Founder Stories: Ben Buelow, Founder of TeneoBio, The Platform that Keeps on Giving

Morning everyone. Thank you to the organizers, and thanks to all of you for taking time out of your busy JPMorgan schedules. I'm trying something new in this presentation and putting up a QR code here, which links to this presentation on our website in case any of you have been tasked with photographing this. You can just save your time that way and just listen. I think the reason FreeMind asked us to talk today is because we've been working with them for quite some time and have successfully submitted a number of, especially SBIR Phase I grants. And so today I'll just talk a little bit about the company, our approach to grant writing, and how that dovetails with our corporate strategy. I'd like to start by giving an overview of Teneo Bio as a company. We're a company of 40 people now, based here in the Bay Area. And we're really founded on two core technologies. The first are these proprietary UniRat animals, which make human, fully human, heavy chain only antibodies. And we pair that with a next generation sequencing based antibody discovery pipeline that has now allowed us to develop a number of multivalent therapeutics for indications in oncology, infectious disease, autoimmunity. And we also partner with a number of companies, especially in creating chimeric antigen receptors for CAR-T cells. So I mentioned that we make heavy chain only antibodies, and you might ask the question now, well, why bother? Conventional antibodies work pretty well. People have made a lot of therapeutics with those. And there are really two answers to that question that are relevant for the conversation today. The first is that heavy chain only antibodies are really good for making multivalent therapeutics. And that's because when you create a multivalent antibody that incorporates multiple heavy-light chain pairs, convincing the right light chain to pair with the right heavy chain is really challenging. But if you don't have any light chains, you abrogate that problem entirely, so that making this variety of very sort of fanciful-looking molecules at the bottom here aren't just drawings anymore. These are all molecules that we've actually created. The second advantage is that we do our antibody discovery by next-generation sequencing, which seems like a pretty obvious thing to do until you realize that if you have to sequence both a heavy chain and a light chain, and then figure out which sequence goes with which one, that, that combinatorial problem gets very complicated very quickly. But again, if you don't have a light chain, that problem solves itself. So when we do antibody discovery, instead of doing initial functional assays, we begin by just sequencing the entire B cell repertoire. And from that, we isolate a few hundred thousand unique antibody sequences, which we then put through a proprietary algorithm which predicts which of those sequences is likely to be antigen-specific. And at this point, the— that algorithm is good enough that about 30% of the sequences that it pulls out actually do bind the antigen with high affinity. So that's an enrichment step of maybe a 500 to 1,000-fold over traditional functional-based approaches. The other advantage is once you have the sequence in hand, you can just plug that into an expression vector, express these in CHO, and do all of your functional screening in 96-well plate format. The other nice thing you can do with this algorithm is you can actually group all those sequences into families based on which, for instance, where one family is a single VDJ recombination. And once you've arranged them as sequences, and the algorithm has told you which families it thinks are likely to be specific for your antigen, you can just pick a few representative members from those families. And do your first round of functional screening on those. So, let's say you do that for 150 families and 5 of them are specific for your antigen. Where it gets really interesting is you can now go back to those 5 families and each of those families has about 50 members. So, you can iterate this approach to find antibodies that have an optimal functional characteristic for whatever you're looking for, viral neutralization, strength of binding, and so on. So, all of this lets us move quite quickly. In the last 3 years, we've looked at about 1.8 billion sequences, done functional screening of about 40,000 antibodies. And why these numbers matter will get— will become apparent in a moment, because one of the problems when you make multispecifics is it's very hard to predict from an antibody that's very well behaved as a monomeric as a monovalent antibody, whether or not it will be well-behaved as a multispecific. So you really want the top of your funnel when you're doing antibody discovery to be as wide as possible, which again is exactly what this pipeline enables. OK, so why did I tell you about all of our technology so far? It really has to do with— because as I was thinking about what should I talk about at the Freemind meeting that I came to realize that our strategy when it comes to non-dilutive funding is really a microcosm of our overall corporate strategy. And as I was thinking about that, a few ideas crystallized about what are really some of the important lessons that we've learned in the grants that we've submitted so far and how we approach the grants that we've been successful in. And I've just put some of those lessons up here. So the first one is, as a small company, we all don't want to have our eggs in one basket, but we're not a Pfizer or a Genentech or an AbbVie. We are small. And so the natural tendency is once you have one really good idea, you just tend to double down and double down and double down. And 2 years down the road, you realize you have 5 projects, but if project number 1 fails, nobody is going to buy into projects 2 through 5., right? So diversification is really important. You want to try and go into different areas if you can. Secondly, focus on things that are your strengths. Play to your strengths. And so this is why I was telling you about our technology, because for us, our strength is making multivalence, making things that require a heavy chain only antibody. If you can solve the problem with a conventional antibody, we're not interested, because there is just way too much competition. Third, collaborate, right? It seems obvious, but when we think collaborate, we usually think with academic researchers, but the reality is that both for grants and for projects in general, we've learned that establishing good collaborations with service providers, with other small biotechs, with physicians, all make you stronger. Fourth, you want to try and pursue a few crazy ideas, a few of those high-risk, high-reward things. Not just the things that are well established and where the data is already out there. And then this last idea I'll just touch on at the very end. I really do think that writing grants, the process of writing and submitting and getting feedback, will actually help you do better science. So now I'll dive into each one of those ideas briefly. In regards to diversification, I just thought I would throw up our pipeline, Tenaille Bio's pipeline. And what you can see is that the things that are most far advanced are T-cell redirecting bispecifics against liquid tumor targets. Very well-established ideas, very well-established targets, things that make venture and pharma feel comfortable with your ideas, because they already know that it should probably work. But when we wanted to get into things that are a little riskier and where the data is not as well-established, solid tumor targets for T-cell redirection, autoimmunity, infectious disease, it gets much more challenging to get money. So grant writing is a great way to start getting into those areas where you may be able to generate great data, but you're not going to get funding to take that project forward into later stages from people like your pharma partners or your venture until you have better data. And grants can really help you get there. So now let me talk about the idea of playing to your strengths. So for us, one of our main things that we've done is made T-cell redirecting bispecifics. And the reason we wanted to get into this is because, as I've told you, our pipeline is particularly good at finding kind of rare or unique antibodies because we get all these family members. And so you can really hunt for the needle in the haystack. And the problem with T-cell redirecting bispecifics to date has been that essentially all or the vast majority of bispecifics that are out there today utilize some variant of just a handful of mouse monoclonals that were developed in the '80s against CD3 that are all very, very strong T cell activators. So they're good at inducing T cell kill, but they also cause a massive cytokine release, which modulates most of the toxicity from these kinds of therapies. So we thought, well, hey, maybe we could discover an antibody that keeps the cytotoxicity but gets rid of the cytokine release. So again, this isn't just some fanciful idea, because there actually is a body of literature dating back to the early 2000s that shows that T-cell activation proceeds by a series of thresholds. So you need to activate only 1 to 3 T-cell receptors to get T-cell killing, but you actually need to activate maybe 10 or more in order to get cytokine release. So can you get kind of this baby bear's porridge in between the 3 and the 10 on average on a T-cell? Could you modulate the strength of your T cell activation to keep the cytotoxicity but not do the cytokine release? Ordinarily, I would spend a lot longer on this story, but the bottom line is, yes, that's possible. And I'm giving you an example here of this special anti-CD3 that we developed using a tumor-associated antigen, CD19, which is a lymphoma target. And our molecule is in purple, and then one of the strong activators is in red. And what you can see in the top row is that we can mediate the same maximum killing as these strong activators, but even when we saturate killing, so when you flatline at the top there for killing, the amount of cytokine secretion mediated by the molecule in purple, which is shown in the row on the bottom, is much, much lower. The molecule also still continues to mediate killing in vivo, which is shown here with TMB486 being one of these molecules that incorporates the— this more weaker physiological activator, and then the positive control being one of these strong activators. So one of the things that you can do now if you have something like this, a platform established, is then we decided we would try to plug and play with a number of tumor-associated antigens. And we've done that for a number of different tumor types. But of course, as I told you at the beginning, even though doing T-cell redirection is playing to our strengths, when it came to raising money for these things, it was much easier to raise money for a liquid tumor target that's well established, rather than solid tumor targets. However, in a number of solid tumor indications, for instance prostate cancer and ovarian cancer, Previous attempts to do T-cell activating therapy have been hampered by problems with significant toxicity. So that was one of those areas where we thought, well, we have something that specifically causes less toxicity. That's the selling point. Why don't we write some grants about this? Because those experts will be really aware of this being a big problem. And that's, in fact, what we did. So for both the prostate cancer project and the ovarian cancer project, by directing our efforts at the particular strengths of the innovation that we had developed, that allowed us to generate some grant funding. And those things are moving forward and will hopefully be in the clinic in the next few years. So here I've just included a brief summary. I already reiterated on the last slide that what I'm really trying to illustrate here is play to your strengths, leverage the innovations that are really specific to your company as the, as the basis for writing your grants. Okay, I'm gonna switch gears entirely here now and talk about our efforts to make a CD38 enzyme inhibitor. Many of you will probably be familiar with CD38 as a great target for killing myeloma cells, but CD38 does actually do something in your body. It is an NADase, and with— as is summarized in these figures here, it actually turns out that CD38 is critical for regulating both extracellular and intracellular NAD homeostasis in the body, which seems important, especially as the metabolomics kind of field is really on a rise right now. And of course, if you're not in an oncology indication, killing all your CD38-positive cells, plasma cells and K cells, activated T cells, probably not a great idea. The other thing that's worth mentioning is that it's been shown that NAD homeostasis is really important in cell senescence and in diseases of aging. So it seemed like modulating CD38 enzyme activity might be an interesting area to go into. So we started playing around with this, but what we discovered quite quickly is that you cannot— we were not able to find a single antibody, a monovalent antibody, that could inhibit CD38 enzyme activity all the way down to zero. We could get it to about 50%, but when we started playing with combinations of antibodies that bound different epitopes, we could get it down to zero. So essentially, only by making a multivalent antibody, again, to this idea, play to your strengths, could we get entire, complete enzyme inhibition. And when we tested this in vivo with some collaborators, we showed that this actually could increase the availability of NAD precursors, NMN, which is shown on the left there, significantly more, actually, than supplementing. These precursors, which is the bar in the center of each at the 3 months and 18 months. If you give oral supplements, it really doesn't stick around at all. But when you give our antibody, then you get a really substantial increase. And this in fact led to an increase in intracellular NAD+ in the tissues of the body. So two really important indicators of functionality. OK, great. So we can increase NAD+ in organs. What do we do with that? It's important in diseases of aging. Well, I guess it should probably come as no surprise to anyone here that diseases of aging includes a vast array of different diseases. So how do you sort through this? How do you decide which of these things to go after? And this brings me to my second point that I might— wanted to make, which is that collaboration is really, really key, not just in grant writing, but I think in grant writing is somewhere where oftentimes we do really want to engage experts in the field. We do really want to engage people who can fill in whatever gaps we have at our company to allow us to accomplish the different aims that we set out in the grant. So for this, this project I think is a great illustration of that. Because when we started getting into, well, what diseases do we actually want to go after? And we realized there were so many. We just started reaching out to all of these different entities where there were experts in each of those areas. And in fact, recently— last Friday, actually— we had a summit with leaders in each of these areas to try and help us narrow this down. And I think this project has just been a really great illustration for me of how working together with people in the field, being collaborative, being open with your data as much as you can be, is a great way to move your grants forward. It gives you a lot of scientific credibility to be working with people who are really well established in the field. So when we— for us, what collaboration really allowed us to crystallize was that we did have a great molecule that for inhibiting CD38. There are some small molecule inhibitors, but those have a number of problems, which is something we hadn't really realized until we started working with our various collaborators. Those collaborators also in their different model systems have tested our molecule, and that gives you a lot more confidence that it actually works if it was working in the hands of other people. And then of course, as I mentioned, it's given us a much better sense of the clinical applications of this kind of therapy. Okay, the last story I'm going to tell today is about our efforts to treat a set of viruses called polyomaviruses. And the way that we've approached this is to make something called domain antibody strings. And the reason I'm bringing up this project is because this is one of those kind of harebrained ideas that I think would have been a very hard sell before we had any data. But we've now been able to generate a lot of data that I think will also generate enthusiasm from other types of investors. So just by way of background, many of you may not have heard of polyomaviruses. They cause diseases in immune-suppressed patients, predominantly of the genitourinary tract, but they can actually migrate from there to other places like the CNS and cause really bad diseases. The problem is there's really no therapy for these viruses other than to reduce immune suppression. And so then these patients get trapped between a rock and a hard place where their immune suppression gets reduced to combat the virus, and then they begin to reject their transplant, for instance, or in the case of PML, their multiple sclerosis comes back. Antibodies have been seen as a promising approach to these types of viruses, and a number of companies have tried to move them forward, specifically because these are slowly mutating viruses that are particularly good targets for antibody therapy. The problem is that the virus replicates in the urinary space, which is a semi-immune privileged space where conventional antibodies really can't go. This is why we got interested in this, because one of the cool things that I didn't mention earlier about heavy chain only antibodies is you can just ditch the entire constant region, and then what you're left with is something called the domain antibody, which is essentially the binding portion of the antibody in a 12 to 15 kilodalton package. So why is that useful? Well, it's useful because you can string a couple of these together and still be under the size cutoff for the kidney, which is about 60 kilodaltons. You can make something that binds or neutralizes as well as an antibody and get it into the kidney this way. And furthermore, because you can put different domains together, you can potentially cover more of the different serotypes of the virus than you might otherwise be able to cover. The problem is that these kinds of molecules, there's only a handful of these that have even made it into clinical trials. And so the comfort level with investors like pharma is low with a molecule like this. They want to see that you can actually create something like this, that this is even feasible. So again, this is a great opportunity for grants, I think, because it is biologically very interesting. Working on such a novel molecule is really interesting. And so this is one of the projects that we did support with a grant from NIAID. And initially what we've done is discovered two very broadly neutralizing families of antibodies that neutralize all the different serotypes of BK or JC that we've tested so far. They have double-digit picomolar IC50s, so potent enough to potentially really help in these patients. They have good developability, and so now we're moving into the phase where we're turning these into strings, hopefully in a phase 2. So again, I just want to summarize that for this project, I think what I really wanted to point out is that sometimes you have these very high-risk, very high-reward projects that you want to undertake. And if the scientific rationale is good and you have some early preclinical, pre-preliminary data, this is something where an SBIR Phase I grant is a really great mechanism to get you the data that can then let you make, get, generate the other kinds of, or later phase investments that you may need to then eventually get it into the clinic. OK, the last thing I wanted to say I think that's been a really great lesson for Teneo in terms of grant writing is writing grants and working with a team like FreeMind for us has been really good for gelling our ideas. And part of that is because a grant proposal really is like a target candidate profile on steroids. You really have to get into the details of the feasibility, the timelines, the resourcing. What do you not have in-house? A good example is for our polioma project. We, when we wrote our grant proposal, we said, "Oh, we'll test all our antibodies for viral neutralization." And then, not surprisingly, we realized that we're antibody discovery experts, not virologists, and nobody has ever done a viral neutralization assay at Teneo Bio. So, we immediately had to reach out to collaborators that were good at doing that. And that collaboration is something that we've now used in other infectious disease projects. So, all of that, just writing the grant, I think, make— has made our projects better, more focused, and more deliberate about what the timeline is going to look forward— look like moving that project forward. I have all— I also mentioned earlier that I think collaboration is key to writing good grants. But, in fact, Grants also promote collaboration, because when you approach a collaborator saying, "We're going to write a grant. There's money specifically dedicated for you in this," that sort of supports your bona fides, right? That means you're committed, and I think for academics especially, money from grants for academic collaborators feels less feel like they're just being approached as a service provider, like you're just trying to buy something from them. It makes them feel more like it really is a collaborative effort. So I think grants also act as a great catalyst for generating collaborations with other people. And then lastly, I think having your ideas passed by reviewers who are likely experts in their respective fields means that you're going to get great feedback about how to turn your ideas into something that can really be successful. Okay, I'm— with that, I'm essentially finished. I just wanted to summarize what I was hoping to pass on to all of you today. I think the lessons that Teneo Bio has really learned from our grant writing efforts are that it's a great way to diversify your pipeline, especially your early pipeline. Secondly, Play to your strengths. Work on projects that leverage the real strengths of your company. But use that to kind of turn those innovations towards ideas that might be higher risk, crazy stuff that you've always wanted to try. And generate some of that preliminary data so that you can then move that forward. And then finally, that grants really thrive on and promote collaboration, which in the end will really benefit all the projects that you are advancing. And that's it. I'll take any questions. Thank you. Yes? [Speaker:AUDIENCE MEMBER] I have a question about— you mentioned that the data are actually aggregated. I'm curious if you went for government contracts as well, within the NAHDA program? [Speaker:CHRIS HAWKINS] We have not done that, but we've started thinking about that more, especially as we start thinking about clinical phase programs. So, for instance, you know, even a Phase 2 grant, even if you get $2 or $3 million out of a Phase 2 grant, a GMP campaign will cost you $6 to $8 million. So, in order to look into generating funds to cover those sorts of things, I know some of the agencies actually do have programs where they will make your GMP material for you. Of course, it's competitive, but Those are the kinds of things where we are starting to look more into that sort of thing. Is the UniRat available for licensing? The UniAbs? The UniRat. The UniRat is so far not available for licensing. We have had some discussions with folks about specific therapeutic areas, you know, outside of oncology, for instance, but so far it has not been licensed to anyone. We're the only ones who have it. It is available for collaboration, yes. Great. Thank you very much.