Amgen Inc. (AMGN) Earnings Call Transcript & Summary

Excellent. Hello, everyone. Thank you for joining us. I know there's a lot to discuss. I'll turn it over to you, Peter, perhaps to kick things off, and we'll jump right into it.

Great, Umer. Thank you so much for inviting us again to be here. We're grateful to you. We're grateful to Evercore. We're always glad to be back to see our colleagues. And I wanted to start with where I was a year ago was with you at the conference, and you opened up by saying, I've been following Amgen for, I think you said about 15 years at that time, and you said, Umer, you said, Amgen's now turned from a capital return story to a growth story. We couldn't agree with you more. And we think this last year really proves that out and the future really proves that out. And so I'd like to spend a couple of minutes on the business, the 4 therapeutic area groups that we're in and why we're so confident, why we're excited about what we see coming in there. We continue to drive innovation at speed and scale leading to a strong long-term growth outlook with breadth and depth across our 4 therapeutic areas. In the third quarter, we generated 23% revenue growth year-over-year. We had 10 products delivering double-digit revenue growth in the third quarter. We had 14 products annualizing at over $1 billion based on third quarter product sales. Let's start with oncology for a minute. The innovative oncology portfolio grew 17% in the third quarter year-over-year. Jay can share with you some of the progress we've made in the pipeline there. I'll just put up xaluritamig, another bispecific T-cell engager moving into Phase III for advanced prostate cancer. Then we turn to our rare disease group. We closed Horizon on October 6 of 2023, we're more excited about rare disease today than we've ever been. We've got 4 great early cycle medicines in there that were up 21% year-over-year to $1.2 billion in the third quarter. Then let's -- and by the way, I'd like to say in rare disease, with UPLIZNA, and I'm sure you may explore that a little bit with Jay even the use of that in terms of autoimmune and the investigation of that. We now have a priority review with the FDA on UPLIZNA for IgG4-related disease with a PDUFA date of April 3. So we're very excited about that, first, of course, for patients. And then, of course, for the ability to continue to work on and deliver that medicine, which we think has other indications coming with it, including GMG. So now let's turn over to inflammation. Inflammation, almost $2 billion in the quarter. And led by TEZSPIRE, a medicine for severe asthma that was up 67% in the quarter year-over-year. And we're also working on initiating the Phase III in that, in severe chronic obstructive pulmonary disorder in the first half of 2025. Finally, let's finish with General Medicine. Repatha up 40% year-over-year in the quarter. Strong volume growth of 41%. EVENITY up 30% in the bone portfolio, a very strong bone portfolio, and we think EVENITY is going to be a great medicine over the long haul, performing very, very well. And then in general medicine, olpasiran in the pipeline. It's in Phase III. We enrolled about 7,200 patients faster than we've ever enrolled in an outcomes trial of that size, and we think one of the fastest in the industry. And so we're very, very excited about what we see in our in our 4 pillars. Quickly on MariTide. I'm going to give you a couple of thoughts. I want to make a few points, and then I'm going to ask Jay to jump in and make a couple of points and flip it back to you for questions and answers. I think on MariTide, what we want to make sure we talk about is it's a Phase II trial. So we always want to keep apples-to-apples here when we're thinking about what's going on. We are going to have a wealth of information we're going to use as we continue to be laser focused on designing the best Phase III for patients that we possibly can do in that investigation. Roughly 20% average weight loss at 52 weeks without a weight loss plateau, patients were still losing weight. This is the first obesity treatment with monthly or less frequent dosing every other month studied in part one of Phase II, quarterly being studied in part 2 of Phase II. We think there's many implications of that, Umer, but one will be persistence. And I think Susan can speak on behalf of that for patients and prescribers when we get to that. I think next, going over to substantial improvements in cardiometabolic parameters 4% low discontinuation rates in patients in the dose escalation arms, 11% due to any adverse event, 8 percentage points to those 11% out of an abundance of clarity were due to GI issues and those dose escalation arms and more than 90% of the patients eligible chosen to continue in part 2. Then in diabetes, 17% average weight loss at 52 weeks, no weight loss plateau. Patients continue to lose weight, lowered average H1ABC -- excuse me, hemoglobin A1c by 2.2 percentage points, really, really strong results in that for patients with type 2 diabetes. Finally, substantially improve tolerability, dose escalation works, dose escalation, resulted in similar magnitude of weight loss compared to fixed dosing. So we think that's very important. One last comment on devices, and Jay, I'll toss it over to you. But we've received a number of questions on devices, and we just want to remind everyone that Amgen has long-standing expertise in combining devices with our biological medicines, strength in protein engineering and our world-class formulation, #2 in viscosity, world-class expertise in high-concentration protein formulations, world-class expertise and use of the excipients in terms of optimizing their use in our biologicals. And finally, proprietary device innovation itself with a clear focus on patient safety and experience. 43 million units in 2024 with devices to Repatha and Prolia patients, 11 million of those. So lots of experience, and we just wanted to make sure we gave everybody a refresher on what we've been doing with devices over many, many years. With that, Jay, you've got a couple of thoughts?

Yes. Thanks, Pete. A couple of thoughts just to fill in on MariTide and then I want to just spend a second, introducing some of what I see as the highlights of right now, the pipeline portfolio, but I'm really excited -- most excited about the Q&A. With MariTide, we have a medicine potentially -- potential medicine with very competitive efficacy, very competitive efficacy. We saw data this morning, head-to-head through tirzepatide and semaglutide, 20% at 72 weeks. We're seeing 20% at 52 weeks. We're seeing sema at 13.7% in that head-to-head study. If you look at the curves, as we presented a couple of weeks ago, they're not breaking to plateau. And so through the 52-week endpoint, patients continue to lose weight. And we look forward to reading out part 2 of that study, 90% of patients eligible to participate in part 2 re-enlisted for another year chance at another year on MariTide as some will be randomized to placebo. And I think that plus the low discontinuation rate due to GI and really any adverse event shows just how attractive this medicine is what -- to patients as a patient experience. Second, dose escalation works. This family of medicines, the incretin family requires tolerization of the area postrema and the hypothalamus. You have to tolerize a patient to this type of medicine, in particular, the GLP-1 activation. And unsurprisingly, dose escalation has significantly already eroded the signal that we saw in Phase I around GI side effects without dose escalation, we'll continue to do more work there, dose escalation works. Third, dramatic improvements in cardiometabolic parameters. I hope you're excited to see those, we sure were. To see LDL-C, cholesterol, fatty acids all directionally down, to see hemoglobin A1c in the diabetic patients down 2.2 percentage points absolute change, 50% of patients in remission from their diabetes on that study. Monthly diabetes medicine has been a long-standing goal in that field, look at all the work to try to make a monthly insulin. And then I would say, finally, in these dimensions, this idea of efficacy translating to drug effectiveness in the real world, requires that patients can remain on this medicine stably, consistently reliable pharmacology, excellent tolerability, durable, maybe even improving efficacy, all of which we will study -- we are studying in our fully enrolled part II of the Phase II clinical trial. So love to fill in the gaps around the knowledge about MariTide. We're very confident in the profile that's arising with this medicine. We think it feels an important void in the -- for the one billion patients with obesity that are right now not being met with weekly injectable therapies. And for those physicians managing the complexity of 24-week dose titration and patients missing or lacking access to doses. So -- but beyond MariTide, there's also a lot going on. And I may understand we might spend a lot of time talking about MariTide and obesity. But the rest of the portfolio is stacked. I'll just introduce 3 domains that I'm particularly excited about. One is T-cell engagers. This is a very exciting class of medicines that we pioneered with the acquisition of Micromet, and further research and development since then, they bind the T cell with one arm of a biotherapeutic and bind the cancer cell with another. BLINCYTO this year, new approvals in frontline use of acute lymphoblastic leukemia with unprecedented over survival and improved cure rate with this medicine. There will be a plenary presentation at ASH in the pediatric study of frontline therapy this weekend. So look forward to sharing those data in the fullness of time. Behind this, IMDELLTRA approved this year in small cell lung cancer, a cancer that has yet to benefit from targeted therapy enter IMDELLTRA targeting DLL3 on the surface of the small cell lung cancer. The efficacy we've seen as a single agent in Phase II in extensive disease, we're very hopeful we'll read through to improve utility and efficacy patient benefit overall survival in frontline and also once used in combination. And then xaluritamig for prostate cancer is our third in line T-cell engager binding STEAP1 on the surface of the prostate cell, binding CD3. We reported median overall survival in castrate-resistant prostate cancer of 17.7 months. And this is the type of medicine because it's an injectable protein therapeutic that is very easy to study in combinations and very -- could move very swiftly into earlier lines of therapy, which we're ongoing -- studying in an ongoing way. The second domain of interest is around CD19. Secrets out that if you target CD19 with a CAR T cell or a protein therapeutic, you can deplete not only cancerous B cells, but also autoreactive inflammatory B cells. And if one could get excited about CAR T cell therapies despite all their complexities and manufacturing and challenges in patient access and pricing, injectable protein therapeutics to deplete CD19 starts to look pretty attractive. And we and others, but we have the most advanced CD19 depleting therapeutics in the world with 2 approved medicines, UPLIZNA, presently used for the auto-reactive B-cell involved in neuromyelitis optica spectrum disorder and that we read out positive Phase III data this year in IgG4-related disease and myasthenia gravis. We've posted to clinicaltrials.gov, a master protocol that allows us to study this medicine as well as blinatumomab itself and other CD19-targeted medicines in our pipeline in severe and refractory lupus nephritis and rheumatoid arthritis. And you could expect those indications to grow over time. And then third and final, I'm very interested in olpasiran. Another modular therapeutic and siRNA, a first siRNA experience for us at Amgen that addresses an siRNA to the liver where Lp(a) is manufactured that we'll read out. It's an event-driven study that is as listed in clinicaltrials.gov is likely to read out late into 2026. But this is one of the last known genetic and modifiable risk factors for atherosclerotic cardiovascular disease. And though it's a quiet period right now, while we're waiting for that Phase III to read out, this medicine has real best-in-class potential. Anyway, there's a lot to love around MariTide and also across really every one of the 4 pillars of the Amgen R&D portfolio and therapeutic area of focus and take any questions on this or anything else.

Outstanding. Well, there are so many directions to go down, but let me just start by saying anyone who wants to ask the question from the audience, just let me know and I'll make sure -- we arrange for that.

But I guess we should start with MariTide even though I actually had a lot of like structural question from a broader Amgen perspective heading into Prolia XGEVA next year as well. But I want to come back to that. But let me -- since everyone is so focused on MariTide right now. Question #1 for me. We can talk about the 20% and the weight loss seen. The curve did not plateau to your point, so it keeps going. However, patients are transitioning to Part 2, and I was just thinking about -- so we technically never get a read beyond that 20%, but then I was also thinking some of the patients will be rerouted to the same arm as well. So can we get a read at least in a small subset of the patients that they got to 20% and did that curve continue from there?

Yes, that is correct, Umer. So first to answer your question, yes. Each one of these arms of the Part 2 has a parallel construction to the dose people were on in Part 1. And that allows us to measure a potentially improved weight loss over time beyond 52 weeks. Second, some of those patients are randomized against that control to placebo. And in those patients, we will study the rebound. You must recall that in the Phase I, we had this very intoxicating signal that after the 3 intended doses of MariTide for about 150 days longer, patients did not significantly regain weight. And that's quite different than the peptides, which are a fast-on, fast-off kinetic. And then third, we're studying a much lower doses, even as low as 70 milligrams in a more maintenance mindset, but we're also studying quarterly MariTide building on the positive experience we had around every 8 week in Part 1. So Part 2 has a lot to learn and persistent or even improving weight loss beyond 52 weeks is indeed one expected learning.

Got it. Maybe a couple of follow-ups on that. Number one, in the obesity cohorts, 280 mg was your best dose, not 420 mg, in terms of dose response. I want to come back to that in a second, but just before that.

I might phrase that as a question, but go ahead.

Okay. But in that 280 mg, looking like the best dose and the obesity course, and again, that was different than the diabetes. So 280 mg gets you to 20% weight loss. But upon rerandomization in Part 2, 280 drops down to 70 mg, meaning it does not move over to 420 mg. So I guess in that cohort that was tracking at 20% peak weight loss. Do we truly get to see how much that will deepen because we technically won't, right? Because they get to a lower dose in Part 2.

Well, first, let me talk about 280 mg. So the 280 milligram dose as graft in the nondiabetic patients with obesity or overweight numerically outperformed the aggregated 4 420 mg arms. But the error bars are all overlapping. And as you know, as a mathematically inclined fella and so he sees a lot of Phase II data, it would be wrong to interpret 280 mg as superior to 420 mg for 2 reasons. We have no mechanistic reason why we would think that to be true. And secondly, statistically, that is not what the data showed. Flip over now into the type 2 diabetes population, and that's really the proving ground for anti-obesity medicines as those patients are so less responsive to incretin-based and actually all obesity type medicines. And there, statistically, 420 mg outperformed both 140 mg and 280 mg. Phase II studies are designed to inform dosing around Phase III. And in -- if you take all the data holistically, we have all the guidance necessary from this trial to interact with regulators to open a Phase III study. So I would not over interpret, if you or anyone else would do this, 280 mg is superior to 420 mg based on these data. They're overlapping error bars. As such, patients that go from 420 mg into 420 mg give us all the guidance necessary to know whether or not we can expect sustained weight loss beyond 52 weeks.

Got it. So 420 mg to 420 mg transition, that's where you want to go above 20%. When can we get that data at the earliest? Just to understand the max weight loss, Justin, is that something that could come by summer next year? Is that going to happen after Phase III is already underway?

Yes. We haven't given specific guidance. I think what you can look to is that we completed the Part 1 of the study in October of this year. And then obviously, there's a rollover period and it goes another 52 weeks. So I think that gives you some sense of where this might land in terms of the finalization of Part 2.

Yes. Just one other thing to add is that you also have an arm that's the 420 mg to 420 mg quarterly. So you'll see that 420 mg pull-through, too.

And Susan, I was actually going to follow up with you on this. So to -- I mean if we just go by the reported, 280 mg look like it outperformed on the obesity cohorts, 280 mg looks more like 140 mg on the diabetes.

280 mg and 420 mg had overlapping performance.

Overlapping...

You have to be very specific. To say that something is better requires that their confidence intervals do not overlap. And so let's be very specific. 280 mg and 420 mg were both had outstanding performance and efficacy on Part 1 of the study.

I'll tell you the direction I was going with is, traditionally, a lot of the trials, which show the weight loss, the max weight loss potential of any drug, they're up to 75% females. Your obesity cohort was 67% female. Your diabetes was actually in the low 40s, meaning that diabetes then technically show where the weight loss could truly go because there were lesser females. And the direction I was going was, is there a male to female imbalance between these arms, which makes them look a little all over the place where you have lesser females in one arm, so the arm looks a little lesser on weight loss and more females than another arm, so it looks deeper. Do we know that off top by any chance?

I don't describe any difference such as was noted statistically on these arms to be related to a male, female imbalance or any demographic imbalance.

Okay. Okay. Got it. Any other questions on efficacy just before you proceed?

Umer, There's one other question that's come up a few times and probably a good opportunity for Jay to address, which is within the 420 mg, how do we think about pooled versus some of the other ones, if you don't mind, maybe Jay could take that.

Yes. No, I'm happy to. And I think actually, it maps a little bit back to what I was trying to tease out about 280 mg versus 420 mg. So I think everybody is probably familiar with table stakes, biostatistics and clinical investigation. But in order to ascertain the difference between the performance of 2 medicines on a trial, we assess the mean or median performance, average response, whether it's tumor shrinkage or weight loss over time. And then we ascribe a confidence interval, a 95% confidence interval that says, we're 95% confident that this is the range that one might expect performing this experiment again. And when those intervals don't overlap, then you can be confident that you're observing something different, a distinct behavior. 420 mg and 280 mg have overlapping confidence intervals. And so we can ascribe no difference. It has come up a couple of times since the presentation, which felt to me at the time and still does, like a lot of information downloaded for a top line analysis. But it has come up a couple of times. I want to just be very clear here with all of you and others that could be interested. We pooled the 420 mg arms to achieve statistical robustness to answer the question around that target dose, how much waitlists can we expect from that target dose going into Phase III. We pooled it of this analysis, and it was informative for us and so we disclosed it externally. There is nothing to read into the fact whatsoever that we didn't break out each of the 4 arms. In fact, mathematically, if you're so inclined, look at the error bars themselves, if there were an outlier in those arms, the error bars will be twice the size. And they're not, they're the exact same size as the 280 mg.

Than the other arm.

Exactly. And that is only goes to show, as we said and disclosed that these arms perform comparably.

Thanks, Jay. Okay. Excellent. So maybe transitioning to...

Easier with a whiteboard.

Sure. I was literally just looking at it as you were talking. In terms of tolerability profile, there's several things I want to dig into. But I think first, perhaps for folks that are still sort of looking into the story, could you remind us where was nausea vomiting tracking in Phase I? Where was it in Phase II, and where is it in the additional titration work in the PK? And by the way, I don't know if we're at a full stop yet. Could it further be optimized in terms of a construct into a Phase III trial?

Yes. I totally agree that it could even be better in the fullness of time. And so we've disclosed but have yet to knit this thread together, Umer, so I'm glad you bring it up, that after the Phase I clinical trial, again, which did not feature dose escalation in the high-dose cohorts, we observed a significant amount of nausea and vomiting as one might expect.

You guys acknowledge that Phase I was significant amount of nausea and vomiting?

Well, by percentage as reported in the Nature Metabolism paper, the nausea rates were north of 80%. What I would like to point out about that though is that from sample sizes of 4 to 6 to 12 patients, that number lacks the significance of a Phase II number but it was high. As we go into Phase II clinical investigation with dose escalation, we already observed 70% nausea, 40% vomiting in the dose escalation cohorts. You could think of that as like roughly cutting it in half around emesis. In parallel to the Phase II study, we designed and executed this year a Phase I study to further explore dose escalation. Here sampling even lower starting doses. Learning that 70 milligram starting dose, which is what was used in dose escalation on the Phase II had such an improvement, why not go to 35 mg? Why not go to 21 mg? And so we designed a 4-week rapid dose escalation as we think there could be some advantages there and observed an even further reduction, let's just say, in emesis, by another 50% to under 20% now. We will continue to explore alternate dose escalation strategies. We have the guidance necessary from the totality of our experience now treating more than 800 patients with MariTide to engage regulators around Phase III design. But we can expect for this medicine, as for all others in the class, that dose escalation works and it will work better with further exploration.

Got it. So there's several layers to that there, perhaps going step by step. Number one, we saw the nausea vomiting on the dose escalation. So I would imagine it's reasonable to assume it's higher than, let's say, a 40% number on the non dose-escalation arms because you're going straight on to a high dose from day 1. Is it also reasonable to assume that within those arms, when you go to higher doses on day one vomiting is also higher on that. Is it kind of like dose related, that early vomiting experience?

I can say for this medicine, dose escalation has a significant improvement in the tolerability profile of MariTide. I'm loath to dimensionalize all the data around all the arms. That's for a podium presentation and a big publication down the road. But from here forward, we are only studying MariTide at target doses as seen on the Phase II with a dose escalation strategy.

But it would be, Jay, it would be reasonable to think just realistically speaking, someone who starts at an arm where patients start at a 140 mg versus a 280 mg versus 420 mg. The higher the dose on day 1, the more sort of your stomach goes into a freeze and there's more risk of nausea vomiting, but similarly, also, an arm where a patient is doing sort of a big peak to trough within every 8-week nausea vomiting would theoretically be a little higher on that than where they're doing it on a monthly basis.

Yes, I would frame it this way, Umer, I think I understand what you're getting at. But let me just say for this medicine as for others in its class, reducing the first dose that patients experience has progressively improved tolerability to the first dose.

Got it.

And this is this aspect of tolerization that I was speaking to.

And I guess this is really for you and for Susan, both, which is one of the things I have been unable to answer for myself from the existing data is -- you're hitting certain thresholds on efficacy. Nausea vomiting still can do more work. There's been an improvement, but it could still do more work, however. There was a third dimension to this program, which was it could be every other month to quarterly. Efficacy suggested could be quarterly. But given the peak to trough swings, when you go from a 20-day half-life to every 90-day dosing, does that make nausea vomiting meaningfully higher than it would be if it was monthly dosing? How do you think about that probability? And you have an early read on that from the every 8-week arm?

I just would have meant one thing you said. We don't have reported as yet any efficacy data around quarterly.

Sure.

Just on the every other month so far, where the performance of every other month and the dimension of efficacy was comparable to what we observed with monthly. I think through the Part 2 of the study, we'll have a chance to get a lot more experience with MariTide used in more of a maintenance setting. I mean these individuals have already lost 15% to qualify for Part 2. But there, I think quarterly could allow us to reach more patients. It could allow patients to have even fewer injections required to achieve durable or even improved weight reduction. And so part one has taught us that this medicine has efficacy with monthly or every other month dosing. And in Part 2, we'll learn whether we have maintenance type efficacy. We'll learn whether we have continued weight loss, and we'll learn more about tolerability because of the peak and trough pharmacokinetic that you rightly referenced.

But is it reasonable to assume that based on everything we know right now, as long as there's a reasonable titration in place and titrations in flux, a less than once monthly is realistically possible from a GI tolerability perspective. Is that reasonable to assume?

I think that we'll need a little bit further study of this to answer the question definitively. And we've not as yet defined for the community what our Phase III design is. It's an ongoing conversation with regulators. It's a very competitive space, as you know. But we have learned from Part 1 that every other month, efficacy is comparable to what we observed with...

And tolerability, we'll figure it out on the titration on how to get there.

The big learning around tolerability is around dose escalation and the dose escalation works.

Even before and every 8 weeks.

Dose escalation, as shown in the schema that we shared did not feature into the every other month dosing scheme for 420 mg.

No, clearly, clearly, clearly. Okay. Great. Any questions from the audience so far just before we move on? Heart rate changes perhaps, I know you guys mentioned they were in line with the incretin class. One of the big themes that stood out at EASD, where I missed you. Was one-off going to very, very high levels. Was there any instances of someone going to 15 to 20 beats per minute, anything along those lines?

There's really nothing more to gather from the heart rate data than what we shared that there's a minimal change at all to the heart rate with -- through the 52-week period of MariTide dosing.

Got it. Okay. Excellent.

And blood pressure. I mean, blood pressure was reduced by 11 millimeters of mercury, which especially in patients with type 2 diabetes for a nonantihypertensive to deliver such benefit around systolic blood pressure is 11 millimeters of mercury would be expected to, based on population data sets as well as randomized controlled studies, most recently, a brilliant study in China performed to further optimize blood pressure management in patients with diabetes. The reduction of 5% to 10% read through to a really significant cardiovascular outcomes. So we're quite bullish on the overall cardiovascular benefit to MariTide observed in Phase II.

Got it. My last one on this, maybe for both of you. I think Peter mentioned at the start about a lot of experience in devices at Amgen. The formulation strength, at least in the patent filings, is 70 mgs per ml. So to me, that implies which is low for antibodies, you could go potentially up to 150 mgs per ml, 140 mgs per ml. So in my mind, since 2 mls could go in an auto-injector, presumably the 280 mg dose could have gone into a very simple auto-injector. 420 mg, I wasn't so sure. I thought it was more an on-body, but you guys always say its handheld. So any color on that, just to make people feel comfortable on what the presentation could look like for all of you.

Yes, I'll start and then Susan, I invite you to share reflections. Here, I'd quote you, Peter. Amgen is made for this moment.

100%.

I'm still relatively new here, been here just the calendar year. And...

It feels like a little bit more than that, maybe, Jay.

No, I love it. Just absolutely love It.

We love having you.

I'm just back Monday from visiting our Columbus, Ohio brand-new finish and packaging, really the device experts in our company. It's a breathtaking fireworks display integrated technology, integrated robotics and Amgen is made for this moment. We're the world leader in bringing biotherapeutics.

So 70 mgs per ml is not the stop?

And so we're not having any challenge preparing this medicine for a patient-friendly, physician-friendly handheld single chamber auto-injector. This is a -- this will be a good patient experience. Susan, what are your thoughts?

Yes. I would say we have a fantastic process development team and I have such experience with working with proteins across the entire portfolio for the entire life of Amgen. And the idea of doing high concentration proteins is right in their wheelhouse.

So Susan, I can see a path from going from 70 mg per ml currently to 140 mg per ml because that's doable with other antibodies. But going to 210 mg per ml to make a 2 ml 420 mg dose happen, is that a realistic possibility with the viscosity and the pressure would kick in?

We feel very confident that we're going to be able to get into a single auto-injector, patient-friendly one injection to move forward for the program. And I think it's important for patients. That's an absolute. Like when we look at requirements of what we need to make sure that we have good patient experience, we want to make sure that the patient is going to have a good experience doing their injection. And right again, they're going to have less frequent injections.

Got it. So the 70 mg is a stop along the way right now kind of like the titration stop along the way, which brings me to my last point on this. I want to move on from this. But my last point on this, remind me, as we think about sort of the titration regimen, the main takeaways for me in the phase -- on the data last week were, a, if you have a 12-week dose escalation, that seems to correlate with an improvement in tolerability. Separately, if you don't start at 140 mg, you start at 35 mgs, that correlates. But in that PK titration, it was only 4-week titration. So is it reasonable to assume you marry the 2, where you have a 12-week maybe even a longer titration and you start at a 35 mg dose. Is that a realistic consideration?

Yes. Here, the mind can run wild. And you can imagine in the conference rooms at Amgen, we have mocked up and schematized every manner and configuration of dose escalation. And I really like where we've landed for Phase III clinical investigation. It's an ongoing discussion with regulators. What I would take away from what we shared, which was integrated data around tolerability from the 2 dose escalation arms is we have a lot of ways to win here. And I think we can further improve in parallel to our Phase III clinical study. But we're going into Phase III with a very strong dose titration approach and target doses that are well informed by the Phase II data.

Jay, if I remember quickly, Lilly went from mid-20s in vomiting in Phase II on tirzepatide to sub 10%. But they pushed the titration up to 6 months. So is it reasonable to assume you guys are taking into consideration there's even a double the period on titration that's possible than what's been done so far?

I'm hoping it won't take 6 months of titration. We want to get patients to target dose before half the year is over. And in clinical practice, many patients receiving such complex medicines actually never even make it to peak dose and protracted titration schemes can also lead to protracted symptoms that follow with every step up of drug administration.

But if they can get to vomiting to sub-10%.

Yes. And I did read your -- or listen to your remarks around the Phase II to Phase III transition. And we're hopeful and expecting that our own Phase II to Phase III transition could come with further improvements and observed tolerability in particular, with respect to GI symptoms. Part of this can be protocol design. Part of this can be preparing patients for clinical investigation. Part of this can be ascertainment bias, which is very high in Phase II clinical investigation. But a big part of this has to be optimized dose escalation. And so we're well informed going on to Phase III.

I realize your -- you guys were guarded in your enthusiasm for the oral molecule. The D5D, I think, was our guest at the time that went into the clinic. How would you characterize your excitement on some of the follow-on stuff in obesity beyond MariTide that's approaching the clinic?

All right. I'll start, Susan, then you back clean up. I think that it's wonderful to be in this field with such an important medicine rising as MariTide. I also think that this is such a vast population of patients that further innovation can only derive further benefit. We're very interested in new mechanisms that map to the incretin pathway, but also beyond the incretin pathway, many of which are informed by human genetic observations with our Amgen Research Reykjavik or the deCODE subsidiary, we have a list of targets that we find very interesting and others through functional biological study. Some of these will be injectable, but some of them will be oral. The oral experience on the incretins to date has not meaningfully advanced that pharmacology. Could there be differences in the ultimate segmentation of the market perhaps, but I don't as yet see around oral incretins a capacity for a real advance when you integrate efficacy, tolerability and cost of goods. That day will come, but it doesn't seem to have come yet. We and others though hope to be very relevant in the development of oral anti-obesity medicines.

Got it.

Yes. The only other thing I would add to that is just, again, as a company that has experience in developing monoclonal antibodies and then particularly with MariTide with that as our backbone. I think we have a unique advantage there of looking at all of the things that Jay's talked about with the backbone of a monoclonal antibody, which can infer a different type of profile for patients.

Okay. Any early observations on 513 that make you just feel a certain way? Or is it too early right now?

That's too early to say.

Yes.

Too early.

Okay. Excellent. Last one, this is really the last one on MariTide. Peter, is there any outside chance on the broader MariTide franchise, whereby you guys decide you need a partner for ex U.S. purposes kind of like you did on CGRP or anything like, or should we model this as a fully owned that will be an Amgen product worldwide?

We think this is such an important message from a...

Tier 1 priority.

We're 100% into this. We have set obesity up as a platform at Amgen. We're using our balance sheet. We have Susan, we've put her in charge. She's top of the house, CEO Staff, Executive Vice President. And we see obesity as a platform. We see MariTide is a key part of that. And we think it goes right after this significant unmet medical need around the world, this global public health crisis. called obesity, obesity-related conditions and type 2 diabetes, and we are all ahead go.

Yes. And add to that, we have a very strong commercial presence throughout the world, and we've really been able to establish ourselves in 100 countries.

And by the way, I forgot when I was opening, speaking to our international presence that we've now launched TEPEZZA in Japan, which we're excited about in our rare disease, so I just thought I'd give you all break for a moment for MariTide. In case you want to get a sip of coffee or water...

Okay, great. I guess the one question I do want to ask is, would -- I realize you guys haven't put numbers on UPLIZNA, but how do you sort of think about broader opportunity in UPLIZNA relative to how the Street is thinking about it, maybe across the 3 indications, including myasthenia gravis, like I realize there's not a peak sales number, but I do wonder your number is probably meaningfully different than Street, but what's your confidence? And even if it's in broad strokes, if you could speak to that.

Yes, maybe I'll start and ask Jay to chime in as well. I mean I think when we shared the UPLIZNA update in IgG4, the headlines were very strong, right? 0.13 hazard ratio, no approved therapies. And then as Jay mentioned, we now have breakthrough -- sorry, priority review and we're full steam ahead there. Myasthenia gravis, when we gave that update, we had on one of the slides, potentially practice-changing. And I know that was maybe somewhat of a debate or controversy of how do we stack up versus other agents, I think the reason we feel very strongly about the offer in myasthenia gravis is one, you see the great efficacy, but two, the fact that that's with a steroid taper and with this every 6-month dosing and this effect that's not waxing or waning. So we think that's another example of a drug that works really well in the real world. So I would say this is probably one where maybe internally, we feel quite strong about the opportunity. I don't know if Jay, if you would add on anything there.

I think you said it very well. I think the myasthenia gravis data are very exciting, and I think we'll be very competitive when you put up with the myasthenia gravis activity of daily living scale numbers like negative 4.2% medicine, works in acetylcholine receptor as well as it works in the MuSK population. I think it's the largest ever prospective clinical study of a medicine in this indication. And then the fact that it's like a set it and forget it, it's like every 6 months of dosing with real symptomatic improvement on that scale. I think this is a very competitive profile to date, knock on all wood. The tolerability profile, this medicine is excellent. People wondered if you're taking out some plasma blasts and your impairing autoantibody elaboration will be the effect -- that effect on immunity and knock on wood, patients have been very safe on this medicine to date. So I think that there's a big opportunity to learn from UPLIZNA to further develop it into new indications. And to expand our impact in the CD19 domain with deeper pipeline assets.

And maybe just one last thought there. I mean, obviously, the other benefit is that we work in both the patient subtypes. And just to remind folks, the data that we reported, that was just 26-week data, so you get a loading dose and you get one more, and then that's it. Those patients received another dose and then we're going to follow them longer. So touch wood, as Jay said, hopefully, we'll see some more results in the future.

Justin, Peter, could there be a huge pricing angle in myasthenia gravis, where your competitors, I think $350,000 for 4 weeks on, 4 weeks off, meaning -- and a lot of people take it continuously, meaning it ends up being over $600,000, $700,000 in pricing. UPLIZNA is kind of pegged to $200,000 to $300,000. So could there be a huge pricing angle that drives commercial interest in your favor? Is that not a realistic possibility with the PBM setup as it stands today?

Yes, it's probably the next step is to see the further data and then go from there.

Okay. Excellent. Maybe perhaps moving beyond, you mentioned CD19 bispecific, there's a lot of interest there. Could you -- I was looking for trials that you've actually put up on clinical trials. And I was just curious do you expect to do a dedicated trial, registrational in myasthenia gravis as well as what are registrational trials you intend to do with the CD19 bispecific?

Well, as you know, the current posted and active studies with the CD19 franchise in emerging autoimmune areas like lupus nephritis -- lupus and lupus nephritis and refractory rheumatoid arthritis are in an advanced signal finding stage meaning these are approved medicines. We know their exact dose and schedule. We're now establishing activity that's disease-specific. And we'll announce at the right time, our registrational strategy in these and other indications.

But these are not -- potential registration trials, not a year plus away, would you agree with that?

We've not as yet we're not in a position as yet to discuss exactly when we'll initiate registrational studies. I will say that in our favor here, it was a 7-year experience to establish dosing schedule with blinatumomab back when we did that work in Phase I clinical investigation, such as the titration required for CD3 activating bispecifics. And having been through all of that now with very precise guidance on dosing as well as a path forward for subcutaneous use. We're in a strong position with both CD19 CD3 bispecifics as well as with afucosylated CD19 UPLIZNA.

Got it. OX40 with a little more frequent dosing, could we get to higher efficacy than what we've seen so far?

Yes. That's a really interesting question, and thanks for following this work so closely as you know and others may be interested to know we're developing a medicine called rocatinlimab, which is a monoclonal antibody directed at OX40. OX40 is on the surface of activated pathogenic T-cells. And so by hitting OX40 and depleting those cells, you take out not the whole of the T-cell compartment, which will be scary for patients owing to conferred immune suppression, but rather rebalance the immune compartment, to take out pathogenic cells and allow healthy naive and mature T-cells to do their function. This medicine is in a very large series of clinical studies called the ROCKET program that will characterize as activity in atopic dermatitis. We have interest beyond atopic dermatitis as well. We've read out the first of these studies and disclose these results showing unambiguous activity of rocatinlimab in atopic dermatitis. Now this medicine was studied and again in Phase II and now into Phase III at 2 different doses, 150 mg and 300 mg, and we've also have an experience with this medicine at a Q4 as well as a Q2. And so even just those 2 dimensions, Q2, Q4 150 mg, 300 mg, we have to take choices as to where we invest to establish efficacy. I do like the choices that were taken. And through this full ROCKET program, I think we will learn whether every 2-week or every month, 150 mg or 300 mg turn out to be the optimal profile for the patient.

Got it. Peter, a couple of very quick rapid fires for you. One, how do you -- do you think you can absorb Prolia XGEVA impact and hold EBITDA stable through 2028 timeframe?

Well, as you know, we don't give long-term guidance. Let me go back, though. Third quarter, think about Repatha up 40%. We are confident in that franchise. Think about EVENITY, up 30%. That's a big player in the bone franchise, and we have high expectations for it. We haven't had a chance to talk today about our biosimilars franchise going really well. We've got some new biosimilars coming out. We've indicated some more coming out later in the decade. We're very excited about that. And Umer, let me go to rare for a minute. All 4 of those medicines, TEPEZZA, TAVNEOS, KRYSTEXXA and UPLIZNA early in their life cycle they grew 21% year-over-year. TAVNEOS doubled in the quarter in the third quarter. We think we've got strong growth drivers in this company. We're fully committed to moving forward throughout the period. We're committed to running a strong operation from an operating margin standpoint. As we indicated with investors this year, we flexed, we moved to roughly 47% or so in operating margin. So we took that money and we reinvested primarily in our development activities on these fantastic drugs in this mid and late-stage pipeline that's gone so well that we've had an opportunity to share with you today. We're going to continue that discipline, Umer, and we look forward to strong long-term growth in the company through 2030 and beyond.

Peter, I felt like you guys probably need some additional support from something close to commercial stage in terms of BD to kind of supplement this period where it's Prolia XGEVA first, but later Otezla, et cetera, as well. How do you characterize that? Or is that part of sort of the routine strategy where BD will continue to be relevant anyways?

BD is always important at Amgen. BD, again, Umer, we are always in the market thinking about what's out there. We're looking -- are we the best buyer, the cash-on-cash return above our hurdle rate. Do we have discovery research in the area. That's #3 and #4, can we integrate it? Can we collaborate? Can we partner with it quickly to effectuate the return? We haven't changed in how we look at BD, we're going to stick to those principles. We think it served us well. And so we'll continue to keep the aperture open there. But right now, we're really focused on what we have in the mid- and late-stage pipeline. We're focused, as we said before. We're on track to get to the pre-Horizon acquisition capital structure at the end of 2025, and we feel very good where we're at.

Got it. My last 2 very quick ones. One on OpEx, SG&A, R&D. Is it reasonable to assume that since you're saying growth, EPS should also continue at least to stay stable as well as growing meaning OpEx will have to remain broadly and sort of the bounds it's in right now. And I realize there's a lot of investments coming out of the pipeline side, too. So I'm always just sort of thinking about that.

Well, take our principle in terms of operating margin and how we think about that, and we try to run a very -- it's a very important -- we've always said in terms of operating margin. If they're cash-on-cash returns that are above our hurdle rate for our shareholders on a basis that makes sense. We'll let our shareholders know we're going to do that. We come back, we'll come back at the end of the fourth quarter with guidance for 2025. But we're confident in how we run the business. We're...

Because Jay wants $15 billion for R&D next year.

You know what? We're raising R&D, just to remind everyone, 25% is our guide year-over-year in 2024. And so we're fully committed innovation is capital allocation #1 at Amgen, and it's going to stay there, both internal innovation and external innovation, the best innovation.

And any update on tax? That's my actual last one. Any update on tax that could come in -- at some point -- just to resolve it.

If you'll head up to Washington, D.C. with me, the trial is underway right now. We continue to be confident in our position. We continue to be confident in our reserves. We think the value of manufacturing in Puerto Rico is enormous. And so we're very committed to winning that opportunity.

Outstanding. Excellent. Thank you, guys, for your time.

Thank you, Umer. Thank you very much.

Thank you.