Antibody therapeutics that uses an integrated wet lab validation and allowing for rapid generation of antibody therapeutics that are in clinical trials at this point. The company aims to generate target hard-to-drug targets as well as therapies to them with their first clinical candidate, ABS101, generating data in the second half of 2025. To discuss the AI-generated antibody pipeline and business strategy, let's get started with Alex. Good morning, Alex. I'm glad to see you here and accepting your invitation. Alex, briefly, for those folks who are new to Absci, could you provide an overview of the way you think about drug creation using your AI models? Also, how do you folks compare yourselves against numerous other tech bio companies that are developing not only small molecules but also large molecules? Yeah, absolutely. First of all, thank you for having us. It's always a pleasure to be here. Absci was founded about 14 years ago at this point, and the idea now is to apply generative design and engineering principles to create novel, differentiated antibody therapeutics for patients. One of the key ways that we differentiate from the start from a lot of companies that are doing AI for drug discovery right now is that we focus exclusively on biologics antibody therapies, not on small molecules. There are a lot of great companies out there doing great work on the small molecule side. When we think about what are the important factors for antibody discovery or what we call drug creation, it's can you really sort of hone in these attributes that would make for an optimal antibody from the start? Can you design the molecule to have the attributes that would make for an effective and differentiated antibody therapeutic? Things like affinity, developability, manufacturability, immunogenicity profile, epitope specificity, and then even smart features we think of like pH-dependent binding and things like that. If we were having this conversation a few years ago, and we may have had this conversation a few years ago, I think some of the real advantages we would point to in our platform is how much faster can we do it, how much more cost-effectively can we do it, and can you increase the POS, probability of success. We still think, of course, that those are important attributes, but we think probably one of, if not the most important feature and differentiating feature of an antibody design platform, including our AI Integrated Drug Creation platform, is does it unlock novel biology. Can you now go after difficult-to-drug targets? We think of this almost as like the fourth dimension of drug discovery where it becomes a question of no longer how much faster or how much better, but can you do it at all. From that standpoint, it's almost infinitely better if you have a target in mind that maybe has been known in the literature for years or decades, a certain GPCR or ion channel, for example, where the biology is known, but it's been intractable. You're able to use a platform like Absci's platform to create an antibody that can actually be a potential binder and potential therapeutic for that. We think that unlocks a lot of potential value for our own pipeline or for partners. Thinking about the pillars of our platform, it kind of works on three legs. One is the data generation and screening capabilities. The ability to generate and screen massive amounts of high-quality, usable, scalable data for protein-protein interactions allows us to train our models in a way that we can then feed the model a target of interest, say target X, and see how it creates a potential antibody that would be able to bind to that. Obviously, more than one, you know, we could think thousands, for example. One of the last pieces that we think is really crucial is the ability to validate everything we're doing in the wet lab. Being able to take the antibody variants that are given to us and test whether or not those actually worked and then further reinforce the model. In practice, the way this works is once the models have been trained to the certain point, we take a target we're working on, such as in our pipeline, our ABS101 program, TL1A, ABS201, prolactin receptor, have the model de novo generate an antibody sequence that then gets optimized, what we see as an ultimate, you know, drug candidate that would be, you know, potentially have a high POS in the clinic and then take that forward from there. We also, as we'll get into, do work with partners. We've done in the past with pharma and biotech companies that'll come to us as a target of interest. Obviously, those deals are all structured differently. That's kind of just an overview of the platform and the business model. I think about a couple of years ago, you've made this strategic shift of going from working for collaborators to developing your own internal pipeline. What led you to make that strategic shift? We'll certainly talk about your pipeline, but I want to get that out. I think, you know, certainly a few years ago, the business model was entirely partnership-focused. Correct. I think it's fair to say for the most part, we still employ a partnership model. The question of what's shifted or what's evolved is the timing of which we partner. If you think about, you know, in the past, we'd always partner with a company like, you know, Merck, for example, or Amaro, where they come to us and we partner at the target phase. That's structured so that we receive an upfront and are eligible for milestones and royalties in the backend. What we realized along the way is that the way that our platform enables us to create these novel, differentiated antibody candidates in a pretty rapid and cost-efficient timeframe, we could potentially take those to a value inflection point ourselves, say the drug candidate phase, or say a phase one, and then partner it with a pharma company or a biotech at that point. It may still be a partnering model, just the timeline of when we partner is different. When we go through that route, we could potentially look to get, you know, larger upfronts and overall higher deal value based on the fact that it's a, you know, more developed program than partnering at the target stage. At the start of that, I was saying how for the most part, it's their model, because as we'll get into what we see as a potential flagship asset for us, our ABS201 product for androgenetic alopecia or common hair loss, we see that as something we could potentially look to take all the way through approval and commercialization ourselves without a partner. It's early stages at this point. We're looking to be in the clinic very soon, early 2026, potentially even earlier. That's sort of one that we see as strong rationale to hold on to as long as we can. Let's start with that molecule, the ABS201. What's the target there, and how is, what are the preclinical work that you still need to do before you take it into the clinic, as you said, in 2026? Yeah, this is a really interesting molecule. It's the prolactin receptor. What our team discovered is that the buildup of prolactin in the hair follicle is responsible for pushing the follicle into a dormant phase of apoptosis and regression, where the hair follicle just recedes and eventually you stop growing the hair there. What the team had noticed, and done a lot of tests on this too, is that by blocking the prolactin receptor and therefore blocking the ability for prolactin to build up in the follicle, you can push the follicle back into the active growth phase, which is called the anagen phase, which for humans lasts anywhere from two to six years. We have this program in IND-enabling studies right now, looking to be in the clinic early 2026, potentially sooner at this point. We're running it as a phase one 2A. Even though we only unveiled this program a few months ago, at the end of last year, we could look to have that interim efficacy, interim POC, second half of next year, which is really right around the corner when you think about it, to potentially see that hair growth on the human participants. Some of the compelling preclinical data we've seen is that there's this species of non-human primate called a stump-tailed macaque that naturally has androgenetic alopecia. This study was not a shaving study. It was not chemically induced baldness. It's a species of NHP that naturally has androgenetic alopecia. A company that had been looking at this a while ago tested a prolactin receptor antibody on these monkeys that had essentially baldness. What they saw was that after about 28 weeks, or even sooner after about 12 weeks, the monkeys just started to grow the hair back. After 28 weeks, they stopped the treatment. I think this is probably the most compelling part. Looking out four years after the treatment stopped, the hair was as thick as it was and still very durable. The way that we're designing our product here is that we're looking for something that would be obviously safe, durable, and convenient for patients. I think with aesthetics and cosmetics treatments in general, things like, obviously safety, but then also convenience are key. If you think about current treatments people typically use, daily pills, daily topicals, things like that, the adherence in that is quite poor versus something that we're looking to design, which would be an injection that you take maybe two or three times over the course of six months, and then you may be set for the course of your entire antigen cycle, which could be a couple of years. The market on this, which I'm sure we'll get into, is quite large. In the U.S., 80 million people have androgenetic alopecia. Again, this is just common male and female pattern hair loss. This is not alopecia areata, the rare disease. This is common hair thinning in people. In terms of the clinical program, you said you're going to start it off as a phase one two. Are we trying to shave off some time even in the clinical development pathway? If so, what is it that you're kind of thinking in terms of the timeline between starting up a study in 2026 and getting to the market? Yeah, no, it's a great point. I mean, we do think that relatively speaking, and I say that as compared to, say, like an oncology trial or even like an IBD trial, relatively speaking, it is a more straightforward path through the clinical development. By doing this as a phase one/two A, we could have that interim POC second half of next year. If everything goes according to plan, could be in sort of an almost 2030 timeframe for a BLA submission, at which point could look to commercialize after that. Fantastic. In terms of the target, is there anything else that we should be thinking about? I believe there's another company, Hope Medicine, which had done some work on this previously. Obviously, since it's been taken up by a large cap pharma, we don't know what's going on at this point. From the data that you have seen with the competitor and with your preclinical data, is there anything that you can compare and contrast and talk to us about? What is it that makes you feel better? Yeah, no, it's a great point. You know, some background on that program. This was a while ago. Bayer had been working on a prolactin receptor antibody for endometriosis. The team there at the time that was working on that is now Absci's current Head of R&D, Andreas Busch, our Chief Innovation Officer, who was with that team at Bayer that was looking into the prolactin receptor for endometriosis. They were doing animal studies with mice where they would shave the mice and then, doing their endometriosis studies, realized that the mice were growing back their fur very quickly and they did not understand why. They looked into it and it was a serendipitous finding that the prolactin receptor is implicated in the hair cycle. Bayer at the time strategically was not interested in pursuing a cosmetic indication and they outlicensed that program to Hope Medicine, as you say. We've looked into some of the data or the top line press releases that have been shared from Hope and have seen that they reported out from a phase 1b, something about 14 hairs per centimeter square, which is pretty solid data. We're not sure where their current path forward on that for androgenetic alopecia is. We think they're looking at it for endometriosis currently. We've done some head-to-head comparisons of our molecule versus theirs. What we see in the tests is that we have superior stability, solubility, and extended half-life, also very high bioavailability. When we look at the receptor occupancy, it looks to us based on the modeling that they may have really underdosed there. The modeling that we have done, we sort of conclude that we would look to formulate ours that would only require maybe two to three doses over six months. The molecule from Hope for AGA would probably require more like six to twelve doses and be about two injections per dose. At that point, you're looking at twelve to twenty-four doses, which from a patient convenience standpoint is not as convenient as the TPP we're looking for. Even from an economic standpoint, the cost of goods sold can really erode your margins pretty quickly when you're looking at, say, twenty-four doses versus the two to three. I think too, one of the nice findings they had in the trial they had done, and as I say, nice for us, is that there was clean safety for what we've seen. They validate the mechanism in humans and also that the hair growth that they saw, even with the lower receptor occupancy, we think gives us high confidence that with our higher bioavailability, more extended half-life, and targeted higher receptor occupancy, we could hopefully see higher affinity or efficacy there. In terms of the market opportunity, I know you just said that there's a large population in terms of this particular indication itself. When you go commercially, there are all sorts of things which are out there in the OTC market and elsewhere in terms of treating alopecia. How do you think about commercialization and also in terms of versus a traditional payer model where alopecia really doesn't come up on the top of the list? Absolutely. We do think that at the end of the day, this would be a novel treatment that patients are looking for now. That's what we hear day in and day out from practicing dermatologists who are on our KOL network, that patients and consumers are looking for something new, not just something that's a little different than what's out there already. To the payment question, this would likely be a cash pay model we're looking at. In terms of existing treatments, what people are familiar with are things like minoxidil, finasteride, Rogaine, Propecia, things like that. What we hear from some of the KOLs who are practicing derms is a small number of patients, a small% of patients will fully respond to something like minoxidil. Part of that is just biologically, some don't have the enzyme to fully metabolize and see the results from that. The other part is the convenience factor or lack of convenience factor, if I had to do a daily topical, is not so attractive to some people. The adherence is quite poor there. For finasteride, there are some pretty well-known, well-documented side effects that people run into. The patient population is much more limited on that side because women cannot take the finasteride. Correct. For minoxidil, it's, you know, sort of a gray area as well. By contrast, what we're looking for with our ABS201 product is that anyone to be able to take this and hopefully see the, you know, durable, effective, convenient hair regrowth. Okay. In terms of validation, ABS101 is the one which will give us the validation sooner because it's already in phase one. In terms of ABS101, which is for IBD, there are obviously other anti-TL1A monoclonal antibodies in development, especially with large cap pharmas. How does ABS101 differentiate itself against some of the clinical candidates out there? That's a great question. We think about sort of the first gen molecules that are out there from, you know, from Merck, from Roche and Hope Medicine, from Tafa Sanofi. We're looking to design our molecule to have higher affinity and potency, potential lower immunogenicity, extended half-life, and, you know, high bioavailability so that we could potentially go straight into sub-Q without needing an IV loading dose potentially. Even with the other potential benefit of the higher bioavailability, there could be some really strong tissue penetration that could lead to potential higher efficacy. The convenient infrequent dosing, looking every, you know, two to three months as a sub-Q, is one of the ways that we look to differentiate there. With the upcoming phase one interim in the, you know, second half of this year, we're looking to show safety, of course. We're looking to show the extended half-life, potentially some signal of, you know, lower immunogenicity or lack of ADA, and then some sustained target engagement like we showed in an NHP study earlier this year. If we could replicate that or demonstrate that in the results as well, that's what we look to show as well. Obviously, the phase one is mostly safety data, but IBD has been a very tough nut to crack in terms of developing a molecule against it. Do you plan to continue development by yourselves, or would you be trying to get a partner? The indication is also large. How do you think about the development path from here? Yeah, no, it's a great question. It comes back to just sort of our business strategy overall. For this ABS101 program for TL1A, we would look to partner that. We're already in discussions with multiple parties about potentially transacting that. When we have that phase one interim soon, that could be a potential trigger point for transaction. Even so, we do have the optionality that we are capitalized currently that if we chose to take it to a 1B, 2A, we are currently capitalized to do so without any additional inflows necessary. I know you have additional molecules in the pipeline, but in the interest of time, which of these molecules, either within your internal pipeline or the partner products, do you think you would like to highlight today? Yeah, I mean, I think the ABS201 product for androgenetic alopecia is really where a lot of the interest that we've had this past year from investors, from analysts, from strategics, potential partners has been. I think people have woken up to the idea this is a very underappreciated mechanism. The fact that we only unveiled it, you know, late last year and are looking to have an interim POC on that second half of next year really is right around the corner. The market opportunity is well understood, and we're looking to have that, you know, efficacy readout quite soon. If we're in the clinic early 2026 or potentially sooner, I think we're well on our way there. Okay. Last year, as you said, in the R&D days when you unveiled this molecule, should we expect another R&D day at the end of this year? If so, what would investors expect in that? Yeah, so there's a couple of things we're looking to do. I think probably toward the end of this year, we may have an event focused really on the ABS201 program, particularly because it's, one, we think of it as like a co-lead asset with our TL1A asset, and that is really accelerating toward the clinic. I think as for a fuller R&D day, that would talk about all of our programs, including some potential new ones, even more like first quarter of next year. You could expect something later this year, potentially on the ABS201 program specifically for the alopecia product. Since your initial business and base business has always been, you know, partnerships and, you know, working with different partners, is that still part of the business strategy? If so, are you kind of growing in terms of market cap when you're working with people now, or is it still, you know, this mid-cap and the mid-cap companies? Absolutely. I mean, we have in our guidance this year that we do expect to have a new partnership with a large pharma company, at least one. That is still in the guidance for this year. We think of that, again, separate from our internal programs, which should be partnered separately. We are thinking more about the drug creation partnerships, like a platform deal with a large pharma company, so that you can expect to see in the coming months as well. Okay. As a last question on the balance sheet, what's your current cash position? What sort of a runway do you think that you could get out of it? Yeah, we ended the quarter Q2 with about $118 million on the balance sheet. In July, brought in about $64 million in gross proceeds on top of that. $14 million of that was from one large order from the ATM, another $50 million from an underwritten public offering we did. We're looking at cash into the first half of 2028 at this point, which gets us well beyond, you know, the major catalysts we see upcoming, that being the TL1A readout of potential transaction, the large pharma partnership that we're guiding to, and then the ABS201 readout for alopecia, which would be second half of next year, all well within the runway guidance. There's also potential upside to that guidance from certain of those catalysts that would come with their own inflows, such as the TL1A transaction or the pharma platform deal. Perfect. Thank you very much. Thank you. Good luck.