IDEAYA Biosciences, Inc. (IDYA) Earnings Call Transcript

Okay. Great. Welcome back, everyone, to day 1 of Citi's Virtual Oncology Summit. This has been running for several years now. So we're pleased to have you back. This is the lunch session, at least on the East Coast, breakfast for Yujiro, out west. It's my pleasure to have Yujiro Hata, who is the Chief Executive Officer of IDEAYA. Welcome, Yujiro. Thank you so much for taking the time out of immensely busy time for you. As a reminder, if anyone has questions for Yujiro, please just e-mail me and I can relay and hopefully see those and relay them over. So obviously, a big year coming up for you, a lot of programs, a lot of studies, both early as well as mid as well as late-stage. So there's a lot to absorb. So maybe just at a first blush, first question, just set the scene for everyone in terms of what to expect, what are the key readouts for 2025, how to think about the company and how to think about the mission and the thesis?

Great. Well, thanks so much, Yigal, for the introduction, and thank you to Citi for hosting us again this year at your oncology event having. Today, I'm dialing in from our San Diego office, so not my typical office background here. So 2025, Yigal, we believe, is set up to be perhaps our most transformational year. And I do think front and center as part of that, as you know, is darovasertib, our lead program, which is in a potential registrational trial in the first-line metastatic uveal melanoma setting. And we have guided towards our full potential approval readout before the end of the year, and that endpoint is specifically the progression-free survival. As you know, we presented at ESMO a year or so ago, where we reported a median PFS of just about 7 months, historical PFS in this indication is roughly 2 to 3 months. The second was a key catalyst for darovasertib and for the company as we're also guiding towards what we think will be a significant clinical data update in the neoadjuvant uveal melanoma setting here. We've enrolled roughly 95 patients, so a fairly large data set. We are also planning to provide a key regulatory update in the first half as well as it relates to locking down the final protocol for the registrational study. And then finally, last for darovasertib. We've also guided to a third readout, which is in the metastatic setting frontline for the combination with crizotinib, adhered to focus on median overall survival. And, Yigal. As you know, we've never reported survival for this program yet, and in terms of our powering assumptions for the full approval study, we're assuming an ability to at least exceed OS by plus 6 months from the comparator arm. I would say beyond that, I would focus on several additional areas. And I think here, Yigal, really, our focus is to build a leading precision medicine oncology company with a truly diversified pipeline going after several key biomarkers. Now I would say next is in the area of MTAP deletion. Here, we have ongoing studies both as monotherapy for IDE397, which is a Phase I/II MAT2A inhibitor currently positioned as a first-in-class agent, where we're setting both lung and urothelial cancer. As you know, we had a late-breaker oral presentation in Barcelona at the end of last year, and we have a core focus on enabling combinations. We do anticipate being able to provide a clinical data readout specifically on the Trodelvy combination with Gilead, here, we announced recently the expansion of that combination effort from urothelial cancer to also now include lung cancer. In addition, Yigal, as you know, we also are applying our own PRMT5 inhibitor in the clinic in the middle of this year and the core-strategic objective to enable that combination as a wholly-owned combination for the company. I would say next is really just a broader thesis on other assets that are in the clinic, Werner Helicase, we anticipate we'll be presenting at a major medical conference in the first half of this year. Here targeting high microsatellite instability of solid tumors. We've also guided towards a clinical readout on the DLL3 Topo-ADC program, also this year we do anticipate at least we're targeting for that program, a data readout at a major medical conference later this year as well.

Okay. And there's even more, I think, that we have to get to beyond those. But let's start with the late stage. So for MUM, you mentioned the PFS at the end of the year. So assuming you achieved a similar benefit over standard of care, just give us the time lines. I mean, when could this drug potentially be approved? How quickly could you file it and approved?

I haven't given the specific timelines to actual launch, Yigal, but as you can imagine, if we are able to have the PFS readout by the end of the year. And some of that will really be driven by the final piece of the enrollment. As you know, we recently announced enrolling over 230 patients in the study. And of course, as an event-driven endpoint, ultimately when the events occur both on the comparator arm as well as the treatment arm. But assuming that we do get it by the end of the year, we do have Fast Track designation, which would accelerate the review process on the time that it would take to assemble the NDA, submit that and then that waiting period. So I think what we can highlight is commercial readiness has begun. We've already hired a Chief Commercial Officer. We have several ongoing searches right now to build that organization out in preparation for that PFS readout.

So help us understand the reasons to be confident in a positive PFS. I mean, obviously, as you pointed out, you've got a very nice separation versus standard of care already. But are there other things that should be pointed out as far as why to be confident? And if there are that, obviously, those risks, but in terms of some things that you believe you need to watch out for to make sure the study being successful?

Yes. We feel good all about the prospects for our PFS readout. I would say primarily because of where we've seen the PFS in the past, including what we presented at a major medical conference, we powered the study where as long as we're above that 5.5-month mark, we believe we'll be in good shape based on historicals that we've seen in the control arm. And at least for PFS, that number has been pretty stable, including for Immunocore study, reported roughly a 3-month PFS. So we feel pretty confident that number as well as from large meta-analysis that have done that have shown that 2- to 3-month range. So we're fairly confident that control arm is going to behave in a similar fashion. And because so far, we've been well north of that 5.5-month mark, again, slightly above 7 months. We do feel good about this endpoint specifically. And in terms of patient population, we don't believe there is going to be any major differences from what we've enrolled in the past. So obviously, here, the focus is on HLA-A2 negative, but as you know, we have not seen a difference whether or A2 negative or A2 positive, at least as it relates to PFS or response rate or other efficacy type endpoints.

I mean, obviously, that will be a conversation with the regulators, but is the expectation that label would be HLA negative out of the gates? Or could it be broader?

Yes. So our base case, Yigal, because the registrational trial enrollment is focused on A2 negative. The focus there would be A2 negative and that we would take a compendia strategy for pricing and reimbursement in the U.S. for A2 positive based on published data. Data we've already published in a medical conference setting and data we will continue to plan to publish moving forward in the future. At the time when we target to have the NDA submission occur, could there be an upside, but it's not in our base case scenario, could there be an upside when we submit real-world data on A2 positives because we'll have a lot of data at that point of that submission. Could there be an upside opportunity where it is in the label. I suppose that's possible as a scenario, but at least right now, we don't view that as a base case.

Okay. Now this -- the Phase II update referenced on the OS is very interesting, especially since it's coming ahead of the PFS for the Phase III. So I mean, what should be the expectations there relative to standard therapy. And assuming that looks really good, I mean I would think that, that should advance or improve the chances on the PFS just because we have a hard survival endpoint. But just help frame what could happen at that OS event, which is, as you point out, we haven't seen any of that before.

Yes. So you go as you appreciate, as we get closer and closer to that PFS readout. I think the obvious question and absolutely a fair question. I think investors and analysts will want to understand is, well, how about OS, and any sense of where OS may be. And so that's where we think this data set will be valuable. Obviously, it's not a randomized data set, typically it's all about a hazard ratio survival, but we do know historical survival here has also been fairly consistent, again, based on other studies that have been done, large meta-analysis, where we anticipate that survival OS will be roughly in that 12- to 13-month range. And here, similar to other studies that have been run that where they have seen approval based on OS was about 6-month improvement, and that's how we've powered our full approval study. So that's the bogey. And where are we relative to that? I think will be the question when we provide this update. So I do think it will be a data point and hopefully an informative data point.

So in other words -- you're looking for something in the high-teens, 18 to 19 versus 12 to 13? Is that correct?

Yes, high teens. 20-plus is probably the range you want to be in ideally.

Okay. All right. So let's talk about the neoadjuvant, which is obviously a great interest as well. So you mentioned you're going to have some updates in terms of the -- well, 2 things, the 95 patients and then the final lockdown of the Phase III, so can we get into a little more detail on that, and some of that's really important as far as understanding the vision endpoint for the Phase II. So what are we going to get in that next clinical update that will help investors get more clarity or understanding on that vision piece.

Yes. So it's going to be a much larger data set. So now we have roughly 95 patients from our company-sponsored portion of the study. I think on the eye preservation enucleation cohort, I think that data, I think people felt quite good about. And at least in our FDA briefing book, the target that we had set was to exceed the lower bound of 10% confidence interval. And as you know, our eye preservation rate was the majority of patients. So we are well north of that objective. So I do think a lot of the focus was on the plaque therapy cohort group and specifically around data on vision preservation. And at the time of the update in the fall, we just didn't have a lot of patients. We had about a half a dozen patients, that data set has now grown. So we have multiples of that number. We also have more longitudinal data as well, that data set is still maturing, but I think directionally, hopefully, it will be an informative data set, with obviously, our objective to put that cohort also on solid footing as well. And then finally, we are considering a possible omnibus type update where we also include the regulatory piece. We are having a lot of back and forth with the FDA currently just to try to fine-tune the last remaining pieces around the clinical protocol. And I would say probably the biggest piece really outstanding right now is, I think, Yigal, as you know, we talked about can we utilize macular edema as an endpoint. And I think there, the main topic is really around, is it a -- can we utilize it as a co-primary endpoint? Or will it be utilized as a secondary endpoint. So we're continuing to go back and forth. And hopefully, we'll have sort of final resolution on that and be able to hit our objective to get the second registrational trial up and started this half.

And then the other aspects are more or less clarified, like using the disease -- the mass metastasis-free survival, if I have that right, and the BCVA?

Yes. So yes. So it will be around a 15-letter type vision readout in terms of time to vision loss, and then for the secondary endpoint for both cohorts will be around no detriment to event-free survival. And that is another piece that we're finalizing what does that look like in terms of no detriment to EFS.

Okay. And then as far as timelines there that as you had articulated previously or that could potentially be modified based on further clarity with the regulators.

In terms of timing, Yigal, study start. Yes. I mean I think we should be in good shape to get that started this half. I mean that's our objective. I think the FDA has been extremely supportive today very communicative. So I think we feel good about our objective to get that study started in the first half. The trial enrollment will likely be in that 400 to 450 patients, which we do think we can enroll that group in roughly 4 to 5 quarters based on the pace we've been going and as well as our objective to get a lot more sites activated, and as we've seen with our existing Phase II study, the enrollment cadence has been extremely robust, as you know.

So anyone, as a reminder, if there are questions, just yigal.nochomovitz@citi.com. I'm happy to read it. So let's move on and talk about some of the mid-stage, early/mid-stage work, of which there's a lot. So I'll try to move quickly here. DLL3 ADC is certainly a hot topic, shall we say, with multiple players. Talk about why your DLL3 is special or differentiated based on payload, linker, internalization, cleavage of the linker, et cetera. There's a lot there, which we could learn as far as why it might be different.

Yes. So I would say that there's a lot of aspects about our DLL3 Topo-ADC that we like quite a lot, and we think it's very well positioned. So first, as you know, in small cell lung cancer, there are several assets or category assets that are being advanced right now, including bispecific cell engagers, you have B7H3 ADCs as well. We -- at least our view is that we believe DLL3 is the optimal ADC antigen: a, it's very highly expressed in small cell lung cancer; and second, it's connected to a key survival oncogene ASCL1,which impacts expression of DLL3, and as you know, one of the key bypass mechanisms for resistance to ADC is to lower the expression of the antigen. So we think DLL3 is very well positioned from that aspect. Now when you look at it at the lens of the DLL3 Topo-ADCs, that competition is more limited. Right now, there is us in the clinic. There's another group that we believe is roughly set by side. And then as you know, Yigal, Roche recently came in with the acquisition of the asset of a company called Innovent, probably about 10 to 12 months behind. Here, we did recently publicly not that several expansion doses have been selected. So we have seen a larger data set from our partner in China, Hungary, and we continue to feel confident about the data set that's emerging. And here the key question will be 2 -- 3 pieces, Yigal. So one is what will the ultimate confirm overall response rate be? Second, what does that durability look like? And of course, third, how does the adverse event profile stack up relative to other ADCs as well as other classes like the bispecific T-cell engagers? I would say the last part really is why we think our asset is unique. So one, you may know that Hungary has used this Topo-ADC platform across multiple assets. It's about half a dozen clinical programs, including that platform that's been licensed to Merck in the past, 2 of GSKs, ADCs therapy B7H3, B7H4 also is from this Hungary platform set up. So it's been at about 4,000 patients. And then lastly, the cleavage happens once it's internalized. I would say there's another asset out there where the cleavage is in the tumor microenvironment. So that MOA is different. I would say our preference is around this internalization, in particular because once you go into combinations, we think that could potentially play out. But ultimately, that data is going to get generated clinically and see where we end up, but we feel quite good about the profile to date.

Okay. And what is this clinical strategy going to look like? Obviously, small cell lung cancer is on the table, but that's not all there is. I mean, there's a lot else in neuroendocrine which you could leverage maybe.

Yes. So I think here, based on the response rate and obviously, we ultimately need to see what the ultimate confirmed response rate and mDOR is. But at least based on the early data, we do believe there should be a potential regulatory path as a monotherapy agent in the later line setting of small cell. And here, similar to what Amgen did with IMDELLTRA, they ran a single-arm study with the primary endpoint of response rate and mDOR. So I think there, we would take that same approach in the front-line setting of small cell, I think here, at least our current thinking is post the chemo platinum doublet to combine with checkpoint randomized versus checkpoint and the maintenance phase. And then as you know, one of our big focus areas is also on this PARG combination IDE161. Our -- one of the main reasons why we brought this asset in is because we believe that's a very compelling, rational combination that could hopefully enhance the durability of this DLL3 ADC. Next, there are other tumor types. And so obviously, neuroendocrine tumors, we think, also has a very high unmet need. We think roughly 100,000 patients, annual patients here, so it is a large population. And Yigal, if you know, I forwarded you that paper the other day that we talked about, which DLL3 expression that was one of the largest studies done to date is being seen upregulation in multiple tumor types. I thought interesting was specifically around findings like fairly large upregulation in tumor types like melanoma as well as others.

Okay. And then the timelines, there will be data this year for that.

Yes. So that was part of our recent earnings release. So we were given an update both in terms of data and also communication from our partner, Hungary, that they plan to present data this year. And we also know which medical conference they're targeting, but until all of that goes through, we'll wait to give that more specific guidance, but at least that's the plan at this time.

Okay. let's talk about MAT2A. So obviously, you have the combo with Gilead. So let's start with Trodelvy. Urothelial certainly makes a lot of sense for the reasons you've explained, and now non-small cell as well. So is the biology -- is the thesis there the same? Or is there some specific differences in terms of the reason to combine in non-small cell?

Yes. So the reason would be identical, Yigal. So here, this is really about the topoisomerase payload and the impact on the purine-pyrimidine and folate pathway specifically. And I know Yigal, we had gone through this before also at the past R&D days that we've held as it relates to the MAT2A mechanism, specifically impacting around nucleotide folds. And as you recall, we've had interest in the past on other chemo combinations like pemetrexed. So that rationale is exactly the same. So at least our perspective, is that the urothelial cancer connection as well as lung cancer should be the same. So that clinical PoC would be the same based on what I just walked through. We have an early better emerging data set in urothelial cancer. That data set, Gilead did review as we made the decision together to expand that effort into lung cancer. So we're excited to get that going. We're getting fairly close to picking the expansion dose with Trodelvy. So hopefully, that will be happening in the next couple of weeks, basically. And we have seen multiple confirmed responses as part of the escalation, and I think importantly, at least based on the data so far, we don't appear to be exacerbating some of the myelosuppression that's been inevitable with Trodelvy monotherapy. So I think that's great. And that was really probably one of the key questions we wanted to answer. So at least right now, we feel good about it. And where now our objective is to build out a larger data set with duration and our goal is to provide a data update on this combo sometime this year.

Okay. And what -- I mean, obviously, in addition to the tolerability that you mentioned, what sort of the rough sort of bar for taking the combo forward? Or what do you -- what would you like to see? Or what would Gilead like to see "resurrect" Trodelvy in UC?

Yes. So we've had quite a few conversations with Gilead. I also know we had a bunch of meetings at ASCO GU recently this past week and spoke to a lot of the KOLs to get their input. And I think the historical response rate here, at least in urothelial for that ADC class is probably in that 20% to 25% range. And so ideally, we're well north of that and can hopefully observe what's going to be attractive durability. So that's really the objective, Yigal. And I think once we have a bigger data set, we can sort of get a sense of where we are. So I think here, our hope is we have a data set that should hopefully position at least us well in that second-line type setting. And obviously, depending on how strong the data is, we could also perhaps consider an earlier line setting beyond that, but we'll just have to see how the data set matures.

Okay. So then obviously, you've been spinning up the engines on your own development of your own PRMT5, and you're advancing that with your 397. So talk about that, and how that study is evolving and just maybe very briefly touch on the choice to move past the Amgen compound.

Sure. Yes. So Yigal, as you know, for several years now, we've been communicating very consistently a core strategic objective for the company is to enable wholly owned combinations in the MTAP pathway. And our program in PRMT5 has been ongoing for probably about 5 years now. So we've been doing chemistry lead optimization for many years. We feel we've identified a great candidate molecule that's really fit for purpose to be combined with IDE397. We got through critical toxicology studies at the end of last year, where we announced the candidate nomination last fall and then obviously profiled the molecule at the R&D Day end of last year. So I think that's really our core strategic objective, and we have a third program, Yigal, as you know, in MTAP where that candidate nomination is also slotted for this year. So we think all 3 of these programs and targets will be extremely valuable pieces of the puzzle to hopefully deliver optimal value to patients. As you know, we've been in the clinic with Amgen for about 5 quarters with our PRMT5 inhibitor. And I think first and foremost, we wanted to mention that we're extremely grateful to Amgen for being our partner for the last 5 quarters. We give them a lot of credit, they really came out. We published, work together preclinically, published that at a major medical conference, actually multiple major medical conferences, the preclinical data, and they were the first pharma company to try to generate that clinical PoC data. So again, we're very grateful on that. I think here, the decision was made to mutually wind down the combination study. Ultimately, our perspective is that we did well achieve the target clinical exposures that we were targeting for both the PRMT5 mechanism as well as MAT2A. And this was based on our preclinical modeling work. So we think we could hopefully be able to utilize those learnings, and implement them very quickly as we get our PRMT5 inhibitor in the clinic middle of this year. So our strategic objective here is to continue to be a leader in MTAP and specifically around this mechanistic combination, and that's exactly what we're going to attempt to do through 2025.

I mean, obviously, there was a very, very strong the synergy of 397 and Amgen 193. Are we going to see that for preclinically? I mean, assume you have it already or you haven't -- I guess we haven't seen it yet, but will we see that before we get clinical data? Or is it essentially going to be the same very, very good story again on the preclinical side with that combo with your PRMT5.

You're talking about our preclinical combo data.

Yes.

Yes. So we did publish some of that work, Yigal. I know all I can say is there's other medical conference that are upcoming that we have some placeholder abstracts on, where we may be able to profile and showcase more of that work preclinically and why we're excited specifically about our PRMT5 inhibitor in combination specifically with IDE397. So we definitely have a lot of different plans for that in terms of publishing that hopefully throughout the year.

And how are you thinking about the prioritization of your combo -- proprietary combo with the original monotherapy work with 397. Where does that...

Yes, great question. We're continuing to interrogate monotherapy, Yigal. We do have a bigger data set. We're getting more longitudinal data. As you know, we're guessing on specifically on lung cancer, high interest in squamous lung cancer as well as urothelial cancer. So that work is ongoing, but as we've said in the past, our central strategy here is around combinations. We are having quite a few other discussions on other combination mechanisms beyond also PRMT5. And so I think those will be hopefully activities we'll be able to bring across the finish line in the relatively near term to hopefully see what our broader combination strategy is in MTAP. So I think that's what you're going to continue to see from us is our focus on combinations is really the central theme to deliver the most benefit to patients.

And as far as the first combo study, are you focusing on specific tumor types? Or is this going to be more of a basket study initially?

For the PRMT5 asset you're going.

Yes, yes.

Yes. I think here, front and center for us is lung cancer. I mean we believe the indication that the greatest opportunity lies for this mechanistic combination is in lung cancer. So that's really #1 by fairly long margin. So I think here, we'd like to remain focused. We do believe there should be an opportunity to see responses across multiple tumor types. But I would say at the outset, we want to be focused as much as possible, obviously, be as capital efficient as possible, and we think they're the best place to focus will be in lung cancer.

Okay. And so that study starts this half?

So the objective for our PRMT5 in the clinic is in the middle of this year, and we hope to be able to enable the combo in the second half. And we've kind of gone through this, including with GSK, with Pol Theta fairly recently. At least our perspective is that we can do a fairly integrated escalation where you clear 1 or 2 cohorts as monotherapy and then starting to sort of seamlessly integrate the -- sorry, the IDE397 into the combo as part of the escalation. We have experience with that with other programs that we've been doing in partnership. So that would be the plan here. Our expectation also is we should see activity very early in that combination escalation. And then second, we know exactly what dose to start with, with IDE397. So I think we should have, at least based on that, a very truncated process prior to hopefully going into expansion quickly.

And to state the obvious, your PRMT5 is wholly owned. It's yours, so you control it, which is a key advantage. But is there anything else in terms of the molecule itself that may provide some additional benefit versus your former partner or is it really more just about the control of the molecule?

Yes. I think there are sort of 3 pieces for us in terms of TPP, Yigal. So first is having that balance between potency and selectivity, I would see there. We think our compound is probably most similar to the BMS compound. Second, we do think this candidate has excellent DMPK properties. So at least our anticipation is that we shouldn't have to have a very large pill burden, at least that's what we anticipate not to have a large pill burden in patients in the clinic. And then finally, third, as you know, we've communicated this in the past. Our perspective is that for this target, in particular, you do not want to have a brain penetrant molecule, and so as I'm sure you know, that's what also -- I guess, similar to BMS here as well. So we think for this combination, that's really the ideal profile.