingrix, we added about 6 or 7 people -- 6% or 7% a year.

So we #nished last year at around 40%, 41% penetration. But what we've said consistently is we expect that accrual rate to drop to a rate of 3% to 5% because you've got less engaged people, less motivated. But I do think we can work through that population. So growth is more challenging, but we can still ultimately capture those patients and vaccinate them over time.

The other thing we're doing is shifting the strategy. This has been very successful in Europe,

in markets like Germany. We launched a broad approach in the U.S. If you're this age group, you should consider a Shingrix vaccine, your risk of shingles and such. What we're now doing is ramping up our interface, and that was really a retail pharmacy-driven strategy.

What we are now doing -- we're not abandoning the retail pharmacy strategy, but what we're doing is increasing the investment behind physician education and really looking at subgroups of patients and looking at physicians that have this high usage of vaccines, but don't necessarily use a lot of Shingrix. And that combination has worked very well in Europe.

And that's one that we're actually launching now in the U.S. We're building a specialty team to target oncologists, dermatologists, cardiologists, groups like that, to encourage them to get their patients vaccinated with Shingrix because it disrupts the primary care that they're trying to deliver themselves, whether it's chemotherapy infusions or stenting a patient and then having the recovery post-stent. And we've seen that resonates really well in other markets in the world. So that's what we're doing with Shingrix.

Europe is an opportunity to grow. Emerging markets, excluding China, is an opportunity to grow.

## Steve Scala Analyst

So all things considered, in the U.S., sales in 2030 are for Shingrix slightly higher than current?

## Luke Miels Executive

I won't speculate because that man in the front row there called Mick will throw his laptop at me or something at me. But what I would say is that our aim is to accrue those 22 million people over time. And we are at a relatively early phase in most markets in the world. And if you look at even Europe, I mean, we're just introducing the product into France now. We've got a good restart and an in\$ection point in Germany.

So yes, this is a very valuable, very e"ective product. It will remain like that for a number of years.

# Steve Scala Analyst

Okay. Let's move to RSV. What data points does ACIP need to see to consider a booster dose for the RSV vaccine?

## Luke Miels Executive

It's di!cult to speculate. And of course, there's a lot of variability in this environment right now. We will have the 36-month immune data, study 004, hopefully presented to ACIP in June. But I think they're going to want to see breakthrough infections. All of our modeling indicates that's going to happen.

This is not a one-and-done vaccine. For simplicity's sake, we modeled 5 years. Historically, ACIP has dealt in prime numbers, so 1, 3 and 5. So I think it would be judicious to model a 5 year vaccination cycle, which is a remarkably e"ective vaccine for high-risk individuals. And so I think there's some strategic patients needed here.

The product is very compelling. We're having good success outside the U.S. in terms of contracts and reimbursement, obviously, di"erent systems to the U.S. But yes, I think in time, the bene#ts of this product will be apparent in the U.S. setting.

And I think in time, we'll see revaccination, and in time, we'll see an expanded population. But it's just not going to happen tomorrow, and it's not going to happen in 2025.

## Steve Scala Analyst

Okay. Let's move on to one more vaccine, and that is, what are the implications of the delay to the U.S. ACIP recommendation for Penmenvy. That was supposed to occur this month. And what's the path forward now?

# Luke Miels Executive

Yes. I mean that was -- we were not expecting that, of course. This initial meeting was not as important as the second meeting that was expected in October. When P#zer's pentavalent was approved, ACIP said, "Look, we'll put it in a context of shared clinical decision-making, but we want to come and revisit this schedule and move it to more of a risk-based." And that's why you've seen relatively low uptake of their pentavalent. And I don't think we would see a dramatic uptake of ours until you move to really looking at segments of adolescents that are going to be in high-risk environments for meningitis.

That being said, we have an outstanding business with the stand-alone B vaccine. We get 76% of the market. It's perceived to be superior to P#zer's product. Remember, we cover about 110 strains. So it's the preferred B.

So we can o"set any delays there because it just means the uptake of the pentavalent is going to be delayed.

The other critical thing is the compliance rates when you get into the B vaccine strains are relatively low. It's about 1/3 of adolescents that get it. Again, a risk base would be, okay, your son or your daughter is going o" to college, they should really get a vaccination. And the pentavalent for that middle vaccine increases compliance, makes things simple. And there's a history of that in pediatric vaccines.

And the B that they use in the middle pentavalent, they have to use the same B for the followup B shot, which again is very helpful for us because they prefer to use our B, which places our pentavalent, when it's eventually approved by ACIP, in a strong position.

## Steve Scala Analyst

And is there a path for preferential recommendation of your vaccine?

## Luke Miels Executive

ACIP is historically very reluctant to separate out vaccines. I mean people forget this, but Shingrix has only got a preferential vaccination recommendation by 1 vote at ACIP despite the fact that it had, I think it was 89% e!cacy, more durability versus Zostavax, which was about 50% e!cacy and relatively short duration. So there's a reluctance to do that. It's in the interest of competition and supply stability. So I would not expect that.

Now in European systems, they are more driven by raw clinical data and relative di"erences. Price is a component. But tender specs and things like that can be more targeted around the evidence that you have versus extrapolating or assuming class e"ects, which ACIP is more typically inclined to do.

# Steve Scala Analyst

Before we leave vaccines, any questions from the audience? Okay. Let's just brie\$y touch upon Benlysta and then we'll go to depemokimab. When you look at Benlysta today, what percent of sales are in lupus versus lupus nephritis?

## Luke Miels Executive

Yes, it's 60-40. So 60% in SLE, 40% in lupus nephritis. If you look at the subcu-IV split, it's actually 60-40 as well. So 40% of those patients are coming in on a regular basis for an infusion, which is important when we look at longer-acting B-lymphocyte type programs in the future. But I mean, Roche obviously presented Gazyva data.

I think that data was pretty much what we're expecting. It's in the range of Benlysta in terms of e!cacy, but you have a far more complicated toxicity pro#le, a black box on pregnancy, et cetera.

I think lupus nephritis will continue to grow. The guidelines in the U.S. have just shifted. We had global kidney guidelines, European kidney guidelines for a year or a couple of years now, favoring the early use of Benlysta. And the U.S.

guidelines have now just shifted last year, encouraging and they're very clear that Benlysta should be used initially. So our market research indicates that physicians see Gazyva as a refractory later-line option and that Benlysta is perceived to be the ideal balance of e!cacy and non-toxicity pro#le, tolerability pro#le, for those patients.

## Steve Scala Analyst

This would not be unexpected, but Roche thinks the guidelines will be revised and that Gazyva will be the drug of choice. But you believe the data is such that Benlysta will not be at a disadvantage.

## Luke Miels Executive

Yes. I mean they missed their secondaries. The e!cacy rate is not dramatically di"erent from -- if you allow for cross-trial, it's really in the realm of a similar pro#le. Rituximab, we know the heritage of rituximab there. So again, our view is at least what we're hearing is that no, Benlysta will remain the favored agent to be used in those early patients.

## Steve Scala Analyst

Okay. Why don't we move to depemokimab. And I think there was some news this morning. So in case anyone missed it, why don't you just recap what the news?

Luke Miels Executive

Sure. So I mean, we are #ling. The initial launch will be in severe eosinophilic asthma, but also in nasal polyps, and that is an interesting area. The bulk of the value, of course, is in severe eosinophilic asthma. And depemokimab, we think, is going to be very disruptive in that setting.

It's a twice-yearly, long-acting IL-5. This is, we think, compelling for a range of reasons.

Firstly, the IL-5 target is well known. You've got mepolizumab, which is ours, Nucala; you have benralizumab from Astra. It's well characterized. The pro#le of the product, if you look at the publications, is typically close to placebo. So it's well-known by pulmonologists.

The challenge is, if you look at penetration rates of biologics in severe asthma, I mean, these are very refractory patients, it's about 28% in the U.S. And there's a range of reasons for that. Insurance, if you look at lives covered, is about 95% or higher. So it's not an insurance issue.

And the other challenge we have is, if you look at the stats, and it's the same for -- and other biologics in severe asthma, typically, someone who started on the 1st of January, you lose about 1/3 of those patients by the middle of the year and you've lost 2/3 of them by the end of the year. So what's attractive about depemokimab is the physician has complete control of the patient. Our intent is to launch it into Part B, B for Bravo, so it's physician administered. And so we also know there's a lot of patients that would be open to a biologic, if they didn't have to inject it themselves. And then you've also got adolescents, pediatric patients where the parent is having to do this.

And so ultimately, the physician, if they want, can buy and build the product, administer the product and then the patient is covered for 6 months, and you don't have to rely on the patient complying.

And then within nasal polyps, we've got a battle emerging here against Amgen and Astra with the TSLP there. But I think that in our market research, with the data, it's very attractive to surgeons because you can do -- most of these nasal polyps patients require multiple surgeries, which is frustrating for the surgeon. And also, of course, the patient is not too happy about it. And so the capacity -- I think it was 88% reduction in surgery with our program is really attractive. And again, surgeons, they're not known for their love of complicated drug regimens.

So the capacity to be able to essentially conduct the surgery, inject the patient, discharge the patient and know that the patient is covered for 6 months, and you don't have to worry about your o!ce supporting all of that or compliance, et cetera, or writing letters to the referring pulmonologist, and then you can bring them back in for a follow-up appointment, typically a few months later, and then you can do the second dose. And you've got 12 months of coverage. So again, it just simpli#es things. Penetration in nasal polyps is about 14%, which is relatively rapid so far. It's about \$1 billion, I think, for Dupixent in the U.S.

So again, these are the areas that we're interested in. We also have EGPA, HES, other indications coming in the next 2 years. And our intent is to launch a COPD experiment as well.

## Steve Scala Analyst

So you mentioned launching into Part B, which has the advantage of the physician administration and so forth, and you cited the advantages of that. But investors are sensitized to other drugs which launched into Part B and really were not all that successful, most notably Novartis' Inclisiran. So why will this be di"erent?

### Luke Miels Executive

Yes. I mean it's a good question, right? And there are bisphosphonates as well as PCSK9. So I think the main element is, if these patients don't comply, the odds of them having an exacerbation, being hospitalized -- this is a severe disease. These are not people just pu!ng on Ventolin.

These are patients that are being managed by pulmonologists, not primary care doctors. So they're at the severe spectrum of asthma. And asthma is obviously a very dangerous disease if it's not managed. So it's di"erent to lipids where there are other alternatives. And again, it's an auto-injector.

It's a very thin high-gauge needle. So simplicity, bring the patient in, have the nurse inject them and then you're much more con#dent that, that patient is going to be under control.

#### Steve Scala Analyst

Okay. On the Q4 '25 conference call, you'll give us an update on access for depemokimab. What would you predict that number will be?

#### Luke Miels Executive

I mean our initial feedback is it's going to have very good access. Again, look, we're not looking for a massive premium versus Nucala, no. And our intent is to target FASENRA and Dupixent, and the TSLP market and the incentive scheme for our organization will be geared that way. What we're o"ering is predictable e!cacy, durable e!cacy. And the other thing we'll run, real-world evidence studies with payers as we've done with Nucala, showing that compliance is important.

And there's an increasing body of evidence, which is being led by Nucala, which is showing that you can actually start to remodel these patients' lungs, if you can intervene early enough. And again, compliance is a real problem here in keeping patients on study.

So the experiments that we have, we have a study that we'll report out next year, looking at people who have stabilized on FASENRA and Nucala, a double-blind, double-placebo, switched across on to depemokimab. Again, that is not an in-the-wild environment. You have a nurse following you up. It's the best case there, but we'll run studies which are actually showing people in the real world and what they actually do to comply. And the #gures I cited earlier, if they're replicated in the study, then over time, the value of depemokimab will be apparent to payers.

### Steve Scala Analyst

Okay. Questions from the audience? Let's spend our last few minutes on camlipixant. What do you think the probability of success of the Phase III trials is?

## Luke Miels Executive

It's good, I mean, for a Phase III program, a novel Phase III program. Why do I say that? Because we acquired that from BELLUS. I was directly involved in that deal as with all our deals. But for me, what I wanted to be convinced of is what have we learned from gefapixant.

And there are chemical pro#le measurement and study design lessons that are very clear there. So from a pro#le point of view, I mean, gefapixant validated the target. It does work to suppress cough. The problem is the primary target is P2X3, but gefapixant is relatively promiscuous and binds to another target called P3X3, which is present in the taste buds. The problem is that the patients get this taste disturbance, which is a bit like quinine.

There are problems with that. Obviously, people don't like that, and it essentially unblinds the medicine.

And so the rate that Merck experienced in their program was around 65% of strong taste disturbance and a lot of dropouts and also put a ceiling on the dose that they could pursue. We have about 6% of people in Phase II that had some taste disturbance. So relatively mild, and we didn't have dropouts because of that. The other challenge that Merck had was these patients are monitored for a year. So they wear a cough monitor counting the number of coughs and you use software to compress that.

Otherwise, you have to listen for 365 days to every patient. And so there were some di"erences with the agency around that. We actually incorporated that agency feedback into our program. And then the #nal one is just looking at the type of patients and cough frequency and using the insights we've got there.

So the unmet need is very high. I would encourage anyone if you're at ATS or other conferences like that and you're talking to a pulmonologist, ask them about the most frustrating part of their practice. There's a good chance that they will say it's chronic cough. These patients typically are coughing for a number of years. They will go on 3, 4 or 5 other specialties before they end up at a pulmonologist.

So typically, it's a disease of exclusion. So no malignancy, re\$ux, asthma. And then when you look at the options that these patients have, once they actually reach a cough clinic, are relatively low. The most e"ective is speech therapy. There's not enough of those people.

Anti-cough medicines make you drowsy and are not good long-term. Low-dose opioids, pretty obvious why they're not a good idea. And when you look at those pulmonologists, the research that we have, only 3% are happy with the options in chronic cough. So there's an unmet need there.

If we take the most severe population, like the most severe, there are people who are coughing hundreds of times a day, it's about 1.8 million people in the insurance databases in the U.S. So severe population, very synergistic with our respiratory business. We saw around a 34% reduction in cough frequency in our Phase II. If we can achieve above 15%, I'd love to achieve that in Phase III, but if we can get above 15%, 20%, then I think we've got a compelling asset here.

# Steve Scala Analyst

Okay. Great. And we actually are out of time, but let's ask one more question. And that is, so we didn't talk about the oncology business. You got some very important assets here.

What do you expect sales of these products to do over the next 2 to 3 years, accelerate, decelerate? So we're talking about Zejula, Jemperli and Ojjaara?

## Luke Miels Executive

Yes. I mean I think with PARPs, relatively static. That's not where our focus is. I mean we have a study for glioblastoma with Zejula, which is, I think, very exciting. We have very high CNS penetration with that.

There's nothing in glioblastoma for the last 40 years. So we did a study there that we're now replicating with a licensing study, pivotal study.

But Jemperli is still growing. I mean if you look at the penetration in endometrial cancer, people are often surprised when I say that we split the market evenly with Merck, with pembrolizumab, which is, I would argue, no small achievement there. And that is continuing to grow. The opportunity for I/O in these patients is very high. And we're also looking at nonchemo regimens and experiments there.

We also have a program in locally advanced rectal cancer and also programs in CRC with Jemperli. So that's exciting.

We just did a deal at Christmas time to buy a product for gastrointestinal stromal tumors. And this is a product which is very, very e"ective at the exons that drive gastric stromal tumors along with imatinib, but imatinib does not work on the escape mutations. And so this is an asset that it works on the most common escape mutations, which are exons 13 and 17. And it's also, like camlipixant, very, very speci#c in terms of its binding behavior. So you don't have as much o"-target issues with it.

We had 200 patients that IDRx had dosed and treated. So we really had a good idea of the clinical pro#le of this. So that's an exciting one.

And then Ojjaara, which is a very similar deal, again, looking for a more targeted, very predictable product that solves a problem that physicians have. That continues to do well in myelo#brosis. We've just actually started yesterday a study in combination as well to broaden that population. We get about 20% of #rst-line patients in the U.S. and 40% of second-line patients against ruxolitinib, which again is, I think, probably above what people expected us to do.

And then the #nal thing is we'll have updates over the year for a program called B7-H3. Merck acquired the Daiichi Sankyo program for B7-H3. We have a fast follower with Hansoh, and that is looking really quite exciting. We have 2 breakthrough designations -- actually, 3 on small cell, which again is, okay, it's 10%, 11% of patients, but an entry point that we could be very close to Merck there; as well as osteosarcoma. And there's a whole other group of tumors that we're looking at.

So it's very targeted. We're humble in our approach here, but we think that we can build this out to be a valuable business and something that's attractive for shareholders and with products that really are doing something over and above what's available for patients today.

#### Steve Scala Analyst

Great. Lots to look forward to and monitor. So thank you so much, Luke. It's a great rundown.

## Luke Miels Executive

Thank you. Thanks, everyone.