Good afternoon, everyone. Welcome to this session of the Morgan Stanley Global Healthcare Conference. I'm excited to welcome the Arcellx team to the stage. Let me just read a quick disclosure before we get into it. For important disclosures, please see the Morgan Stanley Research Disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. I'm Judah Frommer, one of the SMid Cap Biotech analysts here at Morgan Stanley. I'm excited to have Rami, Chris, and Michelle representing Arcellx. Maybe we just start out with a bit of background for those newer to the story. Maybe we spend a minute or two on an introduction to Arcellx, some background on the company, your lead asset, and maybe also your development partnership. Great. Thanks so much for having us, Judah. We're excited to be here. Arcellx was founded about 10 years ago with really the goal of addressing the three primary needs in cell therapy, which are safety, efficacy, and scale and reliable manufacturability that can drive accessibility. We've been able to demonstrate that leveraging our core technology, the D-domain, and our lead asset, A nito-cel, which we believe represents a best-in-class therapeutic option for patients in multiple myeloma. We're also leveraging this technology in our pipeline. We have active trials currently in AML and myasthenia gravis, where we, again, believe that core differentiation of this technology can hopefully open up new fields that have been historically with a significant unmet need. When you pull all of that together, we really believe we're on the road of building a generational company. One of our core goals isn't just to have these incredible assets like A nito-cel in the myeloma space, but to really show that you can build a scale, profitable cell therapy company and hopefully through that avenue attract more investment in the space that we believe can be a forward pillar of medicine. I'm sure we'll get into A nito-cel a lot, and I'll keep it there for now. Okay, great. Like you said, your lead asset is the BCMA-directed CAR-T, Anito-cel. You provided an update at EHA. You had at least four months of follow-up for all 117 patients that were dosed, median follow-up of nearly 13 months, and your iMMagine-1 study that enrolled late-line, relapsed refractory multiple myeloma patients. Maybe just can you walk us through the efficacy highlights from that update, and then equally important the safety side of it, but break it up how you see fit. Chris? Yeah, sure. The data that we disclosed at EHA was actually mostly an update of data that we've disclosed last year at ASH. In those data, what we're seeing is continued deepening of the complete response rate, now showing over that previous number in the 60s, and now we're seeing in the 70s for the CR rate. The PFS rates are the way we've disclosed the durability of the benefit. Looking at the 6, 12, and 18-month PFS rates, you're seeing 92% at 6 months, and then out at 18 months, you're still seeing patients without progression 66% of the time. That was one of the bigger updates because we had heard questions around whether we are seeing that continued durability and will that tail develop on the curve. We've also reiterated that the MRD negativity rate is in the low 90% range. We do see some of these numbers change just marginally, maybe 1% or 2% up or down as more patients get more visits that have happened and more samples have been taken. Overall, I think we're really happy with where the data are in iMMagine-1. It looks, as Rami said, like in terms of efficacy, it's on par with the best-in-class therapy. Where A looks to be differentiated, and this was also updated at EHA, is now all the patients have at least four months of follow-up, and we still have not seen a case of delayed neurotoxicity or a case of immune-related enterocolitis. If you look at the data from CAR-T2-1 or CAR-T2-4, those events have really almost exclusively happened in the first four months. There are some outlier cases, but most of those events are happening actually closer to 60 days or 90 days after treatment. Not having seen any of those types of events is one of the major differentiators we believe from other therapies available. Finally, we have mentioned that in Imagen-3, which was not part of this EHA data cut, but it's an important point. In iMMagine-3, Kite is manufacturing the product for that clinical trial, and we're seeing turnaround times that are actually pretty similar to what you expect with Kite's commercial CAR-T cells, which is really, we think, exceptional for the BCMA CAR-T space. Okay, great. Maybe just to follow up on the evolving safety profile, right? It's certainly something we get inbounds on. I guess maybe just kind of KOL and clinician perspective, once you did share the EHA update, obviously the efficacy profile is very important, and you expanded that with extended PFS with your second quarter earnings. On the safety side of things, I think how have folks in the industry and experts come around in terms of their view that the safety profile is fully formed versus what they might be waiting for? Yeah, I mean, I think I can maybe jump in and answer that in a couple of different ways. I think, one, we shared market research at EHA, which represented a fairly large sample of physicians, around 150 physicians, both across 2024 and 2025. You can see the increase in assumed share for Anito-cel, and that's primarily, I think, an increase in the confidence of the profile as we've continued to share more data, as well as the experience for these physicians with the available therapies right now and some of the challenges that you see with safety. I think it's clear that there's more of a debate, I think, in investor circles around this topic than there is in the hearts and minds of physicians. I think a plurality of physicians who are aware of this data see it as clearly differentiated. Again, as Chris said, when you have a minimum follow-up of four months and the overwhelming majority of these cases happen within the first one or two months, it's clear that this is not something that we expect to be related to this particular agent. Okay. There may be less of a focus now as you continue to generate data, but I guess, you know, the D-domain's potential contribution to the safety profile is a question we get less, but still get. What are your thoughts there? Yeah, I mean, look, I think that's also fairly clear. If you look at the real-world incidents of these delayed neurotoxicities, it's very clear that they're related to one product. I think there's been this debate around, is this a class effect or whatnot, but it's very evident that when you have quarter- over- quarter, 6% rates of Parkinsonism, 6% rates of CNPs, and, combined with the clinical data that we've seen, it's very clear that it's binder related to us, in our opinion. Certainly, when you conversely don't see any of those cases with 155 patients with the kind of minimum follow-up that we've seen, you also feel very strongly that it's not related to Anito-cel. I Anito-cel. I think it's clear, at least in our view, based on the evidence available, that it's specific to the binder. I think more and more information hopefully will come out that mechanistically explains why that is. Okay. Speaking of more information, what can we expect in terms of subsequent updates from iMMagine-1? I guess, internally, what are the focus points that you're looking for as you do balance that safety and efficacy profile, which is more interesting to you at this point? Yeah, look, we're excited for ASH coming up here in December in Orlando. We expect to share more data at ASH. I think you should expect we had a median follow-up of 12 months at EHA. You should expect another five or so months of additional median follow-up, so around 17, 18 months of median follow-up. What I would look for there is, I think, obviously, the constellation of all of these different efficacy markers have trended very well. When you look at MRD negativity again in that low 90s rate, around 93% that Chris mentioned, when you look at an 18-month PFS rate of 66%, I think suffice it to say we're very confident in the profile and the durability of this therapy. It certainly stands out on the efficacy side. On the safety side, you're going to get an even higher sort of floor in terms of that minimum follow-up, and we feel like that can continue to reinforce the safety differentiation. Okay. That makes sense. I would argue that more and more we're getting questions on the regulatory path in multiple myeloma. Generally, you know, obviously, given changes at FDA, there have been some CRLs across the broader space. Can you talk to the regulatory path in multiple myeloma, both late-line and early-line? Some folks point to iMMagine-1 being a single-arm study. Should that be a concern at this point? Yeah, we feel very confident in the regulatory path. It's very well established. I had a chance to attend the FDA CEO Forum with Dr. Makary and other staff from the FDA. I feel really confident in what they're trying to do with the FDA, which is actually reduce regulation, improve the competitiveness of American companies, and help these lifesaving therapies get to market even faster. They were very confident and very reinforcing in terms of the established regulatory paths and choosing the right endpoints and the right approach for the right disease state. We've since seen Regeneron gaining approval with a single-arm study in a setting that I think is analogous for us very recently. We feel very confident in the regulatory path. We feel obviously very confident in the differentiation and the unmet need that Anito-cel can bring to the market and our positioning. I would also say that one thing that has given us increased confidence is that we haven't seen any turnover in FDA staff. That continuity is actually really, really important. We have FDA leadership that, again, we feel committed to advancing these types of therapies and is aware of the unmet needs that exist in myeloma that Anito-cel can address. I guess specifically on unmet need, do you feel confident there's ability to demonstrate unmet need in late-line where you do have CAR-Ts approved? Is that unmet need tied more to safety or efficacy? Yeah, so first let's start with precedent. Breyanzi was approved as a third-to-market CAR-T very recently in the last couple of years here, and with a much narrower differentiation than what A nito-cel brings to market. When you look at myeloma, it poses two actually opportunities for differentiation that didn't exist in that case study. The first one is the safety issue that we talked about a lot. When you have the best agent in class have a, you know, safety issue that, you know, is rounding up to double- digits, call it 10% or in the low double- digits, that certainly presents a significant unmet need. Our market research really enforced that and showed that there is an attenuation and desire to prescribe these therapies, particularly by referring physicians in the face of these safety issues. The other opportunity is access. One of the things we showed at EHA is that despite these two additional agents being on the market for so many years, they're still not covering the majority of the market. There's still an access issue. That really highlights one of the advantages of Anito-cel, both due to the properties of the D-domain as well as the best-in-class capabilities of Kite. Anito-cel We believe we're in a position to be able to deliver at a scale and an accessibility that has not been the case in the myeloma field. Certainly, those two are very clear unmet needs in the myeloma space and are far and above what was even demonstrated in the lymphoma space with Breyanzi, which was able to secure approval. Okay. That makes a lot of sense. I think already you've said is, you know, potential for a launch in 2026 in the fourth-line+ setting. I guess what's, you know, latest communication on timelines for BLA submission? You know, how do you anticipate communicating submission or, you know, FDA receiving submission? How are you thinking about just communication around that? Yeah, so we don't have an answer for you guys on communication yet. That's something we have to talk to our partners at Kite more and align on the communication plan. In terms of the timeline itself, I'm happy to reiterate our guidance, which is we expect to be on the market in mid to late 2026. We feel like operationally we're certainly marching toward that. Chris and his team have done a phenomenal job working alongside their partner, his partners and their partners at Kite to make sure that we're advancing the BLA filing to our expectation, and that's gone very well. We haven't given specific BLA guidance, but in order to enable that mid to late 2026 commercial launch, we would expect the BLA to be filed sometime late this year into early next year. We'll certainly have a more affirmed communication plan as we work that through with our partners at Kite. All right. I figured we'd try. Also, I think at EHA, you also talked about the opportunity for CAR-T across treatment lines in multiple myeloma. In your view and given the recent commercial trajectory for the commercial asset, you know, where do you think CAR-T is headed generally in multiple myeloma as you move into early line? Yeah, so maybe I'll start and I'll pass it over to Chris to comment as well. I mean, look, I think we've seen an incredible, and as you know, I know we talk a lot about sometimes the challenges and the deficits with some of the available CAR-Ts, but I think you have to applaud J&J and Legend. They've gotten their Carvykti up to a $1.6 billion, $1.7 billion run rate here despite a lot of those challenges. I think that's a great thing for the market. It's a great thing for patients. We've always been on the record as saying this is obviously a huge unmet need and there is an opportunity for multiple CAR-Ts and we're happy to play that role and present an option that can help more patients gain access. We have said historically this is a second line plus is a $12 billion market or thereabouts. We shared actually our class share assumptions at EHA, which hopefully are really instructive about where CAR-T fits relative to bispecifics and some other modalities. What we've said is that second line is about half of that market. When you're thinking about us launching Anito-cel next year, you should really think about us certainly launching in that fourth-line+ setting, which is what we expect, which is a $3.5 billion market. Certainly, as we've seen with the launch, there are always patients who will opt to go through one more treatment regimen, get the fourth line and get the CAR-T. It's going to be a fairly expansive opportunity. We're excited about it. I think I'll pass it over to Chris and he can talk about from a physician perspective how they're thinking about the adoption of CAR-Ts into earlier lines. Yeah, you know, Rami made a number of great points there, but I think one of the things that we also see as a real value of, you know, coming behind is that the data with longer follow-up, you know, with all the same indicators of complete response rate, MRD negativity, PFS rates, showing the tail of that curve with CAR-T2-1 actually puts physicians into, or they can actually assess now what is the trade-off of not giving a CAR-T. They can see that they might be giving up a long-term durable benefit for 30+% of the patients, whether that's fourth- line, fifth- line, or all the way up to second- line. It becomes a very different question for a physician when you know that there is something somewhat approximating cure. I know we're all a little hesitant to say cure yet, but that certainly feels for a patient a lot like a cure. If you can go five years not needing any other therapy, not needing to check in with your physician every month or two, getting labs, et cetera, waiting in the parking lot to be able to come in and do your infusion. Those things matter quite a bit. What I believe will happen is that not only will we see relatively rapid adoption of anitocabtagene autoleucel because of the safety benefit that you're going to see, you know, pretty quickly, but when it comes to then thinking about can I adopt in earlier lines, there's going to be more of an urgency to do that. The academic institutions that are going to have success doing that, I think, will become a model for, you know, community oncology centers. Not, you know, single doctors in a, you know, in a standalone building or something, but when we say community now, we really mean these multi-physician conglomerates, right, that are operating outside of an academic institution, but they work like a business. Those businesses are going to see an opportunity where I can give this therapy and I can take care of my patients well. You don't want to be the one that doesn't have it. We think that day is coming sooner than later. Overall, I mean, the feedback we get from our physicians that we work with who've used Anito-cel is this is the product I would use as soon as I can get my hands on it as an approved product. Okay. Yeah, I'd say that's pretty consistent with what we hear. Maybe just to follow up on the initial market opportunities. One question we get is how do you anticipate kind of commercial dynamics shaping up when you're launching with a fourth-l ine+ label against the product that has a second- line+ label? I know you touched on it a bit in terms of potentially, you know, waiting for a CAR-T later line, but anything you've heard, I guess, in your preparations for commercialization then? Yeah, no, actually, it's a good opportunity to highlight one other thing that I think conceptually is really important to take in, which is that this is an incident market. It's not a market where we have to go in and change scripts and make that sort of conversion. There is a certain number of patients, and unfortunately, given the incurability of myeloma, it's a very large number of patients that are presenting with the need for these therapeutic options every single year. Providing Anito-cel in, A, a number of centers that don't currently have access to CAR-T, and B, with what we believe is a safer profile, I think certainly will allow us to garner a significant portion of share. If you saw, again, our EHA presentation, our market research indicates that we would expect somewhere in the range of 80% share. I don't know that I'm going to certainly commit to that as a guidance, but I think directionally it tells you that there's a high favorability for this therapeutic option relative to the existing class of therapies, and it's one that we would expect and we're planning for from a demand perspective, hence why we shared, we expect to have significant coverage of that market right at launch and into the second year of launch into 2027. I think that's a really important thing to remember. Again, the incidence of the market, the fact that there are a lot of patients, the preference that we've seen for Anito-cel in our research and in our conversations with physicians, and our ability to actually deliver on that both through our launch plan and our capacity availability. Okay. To that point, and Chris touched on this a bit, I guess in terms of segments of the market or sites of care, as you know, like you said, I think we would bucket them as kind of academic versus community versus smaller. Which areas might be earlier to adopt? Is there this unmet demand particularly in the community, and how do safety and manufacturing turnaround times factor into that? Yeah, the best way to really think about this is that you really need the community engagement as you're pushing into second line. If you're talking about like third- line+, I think you're largely able to access that market without a ton of effort or push in the community setting. It's something that certainly I think we benefit from the fact that we have two of the largest, most successful healthcare companies in the world in BMS and J&J already working to kind of make those inroads in the community. I can tell you that Kite, being the leader that they are, I've already been in a room through meetings they have facilitated and Chris has as well with some of these leading community centers around the country. It's not that we haven't ourselves started to have those conversations, but that from an understanding of like what it's going to take and what their interest level is in adopting CAR-T more broadly. I think that ultimately in the third- line+ setting, I think you should, you're going to see a lot of adoption for Anito-cel. My guess would be almost right out of the gate. I think over time as we get our earlier line label, as some of this progress is made by some of the incumbents in the space, you'll see a bigger pull into second line. What we saw in our research is that what's actually really hurting right now, particularly among referrers, is that second line access because of some of the challenges that exist with the incumbents that we're optimistic and believe we can address. Okay, great. In partnership with Kite, you're executing on the iMMagine-3 trial in earlier line patients. How does that trial expand the opportunity? Maybe you can talk a bit about trial design and how that will factor into an opportunity for you versus the incumbent. Great. I'll pass it back to Chris for that one. Yeah, some important distinctions from prior clinical trials in the same space. One of the advantages, again, of being, you know, a couple of years behind in development is that you can look and see how the landscape has changed and what questions were raised from prior studies. As we compare iMMagine-3, for instance, to CAR-T2-4, probably the biggest distinction is that the patient population isn't defined by lenalidomide refractoriness. That definition is per IMWG, and patients have to have progressed within 60 days of the last dose of lenalidomide. It turns out that really only about 40% of patients in second line technically meet that definition. Most of them are functionally equivalent to lenalidomide refractory, but don't meet the definition. We wanted to remove that from the definition. Instead, we used exposure to therapeutics as a way to define the population. Like I said, most of the patients are not going to stop a therapy that they're still responding to. Functionally, they end up being refractory, but they don't have to meet the definition. That means that instead of 40% of the patients in second line, that's access to about 80% of patients in second line. That's a remarkable difference. The other thing that's really important is that at the time that CAR-T2-4 was initiated, there was not yet approval of KDd as a second line regimen. KDd really is the best second line regimen available. We were able to include that in our standard of care options for patients. That's something we've heard from clinicians, that one of the ongoing questions they have is, is this therapy really better than the therapy I have right here in my clinic right now? We expect to be able to demonstrate that Anito-cel is better than KDd definitively in second line and beyond. It helps us to answer some of those questions. The other big change that we made to the trial design after the study had actually already initiated was to add MRD-negative CR as a co-primary endpoint. The reason for that is after we saw the FDA ODAC on MRD-negative CR s, we felt like it was something that FDA will be willing to consider as a primary endpoint for approval. That may end up shaving some time off of when we can get the initial readout of the trial. We do think that the field is changing quite a bit and being able to read those changes and then anticipate them is part of how the study was designed. Okay, great. Maybe just for iMMagine-3, if you could lay out kind of base case and kind of best case timeline scenarios for data. You want to take that? Yeah, sure. The way we look at it is that the enrollment timeline is probably really similar to CAR-T2-4. It's a similar size trial. It's a similar population. It should take about the same amount of time. The median PFS expectation we have always said for the control arm should be somewhere around a year, ±10% or so. We think that the way the study is designed, that is still the expectation in that study. We haven't guided to specifically when you would expect data readout, but you can kind of take all of those pieces of information and make a pretty reasonable estimate of when we should have a good inference on PFS or MRD- negative CR. Okay, super helpful. Maybe just moving into a couple other programs toward the end of the conversation here. You're enrolling a phase one study in Myasthenia gravis, like you mentioned. What's the latest on enrollment for that study? What can you tell us in terms of anticipated updates there? Yeah, the study is enrolling, and it's something we think mechanistically makes a lot of sense. This is a disease that's driven by antibodies that cause damage to the neuromuscular junction by binding to the receptor there. Eliminating plasma cells is a really good way to eliminate antibodies. The vast majority of antibodies in a person are going to come from plasma cells. We know that. It's something we feel made a lot of sense from a mechanistic standpoint. The study is ongoing, so we can't really say much about anything we've learned so far other than we continue to push forward to enroll more patients. Our long-term look at this is that this has to be a therapeutic that allows patients to remain treatment-free for a prolonged period of time. We think there's a lot of mechanistic reason for that to be true, and we'll just need to dose enough patients and follow them to be able to give something meaningful. Okay, great. Just touching on the ARC-SparX platform, it probably doesn't get talked about a ton, but maybe it's worth reminding of the potential and differentiation of that platform and what we can expect that'll come out of there. Sure. Yeah. ARC-SparX, just for anyone who hasn't followed our story, is the idea that we can make a universal CAR-T cell, and the CAR on that CAR-T cell actually binds to a tag on the back of a protein that we call SparX. That SparX protein also has binding domains that target the antigen on the surface of the cell. We take the ARC-T cell, and when it combines with the SparX and then binds to its target antigen, it will kill the target. If you take any piece of that apart, if you remove any part of it, then nothing functionally happens. The CAR-T cells are essentially inert, absent the SparX. The SparX are inert, absent the CAR-T cell. We've had a program open in phase one in AML now for about two years. The study is, as you can imagine, mainly challenging as a result of the disease. AML is a really challenging disease because of how rapidly patients' disease progresses. Over the last few years, what we've been working on is dialing in dosing strategy as well as the patient population where we could see the most benefit. That's where we are in development now. What we like to tell people is we wouldn't still be doing this if we didn't think it worked. Whether something works and whether something's a product are two completely different things. We are currently in the process of trying to demonstrate that this could be a real product to help patients. We don't really have a timeline on when that will happen because of the complexities of the disease and trying to optimize dosing. Okay, great. Maybe the last company-specific one before we get into a mini survey we're asking all of our companies, but maybe just remind us of cash runway and what catalysts and milestones will be funded with that runway. Yeah, we ended Q2 with $538 million in cash, which gets us into 2028. That funds us through launch and, you know, substantially in terms of our paths beyond launch as well. Okay, great. Like I said, we're going to spend the last couple of minutes on a mini survey. That's all our management team, so we're not singling you guys out. Biotech seems to be more exposed to external and macro- factors of late. We're asking these three questions to all the biotech companies at the conference. The first is on China. With China's rise in biotech innovation, how are you thinking about Arcellx's competitive position, and will this influence R&D or business development in any way? I'll start on the R&D. Arcellx is certainly an American-made story, and we've found a lot of success running our studies here and executing here. I'll say also broadly, we have not seen a consistent enough level, in my opinion, of translation from clinical data generated in China to the U.S. I think for a lot of American companies like us, the idea of sort of jumping into the full rate accelerator program when not knowing that sometimes the patient populations are different or there are other factors that may impact the translatability of that data, it doesn't seem like a straightforward path for us. I'm also really encouraged, again, I mentioned earlier, attending the CEO Forum, the FDA CEO Forum with Dr. Makary, and the approach of trying to lower the regulatory burdens and increase sort of the acceleration of innovation in the U.S. I'm encouraged by that. For us, I think on the R&D front, nothing really changes. I think on the BD front, and notwithstanding what I said, I think you can't be blind to innovation wherever it comes from. We generally at Arcellx, from a culture perspective, talk about a company about where ideas win, not people. If the best idea happens to be in China or Germany or the Netherlands or wherever, we want to be able to find that idea and get access to it. Obviously, I know there were some negative results this past week, but Summit is a good example of having capitalized on an asset from what happens to be China. I think from a BD perspective, we're certainly open to it. R&D, I'm still a little bit hesitant. Okay, great. The next theme is AI. How are you currently leveraging AI or thinking about AI's future disruption potential? I'd say both positively and negatively for the space. I don't know if a lot of people know this, but I'm actually an electrical engineer. One of the very first things I did when I joined Arcellx is talk about how we're going to leverage AI. We actually are leveraging AI. We've got a great program. We're generating binders. We're optimizing D-domains using AI, generating new scaffold. We're at a point now where we're actually able to solve problems we couldn't previously solve. Sometimes it's hard to find a D-domain for a particular indication, but leveraging AI, we've been able to crack through there. That's an easy one for us because we've been working on it now for about four years, and we feel like we've got actually a pretty good capability. All right, excellent. Last one, we touched on it during the conversation, just on the regulatory side, anything that's generating internal conversations, changes that FDA, pricing debates, tariffs, anything you'd highlight? I know you touched on FDA, but anything else? No, actually, what I will say that is related, but maybe not direct, is that I think there's an underappreciation of the value in cell therapy in that there aren't patent cliff issues. There aren't IRA issues. There's actually a lot of value in what we're doing and in the approaches that we're taking. We haven't found that any of these regulatory or tariff-related things have been really impactful. Aside from volatility in the markets, from an operational perspective, it really hasn't had any impact. Okay, great. I think we are just about at time. Once again, thank you for participating. Good to have you guys here. Thanks, Judah. It's been a pleasure. Thanks, everybody.