Great. Li, thank you so much, and thank you to Cantor for the kind invitation this year. IDEAYA Biosciences, we're a leading precision medicine oncology company. We have a deep, diversified pipeline. Our most advanced program is darovasertib. As Li, as you're aware, we're going through the final NDA submission process under RTOR. The first three modules have been submitted. The final will get submitted here very shortly, and that's for our attempt to get approval in the first-line metastatic uveal melanoma setting. Beyond that, we have several additional studies ongoing in the pre-metastatic setting, including a phase III adjuvant study, which we announced first patient was dosed this morning. Beyond that, I would focus in on two additional programs. First is IDE849. This is our DLL3 TOP1 ADC. Here, we have a significant focus in small cell lung cancer and neuroendocrine carcinoma.
We did recently announce successful Type C meeting with the FDA to finalize a phase III study design in later line small cell lung cancer. Beyond that, we have a significant focus in MTAP deletion. We have two clinical programs here. We've also announced two recent collaborations with Roche, with both our pan-RAS inhibitor, KRAS G12D, with a primary focus on pancreatic cancer. We have much more in the pipeline, Li, as you know, but I think if we cover that, we'll cover a lot of the company.
Okay, great. That was a very good high-level overview. I wanted to start with darovasertib submission. Yujiro, you mentioned you're going to submit the second module very soon. What are the gating steps here? I also know you guys have a pre-NDA meeting scheduled. What will be on the agenda for that meeting?
Yeah. In terms of gating items for the final NDA submission, there's nothing specific that's gating now. It's essentially QC that's ongoing for both the overview of clinical efficacy, overview of clinical safety, and the overall clinical overview for the program.
That is really all that is remaining. That process is going well. We very much remain on track, so we feel very good about what that is. In terms of the pre-NDA meeting, that action meeting has now occurred.
The primary discussion that was had as part of that pre-NDA discussion has been around the label and should we be pursuing an all-comers versus an HLA-A2 negative? I would say here, we have been pleasantly just hearing the enthusiasm around at least a potential opportunity for an all-comers approach. We will have to make that final decision when that final portion of the NDA is submitted, and we will likely provide a public disclosure on what we decided to move forward with. But at least at this time, it is at least good to hear the receptivity from the FDA.
Okay. Yujiro, you mentioned, in terms of the label, the base case scenario for you guys is that you are going to have HLA-A2 negative patients. But it sounds like your dialogue with FDA has been pretty positive in terms of HLA agnostic patient population. Do you feel better about you may be able to get an all-comer label at this point?
Yeah, I would say from when we started the process with RTOR. Which was largely from when the top-line results happened, which as you know was in that April timeframe. We started that RTOR process within weeks of those top-line results to today. Yes, we do feel more encouraged based on the discussions we've had through RTOR with FDA.
Okay. In terms of thinking about this HLA positive patient population, you guys also mentioned you might be able to pursue compendium listing as an alternative. Maybe talk to us about what the next steps are if you're going to go that route.
Yeah. So here, this would be largely based on the published data.
So, as you know here, Li, we're going to be publishing data in HLA-A2 positive at ESMO here shortly. I think, as well, we would hope to be able to put that in manuscript form as well as in the future. Then assuming we do get approval and launch, we would then pursue that NCCN guideline process. Typically, they meet once a year, but based on the high unmet need, we think there is an opportunity to hopefully have an ad hoc discussion on that.
Mm-hmm. So it sounds like in terms of the sequence of events, is that you guys are going to be publishing all the data first, get approval, and then pursue a compendium listing. Is that reasonable?
That's correct.
Okay. Then, Yujiro, you mentioned that you guys going to be sharing some data at ESMO, maybe some OS data. Maybe give us a quick preview of what we should expect.
Yeah. First, for those maybe more or less familiar with metastatic uveal melanoma and also GNAQ/GNA11 mutation, we believe 95%+ of these patients are going to have either an activating mutation of GNAQ/GNA11 or further upstream. The mechanism of action at darovasertib is not connected to specifically HLA-A2 status, so I think that's really the first place to start. From our perspective, we don't anticipate to see a difference in activity, whether that's in the HLA-A2 negative population or HLA-A2 positive. The data we'll share at ESMO will be about 100 patients. About 80% of those patients will be pre-treated, but we will have a subset that are first-line patients. Here, we'll share a very robust data set, including response rate, progression-free survival, and median overall survival.
I think there'll likely be the most attention to what that frontline survival data looks like. Our view is the data should be very similar to what we observed in HLA-A*02 negative.
That data will only be in HLA-A*02 positive patients, correct?
Correct.
Okay. For the NDA submission, you previously mentioned that an early look on OS will be included. Can you say, at this point, have you guys done that OS analysis yet as part of the submission?
Yeah. Once the final NDA submission is sent in, there is a window period, I believe it is 60 to 90 days, you have the ability to refresh that.
So, as part of that, we will provide a refresh of OS events. It is not a specific interim analysis that is being done on survival, but the OS events will be refreshed and provided to the FDA.
Then in terms of the OS data, the interim OS data, I assume that is the official interim analysis that you will be presenting next year. How does that OS readout impact your pricing strategy, especially as we think about you have a competitor in the HLA-A*02 positive patient population, and in Europe in particular, I think overall survival matters a lot.
Sure. Stu?
Yeah. I think it's really more the latter. I think for Europe, overall survival data will be really important.
With respect to pricing and reimbursement discussions. In the U.S., we don't think it will be as important. I think based on the strength of the data that we have shared, and we will add to at ESMO, we think the data package that we'll submit with is very strong and should drive strong adoption, both in the A2 negative and positive subsets here in the U.S.
Mm-hmm. As you guys are just in this process of NDA submission, I wonder if you can share a little bit about your pre-launch activities, especially around payer discussions, pricing strategy, and sales force infrastructure build-up.
Yeah. We're working to be launch-ready in the first half of next year, consistent with some of the timing that I think is out there. What's been a big thrust of that is obviously, as you mentioned, thinking about our pricing strategy. I don't think we're going to comment on that here and now today, but that work is well underway. A lot of the other emphasis has been on building out our sales force and our sales infrastructure. That will look very similar to one of our peers that launched in that setting. It's a relatively modest field force, and all sort of on the order of 25 to 30 reps. Where we are now is to have leadership in place. We have our reps that are being brought on board.
We'd expect to have them in place by the end of the year ahead of that early next year launch potential.
Okay. So it sounds like in terms of the size of the sales force, then maybe the launch trajectory, we should maybe looking at maybe ChemoCentryx as a good comp. Is that reasonable?
Yeah, I think that's obviously a good benchmark for us. They've launched in this space before with a fair degree of success. I don't think you should expect too many differences with how we're thinking about building out our commercial infrastructure.
Okay. In terms of the new adjuvant trial strategy, I know you guys recently have some updated thoughts. I think the ongoing phase III trial faced some enrollment challenges, and now you're weighing the possibility of either do a phase III study or maybe go for compendium listing strategy. Talk to us about how you're thinking about the new adjuvant piece overall.
Yeah. I think, Li, here, there's three things that we would highlight on that question. First, as we've been doing market research in the new adjuvant setting based on the data that we already have, which is about 100 patients of single-arm data, our market research, the feedback that we've been getting is the view that at least our potential prospects for an NCCN strategy, the feedback we've gotten has been quite positive and perhaps more positive than we had initially anticipated. That's the first. Second, as you noted, we've had delays in enrollment, and we believe that's related to a stricter eligibility criteria that's essentially caused more screen failure rates. That's the second.
The third part of this, I think, is connected to the announcement we made this morning, which is now our phase III adjuvant study has now started enrollment because as you can imagine, these pool of patients to some degree do overlap and compete, right?
If you put a patient on the new adjuvant study, they would not then be eligible for the adjuvant trial. I think these are all the things that we're trying to incorporate to make that ultimate decision. I think ideally we're in a position to make a decision on or around by the end of the year, is our hope.
Okay. And in terms of NCCN listing strategy, what needs to happen here? Is it just a matter of accumulating data? Is it just a matter of discussion with the committee? Can you talk to us about timeline?
Yeah. I would say that it would be very similar to what we described for the HLA-A*02 positive metastatic uveal melanoma, if for some reason we did not pursue that all-comers approach.
We would continue to publish data, which, as you know, at ESMO, we'll have about 100 patients' data presented in neoadjuvant, so that will be important. Ideally, we also have data that's put in manuscript form as well. Assuming we get approval in the first-line metastatic uveal melanoma setting, in a similar way that we described it, we would pursue that compendia approach.
Mm-hmm. I wanted to switch to the DLL3 ADC program in small cell lung cancer, which I think is a very exciting program, and we're going to have some data at ESMO and later this year. I think very timely news this week that Amgen announced their frontline trial of tarlatamab succeeded. We haven't seen any data yet, but just from the press release, it sounds like overall survival was very good. I guess, what's the implication of that trial on your DLL3 program?
Yeah. Look, overall big picture, our view is that that data set is great news for us, and I'll go through why we believe that's the case. First is, as you know, Li, small cell lung cancer is a competitive indication, right? There's multiple companies pursuing multiple targets for small cell lung cancer. But we believe now having frontline OS data with a DLL3 target showing a survival benefit just further validates why we believe DLL3 continues to be the most optimal target for this indication.
As you know, that's very connected to the underlying biology of this indication. Second, the reason why we were excited about that result from Amgen is that we think it continues to support why we believe we have the right phase III study design w here we've had this recent FDA Type C meeting, where we're going to pursue and evaluate a post-IMDELLTRA population. One point of feedback that became very clear to us was we were very clearly given this communication around ideally being able to pursue more of a homogenous population where if you did a pre and post-IMDELLTRA, it would be more of a heterogeneous population. Then the question will be, at least for the U.S. approval portion, which really patient data set is relevant.
We think we'll be uniquely positioned. We think that puts us in a good spot from a probability of success. Now with clearly IMDELLTRA looking like it's moving, now we'll be moving into frontline. That would essentially move this hopefully more into a second-line setting, at least for a subset of patients. Then we'll independently evaluate what that frontline strategy may be.
Okay. At ESMO, your partner Hengrui will be presenting some data update as well in about 100 patients, which will be very informative. Can you set the expectations for us, and what kind of data would support a best-in-class claim?
Yeah. First, the data that will be shared at ESMO will be a sizable data set. It will be 100 patients. It will be largely small cell lung cancer, but you will also see, for the first time, some patient data on neuroendocrine carcinoma. The clinical efficacy data that will be provided will be very extensive.
It will include response rate, progression-free survival, and here I would just highlight a much more mature follow-up than what was shared at the World Conference on Lung Cancer last year as an oral presentation. Then I would say importantly, we will share landmark 12-month OS data, which has not been shared, at least from what we understand, in this specific target with modality of a topo ADC. What does a win look like, and what does potential best in class look like? I think, look, directionally, we want to see a higher response rate than what has been published before, and this is, just to be clear, a confirmed response rate, ideally in that 60s range. Second, I would say a progression-free survival, as you know, IMDELLTRA
for their accelerator approval, reported a PFS of four months. I think others reported it in the five-month range around the DLL3 TOP1 ADC. We reported a PFS between six to seven months.
We think that clearly would be a win. On the OS side, we know in that later line study, IMDELLTRA showed, I believe it was an OS of approximately 12.5 months, right? If you can be above that 50% threshold at the 12-month OS landmark, I think that's clearly a win. We believe the topo ADC should bring some clear advantages from an AE profile versus T-cell engagers. That's the opportunity, and we continue to have conviction around our molecule we do think is positioned to be best in class.
You also have your own global trial going on, and you'll be sharing data later this year, I believe in 30- 40 patients, right?
Yeah, that's correct. It'll likely be north of 30 patients. It's going to be largely driven by how much follow-up we have, Li.
We did share all of that data with the FDA recently through the Type C meeting, where we aligned on the doses.
We did get alignment with the FDA that the dose should be either 2.4 mg/kg or 3.5 mg/kg. We know our partner Hengrui has already decided on the 2.4 mg/kg dose.
As you know, our closest peer here has picked a dose of 1.6 mg/kg.
Yes.
Already at 2.4 mg/kg, that means we are delivering 50% more drug, which means we are delivering more payload to that tumor. At 3.5 mg/kg, we would be at double the dose, right? I think that is great news that we had alignment that those two doses were indeed viewed as the two doses we should be evaluating. We will be interrogating that and make that final decision as this phase III study gets off the ground.
I guess for that data update, number one, do you expect that to track with the China data from Hengrui? Number two, how much data are we going to get in the post-IMDELLTRA setting? I think that is very relevant for your phase III.
Yeah. Hengrui has seen patients post-IMDELLTRA. We have seen that data, Li, and what we could tell you is we are clearly seeing responses post-IMDELLTRA. Also based on the biology, as you know here, we don't anticipate DLL3 downregulation as a key mechanism of resistance. We're not concerned about DLL3 being followed by DLL3, and this is, as you know, connected to this key lineage survival oncogene via ASCL1 of which downstream are key Notch pathway, areas including DLL3, impacts the expression. We don't think that's a likely outcome. Lastly, I will mention two more things. Most of our patients that we're seeing in our study are post-IMDELLTRA.
We do have a handle on that. We know we're seeing responses there, so we feel good about that. The last piece I will mention is that we do believe there should be an opportunity with monotherapy for our program in neuroendocrine carcinoma. We're not saying which subtypes at this point yet, but I think that's another hopefully update we'll be able to provide as part of our update, and you'll see some early data on that from Hengrui as well. This is an opportunity that does extend beyond small cell lung cancer. I think this is important to note because, A, it's an area of very high unmet need. It's not clear to us what the standard of care there is form a lot of these subtypes of NEC. Second, that seems to be an area where IMDELLTRA has had less success.
Right? Perhaps that's because it's an IO mechanism where an ADC could be hopefully much more effective there.
Can you talk a little bit about your frontline strategy?
Yeah. We're continuing to evaluate that. I would say the permutations that you want to think about for frontline, so one is PD-L1. We think that's table stakes. You could consider a bispecific, although we don't believe that's table stakes in this conversation. You could also consider plus or minus chemo, although I think one of the big utilities of ADCs is so you can do it without chemo.
There's also the question about do you have TC in there or not? I think based on this recent data, it's likely that you'll want to include it in the control arm, right? So that means that benchmark is not going to be trivial, and that's where we don't want to rush into a frontline study because as you know, Li-
Yeah.
-we think one of our key points of differentiation could be around this PARP combination.
And if that could deliver more durability, which is the hypothesis.
Yeah.
That would position us very uniquely, and particularly in the frontline setting. We want to generate this data from now till next year, and be in a position to hopefully launch a frontline study. But it needs to be data-driven.
Okay. Sounds like you guys may be considering some novel combination as part of your frontline strategy.
I would say that's our preference.
Okay.
Yeah. Versus just going in with PD-L1 ADC. You could do that, and we know others are evaluating that.
Yeah.
But we do think that there's more risk involved there.
And for the phase III trial that you guys are going to start by the end of the year, correct? You mentioned that the interim analysis could support a potential accelerated approval. I assume it's going to be based on response rate and DOR. Anything you can share about timing? When should we expect the top line to read out?
Yeah. We haven't given the specifics, Li. I think once we get the enrollment up and running and we have sort of line of sight of how quickly we think we can finish out the study, we'll give more visibility into that.
But really two key points I would mention. One is we do have fairly aggressive timelines on how quick we think we can enroll this, including in the post-IMDELLTRA setting. We ultimately have to prove that and show we can do it. In terms of the timeline to read out, as you know here, for response rate, that's going to be fast.
So I think that should be a positive, versus, let's say, using a PFS type readout, right?
So I do think there's an opportunity to hopefully get to this readout quite quickly. And that's not probably being fully appreciated right now.
Yeah.
How quickly this could potentially come online if we hit that endpoint.
Especially on the way we set it up, because we think the control arm bar now is very low. We did get alignment with the FDA, what needs to be in the control arm. So it's topotecan and irinotecan.
They felt that we didn't need to have ZEPZELCA in the control arm, so that should be a very low bar in high POS, and now it's this homogenous population. It should be a clean dataset.
Can you touch on the market opportunity for you guys, given we're going to have DLL3 ADC approved in the near term, and then there is a DLL3 ADC competitor ahead of you. Would love to hear your thoughts on how you're thinking about the market size.
Yeah. So we're not going to state a specific percent market share at this point, but I think here, Li, what it's going to be driven by is the quality of the data, right? So maybe there's sort of two portions of that we would highlight. So one, on the general safety side, and I'm going to make some general comments and statements here, but I would say directionally, it does appear the DLL3 TOP1 ADC class so far has shown better safety profile.
We think that's really related to the antigen. It's a more specific target to the cancer type of small cell versus some of these other antigens like B7-H3. Second is we think directionally the efficacy we've seen in small cell in particular, and I would include TROP-2 into that. I do think we've seen directionally better efficacy, whether that's response rate, PFS. We obviously have more data than the public does, but I think that's our perspective. When all of this settles, I think we think DLL3 TOP1 ADC will become the anchor in small cell lung cancer, at least for that modality, in the same way IMDELLTRA is, right?
You can imagine DLL3 the same way you think about HER2 in breast cancer, right? We saw that with obviously, on the antibody side of HER2, we then followed that. We saw that follow with ENHERTU. So we think there could be a very similar parallel here in the indication of small cell and neuroendocrine carcinoma, one using a T-cell engager, another using an antibody drug conjugate approach, but with this focus on DLL3.
Mm-hmm. Another very exciting opportunity that's emerging from the pipeline is the PRMT5. Yujiro, as you mentioned, you recently had some Roche collaboration. So maybe just, number one, talk to us about the high-level strategy in the RAS space. Number two, what can you share about cadence of the data that you guys are going to be generating over the next 12 months or so?
Sure. So I'll take the second part, and we're fortunate to have Daniel Simon, our Chief Business Officer, here. He's been the architect of the two Roche collaborations. So Daniel, you want to take that?
Sure. We are super excited to be partnering with Roche Genentech on these two collaborations. They are reciprocal in nature, so for the pan-RAS combination, Roche is providing that drug to us, and we are running and paying for that study. Then we had a second iteration, and Genentech is running the G12D study, and we are providing the PRMT5 program for that. For the first one, that is roughly 40% of pancreatic cancer patients between MTAP deletions and the RAS mutations. Then within pancreatic cancer, G12D is roughly 40% of the RAS mutations. That is about 15%-16% of the pancreatic cancer population.
Clearly, some super exciting precedent data from Revolution Medicines and Tango. It is great to see a big pharma company so motivated to get that moving. We are excited to get that into the clinic and starting.
Okay.
Yeah, maybe, Li, just the second. Thank you, Daniel. I think the second part of that, just to frame some data cadence and how we are thinking about that.
Yeah.
The way I would frame it is there is really three categories people should think about in terms of what we are doing in MTAP. First is around IDE892 PRMT5. We would like to be in a position to pursue a monotherapy, ideally registrational path forward, ideally, hopefully in a next year type timeframe. We do think lung cancer is a viable indication. We have other indications in mind, but we do not want to disclose that right now for competitive reasons. But that is something that we would like to be able to move forward with. Second, I would say, which is unique to us, which is the ability to combine PRMT5 with MAT2A. There, our biggest focus will be in lung cancer.
I think we will continue to generate that data from this year to next year to hopefully be in a decision to both have a data disclosure on monotherapy for PRMT5, then the MAT2 PRMT5 combination. Ideally, there is a path forward with that. Lastly is what Daniel just walked through. I would say this clear intersection between MTAP and KRAS in pancreatic cancer, and this is a big population, right? We are talking about 40% of the population. In addition to delivering high responses and hopefully durability, we do not know that full answer yet.
We do think a key point of differentiation is around these drug-drug interactions, and we do think in the area of pancreatic cancer in particular, this is important because not only does it impact pan-RAS, but at least that we are aware of, largely the molecules that we are aware of, both G12D, pan-RAS, PRMT5, and MAT2, the key metabolism enzyme is CYP3A4, right? In addition, you have to think about the standard of care chemotherapy. FOLFIRINOX primary enzyme is CYP3A4. Docetaxel secondary enzyme is CYP3A4. So we do think this is a very core part of our strategy to differentiate in pancreatic cancer.
The last part I will mention on this, I think beyond our Roche collaboration, we know there is more interest to do work with us, whether that is on pan-RAS or G12D. I think we are just trying to define ultimately what do we do on that front. The last piece I will mention, which will give you a little bit of a sneak peek of R&D Day in London, and hopefully some of the people in the audience could be there for that, which is a new program in CDKN2A
that is hopefully going into the clinic in the first half of next year. One of the concepts that we wanted to explore is this idea of hitting the key oncogene in pancreatic cancer and the key tumor suppressor loss gene. We know that the really bonafide tumor suppressor loss gene in pancreatic cancer is CDKN2A.
That prevalence is now much higher than MTAP, right? That goes from 40% to up to 70%-80%. You can imagine you may not need to do patient selection there. Combine it
You can combine it with PRMT5. You can also then combine it with pan-RAS in an unselected patient population. So far, the data we've been seeing pre-clinically, which is part of what we'll showcase in London, we think has been very exciting. That's we are differentiated, that we are, you know, we think we stand alone uniquely.
Yeah.
So now with PRMT5, MAT2A, and if this molecule can enter to the clinic, we think we're going to have a lot of optionality.
Yeah. Looking forward to learning more at the R&D Day.
Absolutely.
Thank you so much, guys. That is all the time we have today.
Great. Thank you, Li.
Thanks, Li.
Thanks for your time.