Okay, everybody, welcome back to the 2024 Stifel Healthcare Conference. I'm Dan Arias. I'm the Life Sciences and Diagnostics Analyst here, and we are on the Life Sciences track. Our next company that we have with us here is Absci. Happy to have Alex Khan from IR and Strategy for the company. Alex, appreciate you spending some time with us.
Thanks for having us.
Sure thing. I think a good place to start would just be a little bit about Absci, what you guys are doing, how you play into the AI trend.
Great, happy to. So, you know, thinking about our Integrated Drug Creation Platform, which is an end-to-end platform that we've built, including target discovery, de novo generation, and AI lead optimization, it really comes back to the three foundational technologies of our company. So that's the data to train, the AI to create, and the wet lab to validate. And I think even just taking a step back and thinking about, well, what is it about AI that we think could be very impactful for drug discovery? We think to some of the shortcomings of traditional biologics drug discovery methods. So certain traditional methods are a long, iterative process that involves a lot of trial and error to try to get to an optimal drug candidate.
And oftentimes this results in a maybe four, five, or six-year process just to get into the clinic and oftentimes have suboptimal properties in these candidates because of the trial and error process. What we're looking to do with our platform instead is to apply engineering principles using our AI technology to be able to design better biologics for patients. And the way we do that is that we have the synthetic biology platform that our company was founded on, the E. coli strain that's able to produce full-length antibodies, which to our knowledge has not been able to be replicated by other platforms. What that allows us to do is really unlock this data bottleneck that we see in AI and see in biologics drug discovery AI. And so by having the SoluPro E.
coli create these millions of data points on antibodies that we screen in our proprietary ACE assay. We get these large data sets that we build up on antibody-antigen interactions. So we're screening those for functionality. We have those train our AI models, and we have a very stellar team of AI scientists from top tech companies, top labs, and universities that are working on these problems. And then the AI model is then able to create what we see as an optimal antibody candidate. And we do this in a very rapid cycle time. And the last part of that is the wet lab validation. And we think all three parts of these are crucial: having the data, having the AI model, and having the wet lab. And it's the validation and the cycle that really ties it all together.
We have a facility out in Vancouver, Washington, right outside of Portland. It's a 77,000 sq ft facility. About half of that is lab space, and what we do there is we do all the validation in the wet lab of what comes out of our models, and we do this in a six-week cycle time, going from an AI-designed antibody to a wet lab validated antibody. And in those given cycles, can do about three million antibodies. And so the question is, well, what are the benefits of these? So when I was talking before about those shortcomings of traditional biologics drug discovery methods, what we're looking to do instead is reduce the time to get to the clinic, reduce the cost it takes to develop an antibody drug candidate, and maybe most importantly, design an antibody that hopefully has better properties.
Again, instead of going through the trial and error iterative process, where in some cases you're almost hoping for the best, in this case, we're applying these engineering principles to look for things like better potency, better half-life, better developability, manufacturability, lower immunogenicity, and things like pH-dependent binding, multivalency, epitope specificity, things that we think would be important for potential pharma partners to look at in this. The way that our CEO talks about it a lot, our founder, Sean McClain, he says, it's no longer about trying to look for the needle in the haystack, but create the needle. Use our AI model to create what we see with engineering principles, a potential best-in-class, or in some cases, first-in-class, differentiated antibody candidate.
Okay, that's helpful. When you speak about reduction of time to drug or just the overall cost associated with that, what is an example or how do you frame that for customers or for those that are thinking internally about what your process could be?
Yeah, I think a great example is our lead asset, our ABS- 101, which is a potential best-in-class anti-TL1A antibody. From the time we started that program to when we got to the drug candidate, it was a matter of about 14 months. For context, we've seen some estimates say it's typically about a three-year frame to get from the start of a program to a candidate. The costs, we've seen pretty wide-ranging estimates, but something in the range of $30 million-$50 million is what we've seen from estimates. By contrast, we're able to get to this in about 14 months at a cost of about $5 million or so. We think with the current version of our model, so even further improved, we could potentially get to a candidate in about a 12-month time frame.
So it gives us this opportunity to really get to these differentiated antibody assets in a pretty rapid way, whether it be for our own internal programs or for our partners. I think that's one of the main benefits that we get from the speed perspective.
Okay. Yeah, I want to ask you about some of the assets that you're developing in a minute. But before I do that, maybe just as additional background, what does the competitive landscape for AI-enabled drug development look like? I mean, most healthcare investors are familiar with generative AI, but there are varying degrees of understanding once you get into levels below that. So when you look across what you do and what others do, how do you see yourself as differentiated? What do you see as similar capabilities?
Yeah, no, it's a great question. I think looking across the landscape of AI drug discovery, especially in the public space, there's a good number of companies that are doing great work on the small molecule side. And we look to them, we admire them. I think one of the reasons there's fewer companies that are doing this type of work on the biologics side comes back to the data and the integration. And again, if you think about something like a chatbot or an image generator, you could think of the analog that it needs massive amounts of text data or image data to train the models. And a lot of that is freely available on the internet.
By contrast, something like biologics drug discovery, you need millions or billions of data points on antibody-antigen interactions to be able to train a model to actually have a useful idea of what a created antibody would look like based on a target you feed it. And so we've seen that be, I'd say, a shortcoming of other companies that may be looking to do something similar is having the ability to create and screen these massive amounts of data sets. And then it comes back to, I think, the wet lab validation too, because that's something that really helps to train the models also is being able to validate in the wet lab what the model gave you, find out if it's any good, and then have that iteratively train the model again. So cycles and cycles upon improvement. And I think too it comes back to the team.
We have a very experienced team of drug hunters leading our drug creation efforts. Our Chief Innovation Officer, Andreas Busch, joined on a few years ago. He had previously been the head of R&D at Bayer, the CSO at Shire, and in his career has had about, I think, 10 drugs go from bench to approval, and he and his team that he's handpicked and recruited from top companies are the ones leading these drug creation efforts, and I think he is very, very insistent on that team, the AI team, all of our teams just being very, very collaborative across the space. We talk about being multilingual from that perspective, having people that understand the AI and understand the wet lab side and the drug creation side really working together on these projects.
Just thinking about deploying these capabilities, the company took on a bit of a different direction in early 2023, I believe that was. You still serve pharma companies and the things that they want to do with their own programs, but now you're far more interested in developing the assets of your own. What drove that strategic pivot? Why decide to develop your own therapeutics?
Yeah, no, it's really been a journey since the time of the IPO, over three years ago at this point, and as we were doing a lot of these drug discovery programs with great pharma companies like Merck, like AstraZeneca, like Almirall, we realized along the way that we could potentially create and capture more value, especially in nearer term, ourselves by doing a lot of these internal programs, so in addition to continuing to do these drug discovery, or as we call them, drug creation partnerships with collaborators like the one that is mentioned and others potentially in the future, we do look to start our own internal programs, take those to a certain value inflection point, and then transact those at those points, so it's still looking to be a partnership model. It's just a question of when we partner them.
So instead of partnering at the target phase, it could be anywhere really from preclinical or after phase I, after phase II. It really depends on the program and the partner. But we're very thoughtful and diligent about starting those types of conversations early on. So if and when the time came that we saw a value inflection point, we would potentially be able to move quickly on that.
The lead asset that you guys have is a TL1A inhibitor that you're targeting for IBD. You believe that that can be a best-in-class asset. What is the basis for that claim right now?
Yeah, so we've had some preclinical data out over the course of this past year, and altogether, it shows high affinity and potency, extended half-life, ability to bind both the monomer and the trimer of TL1A, and we're looking to have a lower immunogenicity profile, and based on the high concentration formulation studies and high bioavailability that we've seen in our CMC studies, we're aiming to have a sub-Q dose on that as well, and for the dosing intervals, we're looking to be up to potentially once quarterly on that and through, again, subcutaneous dosing for optimal patient convenience, and the most recent data we put out on that was the non-human primate studies that we did that showed two to three times extended half-life as compared to the first-generation TL1A antibodies in clinical development, those being the ones from Merck and from Roivant.
Okay. I am not a biotech analyst, but I did stay at Holiday Inn one time, enough to know that the TL1A market has gotten more competitive and some household names in biopharma are now in the space. Where is the confidence in being able to develop products on the timeline that these larger companies might do so and with the concentrated effort that a company like Merck and Roche might bring to the table?
Yeah, no, it's a great question. I think it comes back to our business model too that when we look at this asset and really all of our assets, we don't look to be competitive necessarily with the pharma companies at some point. We think what pharma does best is late-stage clinical development and commercialization. We would look to pharma really as a potential partner at a certain point down the line for including this asset, for example. As for why we think we're competitive right now, I think it just comes back to the world-class team that Andreas and his team have continued to build up on the clinical operations side, clinical development side. That gives us confidence to move this asset into the clinic next year, have a phase one readout.
We're aiming for the end of 2024 or have the interim phase 1 readout in the second half of 2025, and then, again, always ongoing discussions on potential partnership opportunities.
Yeah, and you kind of alluded to this, but how do you think about the handoff to a partner as you progress through the clinical development process? Does that vary? I mean, I'm sure it's a little bit of a case-by-case basis, but there are probably also some guidelines that you would put on the situation as far as how far we should think about you taking a particular asset.
Yeah, to your point, definitely a case-by-case, not a one-size-fits-all. But I think the line we've drawn to this point is we don't anticipate ever becoming a phase III company. Again, I think that's what pharma does best, the late-stage clinical development. And from that standpoint, we could look to them as potential partners in the future versus being competitors.
You also have another asset, ABS- 201. I don't think you've said too much specifically about this product, but it is focused on dermatology. Can you talk a little bit about what it is that you're trying to do there, why that as an opportunity is favorable for the company?
Yeah, so teeing up our R&D Day next month on December 12th, that's going to be a big focus of the day there. And you're right, to this point, we haven't intentionally given too much information about that program. To date, we've said it's a potential best-in-class dermatology program where we see an underappreciated target that we could potentially be second to clinic. We see a potentially large unmet patient need there. And at the R&D Day, we do expect to have a presentation on that, including the target of the asset, the potential markets and indications. And we'll have a few KOLs planning to come speak on it, both from the patient perspective, but also the potential market size to really kind of give more flavor on why we see that as a potentially very valuable asset for us.
Yeah. And then maybe just to sort of round out the 301, immuno-oncology. A lot of money obviously going into immuno-oncology. How do you fit in that framework? And then the follow-on question would be, how do you decide what to allocate resources and time to? I mean, these are three, to use your words, big opportunities, potential for you guys to have an asset that's better than others. How do you decide which one in the moment is worth devoting the resources to? To our point before, you are a smaller company than some of the larger players in the space.
Yeah. So for ABS- 301, that is a potential first-in-class immuno-oncology target. They came from our Reverse Immunology platform, which itself is actually a pretty exciting platform. It's based on tertiary lymphoid structure biology where we look at the TLS samples from patients who have been observed to have an extraordinary immune response. We look at the B-cell repertoire in there. And what we do with that is we're able to deorphan and discover these novel targets. And so that's where ABS- 301 came from. So we have both a novel target and a fully human antibody coming for that. So that program is something we haven't given too much detail about quite yet, but we do expect to have the MOA validation studies done in the first half of next year and share some detail on that along the way.
But it's a program and a platform they're really excited about. So not just ABS- 301, but the Reverse Immunology platform in general, I think is something that could have a lot of potential promise in the future. And this is one too that we've shared some information about with some, I'd say, future potential partners, and they see some promise in that one as well. As for allocating resources, it's something that we're always very, very mindful of, which programs we think. I mean, I think the simple way to put it is where we see the most value and the most opportunity is where we would look to allocate the most resources, both human and capital to these certain programs.
I think at the end of the day, what we're looking to do is create this balanced, diversified portfolio where we see our internal programs, our partner programs, and kind of a newer thing for us, our co-developed programs, and where we see the most potential value is where we would look to allocate the most resources.
Okay. And then just thinking about the assets that we mentioned, what should we think about as the next meaningful milestones? I'm sure this is something we're going to hear about at the R&D Day as well. But just loosely based on what you've kind of described up to date, what should we be focused on as far as readouts, partnerships?
I'm excited to say there's a number of those. Looking into the first part of next year, in the first half, we would look to have our ABS- 101 in the clinic, which will be a big moment for us as a company, having our first asset that we designed ourselves going into the clinic, thinking about where we came from a few years ago until then. Then for ABS- 301, expecting to have the MOA validation studies done in the first part of next year and share some data around that at the appropriate venue. ABS- 201, continuing to progress that along. We will have the target IND unveiled at the end of this year at the R&D Day next month.
Then also looking at the end of this year to add a new program or unveil a new program that we're adding to our pipeline. So more on that to come.
Okay. So there are additional assets coming to the pipeline. Do you get to the point where in order for more to come in, something has to sort of go out via a partnership, which is a hard question to answer. I mean, what is the number of assets where it feels like the table is sort of full?
Yeah. No, it's something we're always thinking about. And I think there could be the potential to have a steady cadence of one or two or so, just thinking kind of broadly, programs added over the course of the next few years, not looking to overextend ourselves by adding a dozen or so, for example. I think really just comes down to where do we think the most value can be created and where do we think, and this is key, are we actually creating something that's differentiated and could be potentially valuable to pharma partners? And being thoughtful about which programs we decide to allocate capital to, not just starting a program for the sake of it. Where do we think we have the most opportunity? And that's how we're kind of building out the portfolio.
Have you spoken about, so now that you have your own development pipeline and you're still serving the customers that want you to work on their own drugs, how does that resource or time allocation split? How much of Absci going towards what you're doing internally versus externally?
Yeah. Don't necessarily think we can give an exact split, but fair to say that over the course of the past year, more resources internally have been shifted toward these internal programs. It was only about a year ago that we first unveiled these programs at our 2023 R&D Day, putting out ABS- 101, 201, and 301. Before that, we had been doing exclusively partnered programs, so the ones we have with AZ, with Almirall, with Merck, with now adding on Memorial Sloan Kettering Cancer Center, Twist. So obviously continue to see these ones being added on from the outside perspective or even from co-developments. But certainly a lot of internal work has been devoted to our proprietary programs.
Should we expect you to continue to seek out that work from pharma companies? Or much like I'm asking what represents a full table internally, what might represent a full table externally?
Yeah. Again, I don't know if we can give a hard quantitative cap on that, but we are looking to continue to add pharma partners as much as we have in the past few years. We see a lot of value potentially in these types of partnerships, not just from the economic side, but also the expertise that these partners can bring to a program. So if we think about the AstraZeneca partnership, for example, they're obviously world-class in oncology. And so to work in these types of programs with them, there's a lot of synergies, a lot of domain expertise they bring to the table. And at the same time, it also gives us a lot of opportunity to continue to train our models based on the data generated from these partnerships, which I think is sort of underappreciated sometimes.
And then what about Twist? There's a partnership that's been announced with Twist. How does that fit into the picture for you? If things were to go well there, could you see some sort of integrated offering?
Yeah, no, it's a great point. So Twist, we've been a long-time customer of theirs. Our founder and CEO, Sean, and their CEO have a great relationship. And they even have a facility not too far from us out in Oregon. And so we announced that partnership very recently to be doing this sort of co-development, which again is sort of structured so that we each contribute to the program and then would potentially share in the economic upside from the program. But there's a lot of synergies there where they would be responsible for a lot of the testing and validation of the antibodies that come from this collaboration. And with some of the new technologies that they've unveiled recently where they can do these 5,000 base pairs and could do like a whole heavy chain, whole light chain, could do it really rapid timeframe, that's very important for us.
We see a lot of synergies there because that design-build-test cycle is crucial for us. When we think about our own programs, we have that six-week cycle time from creation to validation. They're integral to that. So to your point on potential expansion, I think I could say for pretty much any of our partnerships or collaborations, we always see potential opportunity for if something were to go well, maybe expanding that with the partner.
What about geographically? I mean, you've got partnership. You just have a new partnership with AstraZeneca, so clearly you're not afraid to go overseas and partner with a pharma company there. But how much globally are you trying to do with the company? Is it really a function of like, let's walk before we can run, go into Europe, go into China, some of these ex-U.S?
Yeah, I mean, I think we'll meet the partners where they are, that if there's potential interest from a large pharma, we're definitely open to speaking with them, and I think it's fair to say we have been speaking with pharma from across the globe consistently.
Okay. Anything that when you think about the way in which you're building out your own internal capabilities sticks out as an investment area or as AI evolves the way that it's very rapidly evolving, becomes more of a need maybe in 2025 than it was in 2023? What are the investment needs, if there are any, as you grow?
Yeah, I mean, we're always looking to continue building up our AI platform. And we've seen just in the general public how quickly a lot of these improvements in the AI space, again, whether it's chatbots, image generators, can go from when we were having this conversation two years ago to where it is now, it just seems like a completely different world. And I think if we do continue to invest in our AI platform, there's potential to bring down cycle times. There's potential to really hone in on these types of specific attributes we're looking to design into antibody programs, for example, or to go in these different modalities or get better at different therapeutic areas, for example.
So I think there's a lot of potential to keep investing on that side of the house, plus adding on the wet lab side, clinical development, clinical operations to make sure that we stay competitive on the proprietary asset front as well.
If Zach were sitting next to you, I'd ask him the cash question, but he's sent you to do his dirty work. So $127 million on the balance sheet. That presumably situates you guys until 2027, I believe you said, first half of 2027?
Yeah, so that's cash into the first half of 2027. And I think one thing we like to point out is that has only very, what we say is modest assumptions of inflows coming in, things like partner inflows and a pretty, I'd say, conservative cadence of new partnerships signed on. So if there were, in theory, anything like a transaction around one of our partnerships or, sorry, one of our programs or bringing in a new partnership that had a larger upfront, those would be potential upside to that first half of 2027 estimate.
Okay. But nothing between now and then that strikes you as changing the quarterly cadence in a way that might surprise people?
Do you mean on the cash spend side?
On the cash, on the burn.
I mean, if anything, it's surprised to the beneficial side. So we came into the year with an expectation to use $80 million cash on a gross basis and had been tracking below that all year and just revised that to think we're only going to use $75 million on a gross basis. Really just as the organization has become more efficient as a whole and the platform has become more efficient as a whole, we're seeing a lot of R&D efficiencies across the board, and that's really helped with the cash profile there.
Yep. Okay, last question for you on revenue targets. Biotech folks don't necessarily focus on them the way that life sciences investors and analysts do, but biotech companies will actually put out revenue targets every now and then. When do you feel like it will be most appropriate to do that? And in lieu of that formal provision of a target, is there any way that you would have us thinking about revenue generation over the next two, three, five years?
Yeah. I mean, I think the business model has really, really been refined over the past year to say we do look to continue to do some of these partner programs that will bring potential, well, will bring upfronts and then potential milestones and royalties. And so if we think about just a few off the top like Merck, AstraZeneca, and Almirall, those three partnerships, six programs combined have a biobucks of $1.5 billion.
Obviously, a little more long-dated, not nearly all that upfront, which is part of the reason that we decided to go into our own internal programs because we think if, and we'll use our ABS- 101 as an example, if we can get to a drug candidate for about $5 million, take it through IND enabling studies, take it through a phase I, and then for a pretty, I'd say, small amount of investment, potentially transact that out to a partner for some pretty eye-watering comps out there for a lot of these assets, especially even some phase I assets these days. We think that's sort of the business model we're looking to pursue.
So sort of relatively minimal investments on these assets to get to a value inflection point, whether that be, again, preclinical or phase I or phase II, whatever the value inflection point is with the right partner at the right time, we'll go off to monetize at that point. We would definitely look to try to get much better upfronts, obviously, and also retain in the milestones and royalties along the way. And that's how we're thinking about sort of the revenue stream going forward.
Okay, fair enough. I think I'll leave it there. Alex, appreciate the time.
No, thank you.
Happy Thanksgiving to you.
Likewise.
Talk to you soon.