IDEAYA Biosciences, Inc. (IDYA) Earnings Call Transcript & Summary

And welcome to our lunchtime deep dive this time with a non-covered company, IDEAYA. The goal of these calls is to kind of help preview and set kind of expectations and kind of set the data into the perspective of our coverage universe and the broader oncology space. And it's the third year we've been doing these fireside chats and hopefully, you find them useful, really appreciate the feedback. And of course, it's [ II ] season so, appreciate any votes. And just a wave of orientation here, it's a 30-minute call. If you've got questions, do e-mail me, I do read them out anonymously. I try not to mangle them too badly, but -- and then you can get me on Bloomberg. And so e-mail address as well, of course, is peter.lawson@barclays and it could be great pleasure to introduce Yujiro S. Hata, CEO from IDEAYA. So time permitting, would love to cover the PKC inhibitor, so the uveal melanoma, the MAT2A inhibitor and PARG.

I guess the first question really around the PKC inhibitor and -- so daro and the combination with crizotinib, so the cMET inhibitor in uveal melanoma, kind of, why does that combination work so well? It drives really deep responses. And if you can kind of walk us through the strategy for that HLA-A2 negative versus the positive in metastatic uveal melanoma.

Sure. So first, Peter, thank you so much for the invitation to participate in your fireside chat series today. So first, your first question on PKC cMET and sort of why we think this is an important combination. So this is work that we had published a few years ago at AACR in a peer-reviewed setting. And the initial discovery at the bench occurred as we were evaluating our clinical samples from our Phase I monotherapy clinical studies. And we saw a clear correlation between patients that observe partial response versus progression based on a met activation signature that we identified. And that was really the first discovery that made this connection in the metastatic setting of the liver, we saw this phenomena. We then followed that with cell-based Synergy Data. We shared that data with Pfizer. And that was the package that was critical that ultimately got Pfizer engaged to work with us on this combination. And here, Peter, just to go into a little bit more specifics, as you know, in the predominance of these patients, so 95-plus percent of uveal melanoma patients, it's believed that GNAQ GNA11, which is an activating mutation is present in these patients. That is the viewed as the driver oncogene for causing these patients to have uveal melanoma, which is known to activate the PKC signaling pathway. And that is also why our view that our agent should be active as well as the combination irrespective of HLA-A2 status. In terms of our strategy in A2 negative versus A2 positive, we did have fairly recently a Type C meeting with the FDA. And as you know, we announced publicly, we are now the first sponsor that has been given the green light to pursue a first-line accelerated approval study in metastatic uveal melanoma. For the registrational trial, although, as I mentioned, we believe the combination will be active irrespective of HLA-A2 status. For more regulatory reasons, which we believe is the likely highest POS as well as the fastest path to approval, focusing on the A2 negative population was the best strategy forward. With that said, we absolutely do plan to target the A2 positive population as well. So in addition to our Phase II/III registrational trial that we're launching, currently, our existing Phase II study, we've decided to modify that to focus only in enrollment in A2 positives. And so here, the strategy would be a compendia type strategy. This is a strategy that other companies have utilized. And in fact, there's been several recent examples in the precision medicine oncology space. So here, as you can imagine, we would publish a certain amount of data, that conversation still is ongoing internally. But just for this purpose, let's say, we increased that enrollment of 50, 60 patients in A2 positive, publish that. The compendia process is typically a 6- to 9-month process. This obviously assumes in this scenario, the drug is approved in the first-line A2 negative population. But beyond that, I would just also highlight, I think, 2 key items here, Peter. First, as you know, we've noted that our view that HLA-A2 negative is the majority of metastatic uveal melanoma. And as you can appreciate, there could be quite a big difference if that majority is 55% or is at 75%. So we believe we probably have one of the largest databases that's unbiased because our drug works in A2 negative as well as positive. What is that actual ratio because we have not seen that data published to date. You may have seen there was a smaller number that was just published at ASCO. You may have seen that Peter, but they actually noted about 65% were A2 negative. And that was just based on the ASCO data that just published.

Would you end up publishing your data?

Yes. So that -- what we saw in the second half of the year, we've guided towards clinical program updates for darovasertib. And we were, I would say, specifically not exactly precise on what that would be. That could be around the status of the program as well as potentially multiple clinical data updates as well, including in the neoadjuvant and in the metastatic setting. And one of the key questions that we are getting is around the HLA-A2 status, negative versus positive prevalence because we are probably one of the few sponsors that has a growing database that's unbiased again because our molecule works involve what is that ratio. So our database is starting to get fairly large. So it's -- now we've tracked probably about 150 patients. So I think it's starting to become a large enough denominator. Statistically speaking, we think is quite relevant. And that's obviously critical because that would hit our bottom line, right, in terms of also maybe complete the picture of why we are so focused on A2 negatives. And I would say a big part of that is some of that data we've been generating.

And kind of how should we think about the flow of data through the next 12 months or so, whether it's existing drugs that are running or your kind of registration trial? How should we think about data flow from that?

Yes. We haven't given the specifics of that guidance with the exception of the program updates in the second half of this year. But what I could tell you is we are being very responsive to key questions that we know we are getting from investors as well as analysts as it relates to the program and obviously, take that into account as we develop our strategy around data disclosure. So maybe there, I can highlight what are some of the key questions we're getting, Peter, to sort of hopefully, give you a sense of how we're thinking about this. And I think this would be largely in that time frame that you noted. So one is on the neoadjuvant uveal melanoma side, as you know, we have an IST that's ongoing. We have a Phase II study that we're getting up and running. We actually just got our first site activated for the neoadjuvant company-sponsored Phase II study. This is a neoadjuvant adjuvant study. I would say a tremendous amount of appetite and enthusiasm to see more clinical efficacy data there. And probably, one of the key questions we do get is, it would be great to demonstrate that you've been -- that the molecule has shown an ability to save eyes as a monotherapy. I think that's fair feedback. And so that's something that we're thinking about on the neoadjuvant side. And then on, I would say, in the metastatic setting, I think here, as I just mentioned earlier, was around this HLA-A2 status and what might the data look like in A2 negative versus positive. What is the prevalence that we're seeing in A2 negative versus positive. Just so you're aware, we've also generated data around ctDNA as well. Obviously, that data set has become important in this indication for reasons you're aware of by other sponsors. I think [Technical Difficulty]

We've lost your sound [indiscernible].

Okay, sorry about that. So just on the ctDNA side, that's also data we've generated as well. So I think those are the type of things that we're evaluating currently. And obviously, at least our perspective is the ideal scenario for an agent as you can demonstrate anti-tumor activity, not just at a ctDNA level, but from CT scans as well, right. I mean that's usually the ideal situation.

And then the data we've seen so far has been very encouraging for response rate, initial PFS data. Do you expect to be able to drive OS benefit as well?

So I think there, what I would just highlight, Peter, is that we've actually reported median OS for this program as a monotherapy agent. And I would just note that, that was in a very -- in a much heavily pretreated later-line setting, and you may know that data, but we have reported a median OS that was 13-plus months. This -- the synthetic control arm that people often reference is Rantala, which comes in at about 7 to 8 months. Here, as you may know, [ TABI ] did do a later-line study, which is their 102 study. I think it was about 125 patients. They did report a median OS from that study. It's one that's not talked about a lot. But I would say it was very comparable. And I think the last point I would just highlight for you is LDH status as a parameter in this indication that cannot be underscored when it comes to metastatic uveal melanoma. So one of the largest meta-analysis that has been done in this indication is the CoGa analysis. So I think it was several thousand patients. And they clearly demonstrate the single biggest parameter that actually impacts OS is LDH status. And in fact, when you look at that data, it spans quite broad. I believe that's down to even 6 to 24 months based on LDH status. So it has a dramatic shift to OS. And we even saw that recently in the [indiscernible] paper that was reported on the Ipi-Nivo combo, where they see -- based on LDH status, the median OS goes down to 6 months, if it's sort of, kind of your typical range, about a year. And this is -- and the reason I'm kind of highlighting this for you is because also, when you look at the TABI study, even if you just look at the comparator arm, the median OS that was reported there was 16 months, right. That's actually at least the longest median OS we've seen in that comparator arm reported at least that we've seen before. And we think that could be partly driven by LDH status. So I know that was a long answer to your question, but we just wanted to highlight 2 points. First is we've already reported median OS as monotherapy. Our anticipation would be the combination, hopefully, should deliver something better than that. And then second, in the context of LDH status where the LDH status of that study as monotherapy was very poor.

And then the OS for the combination, when could we see that? Is that something that's kind of expected to readout? Or should we be thinking about the pivotal for that OS number?

Yes. It's a good question here, Peter. I mean I think at the end of the day, what time to have been endpoints, one has to just be mindful of that, especially from a regulatory perspective, it's about the hazard ratio, as you know. And I think lastly, as I mentioned around this LDH status just because LDH status does make such an impact as we saw even as I just mentioned in the example, the comparator arm in another study. Yes, I think one has to be very, very thoughtful of that. And obviously, that's where the randomization will give you the more, I would say, insightful readout and it does come down to the hazard ratio. So I know I didn't exactly answer your question. But at the end of the day, what matters at this point now is what comes out of the registrational study. We clearly are seeing a signal from our Phase II study by any parameter that you look at, at least we think the data thus far from an efficacy perspective has been unprecedented. And now it's going to -- now it's up to the registrational study and to have us hit our endpoints.

I think I may have lost your sound just now again. But I guess another question would just be around how quickly you can get approval? Kind of what's the time frame you're thinking about for commercialization?

Yes. So we have not given guidance around that, Peter. So but I can answer you with 2 pieces there. First is our analysts that cover us are typically in that 2016 (sic) [ 2026 ] time frame for actual launch. So you obviously have to submit that. Before that there's a review period, but that's kind of generally where a lot of -- most of our analysts are in that '26 launch time frame. The second piece of this, as you know, that answer will be largely dependent on the cadence of enrollment. So all we can say there is that the demand that we've seen thus far to get on to this study has been phenomenal. And even with a very limited number of sites, about 12 sites in the U.S., we have enrolled as high as a clip of even 15 patients a month. So now we're going to be going obviously much broader in the U.S. We'll now be activating sites all across the globe, in particular, in Europe. And you may know, I think our clinical team's view is will likely even have more enrollment in Europe than the U.S. based on how other trials have seen their experience go.

And then just as we think about kind of how these drugs could potentially be used and can they be used -- can it be used after [ chemTRAK ]? Is the response rate you can point to for kind of positive patients or HLA-A2 positive patients post chemTRAK?

Yes. So we haven't done that, that sort of -- we haven't shown publicly that kind of analysis sort of post chemTRAK, Peter. And I think, obviously, the order of agents, we have more work to do here. So obviously, TABI is approved in frontline based on a randomized OS study. We're encouraged by the data we're seeing. Our really focus is to get this approved in A2 negatives. Let the data continue to mature for A2 positives as we mentioned for potential compendia type strategy. Obviously, a lot of that order, I think what your question is getting to, will just depend on the quality of the data. So obviously, PFS, what will that look like. And of course, we will have the OS read off with the negatives. And as we mentioned, at least at this time, we don't anticipate a difference based on HLA-A2 status. So I think that should hopefully, all corroborate and inform a decision of what order oncologists may want to give the 2 possible therapies. The last one is we have seen responses, partial responses post TABI. So that shouldn't surprise anyone just because the MOAs are different. So yes, so that is something that we have observed.

Do you think that will be a combination of agent actually chemTRAK? Is that potential?

I mean, I think that's an interesting -- that's an interesting possibility, Peter. My folks think about, I guess, in that case, it will be a triple combination between darovasertib, crizotinib and TABI. We are not focused on that at this time. But I think right now, we've been focused on the doublet and trying to push that forward. But yes, I suppose that is always a possible scenario.