Good morning. I'm Sam Semenkow, one of the Senior Biotech Analysts here at Citi, and it is my pleasure to be hosting BioAge at Citi's 2026 BioPharma Back to School Summit. I'm joined by Kristen Fortney, CEO, and Dov Goldstein, CFO of BioAge. Kristen and Dov, thank you so much for being here, and welcome. Thanks, Sam. Great to be here. Thank you. Why don't we just start off? It's been an eventful year for BioAge. I think we started off with a lot of highs for NLRP3, and recently we've had some news. Why don't we just start broadly, level setting on the NLRP landscape, and also where BioAge sits today? Yeah, for sure. No, as you said, it's been a super eventful year for us, and for the target. When the year started in January, we were in the middle of our phase I. We published the full... Oh, am I not on? Good. Do I need that for the webcast? Yeah. There. You're on. Is that better? Yeah. Okay, cool. Yeah. When we started the year, we were in the middle of phase I with BGE-102, our lead program. In April, we released the full data set, showing we have a potential best-in-class inhibitor with, again, potential best-in-class brain penetration, full suppression of IL-1 beta, and also CRP results, which is relevant to multiple different indications. Then we started two different trials, one in an obese inflamed population, a phase II, and as we announced just yesterday, a phase II in a diabetic macular edema trial. We've got two parallel indications going on, so we're really excited about that. Of course, for the target, there was the setback a month ago, and I'm sure we'll talk about that. Or, well, it's not a setback for the target, but the IL-6 setback and the way that people are thinking about cardiovascular outcomes trial in particular for this target is, of course, a riskier proposition. There's still a lot of potential that we're excited about. By the end of the year too, we're still on track to file IND for apelin as well. It's a good year for us. We're going to dive into all of that. Maybe we do start with the ASCVD opportunity, and just maybe just help us frame how we should be thinking about NLRP3 going forward, given the IL-6 failure from ZEUS, and then the two other studies that were halted for that particular asset as well. I think one of the major questions is, we thought that the translatability of maybe CRP reduction and IL-6 reduction would translate to a reduction in cardiovascular risk factors, and that didn't seem to be the case. When we think about the landscape for NLRP3, where does that leave us, and how confident are we that this can be a target in ASCVD? For sure. Of course, the ZEUS news from Novo Nordisk just over a month ago, a huge disappointment. Really unfortunate to see that, especially for patients. As you mentioned, Novo Nordisk did indicate that they saw CRP reductions on par with what they were expecting, so we're assuming somewhere in the 80% range, but that failed to translate into MACE outcomes, at least in this patient population. That's really unfortunate. There's still two potential explanations for what's going on, and those won't be resolved really until we have additional data. There's, of course, hopefully more data from Novo Nordisk coming in November, and then the other trial that they have next year, the ARTEMIS trial, will read out in the first half of next year. But the two explanations, as people are aware, is that is it a target problem or is it a patient population problem? Because, of course, the inflammation hypothesis is partly anchored by some very robust data in the CANTOS trial, where an IL-1 beta antibody drove very profound reductions in MACE. As a function of CRP, the biomarker that you mentioned. Amongst those people who normalized their CRP, they saw a 25% MACE benefit, which is super meaningful to patients. It was really disappointing to see a hazard ratio of basically one with IL-6. The two things to be aware of that I think are being discussed right now is, one, that the ZEUS trial population was a lot sicker than the CANTOS patient population, so these people have profound kidney disease. There could be a lot of events that are not related to cardiovascular inflammation that could make it challenging for this mechanism to shine. That's why everyone is still looking to the ARTEMIS trial reading out in the first half of next year, which is IL-6, but taken to a very CANTOS-like patient population. But the second explanation is that this might be the wrong target. We don't really know yet, because again, what worked before was IL-1 beta. IL-6, while it reduces CRP, does not affect IL-1 beta at all. Of course, that's different for NLRP3. IL-1 beta is already validated in this population, and the entire target engagement biomarker for NLRP3 is IL-1 beta, and we showed in our Phase I that doses of our drug could fully inhibit IL-1 beta. It's really quite a different proposition. But the unfortunate thing is that CRP didn't mean what it used to. Before, we were hoping, no one ever thought it was a causal biomarker. But we were hoping that across patient populations, lower CRP would mean improved MACE outcomes, and that's not true. It's not enough by itself. It might depend on the combination of the biomarker with the patient population or with a particular target, like IL-1 beta, but it's just more uncertain. You mentioned that you can get that full IL-1 beta with your drug. You've seen that in, I think, your phase I healthy volunteers. Right and your obese patients so far that you have data for. So you're already reaching the max that you need to be CANTOS-like, but you've got better safety on infection risk. How do you reconcile that? Yeah. There's different pathways that can turn on IL-1 beta, and what you want in your ideal profile is you want to turn off the bad constant background inflammation that's turned up too high, which we believe is through NLRP3, but you still want to, if an infection comes along, be responsive to that infection. Because of course, CRP. The inflammatory response can be very useful, too. It's not universally a bad actor. So what's nice about NLRP3 is that you're turning off the NLRP3-dependent IL-1 beta, but there are parallel inflammasomes that should be responsive to infection. There's a potential for a safety advantage over an IL-1 beta. Right. Okay. We now have the CANTOS data. We know that there was a stat sig MACE reduction. With an NLRP3 doing the same thing, where is your confidence, then, that NLRP3 could still work in ASCVD, and how do we move forward to get confident enough to run a large study that would be required to show it? Yeah, no, that's a great question. Part of the challenge here is just that it's large and long, uniquely large and long. It's not one of those phase IIs that you can run in a year for $20 million. It's $500 million of spend, five, six-year massive outcomes trial. I think strategics are openly still bullish on NLRP3. Novo Nordisk, AstraZeneca, they say, "Oh, we might have to go upstream. NLRP3 is promising." But it's a big bet to make, especially for biotech. That really relates to the rest of the story and for our focus at BioAge, because the focus, of course, has been on ASCVD for a few months because of this ZEUS readout we were all so excited about. But this is not a target that is relevant to only one narrow disease indication. There are a whole number of different indications that are de-risked with human, either IL-1 beta or IL-6, depending on the indication. There's a lot of other places to take it. I think that for us as a smaller company, we want to be ready, in case ARTEMIS reads positive, in case things shift in our favor for ASCVD to move quickly. But that's really upside now. We really want to focus on these opportunities that we, as a small biotech, can move forward that are also really big and exciting. Right. We're definitely going to dive into the ophthalmology piece and beyond as well. But just really quickly before we go there, for the QUELL-CV study that's ongoing, we're going to have data before the end of the year. What can we take from that? What should we be looking for in that? Because it seems like more external factors are driving some of the next steps there. But mechanistically, I think there's probably something we can learn from this study. So what should we be looking for when the data reads out? Yeah, for sure. So our QUELL-CV study, and we announced as part of our ocular press release yesterday that that one's fully enrolled on track for top line by end of year. As a reminder, that's actually not in a cardiovascular disease population. That's a lot broader than that. So it's really closer to our phase I obese inflamed cohort. So these are people who are overweight, who have elevated CRP, and that means that they're really helpful to look at a host of different markers that are relevant in a variety of different contexts. Like for example, we're looking at liver inflammation with MRI because Ventyx saw some really promising data there. There's a whole host of other markers and diseases that are relevant in these patients. NLRP3 has relevance in other metabolic indications beyond the cardiovascular one. I think we're going to learn about, frankly, all of those from the biomarkers that we look at in this patient population that's at risk for numerous different metabolic diseases. Cardiovascular is one of them, and a big one, but not the only one. Importantly, too, it's going to be relevant for dose selection for really any peripheral indication we take forward. As a brief reminder, in QUELL-CV, we're looking at three different doses of our drug, 30 mg, 60 mg, or 90 mg once a day. We're looking at every biomarker you can think of which is cardio, of course, but it's a lot broader than that. As a reminder, too, the reason why we're doing these three dose levels is because in our phase I data, we couldn't really distinguish between the two top doses, like 60 mg once a day might already sort of max out all the inflammatory biomarkers, and we're going to learn that from this study. It'll really let us select a single dose that we'll then carry forward into our other indications, other peripheral indications that we're excited about. Got it. Okay. Well, looking forward to that and maybe some additional indications from that data set in the future. Okay. Maybe we spend a good bit of time then next on ophthalmology. As you said, you initiated or you announced that you dosed the first patient in the QUELL-DME trial. Just maybe level set for us. You're the only one that I'm aware of going into ophthalmology with NLRP3. Just what is the rationale here for this mechanism to work? Yeah, for sure. This is a unique feature of our molecule, too, which is that as an oral agent, it penetrates the eye really well. Which can come along with brain penetration. The blood-brain barrier and the blood-retina barrier are not dissimilar. But we don't know if others are even positioned to go there should they choose to. But we're really excited, the indication is that there is already human de-risking data with IL-6. Unlike in cardiovascular disease, IL-6 significantly improves what is the approvable endpoint in that indication, like it improves BCVA. Roche and Kodiak Sciences have bispecific programs with VEGF and with IL-6 that they're taking forward. We really see IL-6 as more of a potential floor when you think about all the things that NLRP3 is doing. It has the potential additionally to suppress pyroptosis, to have other addition. There's even a VEGF drying effect pre-clinically, so we can see full correction, basically, of the DME deficits in pre-clinical models. We're really excited to bring that to patients. Of course, that's even ignoring the fact that this is a pill. Typically these patients, it's an injection into your eye, so they don't even start treatment until they're quite severe because patients don't want an injection into their eye. There's that whole sort of early patient population that isn't addressed at all. There's a number of them that aren't VEGF responders as well. Or another mechanism can make sense. We're really excited about the potential there. As a final point too, I'll mention that DME happens in the context of diabetes, right? Preclinically, there's evidence that NLRP3 inhibitors can have numerous benefits in diabetic animals. We're waiting for more data from Novo Nordisk to see what they saw in patients as well. The body is not just a convenient way to get to your eyes. There could be benefits there as well, in the diabetes that is really driving the DME. Are you looking for that in the study, the more systemic effects of? Look at peripheral biomarkers as well. Yes. Okay. Yeah. Does that include, I guess, glucose tolerance and things like those? No. Well, all the usual just biomarkers, not like specific sort of challenge tests. I see. Biomarkers. Yeah. Okay. You will get an early feel for potentially what impact you could have in a broad diabetic population. Yeah, additional benefits that could be relevant to the eye as well. Okay. Yeah. Going back to IL-6 in the eye, when you talked about the validation that we have seen from there, when you think about the IL-6 reduction you get from BGE-102, is that on par with the dedicated agents, or is it a little bit less of an IL-6 reduction plus some other benefits? How do you delineate the IL-6 specific effects of BGE-102 versus some of the other effects that it is possibly having? Yeah, for sure. Basically our NLRP3 inhibitor, like others, reduces IL-6 by about 50% peripherally. We would expect a similar amount in the eye. That is about comparable to what we saw in the brain as well when we looked in the CSF. It is not like a full ablation of IL-6, but again, this is one of many inflammatory cytokines that are relevant in DME. We think it is important to go upstream and hit all of those, as opposed to IL-6 being a special cytokine there. We've seen some difficulty in some of these trials with agents in combo with VEGF trying to improve upon VEGF monotherapy. I guess, how do you think about that dynamic, and what gives you confidence that NLRP3 inhibition in combo can perform better than VEGF monotherapy? I think we've all recognized that VEGF is a very effective target. No. Part of that is the data we've already discussed, right? There already is a significant improvement when you add IL-6 on top of VEGF. Then of course, there's also this huge patient segment that doesn't get VEGF because they're not ready for eye injections yet, right? Where a monotherapy oral that has a significant effect would be very meaningful. So both of those are exciting opportunities for us. Is the study designed to test both combo and monotherapy for BGE-102? It is. Maybe just walk us a little bit through the design. Yeah, for sure. This is, as we announced, the last time we spoke about this, we were probably thinking about it as a biomarker trial. It is a true efficacy phase II now. About 180 patients. There are patients who are getting our drug alone. There are patients who are getting VEGF alone, and then there are patients getting the combination. We are going to basically be able to learn about all of those in the same study. The primary endpoint is BCVA, where we have good power to see a clinically meaningful difference. Is this population on the more severe side, more typical DME population, not the earlier stage that you were alluding to as that is a potential future population? Yeah, it is a typical phase II DME design, so it will include both naïve patients as well as ones who have had experience with VEGF and who have washed out. What do you need to see? Just improvement on BCVA. I am sure that there is a statistical, and remind me, how many letters do you generally have to gain to make it clinically? Well, one line would be really exciting. Just one line? Yeah. And that is what has been shown with the IL-6 as well. Yeah. Okay. Got it. If you have that in any of your monotherapy and/or combo arms, that is enough for you to want to move this forward. That is right. No, that would be really exciting. That data is going to come before the end of next year. We could potentially go into a pivotal trial after that, right? This is the clinically relevant endpoint. That is going to be a really meaningful readout for us. As a reminder too, right, we see this as really the first ocular indication for BGE-102. We are really excited as well about geographic atrophy. We went for DME first because we could see proof of concept faster. We are also going to be looking at target engagement biomarkers in the eye, and some exciting data there could support broadening to look at not just DME, but also GA, also MACE. There is a number of opportunities where inflammation plays a really important role in the disease. Of course, it will be like totality of data as you are making those decisions, but let us call it one line on BCVA. Plus, you were saying some biomarkers. Are any specific to GA that you are particularly watching that would give you confidence to move forward in GA? It's a different disease, right? The confidence for GA really comes around the sort of the separate data package. Preclinically, there's some very nice evidence NLRP3 knockout animals are protected. We actually developed in-house, and this is in our corporate presentation, a natural model. I should say too, there's all the usual models of GA where you give the animal some toxin, they get some junk accumulated in the eye, and you can improve that with a drug. But what we also really like is that if you just take really old mice, just like really old people, they accumulate junk in the eye. It's a more natural model of the disease, and we also see a very profound effect there. We're excited about that already. There's also been some recent data. There was an update over the summer from a small company called Inflammasome Therapeutics. They have an implant that goes in your eye. It's not competitive from a modality perspective, but it's an inflammasome inhibitor, and they've seen some really promising early data, strongly outperforming the complement inhibitors there. There's a pretty good rationale. Yeah. Mechanistically, how does it work? How does NLRP3 inhibition clear out the junk in the eye, as you called it, or stop the progression of the geographic atrophy? When you remove the inflammation, does that just allow the body to do the natural healing that it normally would? Help sort of cleared on its own, right? But the various junk in the eye, whether it's drusen or lipofuscin, right? That activates NLRP3, which further weakens things, opens up the barrier, allows it to accumulate more, right? It can sort of stop the things getting worse, and also, as you mentioned, help things clear on their own, and potentially even improve. Yeah. Got it. Okay. Just from a commercial standpoint, this is an oral therapy. We know that ophthalmology is used to injections or eye drops. This would be new, right? I cannot think of an oral therapy for the eye. I do not know if you have examples, but just how do you think the patients and physicians will respond to this? I am wondering if you have done any market research here. Yeah, there is a lot of interest because eye injections are not appealing to patients, right? There is really a lot of demand for something like this. Do you want to take that one, Dov? Oh. Can you hear me? Yes. Great. Okay. Oh, no. Yeah. Oh, no, we cannot hear you. Does that work? I turned it on, pushed it up. What? I just did. It's on. It looks on. Testing. There you go. I think we got it. Okay. Kristen alluded to this earlier. If you look at the distribution of the market, about 40% of- Oh, maybe just share. Oh, I figure it's just at the screen. Sorry about that. About 40% of the market doesn't get injections, so this is the watchful waiting group. The physician, it's generally a little bit of edema without vision loss. The physician and patient decide together to not give injections and just wait. That's an optimal group. You mentioned earlier that's a different treatment group, clinical group that we'd have to address later. There's a group of patients with VEGF who it doesn't work anymore, so they're refractory to VEGF therapy, and that's another group that's optional. This could come in and replace what is basically no option at that point. No doctor wants to give injections of drugs that aren't working for patients, and you'll see that over a serial progression. Those are the two optimal groups as we talk to people and talk to physicians. Just the general desire to treat earlier in the paradigm in that watchful waiting group is there for sure. There's nothing in the development pipeline landscape that sort of fits that exact mold beyond your molecule right now, right? Everything else is injectable, so it wouldn't really work for that watch-and-wait population. That's the big take-home. Everything else is combo-ed with VEGF. If you don't respond to a VEGF, you have potential to respond to a combo, but your combo could also be quite effective. That's right. No, we think there's a big opportunity here, for sure. Yeah. Okay. When you look at the overall competition in the space, obviously there are the IL-6 inhibitors that are probably the most direct competition that we could make for NLRP3s. There's probably other mechanisms out there as well. How do you think about competing versus them? Is one line enough in the context of some of these other mechanisms that maybe could achieve a similar level of efficacy, but perhaps maybe not quite as safe? Yeah, it's mostly the point that Dov mentioned, right? Is that there's these kind of wide-open patient segments that are really only going to be addressed by an oral, especially the sort of early watch and wait. So there's not a lot of competitive intensity there. Yeah. Got it. Okay. Anything else you really want to highlight on ophthalmology before I switch topics? The changes we announced this morning are important also for the timeline of the entire program. This is a true full phase II study in as much as the primary endpoint is a primary endpoint for approval. That will allow us to go immediately into a phase III study, therefore accelerating the program as compared to having to go into a subsequent phase II study. That's a very important point as well. Yeah, that seems prudent. Is top-line data still expected next year? Yeah. Second half of next year. Second half of next year. Wonderful. Looking forward to that. Before I leave NLRP3, I just want to ask again about other opportunities that you could bring forward. You mentioned in the beginning, Kristen, that there's just a lot of opportunities for this mechanism. How are you thinking about prioritizing, and how are you thinking about capital allocation, in particular, for any additional indications or disease spaces that you would want to enter? Yeah, no, we're well-financed with cash on hand, and that includes the potential to support one or two additional indications with NLRP3, and it is our intent to communicate another one, at least one other one, by the end of this year. But I don't have too much more to say about that beyond what I just said. TBD. Again, inflammation is important in so many different diseases, and there are a lot of indications where there are even approved therapies, right? Where it might be an injectable anti-inflammatory, either IL-1 or IL-6, and even just the oral prospect is very appealing. So there's a lot of very attractive opportunities in this space. Yeah. We look forward to hearing more there. Okay, let's spend a little bit of time then on just your APJ portfolio. You mentioned that you're still on track for IND later this year. These are next-gen APJ agonists for obesity. Any update that you could share beyond IND still on track on where these programs stand, or where do you think we'll hear more information on next steps? Yeah, for sure. No, we like to share all our additional data when we do the IND, when we get to that stage. So more coming soon by end of year. But briefly, these are incredibly potent orals and biologics that have the potential to enhance weight loss, but also increase body composition. Now that I think we've seen some of these oral incretins on the market, if anything, there's even more need for enhanced weight loss. So that's a very exciting proposition. That was true for our earlier program as well, that you could have a combination pill that achieves injectable like weight loss. That would be really exciting with an orforglipron or with some of the other drugs that are being advanced here. Then in the injectable space too, I think we are going to see more focus on quality of weight loss and muscle retention, which is where the injectable program has the chance to shine. Yeah. Got it. We will look forward to all of that. I am curious, do you think that NLRP3 still has a path forward in obesity? We kind of crossed it off last October with the Ventyx data. You cannot rule it out. I think we will know for sure by the end of this year. I am assuming Novartis did not see that in their diabetes trial either, or we might have heard about that. As a reminder, our QUELL-CV trial right now, it is an obese, inflamed population. So if something is happening, we will have the power to see it. But given that is purely in the upside category for us, given that Ventyx with their brain penetrant molecule, really just did not see anything there. Yeah. Fair enough. I want to go a little bit into the broader business as well. The foundation of the company is the proprietary longitudinal Healthspan database that you have. It is something that you have leveraged through partnerships as well as your own drug discovery. When we think about where BioAge could be in, say, the next five years, how should we be envisioning additional assets coming out of that pipeline, and how should we think about the internal pipeline perspective of growth as well as partnership? Yeah. Happy to speak to that. BioAge really started off by us creating data sets that we thought would be useful for target selection and ultimately drug discovery. It was the signals in those human aging data sets that gave us initial conviction around NLRP3 and also around apelin, and informed our drug programs there. It is those same data sets that have driven our partnerships and collaborations with Novartis and Lilly. What is unique about our data, it is in a sense very much like deCODE or UK Biobank, where it is with large population cohorts, but with one key difference, which is that it is the same individuals longitudinally profiled over five to six decades, so over most of their life, over the whole period of aging. We have a big collaboration with Novartis, and what excited them about our data was that you really want to go and see what is different about people before they get sick. Like what changes precede cardiovascular disease or Alzheimer's by a decade? What are the earliest drivers that you can modify as opposed to those changes that only happen after you are very sick. When intervention, frankly, might be too late. Separately, you might be able to learn from those particular humans who age really well. There are people in our cohorts who make it to 100 years, 105 years in still relatively good health. So what is different about those people that can teach the rest of us? With Novartis, we have an ongoing target discovery collaboration, and the focus there is on completely novel targets that no one has ever worked on before, that they want to take forward. That is a collaboration where they provided an upfront and are supporting our platform. It is great non-dilutive funding for our science to keep that part of the company humming along. With Lilly, it is a couple of promising targets that we have selected where they are helping us build drugs that we can potentially carry forward. But to your question, yes, both through collaboration and also internally, we are always working on additional programs and like apelin, like NLRP3, we will over time, advance other molecules to the clinic, with a similar kind of data package. Cardiometabolic potential, really nice grounded human data signal as well. Yeah. Is BioAge leveraging AI within the business, particularly within this part of the business at all? I am just curious how it might be used in your business. Yeah, for sure. The everyone question. We were a data play from day one. We have these huge data sets and we've, leveraging AI to find promising targets and to integrate the universe of data, has always been sort of really a core of what we do. Actually, what's been the most exciting thing about the past year actually, is probably using LLMs to help with their regulatory docs. All those stacks of paper. Seems like a good use, yeah. We love it for the data science too, and that's a core part of our business. Yeah. Then just a question on how to think about translating genetic data into a drug. You have a person who has a variant and has a lifetime of exposure to that variant, and it's deriving benefit from it. But when you create a drug, it's a shorter-term exposure. So when you are picking new targets, just across the board for yourselves, for Novartis to whoever, how do you think about balancing the short term versus the actual variant? How do you make that distinction on what's a real drug target? Well, that's a great question. It's something we worry about. Human genetic data is great, of course. People have published papers showing it more than doubles your chance of success in the clinic, but those are still poor odds, right? Like double a very small thing is still a very small thing. I think for us, we insist on having that human signal to know that it's relevant in the human, to know that there is that potential to achieve a benefit with a human. Then we do rely on, like that's what animal studies can be good for, right? If I actually intervene in a model, and in our preference, like a natural aging model as opposed to some very artificial disease construct, can you fix it in a short amount of time? That's also where there's precedent too, to help us, right? What's very nice about our lead program, NLRP3, is that the anti-inflammatory class has been in a lot of human diseases, and you can make a list of things where it does work in the near term and where it doesn't, and that lets us focus our development efforts. Excellent. Okay. I think we're coming up on time, so I'm just going to turn it back to you, Kristen, and perhaps Dov, to just give any closing remarks, remind us of the cash runway, and I think that there is plenty that we can look forward to over the next 12 months from you as well. Yeah, for sure. No, so yeah, it's going to be exciting next year for us with our cardiovascular QUELL-CV and metabolic trial results coming out end of this year. We're looking forward to hearing more news from ZEUS in November as well, right? Then in the second half of next year, we'll have, again, on NLRP3, this really meaningful phase II efficacy readout for our ocular program. Then, of course, by end of year, the APJ IND filing, right? Then we'll have some clinical data for that next year. So, a lot's going on, and do you want to summarize the cash and plans? Yeah. Yeah. At the end of the second quarter, we had $381 million. We've guided to that being cash through 2029. That, as Kristen mentioned earlier, is sufficient funding for us to do one or two additional NLRP3 indications. That runway will come in a little bit, but not a huge amount to do that. Yeah, and that's the cash guidance. All the activities that Kristen mentioned for APJ and NLRP3 are all covered in that guidance. Wonderful. Okay. Well, thank you so much. This has been great. I really appreciate the time. Thank you. Thank you.