Hi everyone, my name is Maury Lacroft, and one of the biotech analysts at Jefferies. It's with great pleasure that I'd like to welcome the IDEAYA team. We've got Yujiro Hata, the CEO, and Josh Bleharski, the CFO. Thanks so much for joining us today. Great, thanks so much, Maury, and thank you to Jefferies for the opportunity to participate this year. Josh and I will be tag teaming today. Thanks again. We're going to do a fireside chat format, so maybe for those who are new to the story, if you can give a one-minute intro to IDEAYA. Yeah, no, so IDEAYA Biosciences, we were founded about a decade ago. We're a leading precision medicine oncology company. We have six programs in the clinic. We're targeting nine by the end of the year. Our lead program is Darovasertib, which is currently in a potential registration-enabling trial. Here we're targeting frontline approval in HLA-A2 negative metastatic uveal melanoma. We are targeting a potential readout to enable that accelerated approval by year-end, and that's going to be based on a primary endpoint of median progression-free survival. Beyond that, we have several activities across several clinical programs, including in the area of MTAP deletion. We have a DLL3 topo ADC. There we are guiding towards a medical conference update in the third quarter. We did recently learn that's going to be an oral presentation, so we're excited about that. We anticipate that'll be roughly over 50 patients. We have a growing pipeline, including clinical programs such as Werner Helicase. We also have earlier programs like Cat67. Overall, we have also made significant investments in our platform, both in the area of target biomarker discovery as well as drug discovery. Got it. Yeah, so extensive pipeline with a lot going on, a lot of investor focus right now on the pivotal study, which is going to read out end of this year, and this is in HLA-A2 negative metastatic uveal melanoma. Maybe talk about what clinical benefit you hope to show in this study, and what's the bar for success there? Yeah, so when you look at metastatic uveal melanoma, unfortunately across several trials and large meta-analyses that have been done, the median progression-free survival in the frontline setting, as well as pretreated setting, has been pretty consistent, which has been in the two- to three-month range. Back at ESMO 2023, we had an oral presentation where we reported a median progression-free survival of approximately seven months. Based on the way the study was powered, we believe as long as we're north of five and a half months and that control arm remains stable, as it has been historically, we feel we'll be in good shape. Got it. For data timing at the end of this year, does that imply you're projecting the top line is going to come in November or December? I guess any more specifics on that, and maybe talk about just the logistics of how long you think it's going to take to lock the database and report. Yeah, so first I would mention for the PFS readout, the target enrollment is roughly 250 patients. We have already exceeded 300 patients. We're probably in that 320 range now. As an event-driven endpoint, it's just going to depend on the timing of those specific events since the enrollment portion is now complete. We're tracking that, but most likely it's going to be towards very year-end is our anticipation. Got it. Okay, and for 320 patients, so you've over-enrolled this study. Is that based on patient demand, or is that to, I guess, boost the sample size for treatment effect? Yeah, so this is an integrated phase two-three study, so we're seeking both accelerator approval and full approval in the same study. The full approval endpoint is median overall survival, and for that we need to add an additional 120, so that would put you at approximately 370 patients. We do anticipate we'll finish off that full enrollment by then, but as you noted, Maury, we have been ahead of schedule. In terms of demand for the study, we've seen significant demand and a tremendous amount of enthusiasm from investigators. Got it. I may have missed this, but did you say how many events you're going to need to trigger the PFS? We have not. We have not. We have that number internally. We've been tracking that. Obviously, it's blinded, but we've been tracking those events. Got it. Okay, and you've mentioned in the past that you'll get an OS readout at the same time as the PFS readout. Do you plan to share how preliminary OS is trending with PFS, and can you confirm whether you're on track for the confirmatory OS readout to occur in mid-2026? Yeah, so just to clarify there, the OS guidance that we provided is from the phase two single-arm study, so that's going to be roughly over 40 patients. There we are guiding towards providing our first OS readout on this combination in the frontline setting, which we have not done before. We are discussing exactly what that timing is. We may have it concurrent with a median progression-free survival readout, and that will either come in the form of several policy scenarios, including potentially a manuscript, as well as a medical conference update. Those discussions are ongoing, so I think those are still to be discussed. In terms of the median OS for the registrational study, when the IDMC does the review for the PFS readout, they will have a look at OS. Our base assumption here is that the OS will not be sufficiently mature at that time, and that we'll need to wait longer for that full median OS readout. Got it. Okay, and when could that full OS readout? We have not given that specific guidance, but I would say probably sometime towards the end of next year. End of next year. That timeframe may be sort of a viable scenario. Got it. Understood. For this update, is it going to be at a company event, or maybe just talk about the format for? For the OS piece? For the pivotal update. Yeah, for the randomized PFS readout. Yeah, I think there because it's going to be top-line results, and we appreciate we'll have some—we're going to be on the clock in terms of the timing we have to get it out. Most likely look for that to be putting out top-line results as sort of a company event. We'd obviously love to follow that up and publish that in a peer-reviewed setting, like a manuscript would probably make the most sense, post that top-line results. Makes sense. Maybe just talk about next steps: post the data and timing for filing an NDA and also commercial launch. Yeah, so we do have fast-track designation in the metastatic setting, and Maury, as you know, in the neoadjuvant setting, we have breakthrough therapy designation. There, once the results are in, we will start assembling the NDA, submit that. We will have expedited review as fast-track, but we have not given specifics yet. As we get closer to that readout and perhaps at the top-line results, we will provide more specificity at that time. Got it. Okay, wanted to talk more about the phase two OS data too that you're going to have second half of this year. For that update, you've mentioned in the past that it could be in the high teens or low 20 months for survival. Is there more you can share on just the cadence of enrollment for this frontline cohort? Wondering also, did all of these patients get the top combo doses? Have you reached the median OS yet, or are you still waiting for the data to mature? Yeah, so I would say there just historical median overall survival in the treatment naive setting we believe is going to likely be in that 12-month range. I know there was an updated meta-analysis recently at ASCO based on line of treatment. They noted a lower number, so for whatever that's worth, but they noted 10 months in the frontline setting just a couple of days ago in a poster. What we've said publicly is our objective to improve OS versus the control arm by ideally six months or more. That's similar target from what we understood that TEBI achieved. Again, as you know, this is all about a hazard ratio at the end. That would put you towards ideally in the high teens or 20-plus month type OS. Got it. Okay, that's helpful. Makes sense. With this update, you don't have to show the OS from this phase two study, so the fact that you're doing it probably implies that you're seeing something good there on the OS front. Yeah, I mean, I think here, Maury, what we feel is that as we get closer to the median progression-free survival readout at the end of the year, we appreciate the natural next question is going to be around OS. I mean, here we would just emphasize that, A, it's a single-arm study. Second, it's still relatively a small data set, but we do think it will be an informative data set. If anything, we'll just provide people a data point on OS since we've not publicly reported OS at this time. I know your earlier question around has the data set fully matured on the OS piece. I think here we're continuing to enroll patients in that phase two study, so it just depends on kind of how large that denominator will be at the end. Our focus right now on the enrollment for the phase two has been now focused on HLA-A2 positive, because as you know, we're planning to submit that data set as part of our NDA submission hopefully sometime next year. Right, yeah, definitely want to talk more about that as well. I wanted to get into the market opportunity. You mentioned Kimmtrak earlier from Immunocore. It's been showing strong growth commercially with about $375 million run rate as of first quarter of this year, and that's in the HLA-A2 positive patient population. You've got some interesting data from the bone marrow registration database showing that that opportunity is a lot smaller than the HLA-A2 negative population. Maybe talk a little bit more about that and just how that implies commercial opportunity. Yeah, for sure. So yeah, as you noted, Maury, in terms of the bone marrow registry, it's based on allele frequency. That's correct. At least what's been reported, of course, highlighting that there are differences that you see across regions as well as across ethnicities. Also, we just highlight this has not been specifically performed in the oncology setting, and I think all of those details do matter. Generally speaking, yes, those allele frequencies that we have seen as majority of the population has been HLA-A2 negative, at least based on the papers that we've seen. I would say most importantly is our trial enrollment and screening that we've done in the clinic, primarily because that's specifically in the metastatic uveal melanoma population. So far, we've looked at roughly about 170 patients as part of that screening, and we've seen approximately 70% as HLA-A2 negative. We'd like to build that data set further. Obviously, we're going to have some screening data from our phase three registrational study as well as the neoadjuvant study. As it relates to other sort of peer market opportunities here, I don't know, Josh, any other comments you'd make here? I mean, look, it's a huge area of unmet need, as you know, Maury. We think ultimately the opportunity in the HLA-A2 negative setting is very large. As you said, we're trying to get a better handle on exactly how many patients there are out there. That work is ongoing. Ultimately, we think with a really valuable drug in hand and data that continues to play out, the commercial opportunity both here in the U.S. and internationally is quite significant. Got it. Makes sense. For the phase two study, you mentioned you're enrolling these HLA-A2 positive patients, which is where Kimmtrak's already approved. You would include that data in your NDA filing. What are the expectations there, and I guess could that be part of your label? Yeah, no, thanks for the question. Maybe if we just take a quick step back, just as a reminder, our mechanism around GNAQ/GNA11 as an activating mutation and activating the PKC signaling pathway, we believe our drug is going to work irrespective of HLA-A2 status. I think that's really the first place to start. As you know, in the pre-metastatic setting, we're evaluating both, all irrespective of HLA-A2 status. Maury, as you mentioned, we will submit all of the data we have in A2 positive. We expect that's going to probably be over 70 patients in the metastatic setting when we submit our NDA, and we are going through the process with the FDA, at least that dialogue around what it's going to take in terms of real-world evidence to submit that. Just to be clear, our base case communication on this is around a compendia strategy. Assuming we get accelerated approval, we would go through the process of compendia to get hopefully pricing reimbursement and the NCCN guidelines for A2 positive. Upside scenario, we think there could be a scenario in which we could get it on the label, but obviously that's going to just depend on what the data looks like and obviously our conversations with the FDA. Got it. Makes sense. I wanted to pivot to the neoadjuvant opportunity for Darovasertib. You've announced pivotal alignment with FDA on March 31, and you're on track to start the phase three. Can you remind everyone just about the design endpoints and granular timelines to starting the study and getting the data, particularly in the plaque blocking therapy and the enucleation cohorts? Yeah, no, for sure. So basically the study design, so it's a randomized phase three study. There's two patient population cohorts: an enucleation cohort and a plaque therapy cohort. The enucleation cohort will have 120 patients. The plaque therapy cohort will have 400 patients. The primary endpoint for enucleation is going to be eye preservation rate. The target there that we provided in the briefing book is to exceed an eye preservation of 10% based on a 95% confidence interval. So ideally, we see an eye preservation rate of 15%. Maury, as you know, in presentations that we've had in the past, including as an oral presentation at ASCO, we are seeing the majority of patients where we're able to preserve their eye and get them off of the enucleation track. So we feel good about that endpoint. In terms of the primary endpoint for the plaque therapy cohort, this is going to be around a vision test. The way this is going to work is that we will look at the proportion of patients between the treatment and control arm that pass or fail a 15-letter vision test. Here, based on the way the study was powered, ideally we need to demonstrate a delta between the treatment and control arm of between 15% and 20%. Lastly, in terms of the endpoints for both cohorts, the secondary endpoint is event-free survival. Importantly here, I would highlight that that's no detriment to EFS. Basically, the way we define this with the FDA is overlapping confidence intervals. We do not have a hazard ratio specific target that we set, so we feel good about this. In terms of timing, we anticipate roughly those 520 patients will be enrolled in about five quarters. We believe the first readout will be the eye preservation data. We anticipate within a year in that cohort, since we know that answer roughly six months when the patient is on. The second data readout we anticipate will most likely be on the no detriment to EFS. That will most likely be approximately two years from first patient enrolled. The last readout would be on the vision visual acuity piece, and that will be approximately 18 months. The first interim read will be 18 months from last patient enrolled. Got it. The eye preservation readout, it is relatively quickly, and that could potentially serve as a registrational filing right there. Yeah, correct. Yeah, no, thanks for that. Yeah, that's exactly right. We did clarify with the FDA if the enucleation cohort data is sufficiently mature, and I think I would just highlight not just the eye preservation data, but if the EFS data has to be sufficiently mature, not just in the enucleation cohort, but across both cohorts. If those data sets are there, correct, we did clarify with the FDA they're fine with us submitting the enucleation cohort first for regulatory review. We anticipate that could potentially accelerate that submission by six months relative to the plaque therapy cohort. Got it. For EFS, you've commented and you've mentioned that you just have to show no detriment on EFS. Based on how Darovasertib works, do you think you could potentially even do better on EFS? Yeah, no, that's a great question, Maury. I mean, Maury, as you know, we've been talking about a potential adjuvant study for a long time. That's something that we've been considering, and the only reason why we'd be considering that is because we know our drug is active in the metastatic setting. Yeah, I mean, that would be our hope. It's a randomized study. We will be ultimately tracking EFS. Obviously, to get full approval, all the FDA is requiring us to demonstrate is no detriment, but we'll see what that hazard ratio ends up being for EFS, and perhaps we'll even see an improvement. We're hopeful for that because we know our drug does work in the metastatic setting. Got it. Makes sense. For the update later this year from this program, you could potentially show some data on visual outcomes. Maybe talk about what you can show there and what expectations should be. Yeah, we think it's going to be a good update. We're guiding towards an update at a retinal conference the third quarter. I believe that abstract has already been submitted or is about to get submitted. We anticipate that'll be 20-30 patients. I think the data that we've lacked last year on the neoadjuvant setting has been around the visual acuity data. I think this will be helpful. Here, what people should look for is what's the consistent level of tumor shrinkage we saw in plaque therapy patients. Second is what have we seen as it relates to impact on vision during neoadjuvant treatment. Second is visual prediction data based on the amount of tumor shrinkage that we've observed. Maury, as you know, there's obviously the key focus here for the endpoint will be what's the impact to vision post plaque therapy, but just to be clear, that data will take longer to mature. The focus will be on the two earlier data sets that I just mentioned. Got it. For the 20-30 patients there, how much follow-up do you think you're going to have? Yeah, it's probably going to be in that 6-12 month range just based on the timing of those patients, sort of within that spectrum. Got it. Makes sense. I wanted to shift gears and talk about DLL3 topical since XiliG had their data at ASCO recently. Your program, you licensed from your partner, Hengrui, and I guess based on the updated data from XiliG, how do you think your DLL3 ADC could potentially differentiate and fit into the treatment paradigm? Yeah, look, yeah, we saw Xi's data at ASCO. I think they're continuing to see good activity, which was nice to see. For us, really, we believe we have a potential first-in-class DLL3 topo ADC. We think probably one of the main points of differentiation is around the linker. So ours is a tetrapeptide cleavable linker, which will cleave once it's internalized based on lysosomal degradation based on enzymes like cathepsin B, which I would say is perhaps the more classical approach in terms of linkers. Xi's linker is based on a linker that cleaves based on a TME linker or tumor microenvironment. There are going to be some differences. Here, I think what to watch for as we give our medical conference update in the third quarter, as I mentioned, it'll likely be north of 50 patients. We will show data as it relates to the escalation as well as the expansion. What is the response rate? Although I don't believe we'll likely have a full confirmed response rate yet since the data is probably going to be too immature for that. Also, what kind of durability are we seeing? Of course, what kind of AE profile we're seeing? We did note that at least this peer company, the dose that they went up to was, I believe, roughly about 1.6 mg per kg. At least that's what we understood from the update. We have been able to escalate beyond that. We're not specifying at this point yet what our target expansion dose is, but stay tuned and more to come on that front. Got it. So presumably if you're going above that 1.6, I mean, safety profile is looking good. I think Gilead had some ILD in there at the higher doses. Yeah, no, I think that's fair to say. I mean, ours is a DAR-8 DLL3 topo ADC. We have been able to escalate sufficiently. I think at this time there's probably several doses that we're evaluating, but that's correct. We've been able to exceed that dose range. Got it. Any more specifics you're saying just on what to expect with responses? In the past, we've talked about the intracranial responses. Anything else you could say on what to expect for durability? Yeah, I think at a high level, we can say we have seen patients on study, not surprisingly small cell, that have had brain mets. We have seen responses in that sub-patient population. In terms of what to expect, I know we have materials on our corporate deck. Obviously, it was a small denominator of 11 patients. We saw a response rate there of over 70%, which is highlight that was not a confirmed response rate. Here, we're going to have a lot more data. Hopefully we'll get more color as it relates to what that looks like with a larger data set with longer follow-up. I would also just mention that we are going to be considering what that data cut will be even from the timing of the abstract. The abstract that we'll publish versus the oral presentation, it will be a much later data cut. We're still waiting for that. We don't have full visibility on that, at least at this time yet. Got it. Okay. You have talked about combo strategy with this asset as well. You have a PARG inhibitor in-house. Maybe talk a little bit about that and what next steps there could look like. Yeah, look, in the small cell lung cancer space, at least this is our perspective. We believe really what's critical here in small cell in particular is all about durability. Whoever can really find that optimal path, especially as it relates to combinations around durability, we think are going to be very well positioned in this area of small cell lung cancer. Within that, there's obviously several different combinations that companies are considering both in terms of platinum chemotherapy, checkpoint inhibitors, and we think the other really exciting opportunity is around DNA damage repair small molecule agents. Within that, Maury, as you know, historically in the topo ADC space, big focus has been on PARP-1. Here, we believe PARG is a much more rational combination. We're also hopeful based on some of the AE profile that we've observed with PARG that we could have potentially a superior AE profile in that combination, but ultimately we just need to test that in the clinic. Lastly, the mechanism we think is highly rational. As you know, as we talked about before, the PARG enzyme is directly involved in the resolution of the DNA damage that's elicited by topoisomerase. We've now validated this in multiple preclinical models across multiple topo ADCs. We have also submitted an abstract at the CA Medical Conference where we'll have an oral presentation on this mechanistic rationale between PARG and the topo ADC. Hopefully, we'll be able to showcase that as well in the third quarter. Got it. Just wondering if Hengrui has dosed any patients with neuroendocrine tumors in the phase I experience. Yeah, good question. Yeah, since they have not stated, I believe, anything publicly, we do not want to comment on that. I would highlight, thanks for mentioning that, Maury, that neuroendocrine tumors is a high priority for us. We are also going to be interested at some point around other potential DLL3 upregulated solid tumors such as non-small cell lung cancer, melanoma, as you know. DLL3 is also upregulated at a smaller frequency in tumor types like RCC and thyroid. These are going to be all areas we would love to explore. Yeah, first we will most likely start with small cell and neuroendocrine tumors. Makes sense. Another part of the IDEAYA story is the MAT2A inhibitor and what you're doing there. You've shown monotherapy data with a 33% response rate at Triple Meeting last year. Can you talk about expectations for the combo data with Gilead Trodelvy and urothelial carcinoma and maybe talk about number of patients and amount of follow-up that are going to be in that data update? Yeah, so we haven't guided specifically towards the timing of the data update with Trodelvy. We did recently, several weeks ago, announce picking an expansion dose with the Trodelvy combination. We're most likely going to consider a second expansion dose as well. I know that the team did have several conversations at ASCO with investigators on this in urothelial cancer. At least what we heard is perhaps the bar is a little bit lower than we had initially thought in urothelial cancer. I would say ideally in combination, obviously we don't want to draw any lines in the sand, but I think hopefully we can see a response rate that's north of 40%. That would be a confirmed response rate with hopefully reasonable durability. As you know, Maury, we have a significant focus on the PRMT5 combination, which will be wholly owned. That is really the context for us is we want to see a very strong signal there just because we have other competing priorities on the combination side. Got it. For the wholly owned PRMT5 inhibitor, that combo with your MAT2A inhibitor, you're going to submit the IND for that middle of this year. Maybe talk just about the strategy there. Yeah, correct. I think within the MTAP space, at least our perspective, we're excited about the monotherapy activity we're seeing at Metsway. We've obviously been very closely tracking the monotherapy activity in the PRMT5 space. With that said, our central strategy in MTAP, and we believe the path forward is going to be around combinations. Within that context, as you mentioned, ID892, that IND filing is going to be happening here fairly shortly. The dosing and the in-life portion for the GLP-tox is now behind us. We feel good about that. We do think it's a potential best-in-class molecule. We did provide a lot of data at AACR here recently. We do believe that combination with MAT2A and PRMT5 has the ability to at least intercept several different potential mechanisms of resistance in MTAP. We do expect to see greater response as well as durability. Our objective is to enable that clinical combination by year-end. Greater responses and durability versus monotherapy, with PRMT5, we see about 30% response rate. Correct. Yeah. We would expect to see greater activity in combo than either agent alone. Got it. Okay. Wanted to ask one question on a Werner Helicase. Maybe just talk about, just read through from Roche's efficacy that they showed recently and how you think you could differentiate there. Yeah. For the listeners, people may know that Roche had an oral presentation at AACR recently. That's the first clinical data that's been presented on the target. At least we believe they've established several important things. One, they appeared to see a therapeutic window, which was obviously nice to see. They also did see monotherapy responses. In addition, when they saw the responses, I would say a few of the responses were quite durable. They in fact had a patient that was beyond therapy beyond 15 months, at least at the time of the presentation. Lastly, where they, I would say probably saw the greatest level of activity was in gynecological cancers. They had four endometrial and one ovarian, if my memory serves me correct. Then, out of that set of evaluable patients, they saw four out of five responses. I think where they lacked was in the area of CRC. I know we had an oral presentation at AACR where I think we felt we clearly demonstrated a potential best-in-class profile. Our molecule does bind in a different binding domain of that D1D2 domain. In particular, we were very cognizant of binding interactions with that cysteine 727. As you may know, Maury, in the MSI high setting in particular, one area you have to be very focused is on point mutations since that setting is very susceptible to that. Also, preclinically, we believe we demonstrated that we do see greater activity in certain CRC models. Our hope is can we flip CRC, continue to see that kind of activity in gynecological cancers. Lastly, I would mention that a significant priority for us on this is also around PD-1 combinations with GSK to startle a mat. Got it. All makes sense. We're pretty much out of time, but we didn't touch on some of the other questions we had about the pipeline. Maybe in closing, Josh, if you want to comment on cash position and where that gets you. Yujiro, just kind of recap key events ahead investors should be focused on. Yeah, sure. Cash as of the last reporting period in March was just over $1 billion. We've guided to that that gives us runway into 2029. As you know, we've always maintained a very strong balance sheet, which we think gives us a lot of flexibility to continue investing across the assets in the pipeline. That said, we are going to be very thoughtful about how we invest just given the environment that we are in. Nevertheless, I think we have a lot of exciting catalysts on the horizon and a lot of flexibility to access the capital markets or raise capital through business development if needed. I think we're in a really strong position to continue executing across this pipeline. Got it. Yeah, Maury, maybe just kind of catalysts from now to the remainder of the year. Some of them we've already highlighted, but it is going to be a catalyst-rich period from now to the end of the year. We're guiding towards four data readouts on Darovasertib. As we mentioned, we have a medical conference update on a substantial data set on DLL3 in the third quarter. It's also approaching our 10-year anniversary. We are considering an R&D day hopefully by the end of the year. Stay tuned on that. That's under discussions right now on that exact timing. In addition, we have three more INDs that we're targeting to hopefully get us nine clinical programs. It will be quite a busy year for us for the remainder of 2025. Great. Yujiro and Josh, thanks so much for joining us today. Great. Thank you so much.