Good afternoon, everybody. Thank you to ORIC Pharma for joining us. We've got Jacob and Dominic here. Not your brother, Jacob. I had him earlier today. But thank you so much for joining us down here in Miami.
Thanks for having us, John.
All right. Now, I'm going to jump right into it. Obviously, the topic for you, the Exon 20, I'd love to talk first a little bit more about the landscape here. Obviously, there have been previous attempts. Several other programs are in development currently. Safety has been a key issue. You have some evidence of early differentiation. What's driving that, and where do you think the key liabilities are across the space?
Yeah, so when we think about the target markets for ORIC-114, which is our brain-penetrant small molecule for various markets within the non-small cell lung cancer space, it's really three populations, John. So it's EGFR Exon 20, HER2 Exon 20, and EGFR atypical mutations. Collectively, those three populations are somewhere between 6.5% and 7% of non-small cell lung cancer. Just to kind of put that in context for you, any two of those populations is about 4%-4.5%, so larger than the size of ALK. The issue in all three of those populations has really been twofold in terms of the programs that have been developed thus far, which is, one, safety, various kinds of safety liabilities, as you referenced. Specifically, if you look at EGFR Exon 20, what you'll see is two types of safety issues that pop up with programs.
One is either EGFR wild-type toxicities. So, in other words, as you're trying to target EGFR Exon 20, some of the first-generation compounds.
Loss of selectivity tox.
Yeah. So what you'll see is EGFR wild-type related tox, which is diarrhea or rash, essentially. So amivantamab, which is the only approved EGFR Exon 20 inhibitor, is classic in terms of the rash that you see.
Lots of them.
That's exactly right. And mobacertinib was classic in terms of the diarrhea that you saw. And that product obviously got pulled. The next generation of molecules in the EGFR Exon 20 space have done, ourselves included, a nice job of dialing down wild-type toxicities. But you continue to see off-target toxicity. So if you look at the kinome trees, they are really dirty. And what that leads to is, in the safety tables, you will see things like liver enzyme elevations, QTc prolongation, anemia. I mean, these are not things that you want to see in the drug profile of a compound, regardless of whether that's wild-type activity related or not. And so these are two of the areas that 114, at least thus far, seems to have differentiated itself on the safety side. And then the other major bucket of differentiation is the CNS activity.
So for all three of the populations I mentioned, there is not a drug out there that is properly CNS penetrant, has put out a compelling CNS data set. And as you've seen in ALK, EGFR, ROS, time and time again in this target program space.
A lot of brain.
Yeah, exactly. And so brain-penetrant molecules are the ones that end up winning the day and ought to lead to the best-in-class profile. So those are the two major areas of differentiation for our program, we think.
Let's talk a little bit about those off-target toxicities. So you've got data so far. You haven't seen some of those super worrisome signals that you mentioned there. Are there any rare, potentially worrisome signals? And I'm thinking about things like QTc, which sometimes only shows up rarely, but severely. Are there signals that you still feel like you have to keep your eye on, that you aren't yet sure you've dodged with the profile of 114?
Yeah, so the last public update we gave on 114 was at ESMO of 2023. It was during the phase one dose escalation, but we, at that point, had already dosed up to 90 milligrams daily as a dosing, and the two provisional recommended phase two doses that we are now in dose optimization on are at 80 milligrams and 120 milligrams daily, so certainly, for the lower of those two doses at 80 milligrams, you've all already seen, yeah, you've already seen data sets that cover that level of exposure, and we haven't seen any evidence of off-target toxicities, at least as part of that dose escalation experience that we've already publicly revealed, and we had seen minimization of the wild-type toxicities as well at that point.
Excellent. Now, you mentioned earlier, Jacob, the three different populations you're running in, EGFR Exon 20, HER2 Exon 20, and the atypicals. Can you talk about the differences between the populations and then also, sort of separately, the differences in drug activity between them with 114?
Sure. So the populations themselves are fairly similar. And so, at a very high level, each of those populations is somewhere between 2%-3% of non-small cell lung cancer. So that's where I came up with the collective numbers of about 6%-7%. There's some homology in the sites that you try to target as you try to address these populations, which is why you see several molecules, not just ours with ORIC-114, that are trying to target multiple of those populations. And, as I mentioned earlier, what you've seen in terms of the profile of the programs thus far has been the same safety liabilities across the populations. And then you've seen the same CNS liabilities across the populations.
Now, I think because HER2 Exon 20 perhaps tends to be a bit more concentrated in terms of the number of mutations that fall under the HER2 Exon 20 bucket, whereas EGFR Exon 20 and EGFR atypicals are a little bit more heterogeneous in nature, I think what you've generally seen in the competitive data sets is slightly better response rates for the agents against HER2 Exon 20 than against either EGFR Exon 20 or EGFR atypicals.
Which reminds me, that's where you are as well.
We're in all three of those populations.
But you saw those similar dose responses.
I would say that, from a dose escalation experience, I'm not really comfortable quoting any kind of response rate for us.
Fair enough, fair enough.
Really, 2025 will be the period. And we'll talk a little bit more about it because we'll have, obviously, data updates in all three of those populations next year in 2025. I think that's the time to really see, are we seeing the exact same pattern?
Now, you've suggested in the past that those updates from those three different expansion cohorts might not be coming in step with each other as they mature across the year next year. How good would you expect read across to be from one group to the other at this point? Any reason safety wouldn't read through well? Any reason efficacy wouldn't read through well from one of those cohorts to the others?
Yeah. So I think safety ought to read through well across all three cohorts and be fairly translatable across all three cohorts. I think efficacy ought to be, if you look at the precedents of other programs that have come before us, efficacy ought to be roughly in the ballpark across the three different cohorts and have some read through from one to the other. But one thing that we're very clear on internally, from an internal decision-making point of view, is that the go, no-go decisions we need to make on those three cohorts next year are independent of one another, meaning that we have one set of competitors for EGFR Exon 20. We have a somewhat overlapping but different set of competitors for HER2 Exon 20 and a somewhat overlapping but different set of competitors for EGFR atypicals.
We don't believe in the peanut butter approach that one sort of generic data set is good for approval across all three. We'll need to basically see, do we clear the relevant benchmarks in each of those individual specific populations before deciding whether we go forward with one, two, three, or zero pivotal single-arm accelerated approval trials starting second half of next year.
All right. Well, I'll definitely have to start using the peanut butter approach in my own life. But beyond that, you mentioned those go, no-go decisions being independent across those different cohorts. What are you looking for in those different populations? Where do you think the relevant bar are for these later line patients?
Yeah, I'll take that one. So, as Jacob mentioned, we're enrolling three cohorts of patients. These are EGFR Exon 20, HER2 Exon 20, and atypicals. I should point out these are second-line patients. And the other thing that's key here is we are allowing patients with active brain mets, whereas some of our competitors have excluded patients with active brain mets. So, with that being said, if you look at the competitive landscape, I'd say in the EGFR Exon 20 and the atypical, the kind of the top-line ORR number range is somewhere in the 35%-40% range. So that's kind of the number we're looking at. And if you look at the HER2 Exon 20, it's somewhere north of that. It's probably closer to 50%-60%. Obviously, you want to continue to see the safety and the tolerability and continued activity on the CNS side as well.
So that's kind of the bar we see. This is kind of the first readout. This is top-line ORR. This will not be mature PFS data. That will be at a later date.
Now, when you set those bars, as you note, many of your competitors, and in fact, the trials that set those bars exclude CNS metastases patients, obviously, having CNS penetrance is potentially a major differentiator. But would you expect to see different levels of response in patients with active CNS mets versus those without?
Yeah. I think the CNS benefit, if you look at precedent CNS active molecule, is more translated on the PFS side. So that's probably something you'd see at a later date, not necessarily on the top-line ORR number.
Fair enough to expect that when you give those ORR numbers for the expansion cohorts, you might be tempted to separate out patients with active CNS mets versus those without so that we don't want to unfairly penalize you for CNS met patients that get better results.
Yeah, that's right, John. For better or worse, the way we do data readouts is we put all the data out there so everyone can look at the exact same thing we're looking at. So I do anticipate we would give you that level of detail to be able to make those bifurcations and comparisons better.
As an analyst, I appreciate that. So maybe moving on to more about the CNS population, do you have a sense yet in these expansion cohorts whether you're going to have enough patients both with and without to get a sense for the delta in efficacy there and to demonstrate a delta in efficacy versus patients with CNS mets on other drugs where there isn't as much data?
Yeah. So let me take the second part of your question first, which is delta in efficacy versus our competition in these three populations will be very easy to see because you've seen no efficacy from our competition in these three populations. So that ought to be easy to see. I think in terms of just looking at, I think what you're driving at with your first question is, with a brain-penetrating compound, you should see very minimal delta in ORR between the systemic ORR and the CNS ORR.
Maybe in an ideal world.
Yeah. And so you will see some delta. Maybe it's up to 10 points of delta. You've seen that with other brain-penetrating compounds. But you really shouldn't be seeing bigger deltas than that. I think the data set for next year perhaps is going to be a little too early to make those kind of comparisons just from the point of view that, as we think about the total number of patients we'll have next year, and then you start to drill down to just the subset that have only active CNS mets, it'll be smaller. And I probably wouldn't be comfortable quoting a specific CNS ORR there. I think what we absolutely are looking, though, to see next year is, just like we already showed at ESMO 2023, do we have at least some patients who clearly have active brain mets who clearly have activity in the brain?
Clearly getting benefit on those brain metastases. Makes sense, and then maybe the earlier question could come down to a PFS curve with the more mature data set.
Correct.
All right. That makes sense. Development-wise, where are the regulatory bars in these populations? And you mentioned single-arm regulatory phase 2s or single-arm phase 2s, potentially relevant for a regulatory accelerated approval endpoint. But how confident are you in that registrational path for all three of these different patient populations? And when would you expect feedback on second-line accelerated approval paths from the receipt of those expansion cohorts next year?
Sure. Why don't I try to cover that at a very high level in about 30 seconds because I want to make sure we have time to talk about our prostate program as well. So from a regulatory point of view, it would be the tried and true path for targeted therapies, which is that we would anticipate that we'll make go, no-go decisions on each of these three cohorts based on the competitive benchmarks by the middle of next year such that we could start our pivotal studies, like I said, for all three of those or whatever number of pivotal cohorts we end up advancing in the second half of next year. These would be single-arm accelerated approval cohorts. So think about roughly less than 100 patients in each of those cohorts, just like other targeted therapies have done. You can enroll those studies.
You can submit the NDAs by second half of 2027. That's a very typical TKI path and be approved by 2028.
Excellent. Well, then, let's move to prostate.
Let's do it.
Obviously, your program here in ORIC-944, which you're calling a PRC2 inhibitor, maybe differentiated from Pfizer's program in EZH2, where maybe that's a more familiar target to a lot of investors. But obviously, that program from Pfizer has been getting more attention. They showed interesting data this year. Seems like they're planning to advance rapidly. Can you talk to us a little bit about the read across from an EZH2 program to ORIC-944 in the broader complex and then maybe where, on top of that, the opportunities for differentiation are?
Sure. So in many ways, there's a direct read across. And the reason actually comes back to the nomenclature that you referenced upfront, which is the reason that we refer to this as a PRC2 inhibitor is because the complex that both we and Pfizer are drugging is the PRC2 complex. So there's two different subunits that folks have tried to drug in that complex. Pfizer's drugging the EZH2 subunit, the catalytic subunit. We're drugging the EED subunit. If you do a little thought experiment and say, just take a hypothetical, two drugs, both exactly the same potency, exactly the same drug properties, one targets EZH2 and one targets EED, they ought to be exactly the same efficacy level in terms of being able to inhibit the target, the PRC2 complex.
Because the complex is the relevant target.
It's the complex as the relevant target, exactly. Now, where there ought to be differentiation in the long term, theoretically, because we haven't seen a clinical data set that anyone's put out yet to really validate this, but in theory, you should be better off with an EED inhibitor in terms of long-term resistance. The reason being that EZH2 inhibitors would be susceptible to acquired mutations in EZH2. They'd also be susceptible to bypass resistance from EZH1. Neither of those resistance mechanisms would be relevant to an EED inhibitor. We'll see if that plays out in the clinic.
All right. Now, obviously, Pfizer has combo data coming next year, pretty highly anticipated for their oncology program. What are the relevant bars in their data for you? What are you hoping to see out of that program to inform your own development path here?
Yeah. So a lot of our development path has already been informed substantially by Pfizer's phase one data sets that they've put out in two different populations in metastatic CRPC. And so just to level set for people that are perhaps newer to the story, Pfizer's EZH2 inhibitor is being developed in combination with enzalutamide, which is Pfizer's AR inhibitor. And they've seen in two different metastatic CRPC populations, so one, a post-abiraterone population, and another one that is post-enzalutamide. In both of those populations, they gave enzalutamide again, but in combination with their EZH2 inhibitor. And in each of those populations, they saw a radiographic PFS that was somewhere between three and a half to 4x as long as what you'd expect from enzalutamide alone. So dramatically longer PFS readouts.
In these mCRPC populations, we obviously expect retreatment with an AR agent to be relatively modest.
Correct.
You're saying the add-on here.
Yeah. So the add-on really, really tremendously outperformed what you'd expect from enzalutamide monotherapy alone in those settings. Pfizer simultaneously has an open-label randomized Phase II study with a control arm of enzalutamide, that I think that's the highly anticipated data that you mentioned. But the key fact there is that was an open-label randomized study. It had always been an open-label randomized study. And so the fact that Pfizer has been making a lot of noise, a lot of positive noise about that program, and the fact that they greenlit two Phase III studies for that program in the last three months.
We already know.
I think, no. Yeah. And so we don't need to see a whole lot more in terms of informing our own development program. And what we're trying to do essentially with our program is just close the timeline gap with Pfizer. So to move that timeline gap from four or five years between the two programs to something that's more like four or five quarters is our ultimate goal.
Now, one thing I will ask you about that data set. There's been a lot of curiosity questions among investors about what the comp arm performance might be in that randomized trial. It seems like that is a real sticking point for a lot of folks. But is there a reason to suspect that EZH2 or PRC2 synergy on top of androgen might be less if the background AR is doing more? Why wouldn't the synergy be maintained even if the background is?
Yeah, the synergy should be maintained is the short answer. I think if you go back, and these populations are so incredibly well studied. So if you go back and look at this population of metastatic CRPC patients who've had abiraterone who then get enzalutamide alone, almost every single data set is going to tell you that you'll see five to six months of median radiographic PFS very consistently. You might find an outlier at four months. You might find an outlier at six months. The one outlier, the one glaring outlier, it was most recently Ipsen with a data set looking at the exact same population, saw a control arm that dramatically outperformed that almost more than 2X.
Double it.
Now, the details sort of tell you the whole story, which is that they enrolled an indolent population in both the control arm as well as the treatment arm. You can validate that by looking at the percentage of patients that had visceral mets at initial entry. You can look at the PSAs of those patients were single-digit PSAs when they came into the study on both the treatment arm and the control arm. Whereas, for example, Pfizer's phase one experience, the data I just quoted you, the PSAs incoming were 50 ng/dL. So very, very different patient populations. And so the fact that Ipsen saw in the treatment arm 17-month PFS, it probably should have been a 20+ month PFS in the population that they enrolled if it was with a better drug.
One of the issues that we didn't get into here just as background for folks, one of the issues with all the EZH2 inhibitors and the one EED inhibitor from Novartis that had been tested in this population previously is terrible drug properties. So you're talking half-life of two hours, things like CYP autoinduction where you essentially are revving up the CYP enzymes that then metabolize your compound. So you'll get, I mean, Epizyme showed this with tazemetostat. They got dose-dependent decreases in exposure. The more you dose the drug, the less exposure you get. This has been the issue with all the drugs that have come before. Pfizer seems not to have that CYP autoinduction. They seem to have a slight.
But they still have a short half-life.
Seems to be a short half-life.
Compared to at least a year.
Double the half-life I just quoted, they seem to have about a four, maybe five-hour half-life. Our clinical half-life is 20 hours with ORIC-944.
Is it possible Pfizer is underdosing with a BID regimen? 20 hours, you clearly are supporting a QD regimen.
Sure. Anything's possible. I don't know. I won't speak on behalf of Pfizer. I guess we'll learn more when they share more data, hopefully next year. But what you referenced is accurate there. Sounds like they're going forward in their phase three study with a regimen of 875 milligrams BID twice a day. So you're talking about gram dose, more than gram dose.
Very substantial dose.
Yeah, in a day, and maybe that's to overcome the shorter half-life.
Makes sense. Now, I would love to, in our last seconds, return to a comment you made about catching Pfizer up a little bit and reducing that timeline gap. So what's your opportunity to do that? And how can you drive the escalation program forward rapidly?
Yeah. So the way that Dominic and myself and the rest of the team look at this is we have two paths in front of us at this point. So we finished our Phase I single-agent dose escalation. We're now doing dosing in combination with both darolutamide and apalutamide. And that's supported by free drug agreements with Bayer and with Janssen because they obviously understand the whole strategic rationale here. We have two paths in front of us. One is we now run a phase two study of our own for the next three years, see what it looks like. And if it's positive, we then go start our own Phase III studies.
Or we can do path two, which is what we've decided to prioritize, which is by the end of next year or early in 2026, we want to start our own one or two pivotal studies, phase three studies in populations of interest within the metastatic CRPC setting. So that's the way to close the timeline gap.
Makes sense. Now, in our last seconds here, obviously the prostate setting is very active. We saw some very interesting data earlier this week from bispecifics. But even beyond the PSMA bispecifics, there's STEAP1, there's other AR degraders, and all sorts of other mechanisms in the works across the board. Are there any biological reason to hope for synergies beyond AR synergies in early line setting?
There are certainly reasons for hope there. There's, in fact, multiple papers that have come out even in the last few months looking at synergies of PRC2 inhibitors with other mechanisms, including in other tumor types, which includes breast cancer, CRC, PDAC, a variety of different tumor types. So our preclinical team is vetting out all of that right now on the hopes that even beyond prostate, there might be other areas of interest to develop this program.
Excellent. Well, we are well out of time. More's the pity. But I think.