Arvinas, Inc. (ARVN) Earnings Call Transcript & Summary

Hey, everybody. Thanks for joining us this afternoon. Glad you could join us for the Goldman Sachs Global Healthcare Conference. I'm really thrilled to be joined by Randy Teel from Arvinas to kind of walk through the story and where things are and where things are headed. So thanks so much. Randy.

Thank you.

All right. Cool. Let's dive right in with a discussion of your lead program, which is the estrogen receptor PROTAC drug, ARV-471, for ER-positive breast cancer. I think going to start, we started with some of the recently announced failures of SERD inhibitors in this ER-positive breast cancer space. So given that, what gives you confidence that 471 is differentiated enough from SERDs to provide clinical benefit?

I guess failure is always a good place to start. But thanks very much for having us. It's great. Regret -- Ron sends his regrets. His flight got canceled. So yes, so 471 is unique in its space since it's the only PROTAC. So there are multiple other SERDs. You just -- you mentioned a couple that you just have news on at ASCO and other places. But 471 is the only PROTAC. So preclinically, we got a lot of confidence that it would degrade better than any of the other competitor compounds. The whole goal for 20 years since fulvestrant was approved by the industry has been to create a better version of fulvestrant oral. We think we have that. So if we look back at our Phase I data, obviously, really perilous to compare across study designs and all such things. But if we look at our patients, much later line treated than many others, all prior CDK4/6 treatment, 80% prior fulvestrant, 80% had prior or 80% or so had prior chemo and still had a really high over 40% CBR. So that gives us a lot of confidence. We have an ongoing Phase II program, Phase II trial right now, which we will have data out by the end of the year, but it's really not gating. We've said that with our partners at Pfizer, we'll start 2 pivotal trials by the end of the year in monotherapy and in combination in metastatic breast and very much still on track for that. So I think the data we've had so far, the difference in the mechanism and even those failures that you mentioned, obviously, not good for patients, always regrettable. But it, honestly, for us, relieves a little bit of competitive pressure in the second line plus space. And even as they had a setback, they were very much undeterred from earlier lines and so on and just like we and Pfizer are. So we think even though they didn't work out in that line, the target obviously is very well validated, and we think 471 could be very successful there.

So on that point around mechanism, is there kind of -- will you walk through some of the data that describes the kind of depth of ER degradation or depth of ER suppression as a kind of demonstrable factor in clinical response in these kind of medium to later line ER-positive breast cancer patients?

Yes, I think -- so I think if we start with the preclinical data, we always showed in our own work that better degradation would lead to better results, better tumor responses. That was true for looking at 471 different doses, that was looking at different molecules, that was looking at competitor molecules. In the clinic, obviously, I don't think anybody has shown a dose dependency on ER degradation for CBR -- or for benefit. That's not because we think it's not there. It's just been very little data. We've only had 10 or I think maybe 12 paired biopsies from our Phase I study. The patients that you get paired biopsies for aren't necessarily the same ones that respond, aren't necessarily the same ones that are -- or even included in the CBR analysis because they haven't been on long enough for example. So I don't think even -- so we haven't shown it definitively, but we think that's just a small number of patients thing. I think the best data for -- to argue that better degradation leads to better results is probably fulvestrant itself, which started off with an approval at 250 milligrams and then was increased to 500 and had enough benefit by making that change to increase their degradation from something like mid-20% to the mid-40s or so. So far in the clinic, we've shown degradation in the mid-60%. Preclinically, it was over 90%. So we think that will continue to improve. Other companies have done studies already as window studies pre-surgery where you can do a bit better, a more direct look at ER degradation. Sanofi just had data on that. Roche has had data on that. We'll start a trial this year and have data on that as well. But we think really the fulvestrant experience is probably the best argument for a better degradation getting to better efficacy.

Okay, great. So kind of following from the discussion of the Phase II trial you guys are running right now, what do you need to see from that study in terms of meaningful clinical benefit versus current treatments in the line of therapy you're testing there in the ER-positive breast kit?

For the Phase II?

Yes.

Really, it's not a gating. It's really just about dose planning. So we've got 200 and 500 milligrams ongoing, but it's not gating to starting those Phase III studies.

Okay. So what has kind of been falling from the kind of Phase I dose escalation data? Like where do you expect clinical benefit to lie, given what you've seen so far and given kind of the doses you're employing in the Phase II trial?

Well, in terms of -- I mean, we'd like to see the data really consolidate around what we've already shown, which is that the tolerability profile has been excellent. We had dosed, I believe, from 30 up to about 700 milligrams, no MTD hit. So we'd like to see continued solid safety. We'd like to see a continued solid CBR rate. That 40% that we showed in the Phase I was already higher than the percent of patients that you'd expect to be solely driven by ER, which is only about 1/3 of patients. So we'd really like to see the data consolidate. But as I said, it's really about helping us pick the dose for Phase III.

It's actually been a good point. So obviously, we are all aware of the data from Daiichi and Astra for their Enhertu trial in HER2. I guess the question whether it's HER2-positive or not has now become an open issue. But like how much flexibility or kind of like fungibility is there now in terms of like these breast cancer states in terms of like how sensitive you are to being ER-positive versus ER-low versus ER-absent in terms of like where each of these different classes of drugs can work? And does that impact how you think about 471?

Well, the Enhertu data was definitely good, right? That's great to see. It's in a later line than we're in. I think those patients are all -- the comparison was choice of chemo. So good to see. And I think for most of the conversations we've had since and before, we think that most doctors will still start with an ER access therapy before moving to that. So we don't really see it as competitive. Now over time, as the practice changes, we'll see. But right now, we see it more as a potential combination partner than we do as a competitor. With respect to HER2 high and low and mid and so on, I'm much less well versed there. But definitely, as we learn more about that data and run a combination studies, we and Pfizer are going to start an umbrella trial this year with multiple potential combination agents and lots of opportunities and excitement by us and Pfizer to consider different combination partners.

Okay. So that's actually a great segue to thinking about the combination studies. You currently have with Pfizer the CDK inhibitor groups. Could you walk us through on how we should think about the combination of 471 with various CDK inhibitors as particularly Ibrance as a starter?

Yes. So Ibrance is the one we have an ongoing Phase Ib in combination with this year, and we'll have data by the end of the year. Again, that's -- we've absolutely been thinking about that as a safety and dose-finding study. So that is a CDK after CDK study. And until just the past few weeks, really very few benchmarks for that. So people ask us what to expect in terms of efficacy. We don't really have an expectation of what good would be there. So it's really for us a safety and dose finding. And like I said before, full speed ahead on getting a combination trial going by the end of the year. With respect to other potential combination therapies, could be Pfizer products, could be non-Pfizer products. They are very much with us on the vision of making 471 the backbone of care across lines of therapy. Certainly open to testing ARV-471 not just with palbo, but with abema and ribo and other things, too. Pfizer has talked explicitly about their CDK -- other CDK inhibitors in their pipeline, certainly could be of interest. So more to come as we decide what to include in that umbrella.

Okay. So then how should we think about toxicity arising from the combination of 471 with Ibrance kind of around the data that's coming out by year-end?

Yes. So obviously, that would be included. We don't have any indication so far preclinically that we should expect any. I think the -- I think it's CYP3A that is responsible for palbo, and we don't hit that. 471 doesn't hit it. So pre-clinic. So we don't have any expectation. I know there was a SERD that had a DDI with CDK4/6 inhibitors. We don't anticipate any. No reason to be worried about it, but obviously, that's why you do the study.

Okay. So there's no like mechanistic basis.

No mechanistic basis for it, yes.

Okay. Cool. So then kind of in terms of the dose-finding aspect of the combination trial with 471 and Ibrance, how should we be thinking about that? Is it mostly in terms of like PK? Will we be able to see ER degradation in that study? Like how should we think about kind of like the dose assessment that can be done in that combination study?

So think of it as the standard dose of palbo with an escalating dose of 471. We don't have to start back at 30 like we did in the dose escalation study. But really, you kind of hit it on its head, which is what's the PK, if there's an efficacy benefit, great. ER degradation, we wouldn't really expect to be affected by the combination. So that would be tricky to use, I think. So I think it will really be based on what we see with PK/PD in the Phase Ib study as well as being informed by the VERITAC study, which has 200 and 500 [ clients ]. So I think if you take all those things together, we'll come up with the doses for the study starting at end of the year.

Okay. So then kind of like in the cadence of news flow from these 2 studies, would one expect them to be separate? Again like how do you think about timing for data at a medical meeting [indiscernible]?

Yes. We've just said second half, and I'll leave it at that. Obviously, we prefer medical meetings if we can do it. But there's also benefits where if one is ready and one is not, maybe you wait and do them together, but we haven't quite finalized how we'll do that yet.

Okay. Great. So kind of moving on from breast to prostate cancer. Could you walk us through the existing data for the AR PROTAC bavdegalutamide, we're going to call it bav for short, in the mutation-specific metastatic castration-resistant prostate cancer.

Yes, absolutely. So bavdegalutamide is an AR degrader. That was really our first program in the clinic back in 2019. And starting from the preclinical data, there wasn't anything to suggest that there would be a higher response in any particular mutation. The only thing we knew about mutations was that there's one L702H which bavdegalutamide does not hit. So we knew that already. As an aside, ARV-766, which is another AR degrader that we have in Phase I, does hit that mutation, but we can come back to that. So when we saw the signal back, I believe, in late 2020, that showed a fairly high 40% PSA50 rate in that 878/875 group, that was enough for us to design the Phase II, to look specifically at that population as well as the thinking was absolutely that if you get to a more AR-dependent population by getting a less pretreated group, which is what we try to do, you might see an expanded benefit there, too. So the 875/878 mutations seem to us to just be marking tumors that are still AR-driven. Those tumors are all pretty heterogeneous. The data we showed at ASCO GU really showed that once a patient has had an NHA enzalutamide or darolutamide and so on, they really look pretty similar in terms of both AR mutations as well as AR independent mutations. So that's why we're moving ahead hoping to start a pivotal study by the end of the year in that 875/878 group.

So thinking about 875/878 group, to what extent are drugs like enzalutamide and others kind of enriching for that population now relative to, say, years prior where they might have been a relatively rare population? What do you think about kind of like the epidemiology of these mutations, and then also like how that's changing over time?

Yes, absolutely. So we get asked all the time, how big of a population is that? And if we looked at the literature or got asked that question a couple of years ago, we would have said maybe 5% each for 878 and 875. And it doesn't have to be both. It's either or both. But in our study, it's closer to 20%. And I think that's pretty consistent with what others are saying as we did that disclosure, talked to doctors and KOLs at ASCO GU and since, they also see a higher percentage of 875/878. And I think that's just because as enzalutamide and other therapies move earlier, including into the cash rate sensitive space, you just get more and more patients that have that profile because it arises as a resistance mechanism based on AR-targeted therapy. So we think it's about 20%. It could be higher, it could go higher, but it's a much bigger population than I think we would have thought a few years ago.

Okay. So kind of given that, can you walk us through some of the next steps for bav in this 875/878 population? I guess can you walk through the key features of the upcoming Phase II expansion cohort you just mentioned?

Yes. So well, the Phase II, we just had the data release in February, and I don't anticipate that we'll do another one soon. We had over 100 patients at that. And really, that was enough for us to get sure that the path we wanted to pursue was the accelerated approval pivotal study. So what we're going to do is meet with the FDA. We have that meeting scheduled by the end of the month, this month, we said first half. And then what we're going to do is take that information and figure out what the next step is based on guidance, based on what they tell us. And then we'll come back when we do our second quarter earnings, which is in the first half of August, typically and be a bit clear with investors about what the plan is. But right now, going into the FDA, the plan is to propose that single-arm Phase II accelerated study and then start that by the end of the year. So that would really be a trial enriched for an 875/878 group. You might have seen we signed a partnership with Foundation a couple of weeks ago. It's exactly what that's for, to help develop their liquid companion diagnostic for the 878/875 mutation for bav.

Okay. Can you walk us through kind of like what are the kind of endpoints you might be using in that kind of single-arm study? How should we think about them? And how we should think about those endpoints relative to the data you've shown so far?

Yes, right. So a lot of the data we've shown so far has been based on PSA50, not an approval endpoint typically. So that's not the thinking. So we've shown OR data as well, but the data that we would have for the pivotal side would be more of a PFS or response rate, not PSA50, no. At least we wouldn't anticipate that.

So then to that end, we've had many conversations with investors about kind of the design of that single-arm study. And I think there's always a view of like, do you believe you have enough data to confidently run out and just say, like, 100 patients today? Do you think there's a kind of 2-stage aspect potential for this trial where you run like a pilot study of 25 patients and then expand out based on kind of hitting some marks? Like how are you thinking about that kind of staging of it versus just a full-on 100-patient study or whatever size patient study?

I think I'll hold off on -- better to answer that question after we get the feedback from the FDA. I'm going to come back and get real clear on the trial design of what we want to do. To date, we've largely looked at the rucaparib trial as a precedent for accelerated approval in prostate cancer. That's what had led us even a year ago to say that if we saw a PSA50 rate of something like 25% plus in that population, we would think that would be good enough to have the conversation with the FDA. And we ended up with a 46%. So we feel pretty good about that. We're using that as the model, but we'll definitely follow up in the second quarter earnings discussion about what the trial plans will be for the end of the year.

Okay. So then kind of given this kind of somewhat late, presumably later line population of 875/878, what do you think is the path forward for treating patients in early line prostate cancer with bav?

Yes. So it could be bav mono, could be combination therapy. We're even doing thinking about the pre-NHA space. And like I just talked about, post-NHA, it seems pretty clear that there's a lot of AR-independent mutations at play. So we may need to do earlier. So nothing to talk about in terms of details about how we might go there. Monotherapy is an option. Combination therapy's an option. We have an ongoing AbbVie trial right now or combination trial with AbbVie. We've talked about dose [indiscernible] or some other chemo as a combination agent as well. But in earlier, we've talked less about how we'll get there, but definitely on the radar.

And would those be 875/878-specific mutations? Or kind of like broadly?

Broad, yes, exactly.

Okay. Cool. So then I guess beyond bav, how are you thinking about the kind of AR? Obviously, you mentioned, 766. You have other AR degraders kind of in queue. Can you walk through some of the other programs you have in AR and how to think about kind of data coming from them later this year?

Yes, there's 2 others. So ARV-766 really started its life as the backup to bavdegalutamide. And the better that bav has done, I guess there's less value to a backup. But the main difference between 766 and one -- and bav is that 766 hits that L702H mutation, which is the one well-known resistance driving point mutation that bav doesn't hit. So what would that mean? Well, that would -- in pure numbers, that's something like we used to say 3% to 9% of patients have L702H. Although I think that like 878 and 875 is also increasing with prior therapy by NHAs. So the initial hypothesis would be, can you get a molecule that looks like bav but that hits that mutation as well? But they're different molecules, right? So even though preclinically the clear difference was that L702H mutation, there may be other differences between the 2 molecules that let 766 have a broader profile than bav does. But we'll have to wait and see when we get that data. We plan on sharing that data by the end of the year for 766 and starting a Phase II there as well.

So is there any kind of -- so to that end and is there any biochemistry data for 766 versus bav that would explain differential activity kind of outside of the specific mutations?

At least not that a non-biochemist could go into. Really, we focused on the L702H mutation is the major difference. But you never know. You have different molecules with different profiles. You may have different PK/PD relationships and so on. So we'll hold off on that discussion until we show the data later in the year.

Okay. So then kind of stepping back, how do you think about the risk of both for the ER and AR programs, the risk of resistance mechanisms to PROTACs generally by various [ mutants ], like lysine mutations and so on and so forth?

Yes. So when we started off, I think the big fear because before we shared the initial clinical data in 2019, there was a lot of fear that you would never be able to make a PROTAC greater either act like a drug, be orally bioavailable or that there'd be resistance. And I think the most obvious thing we worried about was that would the tumor just shut off the promoter for cereblon, which is the -- obviously the E3 ligase that we're using for both 471 and 110 or bavdegalutamide. And so far, that doesn't seem to have happened. That was the worry that it would just shut off cereblon, and you'd lose activity that way. It hasn't been that. We don't have, obviously, a good understanding at all of what will drive resistance in human patients. Obviously, we would expect that to occur at some point. None of these so far were really not curing cancer much. But pre-clinic, it's hard to say what it's going to be. But we've thought about that. Who knows what will happen in terms of other point mutations that could affect binding? But so far, that hasn't been a clear signal for us.

Okay. And you mentioned a third AR drug.

Oh, yes.

You're going to walk us through kind of its utility and then step back and have a discussion on the broader AR strategy.

Absolutely. Yes. So the third one is AR-V7, at least that's what we call it for now. It's back in the research phase. So V7 is a splice variant of AR. So AR has the ligand-binding domain, which is where degalutamide binds. It's also where enzalutamide binds. It's where most other AR target therapies bind. And it's where androgen normally binds in the -- when you're not using a drug. So V7 is a splice variant that essentially laps off the whole ligand-binding domain. So clearly, bav is not going to degrade V7 because it can't bind to it. And that's true for most other AR-targeting therapies, too. So a V7 degrader would be targeted towards the other end of AR and if successful, would really degrade both the V7, the splice variant, but would also degrade all forms of AR. That end of AR is a bit more unstructured. Most companies have had very little success in targeting it so far, but we'll see. But that's the goal. And that certainly would be a further out goal than 766, which is in its Phase I now. So that's really more of a next-generation play that would come along at some point in the future. But it's a preclinical program, obviously excited about it and more to come.

Okay. So to that end, like how should we think about a broader AR strategy about kind of just divide and conquer of like mutation A, mutation B, mutation C versus kind of like moving up the lines a bit? Like how did you know they have a kind of bigger strategy here for individual drugs, individual mutations and kind of that approach?

Yes. So we've clearly gotten beyond ARV-766 as a backup molecule, right? Right now, it's in Phase I. ARV -- bavdegalutamide is in its Phase II, and we'll hopefully start a pivotal by the end of the year. But we really think about them as complementary. So in other words, we're not going to try to make them compete and invest in huge studies in the same population for both. So I could imagine a situation where bav is targeted at late-line therapy and 766 is targeted earlier. I can also imagine a situation where one of them is being used in a situation besides prostate cancer in oncology for other diseases in oncology that are AR-driven. And there are diseases outside of oncology even that are also driven by AR. So over time, I think those 2 things -- I'm really thinking about bav and 766. But I think next year, when we have Phase II data for 766, we're well underway with that. Hopefully, that single-arm Phase II for bav will be in a good situation to compare and contrast the 2 and really ask the question, what's the difference between 766 and bav? Is it really doing what we think it's going to do? Is the difference the L702H mutation? Is it broader than that? And at that point, that's probably what makes sense to make a big choice.

Okay. Can you speak a little bit more about that point around non-prostate cancer AR-driven disease even in non-oncology? Or can you dig a little bit more about kind of where some of those opportunities are for other diseases outside of prostate cancer that are driven by androgen receptor?

Yes, there's a few, none that we've talked a lot about in oncology, triple-negative breast, ovarian or both known to have an AR element to them. And then outside of oncology, there's a few others, polycystic ovarian, benign prostate hypoplasia and maybe a couple of more. So there's a few options out there so that if you've got a successful drug, which is working, improves its concept and is safe and tolerable in humans, could have an opportunity. So over time, we think that will clarify.

Are the existing anti-androgens being used in those contents?

Not very much. Yes, not very much. There's not a lot of data in those areas, which we think is an opportunity.

Okay. Great. So then stepping back beyond AR and ER, which obviously is a big focus for most investor conversations, how do we -- can you tell us more about kind of the timing for clinical entry for additional PROTAC programs? So we're thinking particularly like for KRAS, BCL6, the neuroscience franchise, like how do we think about kind of the timing for additional program rollouts outside of ER and AR?

Yes. Yes. Our guidance there has been admittedly a little bag. So it's -- we've said 4 INDs through the end of 2023 for both -- including both oncology and neuroscience. So that's about as specific as we've been. As we get closer, we'll certainly share data on some of those programs. We had a poster at ASH back in December for the BCL6 program, showing some pretty nice degradation for a tool PROTAC against BCL6. So a lot to come in that space, but I admit we've been a bit bagged there, and we'll try to tighten that up in the coming months.

Okay. Great. I guess kind of like to that end, how much do you think that is kind of underappreciated? I mean should it mean rightly so because you guys haven't said a whole lot about it. But like for [ example ], are those opportunities kind of like under the radar for people these days, given the absence of information?

Well, I think, yes, I think they are, and they're not. So clearly, once you've got a couple of programs in the clinic, that's where everybody is watching, no doubt about it. But -- so from that sense, I think it's not getting nearly as much attention paid to it as it could be. But although like you said, we haven't shared too much data there. I think the one that probably gets the most attention after that would be KRAS only because, obviously, it's a hot topic, having something that wasn't undruggable and is less so now. And on KRAS, we've shown a little bit of data for a G12C program, which we had. But right now, we've definitely pivoted the KRAS program to be still multifaceted, but looking at G12D, G12V or even a pan-KRAS that would hit multiple mutations of KRAS. We just think that the other folks in this space have shown that G12C anyway isn't looking as undruggable as it used to. But we've been able to take a lot of the learnings we had from the G12C program and apply them to G12D and G12E. So a lot of progress being made there, a lot of eyes on that program and look forward to sharing more.

So to that end, like how much -- remind me for the G12C data, how much was the kind of compound you guys had there dependent upon the kind of sustained bond? Is it kind of like key fact the way that the G12C drugs were?

No, that's fully the answer. That's exactly the issue, right? So we showed that we could degrade KRAS G12C very well, but we were doing it with the PROTAC that was covalently bound, whereas usually, when we talk about the benefit of a PROTAC degrader, it's that it can form that complex between the E3 ligase and the target protein of interest from that complex, transfer the ubiquitins and then fall away and then be recycled and catalyze additional grounds of degradation. The PROTAC that we had for G12C was using the C was covalently bound. And so while we could degrade it very well, it didn't have any benefit in our view, at least not enough of a benefit over an inhibitor to be useful. So when we create a program for a PROTAC, we're always wanting to find a place where there's an advantage to using degradation versus inhibition. If you can create a good inhibitor, there's not really much of a need for a PROTAC, for the most part. And so we really would like to pivot that into a G12D and [ G12E ] program that aren't using covalently bound PROTACs but are using reversible like a traditional PROTAC. So that was -- it was a very similar limitation.

Okay. Great. So then kind of stepping back even further, what other areas do you see as an opportunity for PROTACs to be developed kind of even beyond oncology and neuroscience? And where do you see as kind of like the broadest PROTAC opportunity? Obviously, I feel like targeted protein degradation has become kind of a 4-letter word these days. But to what extent do you think there is opportunities in other spaces? And where are the other spaces you guys are kind of like digging in kind of behind the scenes?

Yes. So I would say the broadest possible is something that we already did since I don't think that we initially anticipated doing work in plants. So we have a -- to illustrate that we have a joint venture with Bayer called Oerth Bio, which is in agriculture because plants also have E3 ligases and ubiquitin proteasome systems. In fact, more E3 ligases than people do. So that's an ongoing JV. It's going out working very well. Companies, you'll probably hear more from them in the next couple of years. So I think that really shows the breadth. Again, though not really where we initially planned to go. We're very much focused on oncology and neuroscience. We also got targets in immuno-oncology like HPK1 and immunology. But I also like to say that we don't need to do everything. There's lots of space out there. There's nothing about a PROTAC that makes it a better fit for oncology than anything else. There's lots of diseases out there. I think what you should expect to see us do is maybe further partnerships to expand the value of the platform, maybe not as broad as plants again. But there's certainly other therapeutic areas that we could go into, but I definitely like keeping us focused. Oncology and neuroscience is enough. That's a lot.

Yes. For sure. So to that end, like in immunology, like I guess someone in immunology is we're kind of guiding more towards autoimmunity. Is that a thing where you look to like be a hired gun and like partner out kind of either targets or some level of development? Like how should we think about kind of where you guys would fit in the kind of autoimmune space in terms of like sorting out PROTAC agents?

Yes. Maybe I think the types of deal -- because we do have existing partnerships with Bayer and Genentech and Pfizer, different partnership with Pfizer, that are at that target level that we did over the past few years, between 2013 and 2018 or so. And those are less on strategy for us now, creating degraders for someone else to go and develop. So I think it would be -- as we look at partnerships and we looked at the partnership for 471 with Pfizer the same way, which was that we felt good about where we were. We felt good about our capital position, our resources that we had to progress it. But Pfizer was able to convince us that they could help us move it faster and more broadly. So I think the same math would apply to anything else, which is what could we create internally using a PROTAC in a new area, maybe in immunology. And would it make sense for us to build out all the capabilities in-house for research through the clinic, all the assays that you need, all the clinical relationships that you need. Will it make sense for us to continue? Or would we be able to move faster and broader with a partner? I think that's the question we would ask. But I would see it less likely as a sort of a target discovery deal that we do before we've created a PROTAC. I think for us, it would be we have the ability, let's create them, get them into the clinic, show that they have value and then partner.

That's a good question. That leads an interesting question, kind of like the old joke we'd say for the RNAi companies was because it's kind of genetically determined like in the like basements of all these RNAi companies is an RNAi drug against every single gene. So obviously, PROTACs aren't that simple. So kind of like how many PROTACs you kind of have set up for like the kind of targets that people would look for, transcription factors of various kind of important [indiscernible], how many of those are you going to have like some level of development kind of basally going, always have basally done somewhere down the line, it's kind of sitting there ready for someone who's -- really cares about Target A, Target B, Target C?

Oh, well, that are sitting there waiting? Very, very few.

Okay.

Right. No. But so if you look at our public pipeline, it's something like maybe 10 or 12 lines deep. If you look at the real pipeline, it's something like 25 lines deep. So there's a lot of programs going on under the surface. If your aspiration is to generate a clinical candidate or an IND per year, you need a lot to get there. But I wouldn't say there's anything that's -- that we've gotten to a point, and we're waiting. Everything is moving ahead. And if a partnership has the right approach for it, we'll certainly look to make that happen.

Okay. Great. So then again, at a high level, how do you think about the expansion of the PROTAC pipeline kind of balancing between internal R&D versus external business development? Like where do you think there's kind of room to maneuver on either side?

Yes. Just got asked that earlier today. I think that you'll -- when it comes to actual programs that you see in the pipeline, the vast majority are going to be developed in-house. It's not that we would be opposed to bringing something in. It's not that we think no one else in the world could ever make a PROTAC or another form of degrader. I insist that they're not really that -- A, they're not really very available. If you have an excellent PROTAC that looks like it could be a drug, those aren't generally the ones that you're having conversations about. And we feel pretty good about our own capabilities at taking a PROTAC and turning it into a drug. Anybody can make a PROTAC. That -- I think that, that has ceased to be difficult. Any -- and many academic labs can do it, but actually creating a PROTAC degrader that not just degrades preclinically but will be -- that will work as a drug and has drug-like properties and is safe and is tolerable, all those things, that has proven to be a bit more difficult. So where we're focused much more on the BD side is making sure that our ability to do that stays ahead. So focused on screening. We did an AI partnership last fall, did a partnership with GNS a few weeks ago on their in silico Patient model in neuroscience. So we will continue to look for places to make sure that we stay ahead. We don't pretend to think that we can do everything the best all the time. But I would expect our BD efforts to be much more focused on enhancing the platform than anything else.

So on that point, the challenges are making -- you make a point that anyone can make a PROTAC drug, but kind of it's hard to make good PROTAC drugs.

Well, anyone can make a PROTAC. But not necessarily a drug, right? Not necessarily something that is orally bioavailable, not necessarily something that you can cross the blood-brain barrier, if that's what you want it to do. Those sorts of things are still very tricky, right? And I think we probably have more experience than anybody at doing that, and that means more failures than anybody else are doing it. So more to learn from. So others will get there, right? I mean, there are lots of antibody companies. There will be multiple protein degrader companies. That's terrific. We're big believers in the space, obviously.

So on that end about chemistry and pharmacology versus kind of biology, when you think about the safety and tolerability questions, to what extent are they questions of chemistry and pharmacology versus questions of biology that when you degrade Target X, there are safety and tolerability consequences as compared to a drug that binds other things will have kind of knock-on tolerability questions?

Yes. That's -- so earlier, we were talking about how before we showed the first clinical data, there were lots of folks that wondered if you'd ever be able to get a drug like PROTAC, that looked like a drug. And the answer was, fortunately, yes. That was the other fear. Like that was the fear that if you have a technology that is meant to degrade things, what if it's not specific enough, and it actually degrades things you don't want to. And so far, that has not been an issue at least in what we've shown preclinically looking at proteomic studies. We've shown even back in our S-1 days showing that we're really only degrading the target that we want to. And if you look across the programs that we have, 2 programs in the clinic that we've shared data for and then lots of preclinical data for many PROTACs, there hasn't been a PROTAC-specific toxicity. There hasn't been off-target effects that you might expect to see if you were degrading something besides your target of interest. And I think you're even seeing that, although I won't talk much to it at all in the data from other degrader companies, too, is a lot of times, what they're seeing is related to the target they're degrading. So we at least have not had issues degrading other things. That's not to say there aren't -- we had with bavdegalutamide, we had the DDI with rosuvastatin back in late, I guess, whatever it was 2020. But it was very much an isolated question with bavdegalutamide. You don't see that with 471, right? It's not -- so they have a different profile each one. There hasn't been a -- an off-target or anything else that's been consistent across PROTACs.

Okay. Finally, the question we're asking every company at the conference, what is the reason to own Arvinas stock in the next 12 months?

Well, 12 months from now, instead of being a mid-stage Phase II company with 2 programs with data, we'll be a company with 3 pivotal trials ongoing, hopefully proof of concept for a third clinical program. We've said 4 INDs through the next year, so maybe a couple of programs more in the clinic. And just as importantly, data to support each of those that's more than we have now. So as I've said, the data aren't gating to start those pivotal studies. But absolutely, the VERITAC data will be out. The Phase Ib data will be out. We'll have 766 data. So I think we'll just keep adding on to showing that PROTACs can absolutely be drugs and will be a late-stage development clinical stage company heading towards the FDA, hopefully.

Excellent. Okay. Great. Well, thank you so much, Randy, for your time. And thank you for joining us, everybody, this afternoon. Thanks very much.

Thank you very much.