Everyone to our ESMO KOL event. My name is Dan O'Connor. I'm the President and CEO of Ambrx, and I really appreciate everybody being here this evening to hear about ARX517 and the data we were able to share at ESMO, both clinical safety and efficacy data, as well as PK data. Before I begin, we may be making some forward-looking statements here this evening, so I know this is a long slide, but if you have a moment, and when you have a moment, if you could please read this slide. Thank you. So, here is the agenda, and I'm gonna step through this in a moment, but before I do, I'd like to say thank you. The first thank you I'd like to say is to the panel for being with us and agreeing to be here today to speak with you about our program. We absolutely appreciate that. Several members of the panel are investigators on APEX-01 and have, for the past at least two years, probably more than that, been working on the program to bring us to the point where we are today. So we really appreciate your participation on the study and the work that you've done. The second thank you I'd like to say is to their colleagues at all the clinical sites in the United States that have been organizing and help run APEX-01 over the past two-plus years. And also to their site staff. There's many... As we know, there's many people at every investigational site that operationalizes a clinical study. So without those team members, these studies would not occur, so we really appreciate their support and their hard work and their interaction with not only us, but also our colleagues who are helping operationalize the study. Lastly, most importantly, for that list of thank yous goes to the patients on our study. The reason we are able to be here is because people volunteered and agreed to be on a clinical trial for an experimental drug that has not had, at that point in time, been used in a human being. So that's... You can imagine if you're in the situation of such a person, that's a daunting thing to face, and so we really appreciate that those people actually agreed to help us, to evaluate this program and this drug in, in their lives, as well as, their families and their, you know, their loved ones, which it's, as we all know, when you're, dealing with cancer in your family, it's not just one person, it's a lot of people. So, those are my introductory comments. There's another group I would like to thank. It's our shareholders. When I came into this role, I count myself as really fortunate to be able to be the CEO of Ambrx, and I started in, November, and it's. I can say it's been a pretty wild ride since then, but that ride has been made, it has been enabled by shareholders. And several of the shareholders are here this evening, and you, you all know who you are, but we appreciate the support from our shareholders. We get an, an extremely constructive relationship with them, and again, without them and their support, we would not be here, so thank you to our shareholders. So here's our agenda. That was the first part of the agenda, which is the introduction, the opening remarks. Let's step through the rest of it. We have, again, several doctors who are gonna be presenting here this evening. Several of the doctors... All the doctors have had interaction with our patients, so, you know, you'll have an opportunity to speak with them afterwards. But the first doctor is gonna be Rakesh Dixit. Dr. Dixit is an expert in PK. He comes from Bionavigen, and he's going to be addressing the topic of PSMA targeting ADCs. What are the problems with today's ADCs? The second topic is Dr. Scott Tagawa, who's joining us from Weill Cornell Medicine, and Dr. Tagawa is going to be speaking to ARX517 PK data and the poster that went up this afternoon here at ESMO 2023. Following Dr. Tagawa, we've Dr. Oliver Sartor is going to be speaking to us about mCRPC treatment landscape, and really focusing on how Pluvicto has changed that treatment landscape. And then lastly, Dr. John Shen from UCLA Health is going to be speaking about the poster that went up that relates to the clinical safety and efficacy data on ARX517, a poster review, and some topics that Dr. Shen would like to focus on. Following that, we'll have a brief Q&A session. We've got some prepared questions that we'd like to ask our panel, and then we'll open it up to everyone else to ask some questions. And then after that, we'll have an opportunity to have a reception and interact a little less formally. So, with those remarks, again, thank you very much for being here, and thanks for the folks who are participating online. I'd also like to thank LifeSci Events. Several people from LifeSci Events have helped us tonight and done a really good job of putting this all together. So with that, I'm gonna turn it over to Dr. Rakesh. And this is the clicker for you. Thank you, Dan. You got it. Thanks. Can all of you hear me okay? So good evening, and somewhere, probably those who are online, could be good morning or could be early morning or good afternoon. But in any case... So let me set up the context for the ADC. I'm sure most of you know about ADCs, but from my perspective, it's important to set up the context. So the ADCs have three major parts. First is the antibody, and then the payload, and then payload has to be attached to the antibody, and it has to attach in such a way that it remains stable, doesn't release the payload before it reaches the tumor. O bviously, we start first with the antigen, you know, right antigen, and in this particular case, what we're talking about, PSMA. It's an ideal antigen because expression level in normal, healthy tissue is very, very low or minimal. On the right-hand side, you look at the mechanism of ADC, how they work. So in order for any ADC to work well, I can say this with authority because I've been working with ADC for 20 years, seen lots of ADC, lots of platform. So what we can say here is that ADC, first of all, it has to remain very stable in circulation. That means it cannot release the payload before it reaches the tumor, and then it has to bind precisely to tumor antigen. And then tumor- antigen complex, the ADC antigen complex, has to internalize well, endocytose well. And not only endocytose well, it needs to make it to lysosomes. That's where the enzymes of proteolytic degradation release the payload, and depending on the payload, the payload is released in lysosome. It has to make it to its target. Sometimes it's in target, it's in cytosol, sometimes it's in nucleus. But so all these processes are very synchronized, and they have to work all well. I think most important part, as you're going to hear from Ambrx's ADC later on in the clinical data and some of the other perspective, the conjugation is the key. How do you keep these things so stable in circulation? How can you tie them so tightly? So the three other ADCs that came before this one, the first one was the Millennium-Takeda's ADC. Really, you know, if you look at this maytansinoid, it uses the same DM1 molecule that's used at Kadcyla. It doesn't use a non-cleavable linker. It has a cleavable linker. However, despite the fact it was a cleavable linker, and you wanted to see more, what I call it, bystander effect, it ended up in too much toxicity. Maytansinoid, we know that they are well known to cause peripheral neuropathy and liver toxicity. And here in this particular case, I will say there was too much of the neurotoxicity, and that basically terminated the molecule, just couldn't dose high enough. The second molecule from Progenics, that uses also another microtubule inhibitor, which is MMAE. We all know about MMAE. MMAE is a very powerful, potent payload. However, it has a problem with peripheral neuropathy. It can cause significant peripheral neuropathy and neutropenia. In this particular case, it caused substantial Grade 3 febrile neutropenia, which I don't think I have ever seen with any MMAE ADC. So something went wrong with this molecule. You start releasing the payload way too quickly before it reaches a tumor, significant amount. The third molecule came from my lab from AstraZeneca, in which we acquired this molecule from ADCT. And this was a molecule actually pretty well-designed and preclinical data looks superior. That attracted this molecule to me, you know, and saying, okay, this may be a good ADC. However, because based on tesirine, a dimer, tesirine dimer that sits between two strands of DNA, it caused way too much toxicity at when we are escalating the dose. So because of that very long, very high degree of toxicity, we just couldn't escalate the dose. And if you look at the dose at—we could not dose beyond, 0.15 mg per kg, so that was also terminated. So three ADCs, using three different technology, three different payload, three different linker, all got terminated. And I will tell you why they got terminated, is the answer comes from their PK. So if you look at the PK profile with the first ADC, Progenics PSMA ADC, there you can see how much the MMAE is getting released. A lot of MMAE is getting released. And you might say, you know what? Compared to parent ADC, still the numbers are low, but these are very potent payload. You should not be seeing that kind of payload, 10 nanogram or more release. And the other problem with MMAE is it has a longer half-life, so it can stick around longer in circulation. So that was one of the problem, premature release of the payload. Second one is the same problem, premature release of the DM1, and that caused quite a bit peripheral neuropathy. And same problem happened with the MedImmune's molecule. Loose conjugation, releasing the payload way too quickly, and these are very toxic payloads. Just remember that I'm going to keep on saying this, toxic payload got to hooked up with the antibody as long as they can. Otherwise, they get loose in circulation or in the tissues of high metabolism, they can cause significant tissue damage. So if you look at the reasons of termination, common reasons of all three ADCs that got terminated. First is the Grade 3 toxicity, off-tumor toxicities. Actually, PSMA is very clean target. I never expected I'm going to see this kind of toxicity with these ADCs, because usually you, you worry about if the target expression is away from the tumor in normal tissues. Should have not seen this thing. Then very poor half-life and very rapid clearance. And then the third one is the conjugation. The conjugation is not-- conjugation is weak. That means you're releasing the payload way, way too quickly, and should not be doing that. So what are the solutions? So the Ambrx solution is to make a fit-for-purpose synthetic amino acid, put them in the antibody, and then you have a, what I call it, a linker, which is not that cell permeable. That means it can retain more in the tumor tissues. In this way, when the antibody-drug conjugate go through proteolytic degradation, they will keep more of the payload in the tumor instead of sending them out of the cell. So that allows more high tumor accumulation of the payload, and because this payload is not a substrate for P-gp efflux transporter, you can keep it longer and accumulate more in tumor. That's the number one goal for any ADC, keep as much payload as possible in the tumor and not get inside too much in the normal tissues. So this is the construct. You're going to hear more about this, but I think this is what, this is what the uniqueness of this molecule, is that the way you are conjugating the payload, a potent payload, and using a non-cleavable linker, using this oxime conjugation, and then what I call fit-for-purpose synthetic amino acid. And that's really important. Remember this, these are three properties that makes this ADC so unique. So, so I think this, this conjugation and in the way you're keeping the ADC stable is the secret sauce for this ADC. So what I can say that from this summary slide, so if you look at all three other ADCs, compare this with the ADC from Ambrx. So the first one, look at MLN2704, it's a lysine conjugation, cleavable linker, but significantly peripheral neuropathy. And look at the half-life, 2.5 days. That's not good. That means you're releasing payload way too quickly. All you can... Despite the fact you could dose higher, and yet you don't see any activity and a lot of toxicity, not a good thing. Second is ADC, MMAE-based ADC. Again, same problem, short half-life, too much premature release of the payload, and you're seeing all off-target toxicity of these payloads, which are basically neutropenia and neuropathy, which is very, very common with MMAE-type payload. The third one, PBD dimer-based ADC that came from my previous lab, we had this, we ran into a similar problem. Same problem, very high-grade myelosuppression, the skin toxicity, which is related to the PBD payload and the vascular leakage. I mean, it was... Moreover, we just can't dose high enough. If you cannot dose high enough, you are not going to see activity, and that's one problem with PBD dimer-based ADCs. They are, in my opinion, they just can't dose them high enough for the solid tumor. So if you can look at all three ADCs, what happened to them, very similar fate. But with the. If you look at the molecule from Ambrx, here you have DAR 2 MMAF, very significantly stable. And look at the half-life. That's. I cannot overemphasize. I have never seen any ADC. I've been working for 20 years in this field. The 8 days of half-life. If it continues like this, it will make a great ADC, because the goal of any therapy is to keep it longer in the tumor and keep it accumulated, so that you can keep on sending drugs in the tumor. Because if, if the antibody-drug conjugate disappears very quickly, that means you have a lot of drug holiday going on in the, in the tumor. A drug holiday is bad news for the tumors. I mean, they will develop a distance, and they will, they will come up with the ways that they will. It won't be effective because you don't. You have too much of the drug holiday going on. Similar to chemotherapy drugs, if you don't have enough drug in the tumor, you have too much drug holiday, they will start developing this P-gp efflux transporter, other mechanism to throw out the drug. And that's really important, the long half-life of the drug. So I got only 6 seconds left, and I think I cannot, I cannot overemphasize the importance of the half-life, stability of the ADC, and not releasing the payload so quickly. And I think this molecule does that. So you're going to hear more about this molecule from my colleague, Scott. So Scott, it's your time. All right. Thanks for your attention, everyone. I think most of you, whether you saw the poster or came up to the poster, sorry, I wasn't there the whole time, but I think you've all, all seen this. I'm gonna go over the data that was in the poster relatively quickly, but show you some, some kind of bonus information. This is the exact slide that was left there. I'm just gonna point out the antibody, because I'm gonna show you several slides on what... This, this is a very well-characterized antibody with a lot of data on tumor retention that I'm gonna, that I'm gonna show you. Okay, another exact same slide. I'm not gonna go with it too much, but I agree that when we have a potent payload, we wanna make sure it goes to and stays in the right place. Okay, so this is taken from the poster, and you can think of these, the middle two kind of panels as cartoons. What a very stable ADC with a non-permeable payload should look like in terms of PK, versus one that has a less stable linker and/or a very permeable payload. So just kind of remember what those look like. And this is what is the total TA, total antibody versus the ADC, meaning the antibody that has the payload on it. So you can see it literally mirrors the cartoon example of a very stable ADC. So it's the construct, the theory, the science is all there. B y the way, in human patients with prostate cancer, the PK completely mirrors that. We throw in that, and I'm not gonna go to the methodology. The intense PK was done cycles one and three, which is why you'll see multiple more dots with cycles one and three. But I do think that's important to see what happens with the subsequent cycle, in any human experiment where we're looking at PK. And then the very low levels of the payload in the circulation. So if you remember, with some of those other ADC constructs, where it was logs higher in terms of the free circulation time. So again, all the science actually worked in this particular case in terms of what is circulating. And then there, you know, we are running a dose escalation study. What's the point of increasing the dose if it's not going to circulate for longer? Of course, we're mostly interested in the clinical talk, in the clinical data, efficacy as well as safety, that we'll hear about in a little bit. But this is proportional to the administered dose. Okay. So this is the kind of bonus information. So this, these are pictures that actually show what this antibody is doing, exactly what Rakesh said, is that it is binding, it is internalizing and going into the lysosome, lysosome. So one of the nice things about this particular antibody is that it has been very well characterized visually. So we don't have to make up where is it, where is it? So this is obvious in cells and moving cells into mice. When we look at what is in the blood, I think that's important. It's very important for an ADC to look at the different components because you could run into problems with toxicity, as we've seen with other ADCs. But what is really important, what is in the tumor? So if you look at... These are models of the first one is in blood, the second one is whole body, which is what it, where it's really being retained. Y ou can see over time, it goes up in terms of the tumor. And this is in a mouse. I'll show you humans in a second, where it really is retained within the tumor, and over time, what's in the blood is going down, and it's all accumulated within the tumor. So that's mice. What about people? So this is the same antibody relabeled with Zirconium-89 pictures at a week. So you can see on the bottom, you can see a bone scan. Those little dots that are lighting up are generally tumor, and that's within hours of it. But you can see these pictures, and many of you have seen PSMA PET scans, so we don't need to belabor how much more sensitive they are than the standard scans or old-fashioned scans. But you can see that it clearly lights up, and these are importantly at a week, so it's after it's been circulating. You can see, you know, it starts off more in the blood and eventually lands in terms of residence in the tumor. The bottom is the exact same antibody. This one happens to have lutetium on it, but one of... It, it's gamma emission. That's at two weeks. So we see that the, you know, bone scan versus the, versus the, Ga-68 or PSMA antibody, it is retained still at two weeks. So we know that what happens in the blood, but importantly, this remains in the tumor, and this is, I don't know if it's, how much a deficiency it really is. It's my viewpoint, 'cause there's a lot of ADCs that are approved. But this happens to be one that is very well characterized in terms of where is the circulation and where is the retention within the tumors over time. So conclusions, in terms of the... But, you know, I-- what my summary of the available data that we showed today in terms of, of the ESMO meeting was that the science basically mirrored what is happening in terms of the patients. The missing information you're gonna get in two talks, which is the, the clinical, tox data, which is the preview, pretty clean. And I think that is because of we have a validated target, both diagnostically and therapeutically. Oliver's going to talk about that. And we have a very, clean ADC in terms of, the metabolism. So I think I'll end there, and, I think I'll turn it over to Oliver. Thanks. Thanks for the opportunity to be here, and I'm going to be kind of talking a little bit, maybe on my own. Okay, now these are not my slides—but, I'll tell you what, I'll come to my slides. You know, I think I was charged a little bit with talking about Pluvicto and how it might have changed the landscape. That seemed a little bit boring because you can kind of figure it out. So I'm going to be moving a little bit forward into how can we do better? I've taken a broad view to that, and not just asking how we can do better from the context of radiopharmaceuticals, but maybe just looking at the field a little bit. These are some of the things that I think are maybe worth noting. So anyway, here we go. This is our landscape today, and believe it or not, I kind of have most of the FDA approvals on there. One of the big things, and I think we all understand this, is the novel hormones are going earlier and earlier. If you've got apalutamide, darolutamide, enzalutamide, abiraterone, you know, if you bring it up earlier, you're probably going to have a positive study. That's really been a huge game changer because now what we used to call this castration-resistant prostate cancer was predominantly after ADT had failed. People would get, you know, Lupron, Zoladex, that sort of stuff, have a PSA rise, and then they'd be castration-resistant. Today, the castration-resistant patient is very different than where it once was, and that's because they're failing multiple hormones or maybe even hormones and docetaxel before they get to be castration-resistant. There are a bunch of little small things in the corner, where capivasertib, olaparib, combinations of olaparib, abiraterone, talazoparib, enzalutamide, you know, all of which have had relatively recent approvals. You know, most of those are in the context of not a lot of prior treatment with ADT and novel hormones, or just ADT. But I'll simply say that there's a lot changing, and that's even in the past year. And then we have the Pluvicto and the Radium-223. The theranostics is of course an area where I've spent a lot of time. By the way, Scott spent a lot of time there as well. Scott and I worked together on a number of projects, just as an FYI. Here you have a little different components. You have a cell surface target, you have a ligand that'll be able to bind that target, and then you've got a linker and a radionuclide, and then you can either have a diagnostic radionuclide. We just saw some zirconium scans, and then we can have a therapeutic nuclide, and that potentially could be lutetium, actinium, or whatever. Y ou could actually sort of have your ligand being an antibody, right? So it doesn't have to be a small molecule, but they're small molecules and big molecules and antibodies and things that kind of go in between. The lutetium was important, but that was the last stage study. You know, the truth is that if you wanna look at lines of therapy, people talk about maybe the Pluvicto was, you know, third or fourth line. A ctually, for patients who had already had Abi and docetaxel, 40% had cabazitaxel. I mean, this was really a late-line study, but it was positive, and that's important. But if you look at the landscape today, and this is just PSMA, and actually, I think I may have left one off that I know about. It gets really kinda complicated because you've got the antibody commercial development, that's J591, and you can put an alpha or beta on it. And if you wanna go to Telix and look at J592, that's just a reengineered 591, and they intend to put actinium on it. And then Bayer has a molecule. They had it with thorium, and thorium wasn't the right isotope, so now they're flipping over to actinium, so we'll do that. They get all these small molecules. I don't need to go over them all, but there's a lot of kind of movement here in the field, and you can see the alphas beginning to sort of rise in, in a variety of contexts, but there's also plenty of beta. So right now, we're in a beta world. Maybe tomorrow, we're gonna be in an alpha world. It's hard to know, but there's a, a lot of cool stuff going on out there. The important new trials, so PSMAfore is gonna be presented tomorrow, coming by tomorrow at 5:00 P.M. or something like that. I forget exactly what time, but kinda late in the afternoon, if you wanna see PSMAfore. It's a positive study, okay? The top line rPFS was already announced, 0.41 for the hazard ratio, and then we have an updated analysis where you're gonna have hazard ratio of 0.43, more mature. Y ou know, it's pretty good stuff, and that's gonna be in the pre-chemo space. But there's more to the pre-chemo space. You've got the SPLASH trial. That's the one that POINT had, and they kinda sold it to Lantheus. And then you have the ECLIPSE trial with Curium, and that's moving forward in the pre-chemo space, too. So we've got new trials, and then we have something called PSMA Addition in the castrate-sensitive space, moving forward with the PSMA-617 lutetium. So, you know, lots of studies ongoing here, and one of them is positive. And then when we talk about how can we do better, there's a lot of reason in my mind to think about new targets, because I know that we've done a great job with PSMA. I feel really proud of how PSMA has been able to treat patients in a more effective manner, and AR is still going strong. DNA repair, there's even more DNA repair opportunities. And then AKT is being targeted, B7-H3 Wnt pathway, and then you can look on the right-hand side with things like hK2, STEAP1. Pretty cool STEAP1 data presented with AMG 509, by the way. I thought that was nice. Prostate stem cell antigen, that's actually Rob Reiter's DLL3. You actually heard about it, the bispecific and the Amgen, and the what? Third line, small cell. GPC3 is potentially kinda cool. FAP is kinda cool. GRPR, TROP2, HER2, and there are more. Okay, so this is really, I think, where the future is gonna be. We're gonna have lots of things, and by the way, you can target them with ADCs, bispecifics, radionuclides, you know, whatever. If we want a coordinated attack, you know, we have these kind of small pocket agonist, antagonists. That's where we've been for the last, you know, 50 years. And then we have the protein degraders are new, the ADCs, which you just talked about, the radiopharmaceuticals, CAR T, bispecifics, and then these various sort of immune modulators in way. AR remains a really interesting target 'cause a lot of these tumors remain AR-driven, despite the fact they may have failed multiple types of abiraterone-type drugs. N ow the people are looking at the degraders. I chose one here because it's just a scheme, and I'm trying to push PROTAC, but it turns out that it's kind of cool to be able to degrade a protein, and you can do that. T here are a lot of different protein degraders. There's the Arvinas, the Roche, the Accutar, novel Celgene, and there's a lot of activity in this field right now. And of course, it's not just prostate, it's breast and more. So that's all cool stuff. And then ODM-208, which really inhibits steroidogenesis right at the beginning, and Karim Fizazi has presented some pretty nice data with the ODM-208. And clearly, there's a subset of patients that may have failed Abiraterone, that you can still have additional activity with the ODM-208. Then cell surface molecules, and this gets to what I was saying a little bit earlier. You've got molecular targeted radiation, PSMA, hK2 come to mind, antibody drug conjugates here tonight, bispecifics, CAR-Ts, and all of these are potentially in play. Now, the CAR-Ts have not been particularly successful in the solid tumors, but oh, my God, look what happens in myeloma. It's just amazing some of these things that they're presenting now. ADCs, I don't need to say any more about. B7-H3, PSMA, TROP2, STEAP1, DLL3, HER2. I mean, these are some of the things that are, that are out there. And, you know, I'll absolutely say the ADCs are getting better. And, you know, we have almost a turning point when it came to the Enhertu. The data from Enhertu for the breast cancer was just absolutely outstanding. I played around with the ADCs before, I had a PSMA ADC, and it was, it was one from Progenics. It was a dog. It was just really bad, and people were getting sick, and neuropathies and a tiny amount of activity, but nothing that was worthwhile. I'm so glad to move away from that one. Bispecifics. You know, bispecifics are potentially real. The AMG 509 presented here is a STEAP1, CD3, and again, I thought there was some real activity there. That was interesting. Regeneron, Crescendo, J&J. I hear interesting things about the J&J molecule. I haven't seen it. Somewhere it'll appear in public, but it's just kind of a rumor mill right now. But there's, you know, maybe some interesting stuff there. CAR-T, I'm not that high on the solid tumors or prostate. There's been some interesting data coming out, PSCA-targeted CAR T, and that was pretty provocative, but there's been some pretty nasty toxicity. And one of the things about prostate cancer, just as a reminder, is men are old, and they're castrated, and they can't take a lot. You can kind of take a finger and kind of knock them over sometimes. So a lot of the guys that we treat are not that vigorous. Some things I left off the list, the CDK4/6, BET inhibitors, et cetera, et cetera, et cetera. And, you know, there's more work to be done. So summary, mutation is important, but CRPC, CRPC space, really vibrant. You know, I've been doing this for 33 years. It is incredibly vibrant today and really provocative to work on as an investigator. It's been a lot of fun. The post-lutetium space is going to get bigger as patients get treated earlier. That's pretty obvious. There are a lot of new, interesting drugs in development. Thank you for the opportunity to be here tonight. Good evening, everyone. It's been an exciting day for all of us and the whole Ambrx team, and I'm glad we could be here tonight to share some more of this data, supporting the safety and efficacy of ARX517, an anti-PSMA antibody drug conjugate, in a heavily pretreated patient population with metastatic castration-resistant prostate cancer. I'm John Shen, a General Urinary Medical Oncologist at UCLA. I've treated several patients on the APEX-01 study, along with several of my colleagues, many of whom are here today. I won't repeat this background information, as Dr. Tagawa has already shared the demonstrated stability of ARX517, which has many advantages compared to prior ADCs targeting PSMA, and Dr. Dixit has already highlighted the importance of these features in ADC development. We've also heard from Dr. Sartor about the ongoing role of PSMA-targeted therapy, and more to discuss after tomorrow, I'm sure. So APEX-01 is a Phase I/II first-in-human trial evaluating ARX517 in patients with mCRPC refractory to prior therapies. So I want to do a walkthrough of the poster data that was shared and talk through some of the key updates that you may have noticed already that were different from the text abstract that you may have read earlier this week. The Phase I design included a standard 3 + 3 dose escalation, so we started at 0.32 mgs per kg. We had one patient in that cohort, and subsequently increased. Most recently, the dose expansion component has included the addition of around 20 additional patients into cohorts of interest, namely Cohort 4 and Cohort 6. In order to be eligible for study entry, patients with mCRPC must have received at least two prior FDA-approved therapies for metastatic disease, with at least one being a second-generation androgen receptor pathway inhibitor, such as abiraterone, enzalutamide, darolutamide, or apalutamide. They must also have received, they must also have had disease progression by either radiographic criteria or PSA criteria by PCWG3. The objectives of this study are to determine the safety and tolerability of ARX517 in order to determine a maximum tolerated dose and a recommended Phase II dose. I want to spend a bit more time here on baseline demographics, because you'll know probably considerable updates, again, from that written abstract that only reported on the first 24 patients, right? This data and the data in the poster reviewed tonight are the updated 65 patients. So 40+ patients were very recently enrolled, this summer. At the time of the data cutoff, it was early September. So of the 65 patients, 82% were white. The median age was 68, though there was a pretty broad range of participant. You'll notice that 100 is not a typo. That was actually a patient treated at our site. The median prior lines of cancer therapy was four, again, highlighting the heavily pretreated nature of this patient population. Notably, two-thirds of participants, the 65 sample size, had received prior taxane chemotherapy, which is a bit higher from what you may have read in the text abstract of the initial 24 patients. 64% had received prior immunotherapy. The majority of this was actually Provenge. Sorry about that. The majority of this was actually Provenge, although some received anti-PD-1 or PD-L1, therapies, presumably on other clinical trials. Of note, 17% had received prior PSMA-targeted radionuclide therapy. The median prior lines of ARPI therapy was 2. 75% had prior abiraterone, 69% had prior enzalutamide, and nearly 50% had received both. In terms of sites of disease involvement, 80% had bone involvement, 46% with lymph nodes involved, and 19% with visceral disease in either the liver or lung. Notably, only a third of participants had measurable disease by RECIST, which somewhat limits the scope of assessing radiographic responses, especially given the current sample size and the fact that many of these subjects were very recently enrolled, and we do not yet have these radiographic assessments. In terms of baseline performance status, this study was a bit more lenient than you'll see in some of the other studies and did allow Accrual of ECOG performance status up to 2, potentially a bit more representative of a later line patient. The median PSA was around 47, although again, a broad range was noted, and there were some with several PSAs in the hundreds or thousands. So here's a high-level summary of the safety data, which we'll go through in more detail in the next few slides. Overall, AEs, treatment-related AEs occurred in 74% of participants. Importantly, there were no Grade 4 or Grade 5 toxicities. In terms of AEs leading to treatment discontinuation, there was 1 subject in early cohort that was reported to have an asymptomatic Grade 1 thrombocytopenia after 1 cycle, and another in Cohort 6, which is a dose level of 2 mg per kg, that had a Grade 2 decreased appetite and dysphagia and decided that that was intolerable. In terms of the Grade 3 treatment-related adverse events, of all participants, three subjects had Grade 3 lymphopenia, and two had Grade 3 thrombocytopenia. There was one subject in Cohort 4 that actually had an asymptomatic decrease in LV systolic function. So per protocol, there's a baseline echocardiogram and a subsequent on protocol and continued monitoring. On a subsequent monitoring, that patient completely recovered his systolic function without any heart failure, treatment, or symptoms. Represented here are the most frequent Grade 1 and Grade 2 treatment-related adverse events occurring in at least 10% of participants. So dry mouth occurred in about 28%, and dry eye occurred in 22%. These have also been seen with some of the other PSMA-targeted radionuclide therapies and other ADCs. 20% of participants noted fatigue. Around 15% reported diarrhea, and this was a similar rate for other GI symptoms, such as decreased appetite, nausea, vomiting, and dysgeusia. There was some neuropathy seen in some of these participants. However, at the time of this data cut, it was less than 10%. Notably, neutropenia was also infrequent, and this is important because, as we saw from prior experience with earlier versions of anti-PSMA ADCs, these were some of the limiting toxicities that precluded further development. Now for some of the preliminary efficacy data, and I highlight this is very early data. I will preface this by saying, you know, again, a lot of these patients were very recently accrued, so we expect that this will continue to read out in the coming months. Deep PSA reductions have been seen with increasing ARX517 doses. You can also see the asterisks labeling participants who had prior lines of PSMA-targeted radionuclide therapies, and we are still seeing PSA reductions in several of those patients. A t the putative therapeutic doses, so what we're considering to be 2 mg per kg or higher, which is expanded on the right side, you'll see on the panel, we saw 52%, achieve a PSA50. For those receiving prior PSMA-targeted radionuclide therapies, PSA 50 was seen in 50% at those doses 2 mg per kg or higher, and in 37% in those dosed at 1.4 mg per kg or higher. We also saw a greater frequency and depth of PSA responses at putative therapeutic doses. Amongst Cohort 6 through 8, a total of 23 patients at the time of data cut, we saw 52% of those achieve a PSA 50, and also 26% achieve a PSA 90. The reason this is important is that from prior studies with cabazitaxel, abiraterone, apalutamide, and docetaxel, we've seen that PSA response has been shown to closely correlate with radiographic progression-free survival and overall survival. In addition, ctDNA was also measured using the Guardant Infinity assay, which assesses molecular response to provide real-world assessment, which can be helpful in terms of determining treatment response early in the disease treatment, prior to any standard imaging. The ctDNA changes were measured based on aggregated tumor-specific methylation signal scores. Again, at putative therapeutic doses of 2 mg per kg or higher, 81% of patients had reduction of 50% or more in their ctDNA. And why is this important? Again, ctDNA has been used pretty significantly in other malignancies, such as non-small cell lung cancer, for quite some time. Not as much in mCRPC because we're more used to our more convenient biomarker of a PSA. But ctDNA has been demonstrated to have a wide range of applications in cancer management has the potential to be a powerful biomarker for assessing the long-term clinical benefits. A s you can see in these two studies, one was from, TOPARP-A, that's the left panel. They looked at ctDNA in patients with mCRPC treated with olaparib, and they found a correlation with RPFS and OS. And the right panel shows another study. It was actually just an ASCO abstract, looking at ctDNA dynamics and RPFS with Abiraterone, with or without the AKT inhibitor ipatasertib. CtDNA level changes may also predict for clinical benefits in mCRPC treated with chemotherapy or ARPIs. So this study looked at patients with mCRPC who had either an increase or decrease in their baseline ctDNA. The median time to progression for those with a decrease in ctDNA was significantly longer, as you can see, 6.9 compared to 2.8, in those who had an increase in their ctDNA. And again, we're lagging, in my opinion, in the mCRPC community behind, you know, non-small cell lung cancer, and now probably bladder cancer because they've been doing a lot more of the ctDNA to guide decision-making. At the top of this swimmer's plot, you'll see some durable times on treatment in several of our early participants. But please keep in mind, it's far too early to comment significantly on durability, given the substantive accrual into cohort expansions of levels 4 and 6, most recently. Lastly, here are the RECIST responses for the limited number of subjects for which we had these radiographic assessments. Of course, more data will come soon. From Cohorts 4 through 8, only 9 were evaluable. 56% of those 9 had target lesion reduction on imaging, and of those, two had prior PSMA-targeted radionuclide therapy. One was technically a radiographic progression, that red bar going down by RECIST due to growth in a non-target disease, despite considerable reduction in the size of two visceral lesions in the liver and the lung. One decreased from 38 to 14 mm in the liver, and then the other one in the lung decreased from 18 to 9 mm. So in summary, ARX517 has a strong safety profile at all doses tested up to 2.88 mg/kg on an every 3-week basis. There were no treatment-related SAEs, and there were no dose-limiting toxicities. At putative therapeutic doses of 2 mg/kg or higher, 52% had a PSA50, and 81% had a 50% or more reduction in their ctDNA. Target lesion reduction was also seen in 56% of patients that were evaluable, and two of those had confirmed RECIST responses. PSA responses were observed in patients who had prior PSMA-targeted radionuclide therapies. So in this patient population, without PSMA imaging selection, ARX517 monotherapy achieved favorable safety and demonstrated early efficacy, with deep PSA and ctDNA reductions and confirmed RECIST tumor response in patients with mCRPC who had progressed on multiple FDA-approved therapies. So the PSA, ctDNA, and RECIST results, in my opinion, support further evaluation of ARX517. Expansion of Cohort 8 has begun, as well as further dose escalation into Cohort 9. In my opinion, the future for ARX517 is very exciting based on this preliminary data. More data will come, and we'll see if PSMA imaging can inform the future use of ARX517. Additionally, I think it's important to keep in mind that as medical oncologists, we're familiar with prescribing and treating patients with antibody-drug conjugates. Access is important, and equally important, patients can receive these types of therapies conveniently on an outpatient basis. They do not need to be hospitalized for dosing or monitoring. There are many potential disease settings within prostate cancer where this product may eventually fit. Thank you. Great. So now we're going to turn to the question and answer part of our session. Before I do, I'd just like to thank also Dr. Schweizer and Dr. Pachynski for participating on our panel today, and also participating on our clinical study. So both of you, thank you very much. Before we go on the Q&A, we have prepared questions, actually, that we wanted to ask our panel. So I have to break out the glasses to read these prepared questions. So the readers are now up to the 2 instead of the 1.5, but here we go. Okay, first, if we could start with Dr. Dixit. Dr. Dixit, can you remark on the PK data? More specifically, have you ever seen PK data like this with any other ADC? So I can say that, I have been working with ADCs for the last 20 years. Started my career at Merck and then moved to J&J, and then AstraZeneca, MedImmune. For a long period of time, we had lots of ADCs in AstraZeneca. So most of the ADCs that I have worked with or I have seen, they tend to have half-life anywhere from 1 to maybe 4 days, 5 days, my maximum, but I have never seen any ADC showing a half-life of 8 days or 10 days. And why that is important, because as I said that as we have learned from chemotherapy drug, if you give too much chance, there's too much gap in the drug level over time, tumors will start figuring out how to throw away that drug. And they will start develop resistance. Sometimes they will lose the target, sometimes they will lose the pathway. And if you keep them exposed, as Scott's data also showed, that we are keeping this molecule very intact inside the tumor. And with longer half-life, you're going to keep it longer period of time, and you're going to see more durability of response. That's why it's important. Great. Thank you very much, Dr. Dixit. Appreciate that. The next question is for Dr. Tagawa, regarding prior taxane exposure. Dr. Tagawa, patients in our study thus far have had a median 4 prior lines of therapy, and 66% had one or more taxanes. Is there a reason to think that such patients are healthier than patients who have had prior taxanes? And so the question is, if the patients that have had taxanes are healthier o r multiple lines? Yeah. T he patients that have not had exposure to taxanes. So anyone that is able to get, I'm going to exaggerate, 20 lines of therapy, has been healthy enough to live that long, so that, that's part of it. I think the number of lines of therapy is a worse prognostic factor, whether they've had chemotherapy or not. Most of the, you know, drug that Oliver showed that were approved in the post-taxane era, then showed overall level in the pre-taxane era, they didn't require taxane for that mechanism of action. So, and I don't think this does either. I would say that this drug, even though the payload is a microtubule-targeted payload, the mechanism of action is different than taxanes. T hat being said, we know that cabazitaxel binds the exact same spot on beta-tubulin, yet it works in docetaxel-resistant cells initially, and then patients. So, I think it's, so to be honest, so any clinical trial, people have to be healthy enough to get on the trial. But that being said, median fifth line therapy, you can see that the tox data, I mean, there's nice PSA data, but the tox data are very clean. So despite that prior exposure to many drugs, including cytotoxics, this antibody drug conjugate behaves well in humans. Great, thanks. Thanks, Dr. Tagawa. Next question is for Dr. Shen. Can you comment on the duration being observed today? You gave a little bit of a comment, but can you provide any more color on that? Yeah. So I don't know if we can pull up that swimmers plot, but- Sure, I can. I think that's just representative of time on study. So these are patients that are clinically stable, maybe they're radiographically stable as well and not coming off due to toxicity. I think that's the best we can do in terms of commenting on duration on therapy at this time. Great. Thanks. Thanks. Appreciate that. So, Dr. Pachynski, are you collecting PSMA expression data, and do you have a plan, and do you plan to use it as a biomarker for patient selection, or would you plan to use that? PSMA? PSMA expression. PSMA expression data. Yeah. So I believe now, initially we didn't have the PSMA PET scans in there. Now, we've got them baked in both, pre, and on treatment, and also FDG as well, I believe. So certainly, given the, the VISION data, as using a PSMA PET as a, as a predictive imaging biomarker, you know, hopefully with more patients and enough of that imaging data, we'll be able to start to, to draw some correlations between that. Great, thanks. Dr. Schweizer, how does the data thus far in APEX-01 compare to Pluvicto from your point of view? I think it compares very favorably. I mean, you know, this... You've got to remember, this is a Phase I trial, so you're inherently enrolling patients who, you know, lack a lot of good options, and they're heavily pretreated, and we're in spite of that seeing response rates over 50%. I mean, that's very encouraging. You know, I think it's probably a little unfair to compare head-to-head, you know, the responses we're seeing here versus, you know, those from, like, P hase III trial. But, you know, I think just looking at it, just first blush, it looks pretty good. So I've been enthusiastic about it, and I think it's also notable that patients who've gotten Pluvicto are still responding to this in many cases. And so, you know, it kind of speaks to that these are different drugs. They're hitting the same target, but, you know, for some reason, patients who've failed Pluvicto are still doing okay with this, at least some of the time. So I think we'll have to see what the data shows as it evolves. That's, that's great. That actually dovetails the next question for Dr. Tagawa. Can you comment on the patients treated with prior PSMA-targeting radioligand? Yeah. So there's not enough experience to say that it doesn't matter whether they've had it before. What I can tell you is that there's a large experience of patients getting PSMA-targeted radionuclides, and then subsequently, a lot of them have PSMA scans on progression, and they're almost always positive. So loss of PSMA as a target is not any significant mechanism of resistance. We scan that at a response time; their scans may become negative. Actually, that was seen in the ENZA-p trial presented a couple of days ago. But at progression, the target is still there. And again, a different mechanism of action. So radiation versus microtubule targeting. So there's no... Other than just kind of inherent next line of therapy, having just a little bit more overall resistance than a prior line, there's nothing to make me think that these can't be used sequentially. They might be able to use combination, too, but sequentially, in either direction, I would think that they would work. Super. Thank you. I'd like to open it up to the floor for questions. Yeah, Phil. Thanks, Dan. Congratulations on the data. Thanks. Phil Nadeau from Cowen. One for Dr. Shen, then one maybe more broadly for the panel. First, Dr. Shen, is there any early data on the difference in response between patients who have had a prior taxane and not? Are you still seeing... Are you seeing any difference in response rate between those two groups? Then the second question, broader question for the panel, is where would you develop ARX517? What line of therapy in metastatic castration-resistant prostate cancer would this be most appropriate for? Thanks. Thanks, Phil. Yes, so in terms of the pre versus post, that question's come up quite a bit. I think it's an important question. I don't think we have that granularity yet. I don't know if Dan is-- Yeah. Yeah, good. So one way to answer the question is, of course, we've seen responses in patients prior to taxane experience, so they're included in the data set, and we've seen responses, PSA reduction responses, in those patients. Mm-hmm. How about the second question, Phil? Yes, the second question is, where would you develop this drug? Would you go, I guess, would it go with pre-chemo, post-chemo? I'm just kind of curious what the panel thinks the right environment is. Sorry, can you- Make the jump. That's an open for the panel question, right? Yeah. Okay, good. Yeah, please jump in. Yeah, I mean, I'll jump in. You know, I think I'd like to see a little more data. Clearly, you've got an active compound. It. The concept that Scott was referring to about a microtubule agent maybe being able to overcome some of the resistance that you would see with a Pluvicto radiopharmaceutical is interesting to me. And, you know, we don't quite know how many. There, you know, the few little asterisks here and there, and I think there's activity clearly within those patients who've been pretreated. Now, it gets to be a little bit complicated. So first of all, patients do retain the target. Scott mentioned that. Some of these patients could have been retreated with lutetium and responded there as well. You know, not all the patients who recur after lutetium are resistant to lutetium. And then you may have seen Scott's data with alpha particle and actinium. We know that's a pretty active agent. So I want to see a little more data here. I think it's nice to have a non-radio pharmaceutical in the sense that it makes it a little bit interesting, and I think it's nice to have a different mechanism of action because that really could open up some doors that would be a little bit interesting. Do you want to go head-to-head with Pluvicto? Boy, I want to see a little bit more, you know. Pluvicto is a real drug. It works. W ith the patient selection we've been using, we might be able to get a little bit better. But when you have an active drug, and I think this is, and it's a little bit different in the mechanism of action, even though the target's the same, I'd just like to see a little more exploration before I'd want to answer your question in the best manner possible. Great question. No doubt about it. It's obviously on everybody's mind, but I don't think we can go today. I don't. Maybe others do, but I don't say today exactly where we should develop it. More data is needed. Yeah. I just want to mention one thing t hat, that people probably don't remember. Actually, two things that people probably don't remember. So there's Kuroda et al., lab-based thing, where taxanes, probably because of their partial AR mechanism, will decrease PSA expression in addition to cytotoxicity. And it's PSA 30 that's most associated with near surrogate for overall survival. So you know, it's a lower bar there. I think that head-to-head, this could be taxanes. It's very... I agree with a lot of it. It's very early data. So yes, so could this go head-to-head with another PSA? Maybe. But, you know, cytotoxic versus cytotoxic, makes sense to me. I mean, I think especially given the EV-302 data that we saw today, which was historic, and ADC overtaking old-school cytotoxic chemotherapy, I mean, my short answer would be earlier in the metastatic setting. I think philosophically, in oncology, we've always sort of, we've always studied newer drugs in this late line setting, and I personally would like to see that moving forward. You need to see robust sort of efficacy data in sort of a Phase II setting to sort of justify that, of course. But, you know, if this pans out and translates to long progression-free survival, response rates continue to look good, I, I'd like to see it go sort of in an earlier therapy setting. I probably wouldn't compare it to Pluvicto because I'd worry they'd be very similar, and at the end of the day, that's gonna probably not lead to a registrational trial. But I think, you know, comparing it to, like, taxane chemotherapy, I agree with Scott. I think there's a good chance it could beat it if this, if this data holds up. Yeah, I mean, thus far, in terms of if we just look, you know, again, small numbers, but if we look at the toxicity that we're seeing in this and compare it to Docetaxel 75, q3 weeks, I mean, this is looking pretty nice on the toxicity. We didn't even see the neuropathy come up there. I had several patients on that said, kind of baseline neuropathy grade 1, and you're then kind of worried about, is this really gonna exacerbate? We really haven't seen a bad, you know, neuropathy signal. My experience has been very positive with the tox. I mean, it's been very tolerable. Like you said, these are frail patients many times, so using sort of heavy-hitter agents like CAR T cells isn't always feasible. But this, this drug's been pretty, pretty easy to give and satisfying in that respect. Thank you. Yeah, go ahead. So Joel Beatty from Baird. Congrats on the data. First question is, for the docs, what's the bar for success? What endpoints are you assessing as the data matures from APEX-01 study and assessing whether to advance ARX517? And then, second question is, can you describe the thought process that went into dose escalating to Cohort 9? Yeah, so we saw, you know, early activity in Cohorts 4 and 6, and had not reached the maximum tolerated dose and felt it was a good time to do both, continue to dose escalate as tolerated, because we had not seen any dose-limiting toxicities, and at the same time, expand to get more clinical efficacy data. A lot of those 4 and 6 patients are unfortunately early on, so a lot of those outcomes are still to be determined. You know, this is a Phase I study. So again, the goal is to see what's the maximum tolerated dose and what's the recommended Phase 2 dose. Thanks, Joel. Joel, do you have another question, or was that it? No. Okay. Hey, it's Matt from Oppenheimer. Maybe for the investigators, can you comment on the treatment emergent adverse events? We've seen the nice plots for treatment- related, but any just, what's your experience in terms of the overall patient population, how they're doing? Yeah, that's another question that's come up quite a bit. We initially had it, we just ran out of space for the poster. There was very similar to the treatment-related AEs, and ultimately, this is a late line study in advanced, you know, mCRPC patients, right? So they're gonna have a fair amount of treatment-emergent AEs on the protocol. Thanks, Matt. Hey. Reni Benjamin from JMP Securities. Can you talk a little bit about, on, on average, how long it took to bring your PSA levels down by 50%, and then ultimately, how long it took to get to maybe 90%? I think several of the patients got to PSAs, you know, greater than 90%. And how important, you know, ultimately, are objective responses here? Because we, you know, we're, we're seeing some, but, you know, ultimately, I think that's what's leading to, you know, what could be an overall survival benefit. Yeah. So in terms of the first part of your question, you know, there was... Sorry, could you repeat the first part again? He was asking about how- How long Like, how long it takes- Kind of, yeah. PSAs were checked every three weeks. I don't have the exact numbers in terms of what the average to hit the PSA50 was. I'm sure we could find that, though. And then in terms of the second assessment, in terms of actual radiographic responses, this is far too early to really assess that. One of the issues with prostate cancer is it just, it tends to be so bone-tropic that even late, late-stage patients just aren't radiographically accessible. It makes trials tricky in this space. I mean, I think really what you want to see is sort of a good early signal, and you do say it looks at rPFS. I mean, that's gonna be a better metric about long-term success potential. Yeah, I mean, I would say the majority, at least the majority of our Phase ones at WashU usually don't include a prostate cohort because they wanna see that resist shrinkage, those objective responses, and you can't do that if it's bone only, which happens all the time in prostate cancer. And so that's one of the issues, as Mike was saying, that it makes it tricky. So, maybe just as 2 follow-ups: How important is, you know, a PSA decline of greater than 90%? Like, how much should we be focusing on that, or even following testosterone levels at, you know, less than 0.2 ng/ mL or something along those lines? You know, I'll, I'll comment a little bit on, on Pluvicto. We, we had a poster last year at ASCO, where we looked at PSA decline, you know, like 50% and 90%, and patients who had a PSA decline of 90% did extraordinarily well. Now, that's a PSMA targeted therapy, different mechanism, but I think it may have a correlate. In general, how low the PSA goes in terms of percentage is gonna predict how long the patient will respond. I mean, that's a pretty reasonable relationship for the PSMA targeted Pluvicto anyway. PSA declines are tricky, though, right? Because, I mean, you can look at the experience with a drug like Docetaxel. You often don't see PSA responses for, like, 3 cycles. Yeah. So again, I would say that this is, this is a Phase I study. You're looking for, you know, early signs of efficacy, which I think we're showing, and then ultimately, that's gonna justify moving into larger trials. It's gonna have more sort of concrete endpoints that are meaningful, like rPFS. Yeah. Actually, let me, let me comment a little bit on that. But, I was very careful to talk about PSMA-targeted drugs. When you go from mechanism to mechanism, you don't necessarily have comparable declines in PSA translating into comparable long-term events. I'm gonna point out cabazitaxel. So cabazitaxel does not have the type of PSA decline profile that you would anticipate would be a really good drug. But take a look at the survival, and I don't know if you know about the TheraP trial, the one from Australia, and blasted... PSMA blasted the cabazitaxel out of the water when it came to PSA declines. When it came time to survival, no difference. So you gotta be really careful when you're switching mechanism of action to equate the PSA declines. They're not the same. I think just one last one for Scott. The accumulation of the drug, it's. I think you had shown Cycle 1 and then Cycle 3, and it's accumulating, unless I, you know, looked at the graph wrong. Do you anticipate any sort of a drug holiday as this continues to get developed? So I think there's got to be... You have to reach a steady state, but there was some early data that there may be some accumulation, but that may not actually be happening. That being said, I think that, you know, based upon the preclinical data and practicalities every three weeks there, but I don't know that that's the right, if that's the right answer or not. So, you know, we're still early with this, and I can imagine that there is an induction period of time and a maintenance period of time where there are different dosing, whether it's amounts administered or the interval. But anyway, I think the data are pretty clean for the Q3 weeks, but I don't know that's gonna be the right long-term pathway, depending on how this works out. Great. Thanks. That's gonna be our last question. For those of you who are not jet-lagged, we do have a reception afterwards, and you can obviously feel free to ask more questions. But again, I would like to take a moment to thank our panel. We really appreciate it. Thank you.