Hey, good afternoon, everyone. Thanks for joining our healthcare conference. Next presenting company, which is actually going to do a fireside with me, is IDEAYA Biosciences, and we're joined by CSO Mike White and CMO Darrin Beaupre. Sorry, Darrin.
No, that's good.
I try, yeah. Okay, so for me, I mean, I've covered you guys now for like three years. You are a standout mid-cap oncology company, great science, good drugs. You know, stock really was a darling, I think, into mid-2024, and then it's been a bit bumpy since, and I'd argue probably most of that's just catalyst-driven. But maybe we could just lead off by what do you think is going to get people reinvigorated into the story this year?
Yeah, ma'am. I mean, maybe first I can just jump in and give you a little bit about what we're excited about. First, by the way, thank you for inviting us to speak with you today. We're really delighted to have the opportunity to talk about our programs. And as you know, IDEAYA is a precision medicine oncology company focused on the discovery and development of first-in-class targeted therapeutics for mechanistically defined patient populations selected using molecular diagnostics. That's a theme that runs through all of our programs, and I think that's important for success. And to do that, we've installed a fully integrated target biomarker drug discovery and translational capability platform, which has been important for our productivity.
I would say a central pillar for discovery is an IND engine that's really fueled by structure-based drug design that's dovetailed with a research team that defines target drug and disease mechanism of action across biological length scales from atomic resolution through to molecular networks, cellular subsystems, tumor host interactions, patients, and patient populations. I just want to put that out there because, in short, we're combining good chemistry, good structural biology, good cancer biology, and a judicious focus on battle-tested computational chemistry and computational biology approaches to compress time from target ID to clinical trial. We're very excited about that. We think the investor community should be excited about that. We're targeting three INDs in 2025 alone, which will bring us up to eight clinical programs this year.
And then lastly, before I turn this over to Darrin, I want to add that we recognize as a company tumor heterogeneity as the beast that limits success in our domain at large. And that's because baseline molecular differences between tumor subclones limit response rates, adaptive plasticity limits duration of response. And IDEAYA has a strategic focus to confront tumor heterogeneity with the key targets and drug combinations required to deliver durable therapy. As examples, Darrin is going to talk to you about the darovasertib/ crizotinib combo in metastatic uveal melanoma. Also, our MAT2A inhibitor IDE397 in combination with Trodelvy with our partner Gilead in MTAP tumors. MAT2A also in collaboration with Amgen with the PRMT5 inhibitors in MTAP tumors. Our PARG inhibitor IDE161 in combination with Keytruda with our partner Merck. Also, ADC combinations with that PARG inhibitor that are coming up soon.
Our POLQ inhibitor IDE705 in combination with PARP inhibitors with our partner GSK. And then also, of course, our Werner inhibitor IDE275 in combination with the dostarlimab also with GSK. Darrin, anything you want to add to that?
No, I'd just say from a clinical development point of view, we're obviously a maturing company. As you can see now, we have molecules in all phases of clinical development. And as Mike pointed out, we have a pipeline emerging in the early oncology space with several INDs coming out this year. We have mid-stage development with 397 and 161. And driving forward with the darovasertib in late-phase development with our first registration trial getting ready to read out later this year, a second one getting ready to launch, and a third one in the adjuvant setting being considered. So we cover all phases of clinical development now. We cover both small molecules and ADCs. We have an exciting first-in-class pipeline, and I think we're in great position to succeed. And again, some key milestones, as Mike pointed out, coming out this year.
We'll have more data coming out for the darovasertib. As I mentioned, the registration trial will read out. We'll talk to folks about what we've seen with respect to overall survival in our phase 2 trial. Haven't presented that before, but you'll see why we remain very, very bullish on the likelihood of success in the metastatic setting. Really, another key exciting feature that you're going to see this year is the data from the neoadjuvant study where you've seen some of this data before where we've shown that the darovasertib is effective in preventing enucleation in patients. But what we haven't really shown you much detail is around what happens with plaque brachytherapy patients with respect to tumor shrinkage and radiation reduction and how that translates into vision preservation.
So I think when you see the data this year, you'll understand why we continue to remain extremely bullish on that program. And again, you know, a number of molecules going into IND enabling. We have a number of combinations that are exciting. So from an oncology company at this stage of its growth, you know, to have this many great shots on goal, you can't really be positioned any better than this when it comes to being a medical oncologist, just put it that way.
Yeah, yeah, I think I'd agree. I mean, the pipeline is obviously stacked, but it seems like the common thread is this trying to leverage against tumor heterogeneity. I mean, I think that part, Mike, what you told me resonates a lot, right? Like we have adding crizotinib to daro to target like the liver metastases that are very common in uveal. We've got adding something to MAT2A, whether it's PRMT5 or whether it's an ADC such as Trodelvy to hit different kinds of cancers. We've got, you know, you've recently in-licensed two different ADCs. You're thinking about combining them with PARGs. Is it really just kind of trying to go after what the biology is telling you? I mean, is that really the main message of the company here? Like the.
Yeah, thanks for summing that up so nicely, Matt. I mean, that's also why I wanted to point out how we've installed these fully integrated discovery capabilities exactly for those reasons. For us, it's really about mechanism of action and writing mechanism of action into the clinic for successful trials, 100%.
Let's get into Darrova, as it is still the most advanced program. At least in the metastatic setting, I think you're anticipating a median PFS readout from the registrational trial. We'll probably get an OS readout from the phase 2 trial that's been ongoing, and we'll probably get an update, I think, Darrin, as you mentioned, from the neoadjuvant setting. I guess for someone new to the story, can you just maybe contextualize the importance of those updates and how they fit into the overall strategy?
Yeah, I think the key people need to recognize here is that, you know, if you go back to the roundtable discussion we had with people like Carol Shields and Amy Scheffler, I mean, the idea here is we're going after the driver of the disease, which is protein kinase C. You know, those mutations that drive protein kinase C activity are the harbinger of the disease. And so when you have a drug that gets after that, you obviously can imagine you're going to have a significant impact. And I think what people need to understand is that by doing so, we have the ability to impact the entire patient journey of patients with uveal melanoma from the primary setting all the way to the metastatic setting. And so, as I mentioned, we have the registration trial in HLA-A2 negative patients with metastatic uveal melanoma.
That's the lion's share of patients with uveal melanoma. Based on the data that we've accumulated, about 70% of patients are HLA-A2 negative, so we'll get after that patient population, but also in our phase two trial, we're testing patients who are HLA-A2 positive, and we've already presented data at ESMO in an oral presentation to show that those patients behave no differently than HLA-A2 negatives, so we're going to capture that group likely with the combination of darovasertib and crizotinib. Now, that's relevant because, you know, the patient population there is maybe 4,000-5,000 patients per year, but, you know, in the neoadjuvant setting, we're talking about two to three times larger patient population where darovasertib, even as a single agent, is highly active, and there it doesn't matter what your HLA status is.
We can get after patients, get after their tumors, save eyes, save vision. That's what we're getting ready to launch is our registration trial. So the opportunity between looking at what Kimmtrak is doing with a sliver of the patients, you know, their annual revenue looks to be like about $300 million a year on 30% of the patient population just in the metastatic setting. So that's why we really believe we have a blockbuster opportunity. We're not even talking about what we want to do in the adjuvant setting. We're still considering an opportunity there. So overall, really, we can cover the landscape in uveal melanoma. There shouldn't be a single patient that doesn't see either darovasertib or the combination at some point in their journey with this disease. And with that, we should capture a large fraction of patients.
I think, you know, not only will patients enjoy that, but I think our investors will be very happy what we come out with at the end of the day.
Yeah, for sure. Do you think there's a path for the HLA positive subset in metastatic? I mean, is it possible maybe to just get on NCCN guidelines given how good the data is probably going to be?
That's sort of the default, the baseline case. But what we're also doing is we're collecting real-world data for patients who are HLA-A2 positive. So that way we can take that data set of HLA-A2 positive patients who are treated with daro- crizotinib and compare them to a real-world data set. And, you know, who knows, we may be able to have a discussion with the FDA at the time of the filing to say, "Hey, listen, you know, we have a patient population that's HLA-A2 positive. We have this real-world data set. The data looks very comparable between these two patient populations. Let's have a discussion. We can see how that goes.
I'll start quickly, Darrin, maybe about how you're sizing the markets as we think about adding neoadjuvant. I mean, is this like a double the size or triple the size? Can you kind of put that into context as we think about this pivotal trial?
Yeah, I mean, we're thinking potentially two to three times the size patient population in that setting. And again, even patients with the smaller tumors that maybe are going through a phase of observation, you know, this may be relevant also. We're actually looking to amend our phase 2 trial to start looking at patients with small tumors, those that are typically three or four millimeters in size or smaller. And so we'll be looking at those as well to gather some data so we have an understanding of, you know, what kind of effect we're going to have in that situation.
I think there is still some confusion about who exactly you're targeting for the neoadjuvant trial, right? I mean, you're using a Castle assay base. It's like a genotype assay, right? And that's kind of how you classify intermediate to high, and that's where you're going at. Because there is still some confusion out there about, are you just looking at size or are you inclusive of both small, medium, and large? Or is it really just about the Castle assay tells you this is a patient that's probably likely to progress, therefore we're going to enroll them?
It's actually twofold, right? Because there's two things that we're trying to accomplish, right? Number one is vision preservation, right? So if you want to save the eye, you got to start with a large tumor, right? Because only large tumors are the ones that get enucleated. Then secondly, for the plaque brachytherapy group, you know, even there, you're talking about preserving vision. So if the tumor is very small to begin with, they may not have a big vision problem, right? So we do have size restrictions on the tumors that we are treating, but it covers the vast majority of patients with uveal melanoma. That's the first thing. Then with the risk category that you mentioned, you know, we are looking at event-free survival as a key output, right?
We want to make sure that by delaying the primary treatment of uveal melanoma, we're not causing a detriment to patients, right? Increasing their risk of metastases. So if we want to do a study that's timely, we have to take patients at high risk. And that's where the Castle gene expression profile test comes in. We're taking class two, but not just class two patients. We're taking patients who might have monosomy 3, who also do very, very poorly, and also patients who are stage 3. For those folks who don't do routine testing like GEP or cytogenetics, if not available and you're a stage 3 patient, you'd still be eligible. So there's two different facets there. One is let's get an endpoint that can be seen in a relatively short period of time. So that's high-risk patients. So you have to select those.
But then also we do have to have the size of tumor that could potentially cause vision problems. Otherwise, you can't see an improvement.
So just to confirm, the primary endpoint for the neoadjuvant in the brachy cohort, it's time to vision loss. And in the enucleation cohort, i.e., the larger size or higher risk factor, it's time to enucleation. But then in both cohorts, you kind of have this no detriment EFS as, is it a co-primary or is it a secondary?
No, it's a secondary. EFS is secondary. The first you kind of have, right? For the enucleation group, it's basically save the eye. It's as simple as that. What percentage of patients can you save the eye? So that's the primary. And then I would say for the plaque brachytherapy group, I would call it vision preservation. Typically, we use best corrected visual acuity of 15 letters as sort of the backbone of what people have done in this setting. But we are still discussing with the FDA macular edema and some other facets that we think can make that endpoint more effective.
Yeah, another question I get a lot from investors is just on timelines. How long do you think it would take to kind of get enough data around the no detriment EFS secondary for the FDA to be happy with it and effectively approve the drug?
Yeah, well, I mean, the way things are lining up, it looks like, you know, between the time of enrollment and the time of RIDA, we're looking at maybe two, two and a half years down the road to be maybe in a position to be able to provide that data to the FDA. So we'll have enucleation data by then. We'll have data on, you know, vision preservation by then, and we'll also have EFS data. That would be the idea.
Got it. Okay. That makes sense. Cool. Let's go to IDE397. So that's the MAT2A inhibitor for MTAP-deleted cancers. It's obviously been a very active pathway for years now. The target MAT2A inhibitors in the past have been plagued by either liver tox. You've seen none of that. Frankly, I'd argue that your MAT2A looks better than any of the other PRMT5s out there. This is kind of the analogous pathway that is downstream of it. So my first question is, is there routes for monotherapy approval here, or do you really just think combinations are the way to go? And if it's combinations, what do you think is the ideal partner?
I'd say that, you know, we're still working on monotherapy, looking at monotherapy activity with 397. We still have data that needs to mature with respect to durability. But suffice it to say, we are all in on combinations being the way to go with respect to having longer, more durable, more effective therapies. There's no doubt that maybe if we get a fast track in with monotherapy, that's nice. But really, the heavy lifting is going to be done with a combination. And that's why, you know, we have the Trodelvy combination. We have our own PRMT5 inhibitor coming because we're all in there. We have additional things in our pipeline coming to address this. And so we're about as all in as you can be on 397, but combinations is our leading paradigm here.
Yeah, I can add to that a little bit, Matt. I mean, why combinations? It's not because we can't get responses with single agent. It's because we need durability. And as you know, even the first generation PRMT5 inhibitors, which are kind of like chemo, had acquired resistance mechanisms where the pathway is still perturbed, but the tumors don't care anymore. We've seen that mechanistically in preclinical models with PRMT5 inhibitors, with MTAP cooperative PRMT5 inhibitors, that you get regrowth on tumors and you're still fully engaging that target, fully inhibiting the pathway. The tumors don't care anymore. And MAT2A inhibitor intercepts that bypass mechanism. And PRMT5 inhibitors can do the same for MAT2A. So that combination is really giving us durable responses preclinically. And we know the mechanisms that are providing those responses, and we expect that to translate very, very strongly in the clinic.
So we're really excited about having really a nice duration of response in a combination setting.
Yeah, yeah, that makes sense. A lot of people ask, you know, you've got the clinical trial agreement with Amgen to study your MAT2A with their PRMT5. The preclinical data on that combination is some of the best that I've ever seen when it was presented at ACR a couple of years ago. That trial has been enrolling now in dose escalation. I think we'll probably see some of it this year, at least that's the hope. But then last year, late last year, you've also announced that you're developing, or at least have intentions to develop your own PRMT5. So kind of like what's kind of the thinking behind that? It seems like you're really trying to force Amgen's hand here. I'm sure it's strategic.
you know, I would just say this, Matt, you know, like I said, we're very, very bullish. We believe the combination is going to be outstanding if you're in the right patient population with the right therapeutic doses of those two, one combined. long and short of it is we want to have the opportunity to control our own destiny. We have big plans. Sometimes when you have partners, maybe you don't see things exactly the same way. you know, we have a few indications that we want to get after. We want to get after them fast and soon. we think we can do that best when we're, you know, in charge of our own destiny, to be quite honest.
Yeah, I think that's a great way to put it. We're not trying to force anybody's hand per se. We just want to make sure this combination gets to patients. Our internal program, IND filing targeted middle of this year, is really a bespoke program that was designed expressly for combination with IDE397. We know a lot about this pathway. We know a lot about the properties of a molecule that would combine with 397 in the most effective way. And if we made that molecule, we have that molecule in play, and Darrin can develop it exactly as he described if and when the time comes.
Yeah. Some of the pushback I get on the Amgen drug is the PK properties and the fact that they need really high doses. Is that something that you're trying to work around?
Do you think you can work around that with good med chem or with this pathway, or are we just going to need to have higher doses?
Let me put it like this. The internal program that we have is capitalizing on the fact that the exposure profiles are well understood from a mechanistic standpoint. And we believe we can get very, very exceptional exposure profiles with this molecule together with IDE397 without a dose burden, without a pill burden. We don't think we're going to be limited in that fashion. And that is something that was pretty straightforward to achieve objectively, given that our endpoint was in the combination setting with IDE397.
Gotcha. Do you think we'll get any of the Trodelvy combo data this year?
That's kind of the idea. The idea is, you know, we're getting ready to close to heading out into the expansion. You know, by later this year, our hope is we'll have, you know, enough data accumulated, we'll be able to talk about some of that.
Cool. I might ask Gilead on their earnings call about that tonight. Cool. Maybe switching to the ADC programs, two of which have been recently in-licensed. You've got a DLL3 that has quite a bit of clinical data to support it. And then another B7-H3 x PTK7. PTK7 is another target that I'm getting a lot more inbound interest on recently. I guess maybe, Mike, from a high-level strategy perspective, what was the rationale here behind bringing these two in?
Yeah, great question. So let me say that, you know, first of all, with respect to these ADCs, one of the key components here is, again, coming back to tumor heterogeneity and dealing with that with the effective targets and the effective combinations. The DLL3 asset, you know, Darrin can give you more on this. Very excited with respect to the single agent activity here. That's a fantastic target, DLL3. This is an antigen that is driven by a transcription factor, ASCL1, which is a lineage-specific survival oncogene. So this thing is a transcription factor that specifies small cell lung cancer cell fate, neuroendocrine cell fate in lung cancers. It's a transcription factor that can't be lost in a tumor or it's going to kill the tumor.
So we're pretty excited about the fact that homogeneity of antigen expression should be good for DLL3, as well as persistence of expression of this antigen should be good. We don't expect turning off the transcription factor to lead to a resistance mechanism because the transcription factor is required for those tumor cells to survive. So that's really key for us. And the second piece is, you know, this battle-tested linker payload, which is released intratumorally, and it's a topoisomerase inhibitor. And that plays very, very well with respect to not only single agent activity for response rate, but also what we're really going for here is durability of response. And so we have a shot on goal there for the combination with IDE161, given the very spectacular synergistic mechanisms that we see with topoisomerase inhibitors and IDE161.
And this intratumoral release with this battle-tested linker payload is very tight. It doesn't release extratumorally. It doesn't give you systemic exposure of that payload. It is exactly what we need to have the best optionality for an ADC combination with IDE161. The B7-H3 PTK7 ADC, we're also very excited about. That's a bispecific. Bispecifics are amazing because they really optimize tumor-specific delivery. Our asset, we chose because it's binding to the tumor and its internalization requires double positivity. You need B7-H3 expression and PTK7 expression. And in those double positive cells, you get enhanced internalization of the ADC. This is a DAR8, as is our DLL3. It's got a very nice linker that's been optimized for hydrophilicity so that the ADC has the same PK properties as a naked antibody. And again, it's a very, very tight linker that does not release extratumorally. It releases intratumorally.
It's really going to be able to be deployed with IDE161 because the big limitation for bispecific antibodies is you get more specific delivery to the tumor, but you also are paying a price, and that is the heterogeneity of the antigen expression may give you suboptimal payload delivery within the tumor or across tumors. That's where IDE161 comes in because even if you just get a little bit of that topoisomerase inhibitor into those tumors, we magnify the effect and turn that into an unresolvable lesion with IDE161 to kill those cells. We're pretty excited about that asset. We're also excited about the fact that PTK7, as you might be learning, Matt, from some of your own research, is something that is expressed highly on tumor-initiating cells. That is the cell population that drives adaptive resistance to therapy.
So let's attack that tree at the roots with respect to being able to eliminate those cells right off the bat. So that'll give you the high level there about why we're excited about both of these programs.
I'm curious kind of why the PARG inhibitor is the best partner in terms of like an SL sense for pairing with the topo-based toxin. I mean, you also have a Pol Theta, like Pol Theta and important in translesion synthesis. So like what is it about PARG that's a good target for combinations?
Yeah, I love that question, so there's been a lot of spectacular preclinical work done at the NCI that has been repeated here and that we've extended that really shines a light on that mechanism, and so as many folks on the call know, a topoisomerase inhibitor works by binding to that topoisomerase and causing a covalent link between the enzyme and the DNA, a protein-DNA conjugate. That thing gets recognized by PARP, and it gets PARylated. That PARylation recruits DNA repair machinery, but that DNA repair machinery cannot work until the protein is chewed up and gotten out of the way. That's done by the proteasome, but the proteasome does not have access to that topoisomerase that's glued to the DNA unless it is dePARylated first, and there's only one enzyme that does that, PARG.
So if you put a PARG inhibitor on board, you basically have a sugar-coated boulder sitting on the railroad tracks, a replication fork comes through, and you have mitotic catastrophe. That's a very special mechanism. You don't get that with a PARP inhibitor. You don't get that with a Pol Theta inhibitor. You really magnify the consequence of topoisomerase inhibitors. So it's a spectacular direct synergistic relationship there based on those mechanisms.
Do you think in that case, it's better to use these agents more in upfront settings versus someone that, I don't know, has like chemo resistance where basically like mitotic catastrophe is and they'll just blow right through it? It doesn't matter.
Yeah, that's a really insightful question, Matt. And that's another reason why I'm excited about this mechanism of action because, you know, blowing through arrested replication forks is something that can happen, you know, quite readily with a PARP inhibitor, for example. The lesions that we're talking about here, it's very, very difficult for you to get template switching. It's very, very difficult for you to be able to bypass that lesion and leave behind a single-stranded gap or a single-stranded break because you've got a DNA protein conjugate. And those things are still lethal in the context of cells that can deal with a lot of other kinds of DNA damage and a lot of other kinds of structural variation that's occurring continuously through the mitotic cycles.
Got it. That'll be pretty cool. Where is kind of the PARG in clinical development right now? And how soon can you start to pair the two together?
Yeah, so we're still doing some dose optimization with the 161 program. As you know, we've started our combination with our checkpoint inhibitor, Keytruda for endometrial cancers. That's ongoing, and then with respect to your question, we plan to use a tactic for both of those studies where we can dose escalate, you know, the ADCs as monotherapy and during the dose escalation as we get to higher levels where we think we may be in a therapeutic range, then we can tag on a dose escalation, adding in 161 to seamlessly bring those two together and then expand out both a monotherapy arm and a combination arm to get a good sense of what we're adding when we add 161 to the ADC.
I want to quickly just ask about catalysts for the different ADC. Well, I guess it's really DLL3 at this point. We know that Trodelvy might also come this year, but just in terms of updates on the DLL3 program, and then I'll quickly get to Werner before we wrap up.
Not crazy to think there's some potential there because, as you know, we'll be starting in the U.S. in the, you know, first half of this year. And that way, you know, we'll be getting into, we'll be starting nearly at a therapeutic dose or at a therapeutic dose from the beginning. And so therefore, you might imagine over a six-month period, we may accumulate enough data to start talking about. So there's potential there because, you know, our expectation is with the doses that we'll be starting nearly from the start, in fact, from the start, we should have anti-tumor activity, and there should be enough patients enrolled by the end of the year that maybe there'll be something to talk about.
And then with Werner, I mean, this is kind of being driven by GSK, right? But it sounds like you guys jointly plan to present some data. Is it clinical data at this point?
We're going to, so the data that's coming up soon, Matt, that's going to be on the molecule itself. There's going to be some really key disclosures very, very soon at an international conference. I think that's going to be exciting data in and around why we think we have a Werner's inhibitor that is special, that's important with respect to its ability to lock Werner's into a conformation that is distinct from those that are in the clinic now. And that's super important because MSI-high cancers, that's a high mutation rate. And one of the things that keeps everybody up at night in those tumors is their ability to adapt to therapy by making mutations in the target where the drug no longer binds.
And our Werner's inhibitor, and I believe some of this will be disclosed publicly soon, was expressly designed to be able to lock Werner's into conformations that are distinct from those where we have seen resistance develop against other inhibitors. So that's pretty exciting. And then on top of that, you know, with our partner GSK, I think we're really well positioned to have another combination opportunity here because the dostarlimab is doing spectacularly well in endometrial cancer, also in colorectal cancer. Both of these things are hitting MSI-high tumors, putting them together. We think something special will happen there too.
Yeah. Got it. It's definitely a lot to get excited about. Fortunately, we're out of time. So guys, thanks so much as always. I thought this was super insightful. I learned a lot.
Thanks, Matt. Sorry we didn't get to KAT6. Maybe next time.
Yeah, next time. There's just too much to talk about.
Too much to cover.
All right, guys.
Take care.