IDEAYA Biosciences, Inc. ($IDYA)

Hi, everyone. My name is Maury Raycroft, and I'm one of the biotech analysts at Jefferies. It's a great pleasure that I'd like to welcome the IDEAYA team today. We've got Yujiro Hata, the CEO; Josh Bleharski, the CFO; and Mike White, CSO. It's fireside chat format. To start off, Yujiro, maybe give a brief intro to IDEAYA, and you guys had an announcement this morning with the collaboration with Roche. Maybe provide some key highlights for that.

So Maury, thanks for the introduction, and thank you to Jefferies for the opportunity to participate at your Annual Global Healthcare Conference. So Maury, as you noted, in terms of IDEAYA, we did have an announcement this morning, which we'll cover. So IDEAYA Biosciences, we're a leading precision medicine oncology company. We just came from ASCO where we had a late-breaker oral presentation this past Monday, which was the full complete data set around our top line results for the frontline HLA-A2-negative metastatic uveal melanoma. We have begun the -- under RTOR, the pre-submission process -- pre-submission module 1 has been officially submitted as of a few days ago. So all of that is tracking forward. In addition to the frontline HLA-A2-negative metastatic uveal melanoma with our global partner, Servier, we have two additional randomized Phase III studies that are now either kicking off or ongoing, including in the neoadjuvant setting as well as in the adjuvant setting. Beyond that, we have a very deep pipeline, including a Phase II DLL3 Topo-ADC molecule that is being advanced for both small cell lung cancer and neuroendocrine carcinoma. We also have two clinical assets in PRMT5 and MAT2A, in the area called MTAP deletion with a significant focus in pancreatic cancer as well as non-small cell lung cancer. We did announce this morning a new collaboration with Roche, specifically to do a combination with their Phase I pan-RAS inhibitor. with IDE892, our potential best-in-class PRMT5 inhibitor. And this is specifically going to target MTAP deletion pancreatic cancer, and I'm sure we'll talk about that. Perhaps the last asset I'll mention is IDE574, which is a first-in-class dual inhibitor of KAT6, 7. Here, a significant focus in breast cancer, colorectal cancer as well as other indications.

Got it. Great intro. And we always run out of time at the end. So maybe let's start off with Roche first. Maybe talk about why you picked that compound. Did you see preclinical data or something that gave you confidence that this is the best drug to combine with? Maybe talk about that.

Yes. Look, the parties did your typical diligence as part of the discussion around the clinical collaboration. Our perspective on this is that the real, we think, significant opportunity moving forward is can you deliver a greater efficacy and patient value in the earlier line settings of indications like pancreatic cancer. And we believe a lot of that is going to be through enabling rational combinations. And within that context, Maury, our view is that we have the opportunity to have a potential best-in-class combination with their pan-RAS inhibitor with our PRMT5 inhibitor. Beyond that, I think Roche should be the better group to answer questions as it relates to their specific molecule. But we feel very good about what we're able to observe, and we're excited to get this collaboration going.

And anything more on where you're at with your PRMT5 and anything on time lines for getting to the combo and what that could look like?

Yes. So what I can tell you with IDE892, and we also have Mike White here, our CSO. So I'm sure Mike could also jump into why we're excited about the mechanistic rationale of this combination. The dose escalation has been going very well. We believe based on the human PK data that we've seen, we're going to have a very favorable pill burden size. We also believe we've substantiated some of the key sort of premises of this molecule and why we believe that has a potential best-in-class profile. We will be commencing the combination phase for this program in about a week. I actually just checked in with the clinical team yesterday. We already have multiple patients in screening to start the IDE892 and MAT2A combination, which, as you know, Maury, is a big focus for us in MTAP deletion lung cancer. So that should be within days that combination is going, which implies we've already cleared multiple dose cohorts. The monotherapy expansion will also begin here relatively shortly, we anticipate in the second half. There's significant focus on pancreatic cancer, obviously, to also generate the contribution of components data we would need for combination work we would do ultimately moving forward and then as well as in non-small cell lung cancer.

Got it. And all really helpful. And for your PRMT5, maybe talk about how you think about the dose range compared to the Tango drug and the Bristol drug.

Yes, Mike, do you want to take that? And obviously, we can't specify a specific dose at high level.

Yes. So we're very excited about this compound. When we made this, it was really with an eye towards optimizing specificity for MTAP, making sure that we were MTA cooperative, but also SAM competitive. So we're very clean. That meant that we have a safe starting dose that's very close to our anticipated efficacious dose. which means that we're going to be ready to go into combinations very soon. And we have a PK profile that together with what we're seeing with respect to tolerability and our perceived tolerability with respect to this profile, also a very nice PK, no pill burden, no getting up to high doses where we're going to start to engage other targets. So we're excited about that aspect of the combination as well.

And also, Maury, I did forget to mention for the Roche collaboration. I think what is also unique about our collaboration with them is we also have the ability, and this would require approval by both IDEAYA and Roche, but a combination triplet with the pan-RAS, our PRMT5 and MAT2A. And we do think that could be a key point of differentiation. Obviously, that's going to be further afield, but we think a key potential path to differentiation.

Got it. Really interesting. And then for the combo with MAT2A, how are you setting expectations there for what you want to see on response rate and safety? And I don't know if you're setting expectations for what you want to see with the pan-RAS inhibitor, too, if you've kind of thought about that already. Some other companies have commented there, but...

No. Look, I think for both combinations with non-small cell lung cancer, maybe we'll start with that one. We've already seen some clinical combination data and at least Mike should comment here. But our anticipation is that we will likely be at an efficacious range at the first cohort, the first combination cohort. Based on what we've seen in the past, our anticipation is that we should have the ability to identify several combination cohort doses we should be able to move forward with. Second is that we should see responses at subtherapeutic doses of PRMT5 monotherapy. And ultimately, the real premise of this combination is going to be to drive greater -- both response rate and durability. So I think that's our expectation. We're not going to throw out a specific number, but the bar will be high. So I don't know, Mike, do you want to comment on that or also on the pan-RAS comment?

Yes. So one of the things that's important to us is that combination, the PRMT5/MAT2A combination allows us to really reduce the dose intensity for both of those molecules and get spectacular preclinical efficacy and durability. And the durability piece is very important because what the MAT2A inhibitor brings on board is intercepting escape mechanisms, particularly escape mechanisms due to cell fate transitions and epigenetic alterations. And I want to bring that to the pancreas cancer piece, very excited there because the KRAS combination or the pan-RAS combination in the KRAS mutants, that's exciting in pancreas cancer. All -- virtually all MTAP pancreas cancer patients have KRAS mutations. And this combination, in particular, the triplet gives us multiple mechanisms of action that will combine with each other, we think, to give deeper and more durable responses. So one, you have the co-alteration piece of hit it from two different sides. But two, as was presented by a number of laboratories at the AACR this year, the mechanism of action of MAT2A and PRMT5 to perturb splicing also directly perturbs the machinery that KRAS uses to activate the MAP kinase pathway. So we actually weaken the ability of RAS to drive tumors. And the third piece is that the MAT2A inhibitor restricts the cell fate transitions that occur, which seem to be one of the major mechanisms that drive bypass to monotherapy, mono-RAS therapy.And the third piece is that the MAT2A inhibitor restricts the cell fate transitions that occur, which seem to be one of the major mechanisms that drive bypass to monotherapy, mono-RAS therapy. So that triple combination could do something really important in the pancreas cancer setting.

Got it. All really helpful. And let's shift gears. The daro story has been front and center for IDEAYA for a while, and you guys had positive Phase III data with a 6.9 months PFS, 0.42 hazard ratio, and you just presented this at ASCO as a late breaker. Maybe talk about just the key takes that investors need to know from this program and next steps as well.

Yes. I would first start with metastatic uveal melanoma continues to be extraordinarily high unmet medical need. Unfortunately, for patients, there are currently no FDA-approved therapies in HLA-A2-negative metastatic uveal melanoma. And we believe, based on the data that was just presented at ASCO has the opportunity, as the discussant noted at the presentation, the opportunity to be the new standard of care in HLA-A2-negative metastatic uveal melanoma. So that's extraordinarily exciting, Maury. And what is that based on? That's based on the strength of the data that was presented on Monday. First, just for the listeners that may be less familiar, when you look at the clinical efficacy profile, we demonstrated a confirmed response rate by both central review as well as investigator scoring that was trending towards 40%. Just put in context, the control arm response rate was sadly mid-single-digit percent. And in fact, by investigator scoring, it was roughly 2%, which shows you how challenging this indication is. Next, when you look at median progression-free survival, here, we more than doubled what was seen in the control arm, which was primarily ipi/nivo. So I think the clear win on progression-free survival. Hazard ratio was 0.42, was less than that by investigator scoring and a p-value of less than 0.0001. So all statistical measures, it's clearly mission accomplished. From a safety perspective, just high level, we saw less SAE rates, less discontinuation rates than the control arm. So as well as from a safety perspective, we think significant potential advantages when you look at safety. So obviously, with any kind of new therapy as a potential standard of care, you always want to look at both the risk as well as the benefit in both parameters, we think the molecule is positioned extremely well.

Got it. That's helpful. And you mentioned that the first RTOR module was submitted. Can you just clarify what was in there? And then what the time line could look like for the remaining modules?

Yes. So the FDA with the RTOR presubmission module 1, 2, 3, there's actually some fairly clear defined parameters of what they want to see within those modules. I'm not going to go through that -- all those specifics. You can actually go to the FDA website, and that's all available online. The next presubmission module is a few months away. And then the final third is typically filed with that final submission process. And typically, that would see companies in our peer group try to get that filed within roughly two quarters. And so we're tracking along that. As you know, Maury, we also have Fast Track designation. So we would get expedited review. And now with RTOR, we would anticipate having a further acceleration of that review process since they're now reviewing the data real time. So I think that's going to be great and hopefully can pull in that launch timing.

Got it. And you commented on the potential to include the HLA-positive data in the filing. I guess, where are you at with conversations with FDA around that? Have you already spoken with them on that point?

Yes. So the quick summary is, yes. Discussions have occurred with the FDA on the HLA-A2 positive piece of it. And those discussions will continue, including for the pre-NDA meeting that's already penciled in the calendar. So -- but the preliminary discussions have occurred.

Got it. And so that will be confirmed at the pre-NDA meeting. Are you saying more about timing for when?

We're not saying more about the timing. But yes, that's correct. We would anticipate we'll have further clarity as part of that pre-NDA meeting.

Okay. And would that be included as a rolling -- part of the rolling submission? Or would that be a major amendment for the HLA-positive data?

So one is we would have that conversation as part of this pre-NDA meeting. I think, Maury, here we probably don't want to get into too many specifics on regulatory pieces there. But I would say kind of base view is if that is the path forward that is available, we will submit that as part of the NDA. It would likely be staggered from the full submission.

Got it. Okay. And you've highlighted an early OS trend at approximately 10 months of follow-up with the potential to extend that by 6-plus months. What are the key drivers and options involved with your next OS cut for FDA? And do you see a path to stopping the study early given control patients can't cross over to combo?

Yes. So we are seeing an early trend in OS for the treatment arm versus the control arm. We noted that as part of the top line results. That was with fairly minimal follow-up with just over 7 months, but I think that's encouraging when you see an early trend even with only a fairly minimal follow-up in the study. So in terms of additional updates moving forward on OS, our view here is that the current projection for the interim OS analysis is middle of next year. But you are correct, Maury, that at the time of the NDA submission, the FDA will want to have as part of including the safety database update, where we are on the OS events. And as part of that, because there is no crossover in the study, there is a scenario we may get full approval. But just so this is clear, that is not what we view as base case, but potential upside scenario.

Got it. And for the amount of time of OS follow-up that you could have, are you commenting more on what that could look like, what the range could be?

In terms of -- at the time of the NDA submission, no, we're not. The cutoff for the top line results was end of January. So I think here, it's going to be roughly 6 to 7 months more follow-up, something in that time frame.

Got it. Okay. And you showed -- you got the hazard ratio of 0.42 from the study, which is meaningfully better than KIMMTRAK's 0.76 hazard ratio. How do you think about pricing here? Could you price at a premium versus KIMMTRAK -- or what other variables should we be thinking about?

Yes. So that's work that is ongoing right now, Maury. Obviously, as the data evolves, we want to take that into consideration as we think about pricing. I know as you throw out other competing drugs in the space, KIMMTRAK is obviously one analog. Yes, I think depending on how the data shakes out, we do think there's an argument to be made for pricing at parity or even at a premium to KIMMTRAK. But ultimately, we want to make sure that we take the full data set into account as we make those decisions.

Got it. That's helpful. And for the pivotal study, what are you seeing on median duration of treatment?

Yes. I mean what we've seen so far is 10 months. So that's sort of what our clinical trial experience has shown to date. I think there's a path for that to extend beyond that in the real-world setting, depending on where we are. But that's sort of what the data has told us thus far.

Got it. Okay. And for later this year, you're going to have the HLA positive data update expected at a medical conference. How should we think about the frontline versus second line post KIMMTRAK split within the 85-plus HLA-positive patient cohort? And how are you setting expectations for median OS, median PFS and ORR relative to what you've observed with the HLA negative?

Yes. So the majority of the patients, Maury, will be pretreated patients. So we're not sort of specifying beyond that, but the majority will be pretreated. And as you noted, we'll have a full data set for all comers as well as the frontline patient subset. -- including response rate, PFS as well as median overall survival, which will be important, obviously, because we are going into an important interim OS analysis. And then this will be the second time we're now sharing median OS again, granted it is from a single-arm study, but I think hopefully a useful data point. I would say the high-level takeaway, Maury, hopefully, people will take from the data set is that our view is the data is very robust and also consistent with what we've seen in HLA-A2 negative. And we've said that in the past that the fundamental underlying biology is around the activating mutation of GNAQ/11, and it is agnostic of specifically HLA-A2 status.

Got it. And for patient baseline characteristics, overall, those are pretty similar to the HLA-negative patients.

For -- yes. For the frontline patients, we would anticipate that would be the case, but this is a much more limited data set than what we just shared at ASCO, right? That was at least from a safety perspective, and that was over 400 patients.

Got it. Okay. Makes sense. And you're going to have the neoadjuvant data update at a medical conference second half of this year as well. How mature will the data be across the plaque brachytherapy and nucleation cohorts? And how many patients are going to be evaluable for radiation reduction?

Yes. So there will be roughly just about 100 patients. There will be more follow-up from what was provided in the past. That will be split a bit more, I believe, nucleation than plaque therapy. We're not saying how much data as it relates to actual visual acuity data.But Maury, as you know, even for that, it's still, I would say, early. I would say we need to have continued follow-up. We will have a refresh on the percent of patients that we preserve the eye, which just as a reminder, that is the primary endpoint for full approval in the nucleation cohort. So I think for the plaque therapy, it will likely be more related to this predicted visual acuity tools that we've shared in the past. And I think for EFS as well, we may have some high-level commentary, but I suspect for event-free survival, it will also be still immature. But we are tracking the EFS data from the single-arm study piece. And since we're on the topic of neoadjuvant, one other I will mention is that similar to our strategy for HLA-A2 positive with compendia, we will also likely pursue the neoadjuvant indication as well in the U.S. for compendia based on this data that we'll be publishing.

Got it. Okay. So that could get added to guidelines. You could potentially start getting some uptake there commercially once you're commercial setting. Okay. And for the adjuvant study, I wanted to ask about that, too. So starting first half of this year. It seems like the base case -- base case scenario, the study could take about 3 to 3.5 years. What can you say on powering assumptions for interims relative to Immunocore's adjuvant Phase III, which assumes hazard ratio of 0.55 with an interim analysis at 76 events and 56% probability of early stopping.

Yes. So the study, maybe just kind of base parameters here. So it's 450 patients. Primary endpoint is superiority for relapse-free survival, target hazard ratio of 0.65. So our study is more powered than the peer company that you noted. The randomization will be 1:1. The randomization will be against observation. And importantly, it will be agnostic of HLA-A2 status. So much larger pool of patients will not require HLA-A2 status testing. So I think those are hopefully answered most of your questions there.

Yes. Yes, that's helpful. And let's shift gears to -- well, I guess, anything more about getting that study up and running and how should we think about that?

No, we're ready to go. So we had the Type C meeting with the FDA with Servierrecently. We essentially got everything that we had hoped for and asked for. So now we have obviously a very deep relationship with all the key sites globally. And I think what we can say is there's a tremendous amount of enthusiasm to get this study. And our anticipation is that this study should enroll very rapidly. So -- and again, here, the drug is clearly working in the metastatic setting, including at least based on single-arm data survival. So we think this is a very high probability of success study. We're essentially randomizing against nothing.

Got it. Makes sense. So let's shift gears, talk about DLL3, your ADC there. Maybe starting off with the small cell update. How are you setting expectations on PFS? And could we get other efficacy details based on the patient's prior line and/or breaking it out by dose.

Josh, do you want to take that?

Yes. So we'll have 2 updates on DLL3. The first is actually Hungary's update from their Phase I study in China. There, we anticipate updates in both small cell and neuroendocrine. It will be a refresh of the data they shared last year at the World Lung Conference. We'll see updated PFS, obviously, safety, response rate, PFS. And then importantly, for the first time, landmark OS data. So that will be sort of, to our knowledge, the first look at how a DLL3 type 1 ADC performs from an overall survival standpoint. So that's very exciting. And then in parallel or sort of a similar time frame, we're going to have our update from our ongoing Phase I global study. And I think there, the focus will be primarily small cell. We do have some early look at safety and probably response rate. I don't think we'll have enough to comment on PFS at that point, but we'll see. And we anticipate that will be roughly 30 to 40 patients.

Got it. Okay. Helpful. And should we expect intracranial overall response rate data in patients with baseline brain mets in this update?

Yes. I would expect -- definitely Hungary had that data set originally in their data last year. We expect there'd be a full refresh of that as well. So yes.

Got it. And for the next -- the NEC population, should we expect a similar number of efficacy evaluable patients relative to Zai Lab's update where they had about 34 evaluable? And how should we think about tumor heterogeneity in the setting?

Yes. So the exact numbers, we're not going to specify here, Maury. But look, there will be enough patients on the neuroendocrine carcinoma. This is not for Hungary's data, where I think people will have a good sense of what we're seeing both in terms of response rate as well as progression-free survival. In addition, we've also begun enrolling in the U.S., Europe, outside of China, and we've also -- probably about half of our patients have been in neuroendocrine carcinoma. And really, the quick summary there is we believe we're seeing activity that there should be a monotherapy approval path forward here based on what we're seeing and based on what else is out there. In terms of the heterogeneity of this population, maybe Mike, you can talk about that. But yes, there are several indications of note. Some have higher DLL3 expression than others, but Mike?

Yes. We're obviously focusing on those indications where DLL3 positivity is known to be high. And that's something that we also are excited about with respect to IDE161, our PARG inhibitor. So what does heterogeneity mean? It means for an ADC that you could get less tumor exposure to the payload. IDE161, the PARG inhibitor, its mechanism of action is to amplify the therapeutic effects of what would otherwise be a suboptimal delivery of that payload. So that's where we see a real opportunity in the combination setting.

Got it. And for picking tumor types, that's one thing, but I guess, could you select for patients that have high DLL3? I think Amgen's moving forward with a higher 30 mg dose, and they're seeing better efficacy in patients that have a higher DLL3 expression.

Yes. So for small cell lung cancer, we don't think that's going to be needed. For DLL3, that's something that we're discussing. So I think we don't want to give too much visibility in our regulatory strategy for competitive reasons. But I think here, I think more big picture, there are some neuroendocrine carcinomas that have very high DLL3 where we think that's not going to be needed. And then you can think about following that with, let's say, a pool of NECs where something like a diagnostic could be more helpful. So more to come on that front. And hopefully, we'll be able to share more before the end of the year on that. But we are clearly -- we have our strategy, and I would say we're -- we have a strategy on what we're going to do on that question.

Got it. Okay. Mike, you mentioned the PARP combo. Could we see some data from that by the end of this year?

We're pushing the team. So the dosing has just begun. For those that may not be as familiar, this is a proprietary combination that we have clinically. Our view is that the PARP mechanism may have the opportunity to extend the durability of the topo ADC class. We have very strong preclinical data to support that. We've now begun the combination dosing in patients with DLL3. So we'll push as hard as we can, Maury. I think a lot of it will just be about how many patients do we have. And to really get good visibility on this question, I think durability is going to be important.

Got it. Okay. And at ASCO, I was part of conversations around KAT6. People were talking about KAT6 and breast cancer. You've got the KAT6/7 that you're dose escalating in breast, prostate, colorectal and lung. Can you talk about the strategy here and what improvements you want to see in some of these indications, including breast cancer?

Yes, Mike?

So this is a very exciting asset. This is something where I think people are really starting to appreciate the importance of maximally suppressing the epigenetic mechanism. And to do that, you need to bring KAT7 on board. And so having a dual KAT 6/7 inhibitor, we've shown -- now Pfizer has shown as well that, that's an optimal profile to be able to drive monotherapy responses in a preclinical setting. We're expecting that to happen in patients as well. very excited about the opportunity in p53 mutant MSS, colorectal cancer, especially because that tumor is very heterogeneous because of epigenetic mechanisms that promote diversity. We intercept that with a KAT6/7 inhibitor. If you don't have KAT 7 inhibitor -- inhibition on board, you can't get that phenotype. We're also very excited about large lung cancer cohorts that are addicted to the alveolar type 2 lineage program. That's a large section of lung adenocarcinoma. Again, you cannot get efficacy in models, PDX models without having KAT7 on board as well with the breast cancer setting. Estrogen -- positive estrogen receptor positive breast metastatic breast cancer, we're expecting to see monotherapy activity there equivalent to what you might otherwise get with a KAT6 inhibitor together with an estrogen inhibitor. The reason the KAT6 inhibitor needs the estrogen inhibitor is because it cannot sufficiently suppress that pathway on its own. A dual inhibitor can do so independently of the presence of an RB mutation, independently of the presence of an estrogen receptor mutation. So we're very excited about that one.

And I think we're out of time. So maybe to close up, if you just want to highlight key catalysts ahead for IDEAYA.

Yes. I think the message for the second half of this year is we have a lot of catalysts upcoming. So I know we've covered a lot of it already today. But for darovasertib, obviously, everything that we're doing on the NDA submission in the second half, commercial preparation. We have about 100 patients worth of data in HLA-A2 positive. That's already been submitted for a major medical conference. The roughly 100 patients in neoadjuvant as well has already been submitted at a major medical conference. We have the DLL3 Phase II Topo-ADC. Our partner, Hungary will present about 100 patients worth of data in small cell neuroendocrine, including with landmark OS data. We'll also present on our side from the U.S., Europe, non-China studies. And we're pushing hard also on the bispecific ADC. We didn't cover B7-H3/PTK7. We have a clinical update. We're guiding towards by the end of the year. And obviously, now with our Roche collaboration moving forward, hopefully, we'll be announcing our MAT2A PRMT5 FPI combination start here very soon as well. So look for a very catalyst-rich second half.

Got it. Thanks so much for joining us today.

Thank you so much.