I'm Ben Burnett, biotech analyst at Stifel. Happy to host the next session with IDEAYA Biosciences. Pleased to have Yujiro Hata, President and CEO. Yujiro, thank you.
Thanks so much, Ben, for the kind introduction, and thank you to Stifel for the ability to participate this year. We look forward to the discussion.
Great. As always, if you have any questions for those on the line, email me, burnettb@stifel.com. I'll try to work those in the conversation, or just submit those through the chat. I have that up as well. So, Yujiro, maybe to start off, do you want to just give a quick high-level overview of IDEAYA?
Sure. So IDEAYA Biosciences was founded just about nine years ago. We're a leading precision medicine oncology company. Today we have a significant focus in an emerging area called synthetic lethality. We have four first-in-class programs in the clinic. Our most advanced is darovasertib. Here we're in a first-line potential registrational trial in metastatic uveal melanoma. We've also launched into a phase II neoadjuvant and adjuvant trial as well in uveal melanoma. Here, Ben, as you're well aware, we're guiding towards two phase II efficacy readouts in the middle of the year. First will be from our phase II IST. Second will be from our phase II company-sponsored trial.
In addition, for this program, we're guiding towards getting FDA regulatory feedback, which we anticipate will be a Type C meeting, to get endorsement around what we believe has the potential to be a frontline accelerated approval study in the neoadjuvant setting. The next program is in the MTAP deletion space here, very relevant and a lot of interest from the investor community 'cause the patient selection biomarker is very large, has been reported at roughly 15% of all solid tumors. Here, our central strategy continues to be around rational combinations, including with Amgen's PRMT5 inhibitor, AMG 193. Here we are continuing in the dose escalation. We are guiding towards a joint publication strategy, a communication of that, this year.
In addition, we have a second combination we've launched with Gilead, with Trodelvy, which is a Trop-2 ADC, targeting MTAP deletion bladder cancer. In addition, we do continue to interrogate the monotherapy side for IDE397, and Ben, perhaps, will talk about this. We have been quiet on this activity, but we have been generating what we think is very important data that may inform further a path forward on monotherapy as well. In particular, we've been focused on certain tumor types like bladder cancer and squamous lung cancer. Our third program in the clinic is IDE161, another first-in-class program, hitting a target called PARG, P-A-R-G. This is in the DNA damage repair area, where we're focused on HRD solid tumors. I would say significant interest in tumor types like endometrial and colorectal cancer.
We did recently announce a partnership with Merck to do a combination with KEYTRUDA in the endometrial cancer setting. We are having several pharma conversations on this program to enable additional combinations, so stay tuned for that. Pol Theta is our fourth first-in-class program in the clinic, also with HRD solid tumors here. A significant focus on combination with PARP inhibitors. And then, Ben, as you're aware, we are very much in line to be able to hopefully deliver our fifth first-in-class program in the clinic with Werner helicase. We have multiple more candidates we're targeting by the end of the year, so we anticipate 6-7+, hopefully, first-in-class programs in the clinic here shortly. We do have about $1 billion in cash on the balance sheet.
We have a runway into 2028, so we're in a strong financial position.
Okay, fantastic. Lots to unpack in the next 25 minutes.
Yep.
I want to start with darovasertib in the neoadjuvant opportunity. Maybe talk about the update that you plan to give this year, and I guess what would equate as success when, you know, kind of thinking about the enucleation cohort versus the radiation cohort?
Yeah. So in terms of what to expect, I mean, we anticipate it's going to be a fairly definitive data set in terms of the number of patients. So from our Phase II IST, we anticipate that's going to be 12-15 patients. Our study, we anticipate at least 30 patients. So in aggregate, you know, you're talking 45, 50+ patients, what we believe should be, at a move forward dose, for darovasertib in the neoadjuvant setting. And again, here, we are focused on monotherapy. In terms of measure for success, so here I would say, you know, I would look for several, key, aspects in terms of the Phase II readouts. First is, what does the waterfall look like?
You know, I think here, one of the discussions we're contemplating as we prepare for this potential meeting with the FDA is around a response rate type endpoint as well. Obviously there, the waterfall will be critical. You know, what are the depth of the responses? What's the consistency of tumor shrinkage that we see? That's first. Second, for patients that we're treating that are enucleation patients, so these are patients that otherwise would be getting their eye removed, which is going to be about 20%-25% the patients in the neoadjuvant setting. There, we are also not just looking at tumor shrinkage, but also what percent of these patients' eyes can we save?
At least based on the feedback we've heard from KOLs, we've been told that even a 10% eye preservation rate here could be a game changer. At least based on the early data that we shared publicly, we're just well north of that number. So now the question will be, with more patients, you know, how does that data look, both from a waterfall perspective as well as eye preservation? And of course, safety is going to be, as you know, Ben, you know, paramount anytime you're in the pre-metastatic setting. So how is monotherapy safety holding up? All we can say to date is that, as a single agent, this drug has been generally very well-tolerated.
You know, we've had patients in the metastatic setting that have been on therapy well beyond two and a half years. We even had a complete CR in Europe that was on therapy, to out past five years, with really, very minimal issues.
For the radiation cohort, is it, I mean, or is it too soon to maybe try to look for any kind of vision preservation at this point?
Yeah, great question. I would say right now, it's probably early. I think for those plaque-eligible patients, our primary focus has been around, you know, what's the level of robust tumor shrinkage, you know, are we seeing in that subgroup? So that's been the primary focus. I think from there, you know, ultimately, you would want to continue to track those patients, and then when they get the primary procedure, you know, they'll either get a plaque or maybe not even get a plaque. But then, you know, then you would measure radiation levels, and then ultimately, vision, some type of vision output.
Yeah. Okay. What is your expectation? I know you're going to meet with the FDA and get some clarity on this, but what is maybe a plausible kind of regulatory path? And is it fair to assume that they may be different for those two different cohorts?
Yeah. So first is, yes, we do think they'll likely be different. And, you know, first, I just want to highlight that we are still in the thinking process and discussion process internally. So, you know, kind of what I tell you, what I'll tell you is more high level, and, you know, nothing has been specifically decided. But, I would say high level, the plan is to go to the FDA with two studies. One would be what we believe should be a single-arm, accelerated approval study for these large ocular tumors. And so here as well, you know, Ben, I think one of the concepts we have, as opposed to just focusing just on enucleation, can we, can we categorize this more broadly than just patients that have larger ocular tumors?
And there, you know, what the focus would be is not just eye preservation. I know, Ben, we talked about that before, but can we use some kind of response rate criteria as the primary endpoint, which would then allow us to capture these broader goal of patients? And then a secondary endpoint for these enucleation patients, eye preservation is a secondary endpoint, right? And I think that was based on some of the feedback we got, saying: "Hey, you know, why not be more aggressive? Let's get more of these patients, and that would allow you to do that." And then the full approval studies, where we would try to capture the remaining, every, you know, small, the smallest of the small tumors, medium-sized tumors. And here, the plan would likely be to randomize against plaque.
And then kind of what you just brought up earlier, look at endpoints such as response rate criteria, the same there, in addition to radiation reduction. And I think ultimately, we think we'll also need to have a vision preservation or some vision score readout as well, and do that as in a randomized fashion. So that would be the. That's sort of the current high-level thinking. And our plan right now would be to take both trials, both accelerated approval and the full approval plan at the same time.
Okay, excellent. Is there any... Maybe one last question here, any thought process around relapse-free survival, and if eventually that would have to be some data that would need to be accrued?
Yeah, I'm happy you asked that question. You know, our motivation here in adjuvant is extremely high, and I would say the amount of enthusiasm for our investigators for us to get an adjuvant trial going is also extremely high. So I think our current thought process now is, you know, let's have the discussion on neoadjuvant. We have a growing dataset there. Hopefully get that endorsement, be able to launch into what we believe should be a registrational trial for that, and then come back to the FDA, perhaps later in the year, around a registrational adjuvant trial. Here, we would anticipate that should be a randomized study against placebo because there's nothing approved. Here, just important to note, this would be irrespective of HLA-A2 status, so we think that's also a very important part of this.
I think we're still thinking through the dosing schema. So, you know, at the end, ultimately, you're trying to prevent relapse, right? So, you know, should we start with combination in this setting, and then, let's say, transition to mono? But, I think those discussions are still ongoing. So yeah, so absolutely, very high priority. We're very excited about adjuvant, and it's very much in our plans to get that going as well. There was a session, I don't know if you saw it then, at AACR. I think I was not there, but, I think an FDA-sponsored discussion around neoadjuvant adjuvant, and at least what I heard, was also their desire to see some of those trials actually separated. So that actually fits well.
Mm-hmm
... with what our current strategy is as well.
Okay, that's great. Maybe just one other question on kind of the neoadjuvant piece of this. What is sort of the commercial opportunity? And I think one variable that we have in our model that's a little bit unknown at this point is kind of what the duration of therapy could be. So maybe could you kind of talk around sort of expectations for duration of use?
Yeah, definitely. So I mean, I would say high level here, Ben, I mean, we firmly believe the neoadjuvant indication alone is a bona fide blockbuster indication expansion opportunity, so it's significant. And I think there are several ways you get there, and there's really three parameters we've been focused on, and one is around duration. So one is just purely annual incidence, right? So as you know, in the pre-metastatic neoadjuvant setting versus metastatic, you know, where metastatic has been reported about four to five thousand patients globally, that number doubles in the neoadjuvant setting. Second, key piece is around prevalence versus incidence. And I think this is important because when you look at prevalence, right, the bump from incidence to prevalence is not that great in the metastatic setting because survival is so poor, right?
So these patients, unfortunately, are passing away. In the pre-metastatic setting, that's actually not the case. So you have a lot of these patients on annual incidences that are starting to stack on top of each other. And the reason why this is relevant is because in the neoadjuvant and adjuvant setting, there's nothing approved.... Right? So just because there's nothing approved, they're not on another treatment, and so you should be able to capture that portion of prevalence. So that is a very important part of this equation. We think the pre-metastatic patient population, in terms of prevalence, is about 100,000 patients. And then the last piece is around duration. And so now our phase II protocol, we're treating up to six months.
We are having discussions currently about amending that protocol to enable patients to get treated to a year, and then in certain cases, to even extend beyond a year. And this has frankly been something that's been also very much discussed with our investigators. They're very enthusiastic about us implementing that, a protocol amendment. So, you know, look for that to happen, you know, fairly in the near-term basis.
Oh, that's great. Okay, and just to reiterate, you said the prevalence is something in the order of 100,000 patients, up from kind of 10,000-ish patients in the... Yep.
Correct, correct.
Excellent. Cool. Okay, good. That's, that's super helpful. Okay, I don't see any other questions in the chat on this, so I want to move to darovasertib plus crizotinib combination data that you have [audio destortion] you have in the frontline metastatic uveal melanoma setting. Maybe just very quickly, kind of, can you give us just a high-level kind of overview of the clinical data that you've generated? And then we'll, we'll jump into more details after.
Yeah. No, definitely. So I would say probably the key data set here, Ben, was our ESMO oral presentation, fall of last year. Here, we reported, you know, 45-some odd patients in the frontline setting. And kind of the bigger picture is we saw a confirmed overall response rate by RECIST of, you know, just over 40%. And the HLA-A2 positive setting was a smaller data set, but we saw a response rate that was in that 60% range. So by every measure, you know, we're just so, you know, much higher than what's been reported before from a response rate perspective.
As you know, historical response rates have been from zero to the high end of 10% confirmed ORR, and that includes, I mean, of course, Tebi as well as what, you know, what else is in what we anticipate will be in our comparator arm. Importantly, we also did report PFS. Obviously, it's a single-arm design. However, I think notable in that, in the frontline setting, we saw a median PFS that's above seven months. We think that's significant. Historical PFS, it has been reported in the two to three month range. So right now, you know, at least our expectation is that, you know, we should hopefully have the ability to, you know, at least double historical PFS.
Then last year, as you know, Ben, you know, we have reported, you know, a two year PFS Kaplan-Meier, which I think is quite also encouraging since that's our primary endpoint. And there we see just clear separation versus some of these other historical studies I just mentioned. And importantly, about a quarter of patients we're seeing that are progression-free even beyond two years. So I think that's another point to your model on the forecasting side. You know, obviously, long tails can have a significant impact to also commercial opportunity.
That's great. Okay. And then, so you're comparing darovasertib plus crizotinib versus standard of care, physician's choice. What do you expect to be included in that standard of care? And what are your efficacy expectations for that arm?
Yeah. So efficacy expectation, I'll start with that first. Unfortunately, Ben, are very poor, right? So response rates are in that, you know, we expect are going to be in that single-digit % range, at least confirm ORR. And PFS, we expect are going to be two to three months. There's been several readouts here, and it's been consistently in that range. In terms of what we expect in the comparator arm, it'll be slightly different in Europe versus U.S. In the U.S., we anticipate predominantly PD-1 plus CTLA-4. In Europe, we anticipate predominantly PD-1. We'll also see PD-1 alone in the U.S., also some DTIC, but we anticipate they're probably less than 10%. Sadly, you know, all of these agents have shown not to be effective in the metastatic uveal melanoma setting.
Okay. Yeah, that's helpful. So this study is in HLA-A2 negative patients. I think in your phase I, you included both negative but also some positive patients as well. I guess just thinking about, you know, is there an avenue for the NCCN guidelines to maybe include HLA-A2 positive patients, you know, should this get across the finish line and be approved ultimately?
Yeah, definitely. Yeah, I mean, the feedback that we've received is, you know, if in the scenario that we're able to get first-line accelerated approval in HLA-A2 negative, we're continuing to build the dataset. But as you mentioned, our existing phase II study, so this is not the registrational trial, but our earlier phase II study, we've made a decision to just focus enrollment for that trial in A2 positive. And actually, we increased that number target to 50 patients. And so in this scenario, we get approval in A2 negative, our plan would be to continue to publish that data in the A2 positive, as we've already done, right? So we've already presented it at ESMO, showed a 60% response rate.
And if we build that data set, you know, at least the feedback we've gotten is, you know, we should have a very, hopefully strong path forward, based on those key publications in a peer-reviewed setting to get on compendia for pricing reimbursement. Also, I would just remind, you know, folks that, remember, we're generating all of this data, irrespective of HLA-A2 status, also in the neoadjuvant, as well as in the adjuvant setting, right? And that should only further support, you know, the broad activity of this agent across, HLA-A2 status.
... Very cool. Okay, maybe last question here. Is enrollment, like, how is enrollment tracking in this, in the potential registrational study?
It's going great. You know, look, I, I think really a testament to the clinical team. I mean, they've just done an extraordinary job in execution. I also think, you know, the data speaks for itself, right? And I think that's one of the advantages of being able to present at ESMO an oral presentation, and which is also why our motivation to present, you know, two readouts this year in the neoadjuvant setting. And we've had the benefit of that in terms of enrollment. So what I can say now, Ben, on the registrational trial, I know we've been saying that we're kind of on schedule. We're actually now ahead of schedule on the first-line MUM trial, which is great. And, I mean, the neoadjuvant trial, as we had expected, is going at a just phenomenal clip of enrollment.
And I do think that's just really demonstrates this is, you know, high unmet need. That's what happens when you have nothing approved, and, you know, when you already show encouraging initial data, which we've had publicly, as you know. Small data set, but I think, you know, clearly encouraging to date.
Very cool. Okay, and maybe actually one last darovasertib question. You've shown some interesting data in cutaneous melanoma.
Mm-hmm.
So talk about that, and what is, I guess, the kind of the path there?
Yeah, no, no, great question. So yeah, I mean, really our focus there is all about enrollment. So, you know, we appreciate it's going to be a smaller percentage patient. So unlike uveal, where pretty much every patient-- we are enrolling every patient here, you know, it's going to be kind of a low- to mid-single-digit % of patients in skin melanoma. So it does require more work to enroll these patients. But we know that, you know, basically, we are seeing robust activity here. We think it will about double the metastatic population in terms of commercial opportunity. Ultimately, we want to build a robust data set. If we continue to see a strong signal, we'll likely try to pursue an accelerated approval path going forward as well.
Okay.
But I would say there, Ben, the main focus is really around execution of finding these patients.
Okay, good. Let's turn to the IDE397, the MAT2A program. I guess, kind of question number one, why, why add a PRMT5 inhibitor on top of a MAT2A inhibitor? They're on the same pathway. And then, two, you know, when, when can we expect an additional data disclosure?
Yeah. So look, from a mechanistic perspective, Ben, I mean, we think this combination trial with Amgen is the critical experiment. And, you know, as you know, we've continued to communicate that based on the strong mechanistic rationale. And there's really two key things we're trying to do, and I just want to make sure this is clear. We're not just trying to hit the PRMT5 pathway harder. That's just not what we're trying to do through this combination. First, is we know by inhibiting PRMT5 and MAT2A, the way these agents cause cancer cell death is specifically through a mechanism of RNA splicing defects.
We know that those RNA splicing signatures are different between PRMT5 and MAT2A, and when we bring these mechanisms together, we see true synergy, and I would say an impact on RNA splicing that's much greater than either mechanism alone. So that's the first. The second, which, you know, we don't have too much time here, Ben, to talk about, but I think you and I have talked about this in the past, which is really trying to influence the stoichiometry between MTA and SAM. And so there's been a lot of focus on elevated MTA and MTA, but as you know, as well as anybody here, it's not so simple, right? That there's variability of MTA.
We know there are scenarios where MTA can get pumped out, and MTA and SAM actually compete at the same binding pocket of PRMT5, where these second-generation PRMT inhibitors bind, and they have a similar KD. So it's not just about having elevated MTA, it's also about what are the levels of SAM. So we're trying to really deplete SAM in the elevated MTA setting and really impact the stoichiometry. And by doing that, our hope is we should be able to drive more responses, more durable responses, and hopefully go after a broader set of tumor types. So, dose escalation is going as planned. We continue to be very focused on this.
And maybe just, you know, kind of at the end, I would want to highlight to you, Ben, as I noted in the early portion of this, you know, we've continued to work on the monotherapy. We just want to be clear that our focus of our strategy continues to be combination. But as you know, we've been quietly also generating monotherapy data in very targeted tumor types. So, you know, don't be surprised if you hear more from us on that front, moving forward as well.
Would you... I guess, is it possible that as you kind of develop that monotherapy experience, that we could see that separately from any disclosure with Amgen and the combo side?
Yeah, that's being discussed, Ben, right now. I think we want to just, you know, be thoughtful that our, our central strategy continues to be combination. That's what we believe in, in this pathway. But with that said, we appreciate, you know, the monotherapy data has importance for a couple of different reasons, including contribution of components. Obviously, we think would increase, should naturally increase the probability of success of the combination, more monotherapy activity that you have of both agents. So I, I think the short answer to that, Ben, is those conversations are, are happening internally.
Okay.
I think most importantly for us, is that we're generating that data and refining our thinking, and seeing if there's consistency from what we saw pre-clinically, versus what we're seeing in patients.
Excellent. Okay, great. Let's move to IDE161, the PARG.
Sure.
I guess, so question number one for me is, is maybe just quickly walk us through some of the, some of the responses that you've seen, and I guess maybe can you talk about what precipitated the, the Merck collaboration, Merck/pembrolizumab collaboration?
... Yeah, no, thanks for asking. Yeah, so just as a quick refresher, you know, we're in the part of the dose escalation, dose optimization. We have seen some early responses, in particular, in the HRD setting, in CRC and endometrial cancer. We're continuing to do the dose evaluation. I think hopefully we'll be giving an update on that in this upcoming earnings on just a little bit more granularity on where we are on that front. And one of the key findings around the PD-1 with Merck was based on some of the basic research we've been doing. Unfortunately, here, Ben, Merck has specifically asked us not to disclose that data right now. So hopefully, by the end of the year, we can share that data.
But what I can tell you is that, you know, we've been in conversation with Merck on this for probably about a year. You know, they've been following the data set fairly closely, and, you know, we're excited about the opportunity in combination with PD-1, specifically in endometrial cancer.
Okay.
And as I mentioned at the end, expect more combinations to come on this front. So we're having a lot of different conversations on, I would say, what we think are high-conviction rational combinations.
That's great. Okay. Okay. Maybe just... I know we're a little bit pressed for time, but I do want to get to it. GSK101, the Pol Theta program. What makes Pol Theta a good target? Maybe could you give us just a, I don't know, a little bit of a-- I realize we're short on time, but a very quick teach-in on Pol Theta and why that's a good target, if you-
Yeah. Yeah, I think there are several reasons here, Ben. One is that the mechanistic rationale in combination with, with PARP is extremely solid. So, you know, as, as you're aware, you know, when, when HRD goes down, which is a primary DNA repair mechanism, and you elicit DNA damage through PARP, the backup DNA repair mechanism of microhomology-mediated end joining starts becoming critical, right? And that's really where Pol Theta steps in. It's, it's the first step of that process and, and that backup DNA repair mechanism.
And we also know, based on clinical translational data by a lot of great work that groups like AstraZeneca have published, you know, about a year ago, is that a very large percent of patients, in particular in breast cancer, get PARP resistance through BRCA reversions, which has been specifically implicated with microhomology end joining, and therefore, Pol Theta. So, it's a great story. And then, as a target for Pol Theta, I think what's critical, and I think, Ben, you know, you've been in the oncology space for some time here, you know, I think ultimately the vision is hopefully for these next-generation molecules to have just much safer therapies, right? And Pol Theta has got to be on the top of that list.
Like Werner, these are non-essential genes, so you knock it out, and normal cells survive, and so that's terrific. So I think, GSK has a great molecule. We do think the primary focus will be on PARP combinations. And I know, Ben, when we just spoke recently, and, you know, just to clarify here, not just about addressing resistance, but we should be able to drive even greater durability in the pr- in the primary resistance setting, so pre-acquired resistance. And so it'll be exciting. And I think clearly we'll be in the crosshairs as, you know, different activities like PARP-1 happen. Personally, I think this combination is much more exciting.
Awesome. Okay. Well, Yujiro, a great conversation. I think we are out of time, but thank you so much.
Definitely. Okay, thanks so much, Ben. Appreciate the time today.