Dr. Doug Chapnick, CEO and founder of BioLoomics, discusses the company's mission to revolutionize antibody-drug conjugates (ADCs) for better treatment by improving their targeting and internalization. BioLoomics aims to deliver therapeutic payloads more efficiently to tumors, addressing the major issue of current ADCs. The company is currently validating their technology through partnerships and developing robust data to demonstrate their platform's value while exploring the role of AI in drug discovery and innovative antibody design. Additionally, BioLoomics is excited to announce the hiring of a new Chief Scientific Officer (CSO), Kurt Gish, to lead these groundbreaking initiatives, further strengthening their leadership team and scientific expertise. We had the chance to discuss all of this and more with Doug.

Thank you for taking the time to talk with us, Doug. Can you please start off by providing us with a brief description about yourself, your path, as well as how and why you founded BioLoomics?

My name is Doug Chapnick, and I have been, for 20 to 25 years, a biochemist operating at the intersection of biochemistry and cell biology, mostly in academia. I worked on a DARPA project (which focused on massive rapid threat assessment (RTA)) to design, build, test, and learn how to make molecules and get the mechanism of action quickly. And in that project, I saw an opportunity to turn human cells into living test tubes. Those living test tubes could show us things that we can't see at scale elsewhere. Using this method, we can measure new things, and break through a throughput barrier. Alternatively, we hear about single cell transcriptomics, and other methods like that, and those are other efforts to do a similar thing. Those techniques do not excel at design, build, test, and learn for antibodies; they're more for measuring what happens within the diversity of cells. We founded BioLoomics with the idea that we had already achieved 70% of the goal of turning living cells into test tubes in the RTA project. BioLoomics was then created to accomplish the remaining 30%. Now, we have successfully completed that goal and are focusing on designing, building, testing, and learning about antibodies for the ADC space.

Interesting perspective and story. How do you perceive the ADC market today?

I perceive the ADC market as something that has been around for a while, but that has a lot of key limitations that haven't been overcome yet. A lot of people are pooled on top of the same targets, competing … within the same indications. I see the field as ripe for innovation; there's a lot of time that has passed between the innovations that are in the clinic right now and their use in current drugs in development.

What are some other large unmet needs in the space, noting that you think there's a lot of aggregation on similar targets?

I see the single largest unmet need as ADC targeting. ADCs are essentially the solution for targeting chemo, right? And then you fast forward through many, many years, and you find out that only 1% of ADCs go to the tumor, and the other 99% go to the wrong cell type. We also, historically, couldn't get ADCs to work without putting very, very toxic payloads on them - that was out of necessity. Ultimately, what came out of breaking through that limitation was the realization that when the payload gets processed by the wrong cell, it does terrible things. That’s where all the adverse side effects come from. I would say that the unmet need is ‘targeting the targeting’, so as to actually achieve what we set out to do: to get the payload properly targeted, not 1% targeted to the tumor.

Absolutely. Why do you think this need for increased targeting hasn't received more attention? Or if it has, what have been the drawbacks there?

I think it has definitely received attention to understand where adverse side effects are coming from. Effects are either on-target, or they're off-target / off-tissue, and I think, right now, the field has a consensus that adverse effects are from macropinocytosis. There’s been a lot of investigation to figure out, "What is the mechanism behind payload release at the wrong spot, and where do the adverse side effects that come from it?” I don’t think there has been a solution that has beaten what we have right now, which is basically IgG with some “payload” on it. There's been a little bit of innovation that's also trying to address this targeting in the linker space, where people are trying to get linkers that have "AND" logic, such that the antibody has to be at the tumor AND something else has to be satisfied so that cleavage may occur. This is a pretty new technology. We're not the only ones pointing at this, but in other areas, you see dozens or 50 companies fighting to innovate in the same arena, and I don't think you see that in the antibody engineering space yet. The space is ripe with opportunity.

Speaking of that opportunity, what do you imagine as the next frontier for antibody-based therapeutics?

Well, I usually dumb ADCs down to say that they’re the most complicated molecule. They are a small molecule, a linker that has chemistry behind it, there's an antibody, and so all the things that come with all those pieces are now superimposed on top of each other. Together, those pieces must achieve getting to the right spot. It just gets more and more complicated as you dig into ADCs. What I think is going to happen is it's going to get even more complicated. So, to solve the problems, you're going to have to layer on new innovations. And I think it's going to be in two main spots, the antibody engineering area that we're playing in, and then the linker area where one can leverage new chemistries that don't require enzymes in the lysosome for release. That it'll take a long time, and so we're focusing on the lysosome, and trying to engineer antibody targeting so that the therapy gets where it needs to go faster.

Inefficient internalization (i.e. target degradation) is a significant issue in ADC development that affects all stages of development, from target identification and selection to CMC considerations. In your opinion, why do you believe that targeting and optimizing ADC internalization is a pressing issue, especially when compared to other hurdles ADCs are currently facing?

I think safety is at the center of everybody's mind. I think it's quite logical that if you can internalize and deliver more cargo to the tumor, it comes at the cost of delivering to the wrong cell. I expect that to be true because mass conservation is true. I also think that you have this opportunity to use less toxic payloads. If you can deliver a lot of a molecule, that molecule has less burden to be potent. If we work on this one problem, and can internalize more ADC and deliver more cargo to the tumor, then we can unlock many paths in front of us – using new cargo, or taking the age-old ADCs and making them better to have fewer adverse side effects. That's one direction, focusing on safety. Then there's a third path that's very far behind people's minds right now, which is that you can go after new targets. There are targets out there that are not fast cycling, so they're seen as bad ADC targets. And that's a third path for us. Essentially, if we solve one problem, then three paths open up. So that's why I think it has a lot of promise.

Are there any other avenues, aside from slowly internalizing targets or newer ones, that you think these more efficiently internalizing / highly internalizing ADCs would be better suited for than the traditional models?

Well, I think something very nontraditional could come out. So right now, PK is top of mind for everybody in the antibody world. It is entirely possible that down the line safety is realized with low PK drugs. I think it's at odds right now in people's minds because they don't want patients to have to dose frequently. However, there are companies out there that are exploring this area. I do think being able to target not just the right tissue and cell type, but the right organelles, could unlock a sort of change in the paradigm of how we think about a therapy that must run through the gambit of tests before it's ready for the clinic. Maybe we can get away with low PK and a lot more designs could open up, and maybe even some more therapeutic strategies.

Super interesting. Can you just briefly explain BioLoomics' main offering?

We are engineering antibodies to solve poor ADC targeting. And what I mean by that is not just what target you're going after, but what happens after the ADC binds that target. Antibodies can ride the wave of a target into the lysosome. If that target is fast cycling, like HER2, then the antibody is brought to the lysosome, the cargo is released, and it's a good pairing - it's seen as a good target. Our main offering is to essentially guarantee that type of flux. With the correct antibody scaffold, the right internalization / trafficking, and if we can engineer it to be target-dependent, then there are no longer bad targets because of slow trafficking. And like I said, you can unlock new cargo down the line.

Very interesting. What sort of key components BioLoomics offers that differ from and distinguish it from other ADC engineering tools, whether that's on the antibody side or the linker side?

There's one really clear one - we are leveraging binding to the lipid membrane. And as far as I know, there is not a single company doing that. That strategy has a lot of bang for its buck. Most companies try to get specificity of binding to the tumor through bispecific approaches, and those are all receptor-driven approaches. Receptors fluctuate quite a lot, as does receptor density on the surface of cells, and ADCs can be easily overcome by resistance mechanisms through selection processes. What binding to lipids gets you is a lot of membrane, and it's very hard for a cell to rewire its entire lipid composition. The result of this targeting is that you'll be able to achieve a lot more delivery because there's a lot more of that chemical handle that we're grabbing onto (these lipids on the membrane). This may also allow you to get around some of the resistance mechanisms.

This is fascinating can you elaborate a bit on targeting specifically, as well as different disease areas beyond oncology? Do you believe that your technology will only benefit oncology targets? Which therapeutic areas could benefit from your platform?

I think anywhere that you want to kill specific cell types is where we can bring value. I think oncology is a clear starting point for us. If we want to branch out later, I could see the next spot being autoimmunity where we want to kill the cells that have specific antigen recognition, and it's on my mind, but I think the company-wide mind is laser focused on cancer. I'm talking to a couple of partners that are interested in going outside of oncology. For us, targeting is our value prop. There is a possibility that we would go with one of those partnerships and then we'd be working in a non-competitive space with the partnership, and we'd get a perspective on the success of our modality in another area. At the core of what we're building is a delivery tool for a toxin to kill cells. There’s another avenue where it doesn't have to be a toxin, it can be something like RNA for example. However, I think there's a lot of challenges that can make what's already a complicated modality, just ADCs in general, even more complex; we’ve got to go stepwise through the possible applications of what we are building.

We would love to hear about your progress to date. Which studies have you conducted, what are the next steps, and when will we see the first asset designed with BioLoomics’ technology?

We screened a lot of antibody-peptide chimeras to find those that deliver more toxin to the cell and have faster internalization and lysosomal trafficking. We've been able to achieve 10x and 100x cell death in cell-based assays. We're very focused on KRAS mutant cell lines, where chemo falls short. We've be able to demonstrate these activities in KRAS mutant cells, across three or four cell lines that are all either non-small cell lung cancer or breast cancer cell lines. Now we're taking many of these designs and we're asking a very simple question - "What is the PK of them?” So, we're moving into PK studies, and we'll down-select from there into PD studies. In the next month or two, we should have these studies initiated.

Which targets is BioLoomics validating the technology for, and is that through a partnership or internally-funded work? And what does that landscape look like in the near-term?

Part of getting a partnership is putting together a data package that convinces people that we can add value to their programs. In return, we will be able to demonstrate that risk has been mitigated with what we built for those programs, and we can fundraise on that towards our internal programs. In order to get that data package, we choose targets; rather than taking novel targets, we wanted to pick targets that were already highly validated in the field, so we have HER2 and HER3. For HER2, we're only assessing whether we can make an improvement to what we think is a very good target for ADCs - there's a reason everybody's piled on it as a fast-cycling target. For HER3, we think we can make a dent because there's an element of amplification in our strategy. We’re seeing evidence that we can load more antibodies onto the surface of cells than the target, and that takes a lot to unpack, but that would be in our data package for anybody who's curious. HER3 has a big downside: even though everybody's piled on it and sees it as a good target - HER3 is very low abundance. Therefore, it has a major challenge that we see as a hurdle for ADCs. EGFR is one target everybody's familiar with, it's a tried-and-true target, though it's probably not a great ADC target, and we’re seeing that we can kill cells with our constructs better than cetuximab. Our data will focus on those three targets in addition to cMET and others.

Got it. Do you think that, for future validation, you'll focus on these well-conserved, well-known targets or move into this arena of new targets? And will that be on your own for validation, or would you attempt to move into this new target arena once you have a partnership?

We hope that partners want to do new targets. When we go to the clinic, we want to innovate on the tumor specific antigen, as what we're building requires a tumor specific antigen. We think there's a lot of work out there that shows there are more tumor specific antigens, other than the seven or eight targets everybody's piled on. We haven't finalized whether we will go with a tried-and-true target and compete in the arena with these well-known targets. We want to show a balanced target portfolio where some targets are new, some are old, and we sort of get a perspective to learn where we should point as well.

At the end of last year (2023), you raised a $8.7 million seed round, and you're also backed by an ex-Google investor. We'd love to hear updates since then about your fundraising, how it went, progress since then financially, and then if you're raising now or in the future and what investors should know?

We're fundraising in the future and to close in 2025. We are looking for the right investors that share our vision of de-risking with strategic partnerships in addition to internal programs. We'll be going to the fundraising market towards the end of this year. We're populating part of our data package with the partnership; a partnership will show clear value of our platform. In addition, we are aiming to arrive at and internal development candidate. Right now, we have many initial candidates, and we have not selected one. We're just iterating, iterating, and iterating. And we're looking at a couple of big features - we want to be able to say that we've really gotten ahead of CMC with our development candidate, and we get early indications of what is easy to manufacture and what is not. It's important that we iterate across multiple of these criteria so that our development candidate is robustly vetted.

What does the future hold for BioLoomics that we haven't discussed today?

We intend to publish, and we intend to create a big splash and challenge a paradigm. Many scientists believe that playing with membrane binding properties leads to poor PK, and that it’s sort of a negative thing for antibody design. I think, unbeknownst to us, some clinical antibodies are already doing what we're trying to engineer, with mysterious ways of having better activity, and better spatial targeting in tissues and organs. I believe that, yes, you can't have a protein covered with arginine, but somewhere in the gray area is an opportunity to leverage design principles that are rich in biophysics. I think we'll be able to demonstrate that in a paper - that a lot of the drugs that work have biophysical interactions aiding their activity.

What are the most pressing short- and long-term goals for the company in terms of steps? Is it more experimental or it could be more financial or strategic? What should the scientific community and investment community look out for in the coming months and years?

We’re focused on being able to get a bead on the CMC of our molecules. We can see whether molecules have more or less lysosomal delivery / cell killing power, which we’ll pair with targets that have low receptor density. HER3 is a good example here. We don't know if we will succeed with HER3, but I think HER3 is a difficult ADC target due to low receptor density. Keep a lookout for our paper demonstrating what we call "MEC ADCs," membrane engaging chimera ADCs. That paper will illustrate a bit of the mechanism and we think it will spawn lots of future research.

We recently saw that you have added a CSO to your team. Can you tell us more about his profile, the value that he's bringing and why the company is excited about him?

Yes, so the CSO we've brought on is Kurt Gish. Kurt has been working on ADCs, from the very beginning. At Abbvie, he spent many years working on the early designs and strategies to get efficacy and avoid off-target effects. He brings a wealth of knowledge of all the mistakes of the ADC field. We're quite a young company and he brings that element of the history of ADCs to us, so we can make original mistakes. He also has a lot of expertise in antibody design. Most recently, he had a startup called Trilo Therapeutics, which was toying with design, build, test, learn using DNA-encoded libraries (DELs) of peptide ADCs. We're a peptide antibody chimera company in the ADC space, so he's a sweet spot for us. He knows peptides, antibodies, and ADCs. In his early days, he was also doing target ID and target validation. So he has all these skills. Kurt is unique in that he can sit at the intersection of many disciplines, which is where BioLoomics exists.

Very interesting, there is no question that Kurt Gish will be a fantastic addition to the BioLoomics team and play a major role in the success of the company. Is there anything else that we haven't added that you would like to discuss or that we haven't discussed today?

I wanted to mention that, in the early days of us thinking through our targets, DeciBio helped us to understand the strengths and weaknesses that are perceived in the ADC community, and that we were surprised that people weren't thinking about antibody engineering. Many of my team members went to PEGS two weeks ago, and people were asking, "Why aren’t we pushing on antibody design?” So, I think tailwinds are picking up for people to operate in the same space we are. Things are about to change a lot in ADC space.

Do you see AI potentially playing a role in antibody design or drug discovery, generally, especially in the ADC space?

It depends on what kind of AI. If we could predict what an antibody binds and get affinity measurements at certain temperatures, then yes, it would be a goldmine. I think we're pretty far from that, at least stepwise, maybe not timewise. But I do think the target ID space will experience the first big impact of AI. Not in the protein design space, which is, ironically, where you hear about AlphaFold, and all these tools being built so we can get there one day. I think in the short term, we're going to choose smarter targets where the current limitations are less consequential, so that we may sidestep these limitations by choosing the target wisely.

Any role on the antibody design aspect (independently from target identification, more around the antibody structure, the sites, the affinity)? I'm curious about anything on the role of AI with antibody design itself.

I think if someone fuses a wet lab and dry lab design, build, test, learn, which is what most people are doing right now, but focuses specifically on that, then algorithms will be able to be trained to do that. I think that data is not abundant, and it's not abundant enough to drive model accuracy. That’s where tech bio platform ‘plays’ can bring a lot of value for everybody in collecting the data that doesn't exist. I know a few different companies that are doing something along those lines, and I think we'll see direct validation of that in the next five years. I always stress…although I have bleak things to say about where we're at right now with AI performing well, I the current efforts are all part of a stepwise motion to get us there. Everybody must be very patient, and at the first failure avoid discounting AI and saying it's never going to work. It's data driven, the data is accumulating, and we can create it. We are in the power position to make AI work for us. It's not that it's just going to deliver for free. And hopefully, there'll be some lucky hits in the next five years, where it was just the right time, right place, and there was the right amount of data. That helps us keep all the skeptics from breaking away at the resources that are fueling this. I think everybody must keep an open mind, even if we're not there right now, and it's not coming in five years. It is rational that we can get there and that we should keep feeding it.

I'm curious why people are asking "Why are we not pushing the antibody engineering side?" What is driving the conversation and the push for antibody engineering?

Well, I think it's many factors. One of them being we have infrastructure for design, build, test, learn of antibodies - it's basically phage display or yeast display, and there’s a thousand CROs. When you try to innovate on antibody design, you start to throw a wrench in the turnkey discovery/development processes at CROs and CDMOs. Some people don't want to change the antibody, they want to change the chemistry and utilize the off-the-shelf tools. I think there's a lot of ability to leverage chemistry in multi-well plates and optimize linkers. There's also a push to go where you can use a lot of tools and not worry about immunogenicity. I think ADCs have been dominated by chemistry driven people for a long time, and so the innovations are in chemistry. I think as the ADCs advance through and fail in the clinic, they will show us the limitations that we either thought were there but weren't sure of, or didn't even have eyesight on. I think this targeting of 1% to the tumor and 99% to the wrong cell is top-of-mind, and it's getting more top-of-mind for everybody. Either you're changing the cargo/linker to improve that targeting, or you're changing the antibody to do the same. I think, right now, cargo is top of mind, because most programs have a chemistry focus, but as more antibody design people get involved, it will be more antibody innovations.

Yes, it's time for more engineers to shine, and you're seeing that across not just ADCs, but other modalities as well. This has been an interesting conversation, and we're excited about the future. We are excited to hear updates about BioLoomics progress in the next couple of months and years and look forward to talking again soon.