Amgen Inc. (AMGN) Earnings Call Transcript & Summary

All right. Well, good afternoon, everybody, and thank you for joining us again at this year's IO Summit. I'm Yaron Werber from the biotech team at Cowen. With me here today is my colleague, Gabe Schneider, and it's a great pleasure for both of us to co-moderate this session with Amgen. Today, we have David Reese, who's EVP of R&D; and also Jean-Charles Soria, who's SVP, Global Development Oncology. So gentlemen, thank you so much for joining us. Really appreciate it.

Great to see you, Yaron.

Our pleasure.

There's a lot to talk about. So we are focusing more on the IO pipeline, although we definitely will touch on LUMYKRAS, obviously. But perhaps first we can touch on the Teneobio acquisition, which is actually very important. It's got a very unique and what's been sort of very fruitful pipeline and platform so far, bispecifics and multispecific antibodies that can really complement Amgen's bi capabilities from the original acquisition. So maybe first to David, for you, how does it complement the bi capabilities? What was the reasoning to buying Teneobio? And how does that platform really compare to other self-complementary [ FV or fab ] sort of bispecific platforms?

Yes. So just to remind everyone, in my view, there are 3 important components to the Teneobio acquisition. One is the single domain, heavy chain generating platform capability. We can come back to that in a moment because that has applications across therapeutic areas. The second important component is their CD3 binding technology. Their CD3 binder, now our CD3 binder, has a lower affinity. And so it may generate, for example, a little less cytokine release syndrome in certain circumstances. That, I think provides a real complement to our core technology or platform. And now we can actually choose which CD33 binding technology to use, given any specific situation in the clinic. So it really broadens our capabilities in terms of T cell engaging bispecifics. The third big piece of the acquisition is the clinical portfolio, and in particular the lead molecule, which we now call AMG 340, which targets PSMA. That is in dose escalation. We can come back to that in a moment if we have a chance to talk about the prostate cancer, immuno-oncology portfolio. But those 3 big pieces in what we view really as a complementary technology.

Great. And maybe beyond therapeutic antibodies, there is -- the antibody binding domains that were discovered through their platform really can be used for antigen recognitions relating to CARTs, relating to nanoparticles, ADCs, et cetera. What are your thoughts about -- was that a part of the reason for the acquisition? Or how would you...

That was a big part of the reason. As we have talked about, a big focus of ours is the development of multi-specific molecules. And the Teneobio technology by generating what are called UniDabs, these single domain, heavy chain moieties, we can put those together. They can be strung together like beads on a string to then bind to a collection of antigens. And those antigens may be on target cells, they may be on effector cells or effector machinery. And this, we believe, is a huge complement to our ability to generate flexible multi-specifics rapidly. It's also a complement to our small molecule capability in this regard. We acquired a company in Copenhagen that's now been very nicely integrated into our research operations. That has DNA encoded library technology. Same idea, one can now screen billions of fragments for binders for targets and effectors. So all of this fits together in our broader multispecific strategy.

Okay. Great. Gabe, let me turn it over to you to talk about the pipeline.

Thanks, Yaron. So first, we'll jump into the DLL3 molecule, which is a tarlatamab or AMG 757, which is currently in dose expansion for patients with small cell lung cancer. Could you maybe discuss the design of the current trial, which is I believe potentially registration-enabling this Phase II study, which is planned to initiate this quarter? And how big of an opportunity do you see in small cell lung cancer?

Yes. Thank you, Gabriel. So the registrational tarlatamab trial is a Phase II open-label study that will enroll patients who have relapsed or progressed after a first line of therapy and have experienced also another line of therapy, basically third-line patients. So in support of the Project Optimus, the part 1 of the registrational trial will evaluate the safety and the efficacy of 2 doses of this BITE. And based on the totality of the efficacy, the safety and the pharmokinetics at the interim analysis, then we will move with a single dose for the Part 2. I mean we believe, given what we have seen and what has been the clinical landscape in small cell lung cancer with basically no molecule active in the third-line setting, that tarlatamab can bring real value to the patients starting in the third line setting but then advancing to earlier lines of therapy.

Got it. So next, just to save time, moving to your AMG 160 or acapatamab, which is the PSMA CD3 half life extended bite molecule. That's currently in dose expansion for prostate cancer. And as background, the protocol includes combination approaches with an anti-PD1 AMG 404 as well as XTANDI, which is an androgen receptor inhibitor; and ZYTIGA, which is an engine biosynthesis inhibitor in earlier line prostate cancer. So to either of you, perhaps maybe we could start with what was some of the synergy that you saw preclinically with these combination partners that led you to use to decide to evaluate these cohorts? And then I'll have some follow-ups.

Yes. Thank you, Gabriel. So indeed these clinically tested combinations are based on a preclinical rationale. AMG 160 treatment upregulates the PD-L1, PD-1 pathway and the NHT, enzalutamide and abiraterone treatment, both increased PSMA expression levels. And as you can assume, the more increased PSMA expression level, the more you hope to see activity on PSMA BITE. These AMG combinations are currently being tested in a clinical trial.

Got it. Okay. And then to follow that up, where do you ultimately see AMG 160 fitting into the prostate cancer treatment paradigm what -- based on your current trials but potentially in the future as well?

Yes, I think AMG 160 could enter first as a monotherapy in advanced metastatic castrate-resistant prostate cancer. And we will follow that into earlier lines of therapy, eventually with combinations.

I think one other thing to add here is, of course, we're also [indiscernible] which is in dose escalation. By the middle of next year, we're actually going to have a fair amount of data on both of these molecules, and we'll decide [indiscernible] targeting portfolio as we move forward.

Got it. I think Gabe, you have a couple of follow-ons, right, for 757 and 160 and the dose?

Yes. Just I will --- just yes, so just maybe a little bit further on that tarlatamab response that you gave for the Phase II expansion. Any thought on what doses you're going to be looking at based on your Phase I data, the therapeutic window with perhaps with the higher dose might be a little bit more difficult. But anything more specific you can give us around the dose.

Well, we are really exploring 2 doses -- that are pretty far away each one. And as I say we just activated this trial. And we will look at the entirety of not only the efficacy, but also the safety and the PK to make the call.

Yes. And as Jean-Charles indicated, we chose doses where there really isn't a lot in overlap of exposures, so we could really get a look at the question of therapeutic window here. It's also worth pointing out in this program that we've seen minimal CRS. It's been quite modest compared to many T cell engaging programs. Which really I think underscores an important point, which is much of that profile is target dependent and dependent on the tumor cell mass. So CRS has not been a real clinical challenge in the DLL3 program.

Yes. Thanks for that, David. It's true that DLL3 is very rarely and not at all expressed in normal tissues. So it's a very tumor-specific target.

Got it, yes. And then maybe a similar follow-up for AMG 160. I know that in the initial data, there might have been higher rates of CRS, but in the subsequent data, a bit more encouraging. Anything you can tell us about the doses that you're currently looking at for 160?

We are hopeful to have a better profile of tolerability, and we are currently exploring different approaches.

Yes, including potentially line of sight to outpatient administration at one point here. This is obviously the ultimate goal for -- in particular something like an advanced prostate cancer population.

That experiment is ongoing. As Dave highlighted, we have a specific question about can we achieve outpatient administration.

Got it. Okay. Perfect. And then my last one, just ...

Excuse me, Gabe. Let me just check -- quick questions. So the initial -- maybe just to probe. Because the initial data in the very elderly population had a fairly modest response rate and some CRS, right? And then I think the more subsequent experience actually started looking a lot better. And if I remember correctly, you're also testing some steroids administration, maybe -- I can't remember the full protocol. But just give us a sense kind of what's been the step at changes and how that improved the therapeutic window.

Yes. So as in all of these programs, we've undergone dose optimization. Steroid prophylaxis is pretty typical across the board to mitigate CRS in the first cycle or 2, so that's fairly routine. We have fully enrolled an expansion cohort, and now we just need to get really duration of response and let some time go by to see what efficacy and safety profile looks like over time. We anticipate by the first half of next year or so, we'll have those decision-making data in hand for the AMG 160 program.

Great. And then my last question just before I kick it back to Yaron, Dave, you mentioned the asset, the PSMA targeting asset that you acquired from Teneobio in terms -- AMG 343. So maybe before we move on to the next section, what can you share about that molecule that we -- as we kind of get to know it better?

Yes. It's in dose escalation. It's early days, but it's moving along actually quite briskly. And I would hope by middle of next year, give or take, we'll actually have highly informative data there as well from the dose escalation part of the Phase I trial.

I'll turn it back to Yaron.

And is the attractiveness of 340, to your point earlier on the call, is the lower affinity CD3 arm to potentially enhance the therapeutic window with [indiscernible].

Yes, I think there are 2 -- couple of things we'll be looking at. One is does it change the CRS profile while maintaining efficacy. Obviously, that's a core question. And then one thing that has been a challenge across the board for T cell engagers targeting PSMA for reasons not only understood or antidrug antibodies. And so that's one thing that we'll be taking a look at as well.

A gentler safety profile, of course, will make a lot of sense for a disease that is massively treated by community hospitals.

Yes. For elderly men and to a certain degree, call it, an indolent or slower-growing tumor in cases -- in some cases, anyway. Okay, let me move on to LUMYKRAS. I mean we can't talk to Amgen not asking about LUMYKRAS. So maybe David and/or Jean-Charles or both of you. The -- it sounds like we'll have data next year in the first half of the year, LUMYKRAS and KEYTRUDA. Can you give us a sense -- and you might even be exploring, -- I believe you're exploring other I-O modalities as well. So maybe give us a little bit of a sense sort of in addition to KEYTRUDA, what other modalities are you exploring with LUMYKRAS? And then two, when we see the first data, is it going to be stratified by TPS status or not? What kind of doses are you exploring for KEYTRUDA? I believe they're lower than 9 60? Or are you still exploring 9 60 as well? And sort of what can we expect from the data initially?

Sure. Jean-Charles, you can start there, and I'll be happy to chime in.

Thanks, Dave. So I mean just for clarity, our development strategy for LUMYKRAS is twofold. First, reaching more patients in earlier lines of therapy in lung cancer but also in new tumor types, and you have seen we have reported some recent colorectal data. But the second element of our strategy is a combination strategy to expand the activity. And we have an ongoing -- over 11 different combinations that are ongoing. We are obviously doing a very thoughtful experiment to find the best way to combine this molecule with a checkpoint inhibitor. And in parallel, we are evaluating 2 potential doses of sotorasib in collaboration with the FDA Project Optimus. And we as you say, Yaron, plan to present the key elements of our master protocol related to LUMYKRAS and KEYTRUDA combination early next year.

Okay. And then Jean-Charles, I mean in addition -- in terms of other IO combination, I remember saying -- you obviously have your own PD-1. I believe I saw a [ trisentry ] combo somewhere, but I could have made that up in my old age. And in addition what other IO campus might you explore?

There is no old age. We have an [ trisentry ] combination indeed.

Yes, yes. And I think Jean-Charles covered the landscape of what we're doing here. So the key will be pulling all of these data together. In addition to straight combinations, we are looking at sequential therapy as well. And so pulling all of that together, again, as we've indicated in the first half of the year, we expect to be able to provide informative data on the various checkpoint inhibitor approaches.

And the sequential therapy approach, it sounds like you'll start with LUMYKRAS. Is it for 2 cycles, kind of 2 months and then layer on KEYTRUDA on top of it? Or how are you...

There are different ways to do this, and that's one of the things that we're sorting out, Yaron. There are evidence, for example, that treatment with LUMYKRAS can lead to immune cell infiltration into tumors. We've got preclinical data suggesting that. So that would suggest, for example, an induction period as you're indicating with LUMYKRAS followed by checkpoint inhibitor. But there's a lot of biology and immunology to work out here.

Okay. And then can we ask a question? There are sort of 2 parallel thoughts these days, at least in our head about anti-G12C, and I think frankly in everybody's heads. One is a monotherapy approach potentially in TPS low. I mean you've shown your data has 31% response rate in the double refractory -- double-exposed second line, right, I-O and chemo. I think that same data showed a 58% response rate in about 20 or 24 patients and patients who have failed either/or, right, I-O or chemo in second line. And the data from Mirati shows a 64% response rate in second line Keep 1 -- or STK11 mutant. So there's a -- you begin to kind of see an approach where even monotherapy could be pretty decent in TPS low. I think your own data showed slightly higher activity in TPS low, albeit both data sets were pretty thin. So a question I guess for you, do you think mono can beat KEYTRUDA chemo in first-line TPS low? And then...

Yes, I think so we've got first-line trial open for those with STK11 or PDL1 expression low tumors that I think will give us a sense of what the comparative activity might be. I'd also broaden your question just in our last moment here. Dr. Piro with or Dr. Lito colleagues from Sloan Kettering and colleagues from Amgen, we published a paper in Nature last week on potential resistance mechanisms. That's why we're taking a look. The take-home message here is that there's no common resistance mechanism. There appear to be many ways that the tumor cell, and here specifically the non-small cell lung cancer cell, can escape G12C inhibition. And so we're really trying to work through those data and use that to guide the combination program, but we don't have the equivalent of a gatekeeper mutation that you've got with, for example, tyrosine kinase inhibitors against EGFR or other receptors.

And so the point you're trying to make is as a question, it's not just about response rates upfront but also durability?

Exactly. I think durability and selecting patient populations based on a broader molecular characterization.

Okay. And is STK11 potentially a registrational path? Or you will continue to believe that it's probably not in first line?

Generally, the FDA has asked for comparative trials in the first-line setting. Now of course, we would always look at the data set. And if we felt conversation was appropriate, we would approach them. But one can assume, absent remarkable data, that some sort of comparative trial will be required.

All right. Well, terrific. David and Jean-Charles, thank you so much for joining us. It's always great to see you. We really appreciate it, and we'll continue to follow closely.

Thank you. Take care.

Thank you.