Eli Lilly and Company (LLY) Earnings Call Transcript & Summary

Okay. Great. Thanks, everybody, for joining us. Good afternoon, and welcome to Guggenheim's Fourth Annual Oncology Conference and today's fireside chat discussion with Eli Lilly. I'm Seamus Fernandez, Guggenheim's global biopharma analyst. I'm joined by Jake Van Naarden, President of Lilly Oncology; and Dr. David Hyman, Chief Medical Officer of Lilly Oncology. Jake, David, thanks so much for joining us here today.

So I'm going to jump right in with a question for Jake. Jake, coming from what we all assume is a very fast-moving biotech in Loxo Oncology, you've been on the Lilly platform for several years now. I think many people have this feeling that pharma is a slow-moving kind of plotting organizations that just have a lot of capital to deploy. What's been your experience at Lilly?

Well, first off, thanks for having us. And for those on the line, thanks for joining us. Your question is an interesting one. I actually don't think it's an inherently true statement. In other words, I don't think that large pharma companies are inherently plotting and slow. And I think, candidly, one of the things that was exciting about the initial couple of years of just observing, Lilly is just -- seeing how not true that was actually. Whether it's the work we've done in Alzheimer's or the COVID antibodies or even the Verzenio program that predated me being in this role and David being in his role, there's actually a lot of examples of Lilly -- of moving incredibly quickly. And so I actually think -- I don't know if that's a Lilly thing in particular or if it's really about sort of 2 ideas that Dave and I try and espouse on really a daily basis, and that frankly a lot of our other leaders do as well, which is engaged leadership, so leaders that are really like in with teams, helping make real-time decisions rather than sort of allowing the governance process to slow you down because that is what slows things down, multiple layers of decisions. So that's one aspect. And the other thing is just focus. And that's, I think, a big part of what a small company is sort of forced to do because you don't have that much capital, you don't have that many people, there's only so many things you can work on. So you're forced to focus. And as a result, you sort of optimize every little decision, which I think is actually sort of a good thing. You can apply that same degree of focus at a larger company. The way we do it is just by working on fewer things because in our view, at least in oncology, you just -- you can't have strong biologic conviction about like 25 clinical programs. I can't -- I don't know how that could be possible. There just aren't that many great ideas that can be differentiated products today, unfortunately, for patients. And so one of the things we did was cull the list and sort of hit not a reset button, but hit a little bit of a reshuffling about what was the clinical portfolio going to be that we focused on? What was the preclinical discovery portfolio going to be that we focused on? And even within the commercialized assets, what were the future development ideas we would focus on? And we have skinny down the portfolio to the point where we can literally talk about every single item in whatever amount of detail you want to. And that's not just about us, it's about the team too and how we've organized. So I can't speak to other large companies. I've never worked in any other large pharmas. But I would say I have been impressed actually by the way in which the sort of focused and engaged leadership has been applied at Lilly outside of oncology. And obviously, we've been doing it within oncology. I don't know, Dave, would you want to add anything else to that? You're a little newer to this than I am, but not that much newer.

Yes. No. I agree with everything you said. Obviously, I was an observer of big and small pharma from the outside in my -- first part of my career as a clinical trials and academic. I think the one other thing I would add, and I think where we've been able, just by virtue of the organizational structure to improve efficiency is kind of aligning the objectives of the 3 main parts of the organization, which is the discovery, the R&D and the commercial. And I think that's another area where there can be inefficiencies, where the discovery unit doesn't understand the kind of taste or goals or objectives of the R&D unit. The R&D unit is not communicating effectively with the commercial unit. And I think by virtue of having an integrated organizational structure, it really helps us set like a vision for the entire organization, so that we're not encountering these periods where like we're advancing products that one hand doesn't actually want to receive from the other.

Yes. And just to clarify for, Seamus, you, and the audience, that's a new phenomenon just as of the fall. So when we changed the leadership structure, we took that as an opportunity to do oncology a little differently within Lilly as sort of an end-to-end unit. None of the other TAs are set up like that for a variety of reasons. But it's differentiating within the company. It's also differentiating within industry. I don't think a lot of other large companies are set up that way. But just to even go one step further from what Dave said because I think it's a great point, like can you -- you can imagine what happens when a discovery team spent 6 years on a program, gets it ready for an IND, presents it to the clinical team and then the clinical team says, I don't really want that. Like just imagine how horrible that is for the organization and for all these talented scientists that have worked on it. It's just sort of a toxic thing. That kind of thing happens at our peer organizations. That kind of thing does not happen here.

Got it. So as we just kind of think about that end-to-end integration, how challenging was it to kind of bring that operational structure to bear? Is that a few months in? And once you sort of layered out how you wanted the organization to operate, it was months in and you guys are operating at full steam? Or do you see more opportunity ahead to continue to sort of accelerate that type of integration?

So you're never done. There's always more process stuff to optimize. So -- but in our case, it was not a onetime thing. It was sort of a staged approach, and that wasn't deliberate. It's just sort of how it happened. So first, we took on the task of reorganizing the discovery and investigational clinical development work. Then David expanded his role to include the marketed products from a development perspective. And then my role expanded to oversee the whole thing. And so we sort of have this almost ramped approach to sort of doing things one at a time to the point where ultimately, what we rolled out in the fall internally was meaningful, but it wasn't like as heavy of a lift as it would have been had we done that in like 2019. If we had done that in 2019, it would have been really, really hard. Whereas -- it's been difficult, but it's been good. I think most people feel energized by the changes and we've created a lot of more streamlining of process and stuff.

Got it. Awesome. And just in terms of BD, maybe you can talk about how BD actually operates within that structure. And we'll talk a little bit more about sort of the environment that we're sitting in today.

Yes. So the Head of Oncology BD reports directly to me. She has a team of folks who obviously are out there, including her, on the ground talking to companies large and small. We convene as a leadership team. BD is a leadership team topic. And we convene regularly to talk about the meetings that they've had. We sit in on these meetings pretty regularly with small companies just to learn and -- about what's going on and develop relationships with other executives. And then for the actual transactional work, there's a core function internally that all the therapeutic areas work with to do contracting and financial analysis. But -- so that's a sort of process point. But conceptually, BD is on the table or at the same playing field, I should say, with all of our internal programs. So we -- before we start any new discovery program, we look externally to see whether or not we could get there quicker with less risk. The Foghorn collaboration is actually a good example of that. And we're constantly on the prowl for things that can help us augment the portfolio, whether they're discovery collaborations or actual drugs. We can talk a little bit about sort of the inventory of what's out there and how that plays into the kinds of deals we sort of have and haven't done. But BD is an integral part of the strategy. I wish we could do more because it would mean that it passed our bar, it passed the muster on valuation, and we wanted it in the portfolio. We're not doing more for a reason. It's not because we're not trying. It's a function of largely of sort of assets of interest. That's the main gating item. To a lesser extent, other variables like transactability and valuation of those things. But it's mostly around assets of interest.

Great. I guess just a question for David. When you look at the quality of the assets that you saw across pharma and maybe, to some degree, the quality of the assets that you see coming out of the Lilly organization, the discovery organization, is there a reason to compare and contrast? Is there something unique about what you see coming out of Lilly discovery relative to maybe some of the BD assets that you're seeing? Is there a little bit of a quality of -- we actually think, in a lot of cases, we can do it better and faster here at Lilly?

I don't think we have an inherent bias of not invented here, must be worse. I think that's actually -- I think we work very hard to the extent that's, I think, a natural cognitive bias. And I think we work very hard to correct that in our assessment and really be as neutral as we can in assessing external innovation opportunities that are transactable. I guess I'd say a couple of things. We obviously are further ahead in our small molecule discovery pipeline. I think that's just an objective statement.

Relative to biologics.

Relative to biologics. And so I think like that's just one area where, obviously, we're very keen to look at external innovation because we have -- that's a newer therapeutic area for this new organization. And I think you've seen that reflected in some of the BD deals. For example, the deal that we did with Merus, where we really thought that was an excellent platform and was one that we would be proud to use and we didn't think we could do better even given the time. I think the other thing I'd just say about BD, which is a lot of the real innovation in this space these days is happening at big pharma companies, actually. I think there's been a pendulum back and forth on this. I don't think it has to be that way. But I think there have been periods of time where more innovation has come from biotech and then brought in through BD into big pharma. And I think a lot of the really important molecules in development now have actually originated within big pharma, which I think gets to your original point, that these organizations can make really important leaps in developing drugs. So I don't know if I've answered your question directly. Jake, you could feel free to jump in, but that's some of the ways that we see this.

The only thing I would add is that, yes, we have a lot of conviction in our internal discovery team. And you've seen some of the molecules they turned out and there's more to come. So I think it's an important thing to say because were that not the case, you maybe might see us be more active in BD because we feel more desperate. We don't feel desperate. We feel like we've got a great portfolio of substrates to work on. Obviously, we have a lot of visibility into what's coming as well and what our own teams are capable of. So that's a privileged position to be in, to be honest.

And I guess the last point I'd just make about this, it gets back to the original point of the integrated unit. Again, I've not worked in other major pharmaceutical companies, but you do sometimes get the sense that there have been deals that have been transacted at a top-down level where a deal is transacted, but the development organization isn't thrilled to actually work on the drug. And I think that is a first pass test for us. And for us to bring in a near-clinical stage program or a clinical stage program, it has to be a drug we'd be really proud of and excited to put our time on to through that framework that Jake mentioned because we are intimately involved. We've put a lot of our own time and sweat and equity into these programs. So we have to really believe in them.

Got it. Great. And then maybe just as a follow-up there. The Foghorn deal, maybe you can talk to us a little bit about your thoughts on that transaction. And we are seeing a lot more effort in areas like targeted protein degradation where those -- that seems to be a growing area of high interest. Where do you guys fit on sort of the TPD space, broadly speaking?

Maybe I'll take the second part of the question first, and then I'll top off. I think targeted protein degradation clearly is real. It's a chemistry modality. It's not one where, in our opinion, there's a ton of special sauce in the sense of like intellectual property barriers. I think the barriers tend to be more human barriers. It helps to have people who've worked in those domains before because they sort of instinctively understand some of the SAR around what kind of chemistries to use to make these things both work and have good ADME properties. As you know, pharmacology is really the big limitation of these bulky molecules. We actually have a bunch of those people in-house already. And so we think about those strategies for our programs. And I think, as you know, the Foghorn team has some folks there that have worked in these areas, too, and that's an integral part of the drugging strategies, too. It does unlock a set of targets that you'd otherwise not be able to go after because of an inability to drug a catalytic domain. So it's obviously a new exciting way to drug things. Is it a step change for cancer drug development? I personally don't think so. Again, nor do I believe it's really like a platform science, so to speak, either. I think it's a tool. It's a tool in the med chem toolkit. The Foghorn deal really originated from biologic conviction and the merits of discovering a selective BRM inhibitor, which you can imagine we were thinking about building one. And like I mentioned earlier, in that process, we surveyed the landscape, talked to a lot of companies. We knew this would be an unbelievably hard product profile to accomplish. And we were just very impressed by the Foghorn team as well as the actual science progress they've made against the program. So that was the beginning of the conversation. And then as we learned more about their BAF complex, transcription factor complex, structural biology capabilities, we saw opportunity to do more with them on other targets where we're a big structure-based drug design shop. We're strong, strong believers in that. It's a core tenet of what we do. And there was a lot of kinship with that group in that respect. And what they've done in the -- multiple protein complex structure-based designer on the BAF complex is pretty hard and would be really tough for us to have replicated certainly on any real time line that would matter. So that was a nice nidus to expand the collaboration and make it something bigger. Yes.

Great. And then maybe we'll switch a little bit before we kind of jump into the clinical catalysts and how the programs are advancing. I wanted to just kind of get your sense of how the early breast cancer launch is going for Verzenio. And what -- if there is a positive impact, do you see that having a bit of a halo effect on the metastatic disease setting?

Yes. Look, it's early days, is the first thing I would say and it remains early days. That being said, you guys can see the weekly prescriptions just like we can, and they're looking good so far. Now in fairness, they should look good. So I'm hesitant to do too much high-fiving around that. Although I guess, on some level, it's nice to see things play out the way you sort of expect them to. But yes, we're sort of seeing weekly new highs every week in absolute NBRx, which is great. We need to see that translate into TRxs, of course. And it's too early really to know, and I'm interested in invoking the idea of persistence. And for this agent, the diarrhea is a real phenomenon. We believe that the protocols we've developed to manage it are very effective, and it tends to be a short-term thing. But we have to make sure that all the physicians and pharmacies sort of know about that and are bought in on the idea of proactively managing it, particularly amongst some doctors that have just never written for the drug before. So that's the second thing I'll mention, which is just that as I'm sure you can imagine, there's a fairly sizable cohort of physicians who are big CDK4/6 class users, obviously, in the metastatic setting, who have literally never written a Verzenio prescription before because they were big IBRANCE users, not surprisingly. We're seeing them start to write Verzenio prescriptions. And we don't know if they're metastatic or adjuvant. I presume they're adjuvant for the most part. But that's obviously an exciting metric that we're tracking over time. As it relates to your last question about the halo effect, too early to know. We wonder the same question. But it's just too early to know. Dave, you should chime in, yes.

Yes. I guess on that last point, I would just mention, and again, I think it's just -- it's really early to know. I think the first step is to get a prescriber to get experience with your drug. And if their first experience is with early stage breast cancer and they have a positive experience, I think that at least opens the door to having that experience. So I think this is something that if it's there, we'll see how it play over the long run. It's not going to be a short-term overnight switch.

Right. And obviously, we've got some competitor data that's anticipated either at the very end of this year or early next year, the NATALEE study. David, just -- as you think about the design of the study, the molecule itself, some of the decisions that have been made, can you just maybe offer us your thoughts on some of the pushes and pulls that you think about as those data are kind of coming to market?

Well, I guess the first thing I'd just say is that what we really spend our days thinking about is our own data and kind of our own launch. We obviously have no control over what happens there. We've always believed we have a differentiated molecule. I think you've seen these data now for 2 of the major commercial agents in class, and we'll see that card flip. But I think you're alluding to, obviously, the specifics of that drug as well as that study design. And I think these are not unique insights, but there are a couple of things that I do kind of note about that product and that specific study design. One is we've really heard very consistent feedback with the 3-year duration of therapy they're using in the NATALEE design with ribociclib is just there isn't a lot of support or interest in the community for prescribing an adjuvant CDK4/6 inhibitor for that length of time. And I think that will be a real headwind on that asset even if it reads out positive. I think the second thing, in terms of the study design, is the enrolled patient population. I think one of the things I'm really pleased in seeing kind of physicians adopting Verzenio for early stage breast cancer is they see the clinical value. This wasn't just a stat sig result with a marginal clinical effect size. And one of the reasons that we can see this clinical value, 30% reduction in IDFS and distant relapse-free survival approximately is really because we enrolled a patient population with high unmet need, these high-risk patients. And I think the NATALEE study is a broader population, certainly in a traditional commercial lens that can be viewed as a good thing. But I think really where physicians are looking to escalate adjuvant therapy with CDK4/6 inhibitors, they're looking to direct that in a selective manner to patients that need it most. And then I think the third thing is the safety profile of Kisqali is obviously a concern in a curative population and the kind of need for monitoring and such. And I think that plays out. It wasn't too long ago as a practicing physician, and the reality is even where I practiced at Memorial Sloan Kettering, which had huge health care resources and support for physicians, these kinds of things in terms of need to monitor and monitoring ECGs and other things really do play out in prescribing decisions in the real world.

Great. So let's talk a little bit about Retevmo. I think just as being fair and balanced, we're seeing great acceleration in Verzenio, not quite seeing that in Retevmo at this point. And once you got a better sense of whether or not you think there's a realistic opportunity for the product to kind of accelerate to blockbuster status. And honestly, how much of that is COVID related or a need to really rethink [indiscernible] oncology agents that are going after relatively infrequent mutations?

Yes. So look, I think we're largely having a conversation about expectations. And I guess, I say that because the drug has sort of -- has commercially performed more or less in line with -- so far with how I thought it would. I'm not trying to say like I was right, you -- like none of that, but this is not dramatically different from how I thought this would play out. I always viewed this launch as looking a little bit more like crizotinib, obviously, relative to the different epidemiology than say alectinib. This was the first new drug against a new biomarker that wasn't commonly tested for. We've done the actual math to sort of put this on the crizotinib curve and see what it looks like relative to that drug when you adjust for epidemiology. It's actually very similar. So we feel fine about how it's going so far. Would I like to see more patients tested and more RET-fusion positive and RET mutant patients actually get prescribed the drug? I would. We're obviously doing that work. I think there's a bunch of barriers there that we're working to break down but are real. But -- and the other thing to mention that this is a long duration therapy we expect and hope. That's not going to play out in the first year, 1.5 years of a launch. That takes time for pay -- for the sort of year-to-year prevalent pool of treated prevalence to actually emerge. So -- but that all having been said, of course, it's an opportunity for which we scale our OpEx investment accordingly. It doesn't deserve the OpEx investment or certainly not the OpEx tactics that something like Verzenio gets. So that's okay. And there are a couple of important even clinical development things for that program that have yet to play out that I think will make a big difference, most notably the randomized trial in first-line lung cancer. Let's not forget that I'm pretty sure I'm right about this, EGFR and ALK are the only 2 oncogene-addicted lung cancer targets that have ever generated randomized data. And those 2 classes of medicines, I think, have proven to be big classes. So despite, I think, consensus opinion that the efficacy safety profile for [ Retevmo ] is exciting, I do think the randomized data help. I think they help in the face of PD-1 chemo that's easy just to give. And I think they help in the face of testing nihilism, like should I test? Should I wait to start treatment before [indiscernible]. I think if this trial plays out as we expect it to, the hazard ratio will be significant, and I hope we'll change the conversation at least in lung cancer. And that trial is going well, accrual-wise.

Yes. Got it. That makes sense. So maybe let's jump to pirtobrutinib. You commented, and I think the team has a plan to file an MCL, likely to complete the file in early next year. What's the opportunity? MCL, kind of a small-ish opportunity in the mix. CLL relapsed, refractory CLL obviously, the bigger opportunity and really where we saw Calquence sort of gain its presence. David, maybe, from your perspective, what do you think it really takes to potentially get this filed in relapsed/refractory CLL? Or is that an opportunity where we will really need to see the randomized data?

Yes. I mean I think the short answer is we don't know yet. I think we've been clear from the beginning that CLL really is a disease now for many years where randomized clinical development has been the expectation and the bar for regulatory approval. And I think that reflects the innovation that has been delivered in the space with the covalent BTK inhibitors and obviously venetoclax. We have an interesting situation now where those drugs have been in the clinic long enough that we actually have a prevalent and increasingly large population of patients that have experienced both those classes of agents and are now again without great treatment options. And that's what we hear repeatedly from the physician and patient community. And so that has -- we have asked that -- we are -- it doesn't lead to the natural question of is there an accelerated approval filing opportunity there? And we don't know. We'll continue to engage with the FDA. We'll continue to mature the data set and see. But I think it's fair to say that thrust of our development efforts really almost exclusively at this point is enrolling a series of pivotal clinical trials, which will, we believe, guarantee access both in labeling and payer perspective, and also affirmative reasons to prescribe this drug in addition to the data that we've demonstrated. So I think that's really the best that I could do because it reflects our own best understanding of where things stand.

And what about resistance mutations? I mean I know that they're infrequent, but it just seems like that's an area where there would be a relatively clear opportunity. I know you -- Lilly and Merck both were sort of pursuing that strategy initially. It looks like Merck had to go a little bit back to the drawing board from a formulation perspective, while you guys were able to continue running ahead. Why isn't that a vehicle to really pursue an accelerated approval path?

Yes. Well, I think the first thing -- you're alluding, obviously, to mutations in BTK itself, the target of pirtobrutinib, and the covalent inhibitors and really specifically a mutation at the cysteine binding site, which is necessary for all the covalent inhibitors to be potent and active, but dispensable from the perspective of pirtobrutinib. It's fair to say that that observation, that clinical translational observation was part of the genesis of what became pirtobrutinib from a discovery effort. But I think the short answer to your question from a clinical perspective, and we've shown these data multiple ways from an ORR perspective as well as PFS perspective, the resistance mutations really do not define a subset of patients in which pirtobrutinib is more active. And I want to be a little bit even more clear in my words, pirtobrutinib seems equally efficacious in patients that have discontinued covalent BTK inhibitor for intolerance or progression regardless of the presence or absence of this mutation. So I personally would feel good about advancing a drug based on a biologic thesis that hasn't defined an enrichment strategy in the clinic where the drug is active more broadly. I think we should be developing the drug best -- reflective of our best understanding of how it helps people.

It would also introduce a testing component that is sort of -- is unnecessary for the health care system and for patients and doctors.

Yes. We heard a lot about these mutations at the outset of our program, and they were on our mind too, for sure. We don't hear about this anymore from our investigators. It's just -- it's not a question we're getting. It's not -- they're not seeing any enrichment of these patients on our clinical trial. They've kind of internalized based on the data that we've shown that this drug has the potential to help their patients regardless of the mutational status. And it's not part of routine clinical testing for that reason.

Got it. I mean it also sounds like, to some degree, commercially, you just -- it's not a great idea to pigeonhole the asset when it's performing as well as it does.

Yes. I think if that was -- if there was an enrichment strategy, you'd do it. You would do the right thing for patients. But in this case, there's no clinical rationale. So it would only be really limiting in many ways.

Got it. Okay. So let's go to the -- I guess, the topic of the day, which is yesterday's briefing documents. I'll just, Jake, offer you an opportunity to give us your brief thoughts because I know that this is probably not an extended conversation. But yes, maybe we'll just start there. And I might have one follow-up question.

Yes. Well, I don't have a ton more to say on this that hasn't already been said by many people in the past couple of weeks, I guess, myself included. I think we know how tomorrow is going to go -- or I should say, actually, we don't know how tomorrow is going to go. I think we know how the PDUFA date is going to go. I think that's been fairly broadcast at this point, fine. At the end of the day, by the strict regs themselves, we believe that the study merits consideration, the data merits consideration in the application. And that's in part why tomorrow is still happening, to be honest. The other reason is that we think there's an important conversation to be had. I know that FDA can't take price and costs into consideration, and that's fine. Of course, that's fine. But we're allowed to say it and whether or not it comes out tomorrow or not, different story. But we've said it publicly, we've said it in the months leading up to this. It is and remains a core part of the idea that we decided to do this, to innovate on price, disrupt on price. And I think that's an important conversation for health care practitioners and sort of society to have. Whether or not that conversation happens tomorrow or is allowed to happen tomorrow is not really our prerogative. So we'll see. But yes, I think the ultimate outcome is sort of known at this point, which again is fine. This whole topic, as you know, matters a lot more to a bunch of other folks than it really matters to us. And as much as this product was -- even in a successful situation, was never going to be that important or significant financially to us. We just happen to be sort of the first people here because of the prior relationship with Innovent and the study that was already conducted. I don't know if Dave, you want to add anything else on that?

No. I mean I think we've proceeded in this application in good faith, and we'll see it through. And ultimately, it's not really our decision. And I think there's no information asymmetry on where things stand here between the public and us at this point.

Yes. That's right.

And in terms of information asymmetry, what about other [ OUS ] markets, not just the U.S. market, but other opportunities, whether it be in Europe or other individual markets. Is there an opportunity there that may be underappreciated?

There might be. I don't know is the short answer. We had been -- well, obviously, it goes without saying you did mention it, but our business in China is still very robust, and the relationship with Innovent is really important in that market. And we -- back on the NRDL, we have more indications there. That will be another important business for us this year. Outside of China, we had been focused primarily on the United States. That's going the way it's going. And in terms of other markets, we'll sort of get there and figure out what we want to do. It's not been part of our strategy or tactics to date, but I think might there be opportunity in certain countries that are open to this? There might. But I don't -- I literally don't have enough information yet to have any definitive answer on that.

I just want to say one last thing, which is that I think it's important as we proceed into tomorrow and the days that follow just to remember really what the topic is here. The clinical trial itself is sound and we stand behind it. I think really, it's not -- we've made an affirmative case about why it's applicable to the U.S. population. Ultimately, regulators have the final word on that. But I just want to separate the applicability issue from the core quality of the data itself. We're very proud of the work that Innovent did on this clinical trial. We think they run great clinical trials. They have a great molecule.

Yes. It seems pretty evident from the result and obviously, some of the other trials that we've seen coming from there. Maybe let's shift a little bit to the pipeline and go into some of the earlier stage products. Imlunestrant, you guys made an interesting decision to maybe shift gears a little bit and hit the accelerator to bring forward a second-line study after we got the EMERALD study results. So I'd love to just kind of get your thoughts on the EMERALD study itself. And then separately, what can you guys do differently that perhaps -- or what is different about the imlunestrant itself that you think can bring a little bit more to the table for the second-line patient population. And then we'll kind of talk a little bit more about the earlier stage.

Dave, do you want to take it?

Yes. I just want to kind of correct the record on one thing, which is that we advanced our second line imlunestrant randomized Phase III study, which we call EMBER-3, well in advance of the readout of EMERALD. So there was no acceleration kind of related to that. We obviously knew that that study would readout and have some bearing on that clinical experiment, but we launched that well ahead of the EMERALD study. In terms of -- let me take your questions in turn. In terms of the EMERALD study itself, I mean, obviously, it's nice to see initial proof of concept of the oral SERD class. This has been a class of molecules that have been circulating around for many years. It's -- this is an initial proof of concept. I think expectations were always low for the second and third line kind of population given the historic outcomes in that population even before the introduction of CDK4/6 inhibitors. And now after the introduction, kind of even worse outcomes in that patient population and by virtue of that more unmet need. So I think I just want to pause and recognize them and also congratulating them on being the first ones to show that. I think, obviously, you see the biggest effect size in that particular study in patients that have ESR1 mutations. And I think for this class of molecules to really reach its full potential, you'd like to see some clear evidence of benefit in patients that don't have that mutation. I think the one word of caution there is that that study, in particular, enrolled very heavily pretreated patients, including third-line treated patients. And I think we just need to be careful to kind of recognize that ESR1 can be both potentially prognostic and predictive and also maybe identify breast cancers that remain reliant on estrogen signaling through multiple lines of therapies. So I think we really need to see whether this pattern repeats itself in more clinical trials, including our own. I think the -- there are also some differences I should just mention between the EMERALD study and our study. We focused on a more pure second-line population, where we think these drugs have more potential to really show their benefit. I think your broader question is about how we view the space and how we view the agents within the space. And I guess I'd say a couple of words on that. All these drugs are a little bit different, and we do see different toxicity profiles at the dose escalation of these studies. These are now all matter being in the public domain. We kind of alluded to this in the ribociclib discussion. There are drugs, for example, that have QTc effects, vision effects, bradycardia effects. Even if the doses have been moderated to tune those out as a frequent adverse event, I think that will commit some of these molecules to monitoring that makes it challenging to commercialize in the earlier line setting or even the curative intent setting. So I think that's one area of differentiation. But I think it's also just worth saying that these drugs are probably more the same than different. And really the biggest area of differentiation is in your clinical development plan and kind of how you design and execute on your clinical trials, and that's a huge focus for us. So we can talk more about how we see the space and how we're thinking about developing the agent. But I think I've answered most of your questions. I should probably take a breath.

No. That's great. I guess as we think about not just the second-line opportunity, I think you guys have said you see the first-line metastatic setting as an area that would be particularly challenging to really differentiate and that early breast cancer adjuvant setting is really where you see a combination setting itself apart. Why is that? Maybe you can just help us understand that discussion a little bit better.

Jake, do you want me to jump in or...

Yes. Yes, keep going. Yes, keep going. Yes.

Yes. I mean we -- as we got to understand the characteristics of imlunestrant and went through the Phase I/II clinical trial, we obviously benefited in a way from being a little bit behind study starts to our competitors to see more data disclosures in the field. One that has in the margin our mind is the PARSIFAL study, which looked at palbociclib first-line metastatic patients, palbociclib with fulvestrant versus aromatase inhibitors. And not only did the fulvestrant not outperform the aromatase inhibitors as was the expectation, but numerically, it was a little bit below. I don't think much of that from a statistical basis. But really suffice it to say there was no evidence that fulvestrant was superior to AI in the presence of CDK4/6 inhibitors. And that has driven a thinking in the field, which we spend a lot of time thinking about as well is that the effect of CDK4/6 inhibitors, including Verzenio, is so significant that it may wash out of the effect of minor endocrine therapy optimizations that you would have been able to perceive before that innovation was brought to patients. Now obviously, core to the thesis of oral SERD is that they're better than fulvestrant. So -- but really, it's a magnitude of how much better you need to be to be clinically significant given the good outcomes in patients with first-line metastatic disease. So that really caused us to kind of deprioritize that. I think many of our peers launched their first-line studies or began enabling those studies prior to that readout. I don't know. I can't really speculate how that readout impacted our own assessment of those clinical trials, which are underway. But I think more -- even more broadly before PARSIFAL, I think we've always held the view that oral SERDs or SERD in general is the best endocrine therapy class. And really where you want to put your best medicines is in the curative setting. That's where you can have your biggest effect size, and it's also where fulvestrant has been prevented from going [indiscernible] due to its route of administration and just the impractical nature of administering that indefinitely in curative intent patients. So that's really where we spend the focus of our time. Obviously, we know we need to develop data in the metastatic space. And the one thing we haven't talked about is that in EMBER-3, not only are we looking at imlunestrant versus standard endocrine therapy, but we also have an arm looking at imlunestrant with Verzenio. And that's another area as I speak to the ability to differentiate not just on the molecule itself, but through clinical trial and development and innovation. That's another area where we think we can innovate by using combinations, in this case with Verzenio. And I think that has the potential to be really impactful in that space as well. So that's really how we're thinking about it. Jake, do you want to jump in with anything else?

No. I agree. On the last one, I would just add that CDK -- abema, given after a prior CDK4/6 inhibitor, with hormonal therapy is done today. It's just not -- it's done sort of against not a lot of data. It's just done. Obviously, we'd like to see abema written in the first-line setting instead of palbo, but fine. There's a lot of patients who are getting palbo and when they relapse, abema is being written a fair amount. So we're now just -- we integrated that idea into our SERD study as a means of generating data on both compounds together in a setting which abema is already used.

Makes sense. Great. So let's jump to the KRAS inhibitor, so 7982. You had a prior KRAS that I think needed to be shelved. You have a new one. Feedback that we're getting is that this looks like a really compelling compound. As you just think about it, first, what's unique that warrants the incremental attention from thought leaders? And then separately, where do you see it differentiating? Is it in metastatic CRC? Is it in lung cancer? Where do you see your KRAS potentially capitalizing on a better profile?

Yes. Our KRAS program is an interesting one because it's a big upside lever for the portfolio and yet one that I think today still has a fairly low PTS associated with it. So let me just tell you how we're thinking about it. I think the molecule has the potential based on its potency selectivity and ADME properties to differentiate on efficacy and to differentiate on safety. And those are 2 different ideas. I think the efficacy idea is would we have a meaningfully different monotherapy response rate and duration of therapy in a comparable patient population? We've always thought that was sort of a tough thing because unless they're meaningfully different, I think you're in the gray zone across trial comparisons, small numbers, et cetera. So we'll generate those data. And if they do look meaningfully different, that will be amazing. But I think that was always going to be sort of a hard thing to prove. The safety idea, I think, is actually becoming more important to our overall thinking, not because we've generated any data yet but because of the changes in the landscape I think we've now seen. And this has been our view all along that first-line lung cancer is really the main opportunity for this class of medicines. It's not the only. There's obviously an opportunity in second line. There's opportunities in colon and pancreas and other places. But the bulk of the opportunity really is in first-line lung cancer. And to access that space clinically, you really need to be able to combine well with PD-1. And I think we now know that the 2 leading agents in this space just really can't do that. And so it creates a big white space for an agent that could do that. Now maybe none of these agents can do that. In which case, this whole class is unfortunately not that important. But we haven't proven that to be the case yet. And I hope we don't prove that to be the case. So we're going to test that in the context of the clinical trial. I think if that turns out to be real for this drug, if our drug can safely combine with PD-1 at a dose that we like, that's a big upside lever for the portfolio and for patients. So we'll see. We'll be generating those data, hopefully, over the course of this year. We'll have to figure out the right time to present them at a medical meeting or something, whether this year or next year, we haven't committed. But that's how we're thinking about it.

Got it. And so the differentiation on safety as we kind of think about it and the combinability is really...

Again, there could be an efficacy differentiation too, but I just -- again, I struggle with it. If -- I struggle getting there just based on -- it would have to be really outsized. And I think that is -- it's not impossible. It's just a hard thing for even us to believe. Otherwise, again, you're just end up 40 patients, [indiscernible] -- I don't like leaning into that kind of thing too hard. You can often get...

And just to be clear, I think Jake is making more of like an interpretation of clinical data and statistics point. It's not based on anything we're seeing specifically, but just kind of in a first principles basis of trying to sort out the difference between response rates in different single-arm clinical trials scaled at the size that we've seen.

Yes. That's well said.

And we've seen sort of these basket combination trials emerging outside of [indiscernible] potential combinations. Are there combinations outside of PD-1 that you guys are interested to see incremental data on? Or that you think are targets worth pursuing in combination with G12C inhibitor, whether it be SHIP2? Do you not have a lot of hope for SHIP2? Are there other RAS/RAF mechanisms that you think are worth paying attention to?

So I think EGFR antibody combinations in colon cancer, I think that -- right, that we -- that's sort of rearview mirror at this point, right? That's real. That's going to be real. We're obviously in a good position to take advantage of that with our portfolio, and we will. I think outside of that one, we remain -- and PD-1, which we talked about already, we remain as a field in the learning mode. So our study contemplates a bunch of combinations. There's rationale biologically to combine with CDK4/6, so we'll combine with abemaciclib. There's rationale to combine with MEK-ERK, so -- and we have an in-house ERK inhibitor where there's rationale even to combine with Aurora kinase and we have an Aurora A kinase inhibitor that will test that thesis. I don't know how to be bullish on any of these things in advance of generating clinical data. Doesn't say I'm bearish on them either. It's just the preclinical data don't do a great job of predicting this. So it's an unfortunately somewhat empiric exercise.

Yes. Got it. Okay. I guess one question that we got from the folks listening in is the conviction that the other KRAS inhibitors can't necessarily be successfully combined with PD-1. What were the data that really informed that? Was that recently presented? Or is it sort of tested behind the scenes at Lilly, to some degree, looking at different compound combinations? Just love to know where -- the thoughts along those lines.

Yes. Well, I think it is a point of fact to say there's almost no data in the public domain on this topic after multiple years of this being experimented with. So that's just a statement of fact. I should be careful here because we don't have like confidential information of others. And frankly, this question is better posed to those sponsors. That having been said, we're out there with a protocol that contemplates combining our drug with PD-1. We're out there talking to investigators and sites about opening the study and enrolling patients to it. So we hear feedback on this idea conceptually. And maybe I'll just leave it at that.

Got it. Understood. And then in terms of some of the other programs that you and David revealed at the R&D Day, we're coming up on time, so I just want to make sure that we don't give the rest of the portfolio short shrift, but we do only have a few minutes left. So maybe David, your thoughts between the Aurora kinase, IDH1, FGFR3, the [ PIK3 ] assets, which are you particularly excited to sort of share with investors over the next 18 months or so?

Yes. Great question. I think I'd highlight 1 or 2 in the last 2 minutes here. One is our -- what we call LOXO-783, which is our mutant selective PI3-kinase inhibitor specifically for a mutation called H1047R. We are really excited about that drug. It's conceptually entirely different than the entire class of PI3-kinase inhibitors that have been developed to date, which I think have shown initial proof of biologic clinical concept. There's even one approved drug. Obviously, Piqray, alpelisib, from Novartis, but those drugs really haven't moved the needle on an efficacy or safety standpoint. We think really it's the therapeutic index and the lower target coverage that that drives that has been the limiting step there. So we really set out to build something clearly different, which was a drug that really allowed kind of no regrets inhibition of that specific mutation and completely spared wild-type. We think we've achieved that. We made a very opinionated decision in the way that we design that clinical -- that asset, sorry, to deliver on that. We decided to really target that mutation in no regrets manner, rather than try to be kind of somewhat mutant selective across a broader range of alterations in that gene. So that's the driver we're putting in the clinic this year. Really excited to see what it produces and kind of sharing that to the investor community. And the other one I'll just highlight quickly, which is really very similar in thesis is our FGFR3 inhibitor program. Again, clinical validation of FGFR3 as a driver primarily in urothelial cancers, but the approved agent has real toxicity associated with it. And we believe inadequate target coverage as a consequence of that. And a selective FGFR3 inhibitor was incredibly challenging build due to the homology of 3 versus 1 and 2, and we really think that we've achieved that and think it's going to deliver a clinically differentiated profile for patients. We spent a lot of time on these programs. These are programs built over years. They're hard, which is why we feel they're differentiated. And so we'll have to prove that clinically.

Jake, maybe just to wrap it up, anywhere you don't agree with David?

We deal with those outside of the public's eye. I'm kidding.

I'll get a talking to later.

No. No. No. I'm just joking. Team alignment is an important concept. This is an -- I mean, I am serious. This is an incredibly multidisciplinary sport. And frankly, one of, actually, our cultural hallmarks is we encourage dissent and debate, and we oftentimes have pretty robust arguments amongst both our leadership team as well as the sub teams beneath that work on the projects and it leads to better outcomes. So it's -- I joked, but -- and generally speaking, we have the good fortune of oftentimes being in alignment across our leadership team, but dissent is good.

Yes. Understood. Well, thanks again, Jake and David, for joining us. Great discussion. Really appreciate it. And outside of tomorrow, I'm sure we're going to see a lot of really positive data sets coming from Lilly over the months and years ahead. So thanks again for spending the time with us and really look forward to our next engagement. Thanks so much.

Thanks, Seamus. Thanks, everyone, for joining us.

Thanks.

Appreciate it. Bye-bye.