Good day, everyone. Welcome to day three of the Cantor Global Healthcare Conference. My name is Prakhar Agrawal. I'm a biotech analyst here, and for our next session, we are very excited to host ORIC Pharmaceuticals. Representing ORIC, we have CEO Jacob Chacko and Dominic Piscitelli, CFO. Gentlemen, thank you for joining us.
Thanks for having us.
So lots going on in ORIC. Maybe just start off with a quick overview of the company and where you are with the different pipeline assets.
Sure. ORIC stands for Overcoming Resistance in Cancer, which, in a nutshell, is the mission of the company. We are focused on small molecule drug development, primarily in solid tumors, and just given the expertise of the team internally, that tends to be almost exclusively in prostate cancer, lung cancer, and a little bit in breast cancer. We've advanced three programs through dose escalation as of the first half of this year, and then I think as most folks know, we've chosen to advance two of those programs into later-stage drug development. So ORIC-944, which is a small molecule inhibitor of the PRC2 complex that we're developing in prostate cancer in combination with various AR modulators.
And then ORIC-114, which is a brain-penetrant small molecule inhibitor of EGFR exon 20, HER2 exon 20, and EGFR atypical mutations, which we are developing as a monotherapy, potentially in the future, as a combination therapy as well. Both of those programs, as I mentioned, have completed dose escalation. We're in the dose optimization phase of drug development right now for both of them. And the plan is to start registrational studies for both programs in the second half of next year, so exciting times.
Okay, so maybe let's just start off with 944 in prostate first. What makes PRC2 inhibition an attractive target in prostate?
Yeah, so, people who have been following this PRC2 target for a while probably know that, large pharma, and some small biotechs have been developing PRC2 inhibitors now for many, many years. There's usually four or five different tumor types that have been implicated based on the underlying biology of PRC2 as an epigenetic dysregulator, and modulator. In particular, in prostate cancer, there's a huge body of biological evidence that would suggest that PRC2 inhibition can help essentially resensitize the tumors to AR modulators, so basically, they keep tumors in an AR-dependent state. And so essentially, the therapeutic thesis would be that if you could have a next-gen AR modulator, so for example, enzalutamide, darolutamide, or apalutamide, all of which have done phenomenally well for prostate cancer patients, but eventually those patients develop resistance.
If you could take one of those ARPIs and combine them with a PRC2 inhibitor, then you ought to get longer, more durable responses. Now, the issue with that therapeutic hypothesis, at least testing it in the clinic, has been really subpar drug properties of the drugs that have come before. What I mean by that is short half-lives on the order of two or three hours, CYP autoinduction, where you literally get a phenomenon of dose-dependent decreases in exposure, high interpatient variability. These are all the issues of the drugs that have thus far been tested in prostate cancer with PRC2 inhibitors, with obviously one notable exception, which is a drug that Pfizer is developing that has had some pretty phenomenal data already just in the phase I studies.
And I'm sure we'll get into some of the specifics of ORIC-944, but suffice it to say that we think it's got better drug properties than all of them, including Pfizer.
Okay, so remind us of what Pfizer saw with their EZH2 in prostate, and what are the typical benchmarks in those settings?
Yeah, so Pfizer has studied their EZH2 inhibitor in combination with enzalutamide, which is an AR modulator that obviously is near and dear to Pfizer's heart in two different settings within the metastatic CRPC setting. So one is a population of patients who have progressed on abiraterone and then are now getting enzalutamide for the first time, but in combination with Pfizer's EZH2 inhibitor. There was a second population, which is patients who, in the metastatic CRPC setting, who had progressed on enzalutamide and were now getting enzalutamide for the second time, essentially being rechallenged, but again, in combination with their EZH2 inhibitor. Their p...
And, essentially, in both populations, what they found was a PFS that was somewhere between three and a half to four times longer than what you would expect from enzalutamide alone in those settings. So in the case of the post-abiraterone population, this is a very, very well-documented population, where you will see almost all the studies would suggest that if you give a patient enzalutamide after progressing on abiraterone, you should see somewhere between four to six months of radiographic PFS. What Pfizer saw was a median radiographic PFS of 17 months in that population. And then in that second population, the Enza-experienced population, if you give Enza after Enza, you essentially expect to see about two to three months of radiographic PFS, and what they were seeing in that population was almost a year.
So in both settings, albeit without a control arm of their own, they saw phenomenally long PFS relative to what you would expect to see.
Right. And so remind us about the drug properties of ORIC-944 and Pfizer's mevrometostat, how does that differ, and what you saw in phase I in terms of the PK/PD?
Sure. I'll get into the drug properties of ORIC-944 specifically, but one thing I think it would be helpful to just level set on at the outset, because I've used a couple different terms here, is the PRC2 complex. So the PRC2 complex is comprised of three different subunits. Now, I mentioned a number of different programs, Pfizer's included, that have looked at EZH2 inhibition. That's essentially focusing on the catalytic subunit of the PRC2 complex. Our program, which we acquired from Mirati several years back, is focused on the EED subunit of PRC2. So essentially, it's allosteric inhibition of the PRC2 complex via EED. One of the most basic questions we get, but an important question, is: Is there any difference between EZH2 inhibition and EED inhibition?
The short answer is no, not at the outset, but there is reason to think that there's a benefit to targeting EED from a longer-term resistance point of view. And what I mean by that is, if you have two drugs that have exactly the same potency, exactly the same good drug properties, one targets EZH2, one targets EED, they should both be equally good at inhibiting the PRC2 complex at the outset.... now that's a big assumption for two drugs to have exactly the same potency and drug properties, so I'll come back to that in a second. But over the long term, there's a biological hypothesis that you should see resistance come about more quickly with an EZH2 inhibitor than with an EED inhibitor.
The reason being that EZH2 inhibitors ought to be susceptible to EZH2 mutations of the PRC2 complex, and also bypass resistance from EZH1. Neither mechanism would be relevant to an EED inhibitor, so that's why we think it's advantageous to be targeting EED. Now, back to the drug properties piece of the question, 'cause that's an important one, which is what Pfizer looks to have improved upon the drug properties of the drugs that have come before, but still seems to be on the order of about a four-hour half-life. They haven't commented on their level of CYP autoinduction, whether they see that or not. Our evaluation of their PK curve seems like there's not CYP autoinduction for them, which would be a good thing.
Now, in the case of ORIC-944, we have a 20-hour clinical half-life, no CYP autoinduction, and really tight interpatient variability, so what I mean by that is very little interpatient variability, whereas you do see that with the Pfizer compound and with others, so those are all things that came out of the phase I experience for us. Then obviously on the safety side, we also, I think, have commented extensively on this in the past, that we've seen a very good, well-tolerated safety profile, where we had to push the dose up to 50% higher than essentially our go-forward dose as a single agent. So up to 900 milligrams QD to see the very first grade 3 AE. Up to and including 800 mg QD, we saw only grade 1 and grade 2 AEs.
Those were on-target AEs that you'd expect for a PRC2 inhibitor, so essentially heme and GI AEs, and then, as I mentioned, we took that all the way up to 800 mgs QD with only seeing grade one and grade two AEs. The go forward dose for the combination testing that we've started at is 600 mgs QD.
And so what was the grade three AE?
Heme tox. Yep.
That's on target?
On target, correct.
Pfizer also has ongoing phase II, and they started some of their initial phase III trials as well. The design is up on ClinicalTrials.gov. What are the implications of that trial design that you have seen in your program?
Yeah, I'll take that one. So that's, I think you're referring to the MEVPRO-1 study. This is basically what they're studying in a post abiraterone setting. So they're doing Xtandi plus mevrometostat in this post abiraterone setting. What we've noticed there is that a couple things. One is the study's powered for 600 patients, which is relatively small in a prostate cancer landscape. They are going forward with a 875 BID. If you look back at the phase II study that they do, the open label phase II study they have ongoing, which they haven't shared the data yet, there they use a 1250 BID. So this could be a function of just better drug formulation.
They got better exposure, or they reduced the dose because they did see a relatively high degree of diarrhea and dysgeusia in the phase I data that they showed at ASCO earlier this year. And the last thing that if you look, the control arm, they are using Xtandi or docetaxel. So I think everything that they've shown so far is indicating they have strong confidence in the experimental arm in beating that. Again, if you look at the randomized study that everybody keeps asking us about, this is the ongoing study. And for those that are not familiar with it, this is a study comparing enzalutamide alone versus Enza plus mevrometostat. This is a study started in December of 2021 .
The study was powered to show a hazard ratio of 0.5, and this is an open label study and they've said on their Pfizer Innovation Oncology Day that the decision to move forward into two phase 3 studies in 2024 was based on both the phase I single-agent data as well as the randomized data so everything that Pfizer has been saying and doing since, I'd say, late 2022, has shown strong conviction in their program.
There's one more competitor, Ipsen, recently, last week, presented some data for their EZH2, tazemetostat and prostate cancer. The control arm looked a little bit variable on the high end versus the benchmarks you have cited. So thoughts on that?
Sure, I'm happy to talk about that. I think you're being very generous when you say it looked a little variable on the high end. That's a more polite thing than I would say. I think that the benchmarks I cited are, you know, if you go back and look, you'll find a dozen studies, basically every study in this space. When you look at the performance of enzalutamide after abiraterone, you will see that it falls in that four- to six-month radiographic PFS window, very tight, with pretty much no exceptions. What they showed in that study was over double that radiographic PFS, and a treatment arm that did even better than that. And so obviously, that was of great interest to us to try to dissect that and understand it.
I think the bottom line is the poster was very much lacking in terms of details of the patient demographics that would allow for that kind of a population. Now, there was a couple of tidbits that were buried in there, which I think are notable. So as you compare that population to, for example, the Pfizer phase I population, what you'll notice is that, in the case of Ipsen's drug, tazemetostat, they saw 0% of the patients had chemotherapy before. So chemotherapy was disallowed, meaning they were enrolling a quote, quote-unquote, "healthier population." There was less of an incidence of measurable disease, as well.
And then the other big factor is as you look at the PSAs of the patients that were enrolling in both the treatment arm and the control arm were mid-single digit micrograms per liter, which for folks that follow the prostate space will know is very, very low PSAs to be coming into the study. As a cross-reference, in the Pfizer phase I study, the median PSA at study entry was 50 micrograms per liter, with a range that went all the way up to 10,000. So very big differences in the patient population. So the best we can tell, obviously, outside in, is that it was an indolent population that they enrolled in both the treatment arm and the control arm.
So it's hard to make much of it. Now, just to level set everybody, you know, in my introductory comments about the first generation of PRC2 inhibitors and the poor drug properties, tazemetostat is probably at the very top of that list in terms of poor drug properties. So a two-hour half-life, CYP autoinduction. Tazemetostat, which was originally developed by Epizyme. Epizyme put out their own data that showed dose-dependent decreases in exposure. So in other words, as you push your dose, you get lower exposures of the drug. So in all ways, it was sort of a flawed experiment from start to finish, including the initial drug.
Right. And some of the secondary endpoints, like PSA and response rate, also did not look good for Ipsen's drug. Like, what's the efficacy of EZH2 on PSA?
We only have half an hour today, so what I'll summarize is, there hasn't been a lot that has been published about the impact of PRC2 inhibition on eventually what you want to see for PSAs. So I think people for a long time have been trained to look at PSAs as an early potential proxy for what a drug is doing in the prostate space, and I think that absolutely holds true in earlier lines of prostate cancer and certainly single-agent mechanisms of action. I think what is still unknown, and you know, ORIC will be at the forefront of figuring this out, is whether you ought to see moves in PSA as a result of PRC2 inhibition.
PSA, remember I said, is a good proxy for what prostate cancer drugs can do in an earlier line setting. The reason for that is, as you shrink a tumor, you should see PSA levels go down. On the other hand, PRC2, as I mentioned, is an epigenetic modifier. One of the genes it modifies is called KLK3, and so inhibiting PRC2 via KLK3 activity ought to push PSA up. And so you have these competing factors of hopefully a tumor shrinking, causing PSA to go down, but KLK3 causing PSA to go up, and that's why I say that it's quite complicated.
Okay. And so what's next in the development of ORIC-944 in prostate? You announced some clinical trial supply agreements with Bayer and J&J. Just talk about the strategy there.
Yeah, we announced those agreements just within the last couple of months, that Bayer is providing darolutamide and Janssen is providing apalutamide into two different combination studies that we're doing for ORIC-944 in combo with those AR modulators. So as people who follow the space know, the big three ARPIs are, at this point, enzalutamide, apalutamide, and darolutamide. All three are blockbusters unto their own, even though apalutamide came years after enzalutamide, and darolutamide came years after that. Daro, for example, is growing at 40% to 50% per year, having already achieved close to $2 billion of annual revenues on an annualized basis. So in other words, these are all three very good drugs.
Obviously, Pfizer's drug is enzalutamide, and so we went and started conversations right after we saw the interesting Pfizer data. We started conversations with both Bayer and Janssen. Both companies immediately saw the strategic value here to their own franchises, both in terms of the lines in which they're currently approved, but also other potential lines in which they're not approved today, and so we were able to strike up those collaboration deals. Now, at the simplest level, that provides free drug into our two different combination trials. At a more complex and important level, what it does is get two strategics who know the prostate space very, very well, helping us as we think through our own phase III trial design and designs for the multiple phase III studies.
So, it's quite helpful for us from a strategic perspective, and it's probably not a surprise to anybody that those two companies would be interested in doing a collaboration here.
What's the timeline of the, of the readout in terms of the combination cohorts?
Yeah, so, as you know, we just started those combination studies in the first half of this year, pretty much Q2, and again, that's a combination, combining nine four four with both apalutamide and with darolutamide, so at this point, we're not giving any specific guidance, as you know, our practice has been to kind of provide updated guidance early in the new year, so stay tuned for that. You know, typically in the January timeframe, we'll provide a little more color on potential timing for that combination data.
We can talk about prostate for a while, but in the interest of time, let's move on to ORIC-114, so maybe just level set expectations on why is brain penetrant activity important in the EGFR exon 20 space or even broadly?
Yeah, I can take that one. So there's three different populations in which we're developing ORIC-114: EGFR exon 20 mutations, HER2 exon 20, and EGFR atypicals. Like pretty much every target within the targeted therapy space, brain penetrance is important for therapeutics that are being developed there. Now, I prior to ORIC, I was at Ignyta. Back then, ten years ago, there was a debate about whether you wanted a drug to be brain penetrant or not for lung cancer. Now, it's dogma that if you can have a brain-penetrant drug, you ought to have a brain-penetrant drug. The reason being that about 35% of these patients at initial presentation have brain metastases. Those are brain metastases you can see on imaging.
There's actually a big chunk of patients who have what are known as micrometastases that you can't see on imaging, but are there. And then all patients, even the ones that don't have brain metastases initially, the first site of progression often is in the brain. So, for example, in EGFR exon twenty, amivantamab is a drug from Janssen that's approved there in the EGFR exon twenty space, has done very well for patients, but is not brain penetrant. And if you look at the PAPILLON study that read out last year at ESMO, the Phase III study that got amivantamab its first-line approval in the space, what you'll see is very good results.
But if you look at the sub-cohort data, what you'll notice is a pattern that has played out time and time and time again in the ALK, the ROS, the EGFR spaces, which is their performance in patients with versus without brain mets is dramatically different. And so what you end up having in these targeted therapy spaces for drugs that are not brain penetrant, is very big discrepancies in the PFS of those drugs in patients with versus without brain mets. And that's ultimately what leads to worse PFS than could possibly be achieved if you're using a like a brain-penetrant drug for those targets. A really good analogy for this is if you look at the ALK space and you look at alectinib.
So alectinib is a blockbuster drug in the ALK space, has done phenomenally well for patients. If you look at how alectinib, which was the third drug to market, compares to the first two drugs that got to market, which were not brain-penetrant, while alectinib is brain-penetrant, what you'll see is not a dramatically different ORR for alectinib than those earlier drugs, but you'll see a dramatically different PFS. And then when you dig into the data there, you'll see the same pattern emerge that I mentioned, which is for alectinib, the PFS for patients with versus without brain mets is not dramatically different, whereas for the other drugs, it is. So that's why it's important to have a brain-penetrant drug.
Now, what we like about these three spaces that I've talked about, these three populations for ORIC-114, is there is not an approved or late-stage drug that has shown definitive CNS activity in any of those spaces. And that's continues to be a real big advantage for ORIC-114 as we develop the drug in the three populations. On top of it, and I'm sure we'll dig into EGFR exon twenty as well, we seem, at least from the early phase I results we've presented thus far, to also have a very well-tolerated profile on the safety front, which has been another issue for the drugs in these three populations.
Right. So just expand on the safety profile, both on-target and off-target, why, why you're confident that you will have a differentiated profile?
... Yeah, so, you know, in the case of, let's take EGFR exon twenty, for example. So, the first-gen compounds, in which I would include, you know, mobocertinib, for example, poziotinib, those are drugs that, you would see one of two kinds of, on-target tox, if you will. And, what I mean by on-target is, any drug that targets EGFR exon twenty is gonna have some degree of EGFR wild-type toxicity. And so what you wanna do is dial down the EGFR wild-type potency, so you don't see too much diarrhea or rash, which are the two major kinds of toxicity. In the case of mobocertinib and poziotinib, you saw lots of those two types of EGFR wild-type toxicity.
For the next-gen compounds that have come, ours included, we've all been able, it seems, to dial down the levels of diarrhea and the rash the patients are seeing. So in other words, EGFR wild-type toxicity. But the real advantage for ORIC-114, even on top of that, is a very clean kinome tree. So as you look at the kinome tree, our kinome tree is exquisitely clean, very different than the kinome trees of the competitors in that space. And so what it leads to for the competitors is a whole host of off-target toxicities, things like QTc prolongation, liver enzyme elevation, anemia. These are not things you should be seeing in a clean drug, and that ought to be an added advantage for ORIC-114, that we don't see those types of off-target toxicities.
The ESMO data that you presented last year looked good in the refractory population, but just remind us of where you are now with one one four and the different dose expansion cohorts that are being enrolled.
Early this year, we did announce that we started dosing three expansion cohorts. The three patient populations, there were EGFR exon twenty, and these are post-chemo patients. They're naive to a prior exon twenty. The second one is a HER2 exon twenty, and again, these patients are second-line or later, so they could have seen a prior treatment. The third bucket of patients are the EGFR atypicals. Again, these are patients that could have seen chemo, could have seen mobocertinib, as well as afatinib as well, so kind of second or later-line patients. We started this in Q2 of this year, and our current guidance is to provide an update in the first half of twenty twenty-five. With regards to the guidance, what we've guided to is to really focus on the expansion cohorts.
The goal ultimately is in 2025 to, A, obviously evaluate the data, determine whether it's a go, no go, select the recommended phase II dose on that, in each of the cohorts. Then thirdly, assuming it's a go decision, we would look to start a potential registrational study in the second half of 2025. Given the unmet medical need with CNS active agents in this space, we do believe there's an accelerated approval path, for each of those cohorts.
Okay. So realize the update is a little bit far off in the first half 2025, but any initial comments on how many patients to expect? What are the efficacy benchmarks you are looking in the different cohorts?
Yeah. So we haven't given specific numbers. We'll provide a little more details again early next year. What I will say is, we feel we need about two to three dozen patients in each of those cohorts to make that go, no-go decision, so we feel like we can get there in twenty twenty-five. And then when you think about the benchmarks there, if you look at... Maybe I'll take EGFR exon 20 and atypicals together. If you look at the competitive space there, I think the benchmark generally is that 40% ORR. Again, we think there's room for improvement on the safety side and definitely room for improvement on the CNS side, and that's where we'd like to differentiate. And then with regards to the HER2 exon 20, I think there's been some evolving data in the last few months.
The benchmark there is probably in the 50%-60% ORR. But again, if you look at that data set, a lot of on and off-target toxicities, so we still think there's room for improvement, and we feel like the data there, again, we've not seen good CNS activity in any of the agents to date.
Would you have data on durability?
Say that again?
Would you have data on durability of responses on that type?
No. I think initial data readout will be focused on the top-line data, safety. Durability, obviously, we'd want longer durability. With a CNS active agent, hopefully, you get to see longer durability, so hopefully, we will probably not have mature durability at that time.
What could a registration path look like for ORIC-114 in second line and front line?
Yeah. So in the second line, it'd be the standard TKI playbook, which essentially is single-arm accelerator approvals. You know, so in a patient population that's post-chemo but hasn't had an agent that's directed to the target of interest, those tend to be very small studies that can enroll quickly and get an answer quickly. So that's the good news there. So if we can start one, two, or three of those populations in single-arm accelerated approval cohorts in the second half of 2025, we'd anticipate that the NDAs would be submitted two years later, second half of 2027, for 2028 commercial launches. So that's the benefit of drug development in the targeted therapy space.
And like I said, that's a playbook that has been tried and true for all the various targets in the space. And the fact that we have CNS activity with our drug and what's come in front of us has not is what leaves that door wide open for really moving quite quickly there. But importantly, you know, both because of FDA's guidance, but also frankly, just because of our own desire to get this patient - this drug to as many patients as possible, we would want pretty quickly after that to start the confirmatory trial. So as an example, let's look at EGFR exon 20. We'd like to start the confirmatory trial there in 2026. And that's a space where there has been two flavors of confirmatory trials.
You've seen some companies do a combination of their drug with chemo versus chemo in a treatment-naive population that's never had even chemo or any of the directed therapies. There's a second flavor, which is to do head-to-head, where you take your drug and just compare it straight away to chemo. Now, mobocertinib, for example, went head-to-head with chemo and lost. Amivantamab combined with chemo and beat chemo, but there's three other drugs now in development that are doing some flavor of the two paths I just told you, and so those are both under debate internally at ORIC as to which path we would choose. I think as we have more data under our belt, you know, we would choose, we could choose one or the other.
Now, there's a third very interesting path that is pretty much open only to ORIC as a brain-penetrant drug, which would be the potentially combined with amivantamab. So, if you think about the analogy of an HER2 to EGFR in the HER space, a really interesting idea here would be a combination of amivantamab, which is, as I mentioned, approved in EGFR exon 20 in the second line and the front line. Ami is a bispecific that would cover EGFR exon 20, but also MET, which is a potential resistance mechanism for exon 20 inhibition. Ami is not brain penetrant. ORIC-114 is brain penetrant. So there, it'd be a very analogous to even to what Janssen has done with their amivantamab lazertinib combination as an example.
So that is a third potential regulatory path for a frontline study that we're debating. Obviously, the first step there would be to do a combination dosing with amivantamab and just ensure that there's not too much overlapping tox from the two drugs and that they behave well together. But stay tuned on that front as we learn more ourselves internally and then decide what the path is.
And you're testing 114 in atypical mutations as well. Maybe just lay out the landscape for us in terms of how big is this segment relative to exon 20?
Sure. So if you look at the non-small cell lung cancer, so again, we're doing three patient populations. EGFR exon 20, we estimate about 2.1% of non-small cell lung cancer. HER2 exon 20 is about one and a half, and atypical is actually the biggest. It's actually almost 50% higher than the exon 20, so it's about almost 3%, 2.9%. So we do think that's obviously a sizable patient population. We are pursuing all three of those patient populations in the expansion cohorts. And again, the, you know, Jacob did the analogy to the ALK inhibitors. ALK inhibitors are about 4.something% of non-small cell lung cancer.
If you take any two of those patient populations that we just outlined here, that's actually close to or actually larger than the ALK inhibitors, which, you know, that's a multi-billion dollar opportunity. Now, obviously, you'd have to take into consideration durability. You're seeing great durability with these ALK inhibitors, but with a CNS active agent, we do think you should translate into longer durability as well. So we feel that this is a large patient opportunity. Again, I think people just kinda balk on it based on some of the early, smaller patient population and launches in the RET and ROS.
But I think here, you know, if you look at the space, you look at the TKI space, you look at any of these spaces, and you look at the CNS active agent, you typically see that that agent ends up winning out in the marketplace.
We saw some data from competitors in the atypical population, from ArriVent's, from furmonertinib. What was your take on that? How can you differentiate, given they are further ahead? And anything preclinically you can say about the activity for ORIC-114?
Yeah. So they had data in basically two doses. They tested 160 and 240. I think the response rate on the 160 was, you know, less than desirable, even for them. At the 240, they did see a good top-line ORR number. But again, I think you have to take that in the context of the safety and tolerability that they saw that as well. So I do think they saw some high levels of both on and off-target toxicity. And also, you did see a high percentage of both dose interruption and dose discontinuation. Durability data was, you know, immature at the time of the release. But I do think you have to see if that dose interruption, dose toler...
You know, the interruption in the doses or down dosing does have impact on that durability. On the CNS front, again, they showed some activity there. We have some questions around that overall. I don't wanna get into the details, but I think, again, there's room for improvement there.
Okay, and last question, just remind us about the cash runway and what does that guidance envision?
Yeah. So we've ended the second quarter with $309 million in cash and investments. Based on a current operating plan, that gives us cash runway until late 2026. And the way we provide guidance is, I'd say, somewhat on the conservative side, that assumes full success for ORIC-114 and ORIC-944, so moving both programs forward on our own. And that does assume that we're starting registries-