All right, good afternoon, everyone. Day one of our global healthcare conference. I'm Andy Berens, Senior Biotech Analyst at Leerink Partners. Thank you for joining us. We're really excited to have IDEAYA with us. We have Yujiro, the CEO of the company. Thank you for joining us.
Well, great, Andy. Thanks so much for the introduction, and thanks to Leerink for the kind invitation to share.
Great, great. Yujiro, for those that don't know your company and the story, why don't you give us just a brief overview?
Sure. So IDEAYA Biosciences is a leading precision medicine oncology company. We were founded almost nine years ago. We have a significant focus in the area of synthetic lethality. Today we have four potential first-in-class programs in the clinic. Our most advanced is darovasertib . We are now in a first-line potential accelerated approval study in the metastatic setting for this trial. In addition, we've launched a phase II study in the neoadjuvant uveal melanoma setting as well, where we have key guidance towards phase II clinical efficacy readout, which would be two separate readouts by the middle of this year, as well as our target to have guidance from the FDA on a potential accelerated approval study design in the neoadjuvant setting. Our second program in the clinic is IDE397. This is a potential first-in-class MAT2A inhibitor.
Here we're targeting a very large patient selection biomarker called MTAP deletion, which there has been very high interest in, partly because it's a very large patient selection biomarker has been reported as high as 15% of all solid tumors. Here we have shown early phase clinical efficacy as a monotherapy. In addition, a significant focus on rational combinations, including a key study we're doing with Amgen with their MTA-cooperative PRMT5 inhibitor, AMG 193. We dosed our first patient there last August. We're guiding towards having a publication strategy with Amgen this year. And then second, we recently announced a new partnership with Gilead for the combination with Trodelvy, their TROP-2 ADC, here specifically targeting MTAP deletion bladder. Beyond that, we have two more in the clinic, PARG, also in phase I, targeting HRD solid tumors.
Here also a significant focus on rational combinations, including to get us outside of the HRD space. Pol Theta Helicase also in phase I and dose escalation. That is focused on the combination with GSK's PARP inhibitor, niraparib. And then lastly, we're targeting to have Werner Helicase IND filing this year, hopefully in the clinic this year. And then beyond that, we're targeting two to three candidates in this calendar year. So, you know, from our perspective, we hope we'll have an opportunity to deliver seven or more potential first-in-class programs in the clinic in the next several quarters. We do have about $1 billion in cash. Prior to the financing we did a couple of weeks ago, we already had a runway to 2028. We're not updating that guidance, but, you know, we're in a very strong financial position.
Great. Well, thank you for the overview and congrats on all the progress. Why don't we start with the PKC program and uveal melanoma? Why don't we start with what you've shown in the metastatic setting? You know, let's talk about the data and then the commercial opportunity, and then we'll move into the earlier settings that you're testing now. So what have you shown with the drug and the program so far?
Yeah, so there's really two data sets we've shared. So one is we had a very important oral presentation at ESMO fall of last year where we provided a phase II efficacy readout in the metastatic uveal melanoma setting. And really there the quick summary is that we saw a response rate roughly 45% confirmed ORR by RECIST. For just the listeners, historical response rate has been typically in the 0%-10% range. So we're substantially higher than what's been reported before. And then in terms of median progression-free survival, which is our primary endpoint for potential accelerated approval, here we reported a PFS greater than seven months in the frontline setting. And if you look at the hepatic-only setting, so when you get metastatic disease in uveal melanoma, it's 90%+ of the time in the hepatic setting.
For single site metastasis, hepatic-only, we saw a PFS that was approaching a year. So, and then lastly, we did also share a two-year PFS Kaplan-Meier curve, you know, significant, I would say, delta versus other studies that have been run here. And then lastly, we've seen almost a quarter of patients that have been progression-free even beyond two years. So that's really the summary. From an AE perspective, I would say generally well tolerated. You know, we see very few sort of higher-grade SAEs, discontinuation rate, which is in the mid-single digit percent. So I think overall very favorable in nature. And then the next efficacy data we've shared is in the neoadjuvant setting. It was a small data set of about six patients. Here we showed in about half of the patients we were able to preserve their eye and to prevent the enucleation procedure.
We have seen consistent and robust tumor shrinkage. This is being studied as a monotherapy, which, we know, has a very favorable safety profile from monotherapies.
Oh, okay. Well, why don't we talk, I guess, about the neoadjuvant setting? That would be given prior to surgery. How much before surgery do the patients start getting the drug? And then, you know, what do you see clinically? How much does the tumor shrink? Does it make the procedure easier for the surgeon?
Yeah, so there's sort of, I would say, three buckets of patients I would think about here. So one bucket are patients that are on schedule to get enucleated. And so in many cases here we are intervening even just several weeks before their scheduled enucleation. And so here the, you know, the physician is really looking for a quick response. And so, you know, our patient's tumor shrinking, what's that consistency and level of tumor shrinkage? I would say, you know, we did share a waterfall plot. And, you know, the kind of patients where we saw greater than 30% tumor shrinkage was in that, you know, 40%-50% range. But I think here, you know, we still need more follow-up on these patients. Will these, you know, will the depth of that shrinkage become deeper with longer treatment duration?
Three out of those six patients we did prevent the enucleation procedure. And that was observed in a couple of months. So right now our phase II study design, we're treating up to six months. I think this is important. We think that could just be the start. So ultimately we anticipate we may have the ability to even amend the protocol to treat up to a year. And perhaps even in certain cases patients would get treated well beyond a year. The other set of patients are patients with small to medium-sized tumors. And here I just want to also emphasize that our ultimate vision here is that we would treat all patients with all sizes of ocular tumor, the smallest of the small and the largest of the large tumors.
Our current thinking is that we would likely randomize against plaque therapy and look at other endpoints such as potentially a response rate criteria, which we would still need to get agreement with the FDA, looking at other endpoints like radiation reduction as well as vision preservation.
Okay. And what about there are some patients that have indeterminate lesions that are being observed? Would that be a potential?
Yeah, so exactly. That would be the situation of these sort of smaller wait-and-see tumors. So we do think ultimately we will also be pursuing those tumors as well. Because at the end of the day, you know, from our perspective, we believe our therapy is generally well tolerated. It's an oral tablet. If you have an ocular tumor in your eye, you know, we think you should be treated at the end of the day. We think right now there's a large subportion of patients that aren't doing anything because alternatives are not great alternatives, right? You know, plaque therapy is not a great procedure. They're stitching a metal clamp device to your eye. These patients often have, you know, significant complications as well as a majority of these patients ultimately going blind.
So we think if once there is an intervention that pool of patients that are currently a wait-and-see that are these smaller tumors, absolutely, we think they're going to want this treatment. In fact, it's very possible that those subset of patients may even respond the best, right? Because those are smaller tumors. And so I think that's usually how it goes in cancer, right? The earlier you treat, usually more robust activity, you know, typically you have the opportunity to deliver.
Okay. And then in terms of, I mean, you probably haven't spoken to the FDA yet, but, you know, what would a regulatory trial look like in the neoadjuvant setting? Would you have to have an endpoint like survival or would it be visual acuity or, I mean, I guess it sounds like there'll be different buckets of the patients, but would you run a separate trial for each of those buckets?
Yeah, great question, Andy. So maybe think about this in three specific categories or buckets. So first is neoadjuvant setting, large ocular tumors. We think there those patients are likely to get eye resection. And so there, at least our current thinking is we should have an opportunity with the FDA to talk about a frontline single arm accelerated approval study. Where here, at least our current thinking is the primary endpoint being eye preservation. So that's obviously a very black and white result. We think we should know it in several months. And a single arm design because there's nothing to compare it to, right? You're either getting your eye removed or you're not. So that's really the first part. The second group in neoadjuvant, you know, are these more small to medium-sized tumors. And I think there our current thinking that would likely get randomized against plaque.
Here, as I mentioned, you know, endpoints such as vision preservation, radiation reduction, response rate. There is precedent around the vision preservation endpoint as an approvable endpoint by the FDA. So that's not, we're not paving new ground there. I think here the concept is if you shrink the tumor in the eye sufficiently, you can reduce the amount of radiation provided, that should translate to a more robust vision preservation. The final third group, which you're mentioning related to relapse, there we would plan to tackle that in a separate independent adjuvant trial. Here we would anticipate that would again be a frontline accelerated approval study, most likely randomized against placebo. Placebo because there is nothing approved in the adjuvant setting. Here we would do this irrespective of HLA-A2 status, which we think will have significant advantages because of that. It's also an oral tablet.
We know that the drug is also, again, well tolerated, which is going to be, as you know, very critical in the adjuvant setting. Here we would look at endpoints such as relapse-free survival, one, two, three -year type endpoints. We are seeing activity in the neoadjuvant setting. We've obviously seen activity in the metastatic setting. We would be hopeful that we should also see activity in the adjuvant setting.
Okay. And I just want to make sure I understood you. So in the large tumor, the idea would be that you're going to, patient would not have to go enucleation of the eye at all?
Correct.
How, I mean, how much tumor shrinkage would you have to have to, would it have to be a complete response or what, you know, I mean, yeah, I mean, what would make a doctor not decide to do the enucleation?
Yeah, and we've already seen evidence of this, including from the data that we share publicly from our IST. And, you know, it is dependent on the patient. So, you know, typically the enucleation procedure, there's a specific threshold of size that they have to exceed to be eligible for, as a candidate for enucleation. So one is get below that threshold size. Second is there also could be driven based on where the tumor's located relative to the optic nerve, the fovea, et cetera. So even if 20% or 15% shrinkage could be sufficient to get them off of the enucleation track. But, you know, I would say right now, you know, half of three out of six patients in this small data set, we've been able to preserve their eye.
You know, pretty much every patient we're seeing tumor shrinkage from based on the public data set that's out there. So now the question will be with a larger denominator, we're more follow-up, you know, what that waterfall would look like. But at least the threshold that we've heard from key opinion leaders is that even a 5%-10% eye preservation rate would be viewed as a game changer for this indication. So we think the regulatory bar could be potentially low.
Right. And then it seems like the commercial proposition would be very high if you can achieve that.
Yep, exactly.
Okay. Why don't we talk about the metastatic setting since that's the most advanced? So where are you in the process? You know, what does it take to get it across the finish line commercially?
Yeah, so we had a Type C meeting with the FDA about the middle of last year. What we got agreement on is to execute on an integrated phase II, phase III design. So through the same study, we would have the ability to get both accelerated approval through PFS as the primary endpoint. That trial would then continue on to a full randomized OS readout to get full approval. Here we're targeting to enroll, you know, roughly 200-some-odd patients for the first readout for accelerated approval. We would add additional 120-some-odd patients for the full OS readout. A lot of our analysts are projecting a late 2026 commercial launch. So we have pegged our target enrollment to basically hopefully enable that key objective. Here we dosed our first patient in that registrational trial, sort of third quarter last year.
We did note in January that enrollment is on track. We've already, in January, we noted we've already achieved double-digit enrollment for the registration, potential registrational trial. We have also achieved international site activation as well as international enrollment.
We've heard the FDA, and obviously the FDA, there are different divisions, but sometimes now they're preferring response rates for accelerated approval at the interim. What's the rationale for this division wanting PFS? Is it the disease, the course of the disease, and that's a better predictor?
Yeah, I mean, yeah, I mean, Andy, I think you definitely have a much broader purview than I do on this one, on sort of accelerated approval and what other maybe sponsors are hearing. But I would say as a general rule of thumb with the FDA, I do think there's a strong desire to see data in a randomized fashion. And, you know, I think that's, you know, I think that's a great thing to do to have a controlled study. I think from our vantage point, we had probably both options available to us, both considering a single arm ORR type readout or doing it as a randomized fashion and PFS as well. So I think we had both options available to us. We felt very good about the PFS as the primary endpoint based on the data that we saw.
You know, we're so far right now, our PFS is trending, you know, roughly double or more historically reported PFS. So, you know, as we had the discussion with the FDA, there were really two primary pieces. One was, you know, as it relates to this question on, you know, what is the right design and endpoints. And then the second was related to contribution of components. And maybe that's also partly goes back to your first question, which is, you know, this is in combination with crizotinib. So I think there, you know, perhaps, I don't want to generalize, but perhaps a little less precedence with response rate on that front. So, but, you know, we're very happy with this endpoint. Historical PFS, again, is two to three months. So it's not going to extend the readout timeframe, whether that's an ORR endpoint or a PFS.
Okay. No, I just bring that up because you had such a, the magnitude of response rate, so much more than what's been seen historically.
Yep.
So that would be a pretty, it seems like something very easy to achieve.
Yeah, I think, look, I think both would be, you know, terrific outcomes, but I think being able to report, if you can report a robust PFS readout in a randomized fashion, I think that's going to go a long way also with the medical community.
Okay. And what do you see as the size of the metastatic setting? What's the commercial opportunity?
Yeah, so from our perspective, we believe the annual incidence is roughly, this is primarily in the U.S. and Europe, also partly in Australia, is about 4,000-5,000 patients annually. Our first line trial is targeted in the HLA-A2 negative setting of MUM, which we believe is roughly two-thirds of that population. Also based on data that we presented in a peer-reviewed setting at ESMO, as well as other reports that we've seen in other peer-reviewed settings. But with that said, you know, I would just want to emphasize, ultimately we believe this therapy should be available for all patients irrespective of HLA-A2 status. You may have seen at ESMO, we reported approximately a 60% response rate in the HLA-A2 positive setting. So there we would in parallel pursue a compendia strategy for A2 positive. So we've already published data in the positive setting.
Assuming we get approval in our first line existing trial, we would then ultimately target to get this on compendia through further publications in a peer-reviewed setting.
Okay. Why don't we switch now to the MAT2A program? And you guys are trying a novel approach with Amgen. Can you just give us a guess before we jump into what the scientific rationale there? Can you talk a little bit about the Amgen relationship? You know, we've seen some of their data as a monotherapy. You know, what is, is it a collaboration agreement or is it more comprehensive than that? And how much, how committed do you think Amgen is to this program?
Yeah, so we put our relationship with Amgen and, you know, roughly a year ago, there's two parts to the collaboration. One is a research-based aspect. So the teams are doing quite a bit of characterization of AMG 193 and IDE397 through various preclinical models. We actually just expanded that scope of research work that we're doing. So here now we're looking at a much larger number of in vivo models, looking at different tumor settings and sort of doing that work. Also there's a piece around continuing to reinforce our conviction on the mechanistic rationale of combining MAT2A with PRMT5. The second part is around clinical. Here, just so people know the construct, Amgen is the sponsor of the IND. So a decision was made to actually follow a separate IND on this combination, which took a lot more work.
But ultimately, if we did, for example, start a registrational trial in the future, having a separate IND would more seamlessly enable that. Next is Amgen is running the trial 100% themselves globally. And so I think that's also significant from their commitment perspective. There's both a joint development committee that meets monthly as well as a joint operating committee that meets quarterly. We do have data sharing, joint publication, joint IP provisions as well. Although the companies are sharing in cost 50/50 for external cost, because Amgen's running this trial, all of the internal costs Amgen is covering, we're targeting to roll roughly 200-some-odd patients. So it will be a robust data set. The expansion focus here is in non-small cell lung cancer.
In terms of, you know, what we think the enthusiasm for Amgen is, I think here, you know, Andy, what I'll just highlight is that, you know, this is in Amgen's materials, both in their earnings updates, both in their presentation, as well as their written materials. Right now, this is the, you know, the only novel, novel combination that they're executing on today. We've already published data together in a peer-reviewed setting at AACR last year, which was largely Amgen's data. And at least guidance for 2024 is that we're going to develop a joint publication strategy with Amgen this year. So I know the two organizations are talking right now, both on what that publication strategy would be from both a research perspective as well as a clinical perspective.
Our view is that we think this combination study that's being run with Amgen is the most potentially important as well as exciting trial that's being run in MTAP deletion today.
Okay. Sounds like a big commitment on their part and your part. Is it exclusive if you wanted to run a combination study with one of the other PRMT5s by Mirati or AstraZeneca has one now, Tango has one? Are you able to do that?
Yeah, so there's nothing in the agreement that would prohibit our ability to enable additional combinations. You know, I would say with that said, we're doing a lot now in the organization. You know, we just recently formed this new partnership with Gilead. We've noted publicly that we'll be anticipating enabling new combinations on PARG IDE161 clinically as well. We're having several conversations there. As you know, we're doing this combination with Pfizer on crizotinib. So we do have a fairly full plate. And I think next is we feel very good about the Amgen molecule. And we're getting much further along with them. So, you know, I think there for us to enable additional, I think it just would have to make, you know, strategic sense for us.
Okay. You know, I know some of the other PRMT5 companies, based on their preclinical work, felt that adding in a MAT2A was not necessary. So, I mean, obviously you guys have done work that suggests otherwise. You know, what is it that gives you conviction that, you know, two is going to be better than just the PRMT5 alone?
Yeah, I think there, Andy, I mean, at least our kind of opinion on that is it's not following the data that's already been published in a peer-reviewed setting at AACR, not just our data, but Amgen's data as well. We think the, this is not just about hitting PRMT5 harder. That's just, that's not what we're trying to achieve through this combination. There's really two key aspects of it from a mechanistic perspective that's distinct from just hitting PRMT5. One is from the perspective of how these agents are causing cancer cell death in MTAP, which is around an MOA of generating these RNA splicing defects. And we know that those signatures of how it impacts RNA splicing are different between PRMT5 and MAT2A.
And then when you bring these two mechanisms together, the synergy and much greater effect in RNA splicing you see is much greater than either mechanism alone. So that's the first. The second, which also I think is very important and has been an important focus with us with Amgen, is that it's not just about elevated MTA, right? We have this perspective that elevated MTA is what's driving the synthetic lethal in MTAP. However, there's another key cofactor here called SAM. We know with these second generation inhibitors, both MTA and SAM bind to the same binding region of PRMT5. And if SAM occupies that pocket, these second generation inhibitors will not bind and do what it needs to do. So it's not just about having elevated MTA to bind PRMT5. You really want to deplete SAM levels.
And so what we're fundamentally trying to do through this combination is to leverage the stoichiometry between MTA and SAM. So now you can imagine, do we see variability across different tumor types? The answer from our data says yes. Now you can imagine in the clinic, are you going to see variability across patients, right? Which is going to be even a much more complicated setting to be in. So we're really trying to achieve two key aspects. And I think last piece I would mention, at least based on our preclinical data, we've shown we can dose reduce both mechanisms by about one-tenth. And even by doing that, we can drive greater and more durable responses. So here, you know, Andy, from our view, this is not just about response rate. Ultimately, you have to drive greater durability.
If you look at the swimlane plots, I won't throw out a certain number. You can all look at the presentation maybe now or after this fireside, but you'll see, you know, at least it's a small number, but the PFS is, I think, pretty borderline right now based on the data that's been published. So how do you really drive greater durability? We don't think that's going to be just by hitting PRMT5 harder.
Okay. So let me just clarify. I mean, other than, I mean, the RNA splicing part makes sense. What you're saying with the SAM, MTA, is SAM and MTA, are they competing? Is it a competitive or is it a covalent process that they bind to the?
They're both binding to a similar region of PRMT5. And so for these second generation inhibitors to work, MTA needs to be bound. And that's how then PRMT5 is able to do what it needs to do. If SAM occupies that pocket, where the rotamer is, does not enable that sufficient binding. So this is not just about having elevated MTA. It's also about how do you deal with this important cofactor of SAM. And we know that the KD, the binding constant between MTA and SAM are similar. So they compete. So, and this has been published by others in the past as well. And the main distinction between PRMT5 first and second generation is the second generation is leveraging with this binding of MTA, right? And that's really the key piece.
So if you have, there's also been whole CRISPR genome screens that have been done where this has been published. One of the key, you know, compensatory pathways that have been shown is that MAT2A overexpression actually causes resistance to these second generation PRMT5 inhibitors. Why is that? It's because if you overexpress MAT2A, that means you're pumping more SAM into the system.
Okay. And so SAM and MTA, that's a competitive process in the binding.
They have a similar KD.
Okay. And so 397, what have you shown? How much does it lower SAM?
I mean, we show an ability in the plasma to very robustly deplete SDMA even in our first cohort. So we published that data. Preclinically, you know, we typically dose at 30 mg/kg, which is already a low dose. I mean, we can reduce that to 3 mg/kg and still see this tremendous synergy with these second generation PRMT5. And we can actually interchange these second generation PRMT5 as well. So it's, you know, and I think the last part, Andy, I would just mention is, you know, people also shouldn't assume that we can't do this even with the first generation PRMT5 inhibitors.
Interesting. Okay. No, I mean, it makes a lot of sense.
Because if you can dose reduce, right, this whole concept of the therapeutic window is different. I mean, ultimately for us, right, we want to be combining with the agents that have the widest therapeutic window, period, right? Because that's ultimately where you're going to be most competitive. But, you know, I think that's another possibility here as well.
Has Amgen shown data suggesting that the resistance to their PRMT5 is upregulation of SAM? Is there data that shows that?
They haven't specifically published that, you know, without kind of giving specific names. That was published at AACR. I can send that to you after this. That was published at AACR in a peer-reviewed setting. We've also done separate work independently, you know, on this question.
Okay. Let me open it up with the last minute and see if there are any questions from the audience. Any questions for Yujiro? Okay. Maybe with the last minute, Pol Theta, a lot of companies have, you know, suggested that's a really difficult drug to create. What are some of the challenges and how have you guys addressed those?
Yeah, look, you know, we're the first company to get a Pol Theta Helicase molecule in the clinic. You know, chemistry for Helicase target classes in general has been very challenging. You know, as you know, we also delivered a Werner Helicase development candidate, you know, a couple of quarters ago. At least that we're aware, we're the first biotech or pharma that has delivered back-to-back Helicase development candidates. So I would just say structurally, you know, we can't disclose too much because of our relationship with GSK. Sort of, you know, soup to nuts from the very beginning, you know, it took us over six years to effectively drug this target. So I know groups have both targeted the Helicase and polymerase domain. There are very specific reasons why from both the biology as well as a drug discovery perspective why we prioritize the Helicase domain.
But I think a very exciting combination, you know, I think here that we think at least the primary focus will be around combining with PARP and specifically trying to address BRCA reversions, which we know functionally is largely driven by a backup DNA repair mechanism called microhomology-mediated end joining, which Pol Theta is the first step of that process. And we believe this is really the key Achilles heel. Recently, we know there was a paper, recently reported, I believe, by Chris Lord that noted in breast cancer, you know, now they're reporting even over 50% of patients with, I believe, BRCA being driven through these BRCA reversions. So if that's true, as you know, Andy, a lot of the issue with PARP inhibitors has been actually their time-to-event endpoints like PFS and OS.
And so a larger percentage, this is really the acquired resistance is being driven by this mechanism, you know, is you're going to have a much greater ability, at least we think, to really impact these time-to-event endpoints like median OS. So our expectation is high here. I think we're fairly far ahead. And dose escalation is going well. They have a very good molecule there at GSK. And I think this is going to generally set a very high bar in HRD.
Okay. And the idea would be to preempt the reversion mutations or to treat patients that get them and resensitize them to PARP?
Yeah, great question. I think that will be, you know, as the development goes on and, you know, as you can appreciate, later line setting you're in, right? You're going to be probably more in that setting where you're dealing with the BRCA reversions that are already in place. I think ultimately I would hope the vision for this will be around doing a frontline head-to-head trial and, you know, really delivering an outstanding result. So I think ultimately the opportunity should be both, right? You should be able to treat these patients that have these BRCA reversions, but then ultimately hopefully push into the earlier line settings where those reversions are not in place yet.
Okay. Well, congrats on all the progress. We look forward to seeing these molecules as they continue to advance.
Yeah. Well, thanks so much, Andy. Again, thank you for the opportunity today.