All right. Welcome, everyone. For the next session, we have Zach Hornby from Boundless Bio joining us. Zach, welcome. It is the first time for me to host you at the Goldman Sachs Conference. Maybe I will just turn it to you for opening remarks before we kick it off with questions.
Sure. Good morning, Rich, and thank you for hosting. Nice to be here in Miami on a Monday morning. Yeah, I'm CEO of Boundless Bio, a company that was formed about six years ago to address one of the largest unmet needs in all of cancer, which is patients who have tumors driven by oncogene amplifications. This counts for about 25% of all cancer patients. Unfortunately, patients with this type of tumor generally do not benefit from targeted therapies or immunotherapies or the classic standards of care, and the consequence is that they have worse outcomes and inferior survival. Huge unmet need to try to bring new options to these patients.
Great. I mean, obviously, this is a very different way of targeting and treating cancer due to the oncogene amplification. How do you think, I mean, can you tell us a little bit more about these types of cancer, and then what's the unmet need there that you're trying to address?
Yeah. Yeah. Patients with amplifications, unlike patients with fusions or drivers, instead of having a mutation of the gene sequence, they have too many copies of the gene. It was classically thought that that always occurred on chromosomal DNA based on the human genome map. Our scientific founders made a discovery over the past decade plus that frequently when amplifications occur, it is not in the normal chromosomal locus where you might expect based on Mendelian genetics, but instead it is based on these cancer-specific units of DNA that are circular, and they are called extrachromosomal DNA or ecDNA.
These little circles of ecDNA that are unique to cancer, they reside within the chromosome or within the nuclei of the tumor cells, but they have this whole set of properties that leads to oncogene amplification and leads to genomic heterogeneity that can allow cancer cells to evolve real time, particularly under therapeutic pressure, and drive resistance. This presence of ecDNA helps explain why targeted therapies do not work in the traditional sense and why these tumors are so aggressive and refractory to care.
I see. So there's a lot of buzz in this area scientifically, but this is still very early and very new. How would you target these types of DNA, and how is the company built around doing that?
Right. It's funny, you speak to the buzz in the field. I'm actually getting on an overnight flight tonight to London because there's the first-ever international ecDNA conference hosted by CRUK for the next three days. It builds on ecDNA being the opening plenary address at AACR this past April. It really is becoming a very hot field. Boundless has been at the cutting edge. We were the first company formed to address this biology. The way that we identify targets is we have built out a unique and proprietary platform within Boundless. It's called Spyglass. It consists of hundreds of tumor models that have oncogene amplification that we've characterized for either having these ecDNA and relying on these ecDNA or control models that do not implement ecDNA.
By having those two different sets, we can run all kinds of target ID screens where we compare and contrast the two sets of models to try to identify which targets that, when inhibited or genetically knocked out, they selectively kill the ecDNA-reliant models, and they spare the control models. Because when we ask the question in that way, we can identify targets that are exquisitely sensitive in the ecDNA models and therefore must be somehow fundamental to biology. That is really the premise of our platform, Spyglass.
I see. Maybe just a little bit about Spyglass. I mean, you mentioned this is your discovery platform. How is this platform unique? Compared to what other companies are doing in this field, so far, what have you been learning about the ecDNA and about these types of cancers from this platform?
Spyglass, we really custom-built because we were looking at a new set of biology that no other company had ever worked on before. It had to be unique just by its founding principles because we were trying to do something that had not been done. It really consists of three components. One I alluded to, which is these hundreds of wet lab models as well as thousands of models in the cloud where we can access publicly available data. All of these have been exquisitely characterized by our scientists for their ecDNA status. That is one component. A second component is we had to build a whole library of tools for detecting ecDNA, monitoring them, quantifying them, characterizing them, perturbing them. Third is we built a lot of unique computational or bioinformatic solutions to analyze the data.
When you put all those three things together, the ultimate goal is to identify targets that are synthetically lethal in patients with oncogene-amplified cancers. That too is differentiated because we do not really think any other companies are exquisitely focused on amplification-driven cancers the way that we are.
I see. So let's maybe move into the pipeline. You guys have three candidates in development. Why don't we just start off with the first two, the two lead, the 355 and 825. Both of these came from Spyglass, and I believe that you guys made some changes in your development plan. Maybe walk us through that as well.
Absolutely. We talked about how Spyglass works to identify targets that are synthetically lethal in oncogene-amplified tumors. Collectively, the efforts that we've been applying over the past half dozen years have honed us into three facets of the biology or three nodes, let's say, of the ecDNA life cycle. The first is the replication stress that occurs when tumors have high copy number amplification, particularly when due to ecDNA. They have a lot of replication stress. If that replication stress goes unmitigated, it will lead them into mitotic catastrophe and death. These cancer cells employ some techniques to self-regulate their replication stress, overcome it, and then move on in the cell cycle. One of the targets that they upregulate to orchestrate through replication stress is called CHK1 . It's been a well-known cancer target. Many companies have tried to work on it.
Yet despite that, there have been no approved drugs. There are no approved CHK1 inhibitors, and nobody's looked at it from this angle or with this application as us. Our first program is an internally discovered, wholly owned oral selective CHK1 inhibitor. That's in the clinic, and I'll talk about its development in a moment. The second area of biology that we discovered is the assembly and repair of DNA, including ecDNA. We found an enzyme that is rate limiting for a cancer-specific pathway of DNA assembly. There are two pathways that cells use. One is called the salvage pathway. It's more energy efficient. That's what healthy cells use. There's a less energy efficient pathway called the de novo synthesis pathway for DNA assembly. That one, while it's less energy efficient, tumor cells rely on it because they have higher needs.
An essential enzyme in that pathway is called ribonucleotide reductase or RNR. We have found that when we inhibit RNR, we can selectively kill amplified cancer cells. Our second program, internally discovered, wholly owned oral selective RNR inhibitor, BBI-825, also is in the clinic. To your question, we brought both programs into the clinic recently. We had some initial development hypotheses where, as we generated data for the first program with the original kind of approach and dosing regimen, we were bumping up against a limited narrow therapeutic window. We were seeing activity, signs of tumor response, including RECIST response, but we were not getting robustly to the level we want in balance with the safety. For the second program, 825, we were getting good initial dose-dependent exposure.
However, what we were seeing was that after steady dosing, we're seeing a CYP3A4 induction and induced metabolism and a decrease in exposure. That was limiting to our original development plan for that program. However, over the past year, we've been doing some preclinical work based on some clinical precedents where if we combine programs, we can actually administer both of them much less frequently, which potentially overcomes the safety issue of the one, the metabolic issue of the other. Their mechanisms potentiate each other such that with weekly dosing, we see profound anti-tumor activity in vivo, and we see lack of overlapping toxicity. We believe, based on what we're seeing, there is a chance that this combo will achieve the sweet spot of giving the requisite anti-tumor activity, but without the overhang of the safety. That is what we intend to study in the clinic now.
I see. Has there been any historical clinical evidence that supports the potential synergy between the CHK inhibitor and RNR?
There actually has been. There have been prior companies who have tried to develop CHK1 inhibitors who, in these earlier cases, their CHK1 inhibitors were a couple log fold less potent than ours. They were not potent enough to see single-agent activity. They did combine them with non-selective RNR inhibitors, particularly the compound gemcitabine, which was not originally discovered as an RNR inhibitor, but it turned out that it has potent RNR inhibitory activity. However, it is non-selective. It is part of a polypharmacology. One of the things about gemcitabine, it is a nucleoside analog. It kind of looks like it resembles the building blocks of DNA. It gets actively ingested into heme cells. This will speak to the toxicity, which we will talk about in a minute.
Nonetheless, adding low doses of gemcitabine as a non-selective RNR inhibitor to CHK1 inhibitors took multiple different CHK1 inhibitors and brought them from having no clinical activity, no recess responses, to multiple clinical responses. That was across two distinct prior CHK1 inhibitors. I would say there is robust evidence that adding an RNR inhibitor can potentiate the activity of a CHK1 inhibitor. In a minute, we can talk about the limitation of gemcitabine versus what we are going to do.
I see. Okay. What's giving you confidence that combining 355 and 825, and will you explore different doses between these two, different combinations, and how that would substantially lower toxicity that you've seen with these other historical trials and in terms of any increase in overlapping toxicity?
Perfect segue to what I was just trying to answer. As I mentioned, gemcitabine as a nucleoside analog gets actively transported into heme cells. Any pharmacology it has is going to be particularly acute in the bone marrow. Gemcitabine on its own has hematological toxicity. When it was combined with prior CHK1 inhibitors, while it led to increased tumor activity, it also led to increased hematological toxicity. Therefore, it did not expand the therapeutic window of the CHK1 inhibitors, it simply made them more potent. In our case, our oral selective RNR inhibitor is not a nucleoside analog. It does not resemble dNTPs. Importantly, it does not get actively transported into heme cells. Therefore, the pharmacology that we have achieved, that we've observed in vitro and in vivo, seems to be much more preferential to cancer cells.
Therefore, when you combine the CHK1 inhibitor with the RNR inhibitor, that effect also to date has been preferential into the cancer cells. In fact, because we really wanted to be rigorous about this, we actually conducted a dog heme tox study just to confirm for ourselves that we were not having the heme tox that's observed when you combine gemcitabine with the CHK1 inhibitor. Indeed, in that study, we did not see overlapping heme tox. It does appear that with our regimen, we have a profound tumor activity at levels that do not lead to toxicity, at least in animals. Once again, that's what we need to test in humans. The final thing you talked about was dosing regimen. With these two things combined, we can actually dose each of them once a week instead of daily or BID or Q2D.
This is a much less intensive dose regimen. This gets back to the synergistic mechanism of these two compounds.
I see. So maybe just to kind of follow up on that, I think some human PK study for 355 monotherapy showed that the 60 mg Q2D is a minimum dose that's required to achieve sort of the exposure in that therapeutically active range. And then the minimum dose for 825 is around 800 mg BID. Now switching to a weekly regimen in this combination setting, what do you think is sort of the minimum clinical dose that you need for both to achieve that same therapeutic active range?
Everything you said is accurate, but there's one caveat, which is the 355 data that showed that 60 mg Q2D was the level associated with exposure in the active range. That was based on a specific setting, which was in the combination with targeted therapies for certain oncogene-driven tumors like EGFR amplifications or FGFR amplifications. Then with 825, the exposures that were in the active range were for a setting of MAP kinase resistance. What turns out when you put these two compounds together in our in vivo models, the exposures that are required are lower than what was required in those other settings. Once again, this is because they have synergistic mechanisms where they potentiate each other. We do not necessarily think we will need to get to that same dose level that was required for those settings.
Nonetheless, as you mentioned, we're going to be dosing both weekly, and we're going to be starting, subject to FDA feedback, probably around 100 milligrams or so of 355 once a week, which is a significantly less intense regimen than the 60 milligrams every other day, which was already well tolerated. Because if you think about it, that equates to 210 milligrams per week. And for 825, that was already tolerated at 125 milligram BID continuously. In this case, we're hoping to start at 1,200 once a week and with 355 at about 100 or so once a week, and then hoping to get up to maybe around 200 once weekly with 355. This will all be ultimately dependent on what we say with PK, PD, safety in the clinic. But that gives you a sense of what we're intending to do with the escalation.
I see. This combination, I think you mentioned that you guys have done some tox study in dogs.
Correct.
How many days and how many treatment cycles were tested in dogs? How robust is that data, you think, in terms of translating to humans?
It was a 14-day study. The dosing regimen for the safety was actually more intense than what we required for efficacy in the xenograft models or even than what we are going to be doing in the clinic. In that particular study with the dogs, we dosed twice weekly. Like I said, we followed them for 14 days. I think they got four treatment administrations, two per week times two weeks, whereas in the clinic, they will only be getting one per week. 14 days, based on both preclinical and clinical experience, is sufficient to see hematological impact. We do think it was a well-designed study to assess what we are going to be looking for in the clinic.
I see. Okay. In your previous first-in-human study for 825, you guys did not see, in terms of after your BID dosing, you guys did not see any signal in the therapeutically active range. You mentioned likely due to the CYP3A4 induction. Maybe talk to us about what's going on there with the CYP3A4. What's the highest clinical dose for 825 that you can use that meaningfully does not induce CYP3A4?
Right. In the initial study of BBI-825, it entered the clinic last year in a first-in-human study. It was called the STARMAP study. We dose escalated through four dose cohorts. It's administered orally twice a day. We tested 200 milligrams, 400 milligrams, 800 milligrams, and 1,200 milligrams. There are several pieces of good news. It was very well tolerated. We did not establish a maximum tolerated dose. On cycle one, day one, the first day of administration, we did see a nice dose proportional increase in exposure, including getting up into the active range. The problem is that when we measured PK at steady state at day 15, those dose levels were reduced. This is a classic hallmark of a CYP3A4 induction and auto metabolism. It is what it is.
Ultimately, that was limiting the steady state exposure, and it was not where we wanted to be. Yes, it's true. We didn't see any tumor responses, but nor did we see CYP2 as a single agent. This was really a combination play. The reason why our new approach is different is that CYP induction was observed after continuous twice-daily dosing. CYP induction takes a couple of days to set in. If you're continually dosing, it's going to become a consistent effect. What's different about what we're proposing to do now with the 355, 825 combo is we're dosing for one day. Even if any modest CYP induction occurs, then the dosing, there's a break for the next six days so that CYP could dissipate prior to resuming the dosing.
Really, the effect should be negated based on the dynamics of CYP3A4 induction. Once again, I have to confirm that in the clinic. To your question of dose, I think I mentioned a few minutes ago, we intend to start at the 1,200 milligram dose level of 825 because that was well tolerated when it was administered continuously. Now we're only going to be administering it once a week. We have no reason to believe that that would not be well tolerated in this regimen.
I see. Have you guys looked at potential DDI between the combo with the 355 and 825 because of the CYP3A4?
We have. We don't anticipate any major DDI issue. The only possibility is that we do know that 825 induces CYP3A4 and 355 can be a substrate of CYP3A4. There is a possibility for a minor DDI there. I'll still come back to the same theme. We're dosing them once a week. All of these effects get exacerbated when you're dosing continuously and there's a continuous pharmacology at play. When you're doing more of this hit and run, there's just less of a risk of things like CYP induction and DDI. We'll see in the clinic, but we're not modeling or forecasting a major DDI issue.
I see. So with CYP3A4 being such a major metabolic pathway, is there any concern about that interfering with any other drugs that the patient would be taking?
We have the standard inclusion-exclusion criteria that I might expect in terms of physicians being aware of the CYP properties of each of these compounds. It is nothing that we see as a deterrent to enrollment or overhang of the conduct study. We just came from ASCO. We had a nice investigators' meeting with our investigators. They are all excited about participating in this study. They shared no concerns about the protocol design.
I see. Okay. What's the next steps and timeline for this program?
Yeah. The amendment has been submitted to the FDA. So we're just awaiting FDA feedback or clearance. But assuming everything's clear to go, then we look forward to initiating this new arm, this 355, 825 combo, hopefully dosing the first patient in the next couple of months here and then escalating as rapidly and efficiently as possible. Really, next steps are FDA clearance, starting the study, and then escalating. I mentioned to you that we're going to start with a relatively high dose of 825 because there was no MTD. We're going to then escalate 355. We project that after about three to four escalation cohorts, we should be in the range where we'd like to be, at which point we would go into expansion cohorts. That's the plan.
I see. Okay. Great. And then you have another program heading into the clinic, the 940.
I do.
That's a kinesin oral degrader.
Yes.
Can you tell us a little bit more about this target and what makes it unique and also why do you want to approach it from a degrader perspective?
Yeah. We're really excited about this target. It's a novel target. From everything we can see in the literature, patent databases, it's never been worked on by any company in the world. And we discovered it directly through the Spyglass platform. It plays a critical role in the congression of DNA and chromosomes during mitosis. Chromosomes have centromeres. So there's something called the kinetochore that during mitosis, they align at the mid pole and then get evenly divided into daughter cells. The ecDNA that we're focused on do not have centromeres, therefore do not have a kinetochore. And therefore, they rely on different kinesins to bind the DNA. And they basically hitchhike with the chromosomes during mitosis. And that's how we discovered this kinesin. The beauty of it is it's non-essential in healthy cells, but it appears to be essential in amplified tumor cells with ecDNA.
We discovered this kinesin. Because nobody had worked on it, there was no chemical starting place, nothing we could work off of. We had to design all of our own assays, do high-throughput screens to discover compounds. We screened over 1 million compounds. We found a single hit. It's like 50 micromolar, not very potent. We did a tremendous amount of work in medicinal chemistry, SAR, to optimize to the point where we were getting down to single-digit inhibitors, even picomolar. Still, those compounds were not achieving robust cellular pharmacology. Our hypothesis is that this kinesin, this molecular motor, which binds to microtubules, was probably playing a scaffolding function where inhibiting it alone was insufficient. We had the idea to try to degrade it and see if that would kind of overcome.
Sure enough, we started making degraders, and we suddenly started seeing really robust cellular pharmacology. We were able to optimize those degraders, very importantly, to the point where we're achieving robust oral bioavailability in multiple species. We're excited to have a potent oral selective degrader that's achieving very high % degradation. We just declared the development candidate at 940 about a week or two ago. We're now in IND and enabling work and hoping to dose the first patient first half of next year.
I see. Can you tell us more about maybe the in vivo work that you've done for 940 and also what is the development plan for this?
Right. In vivo, we've done what you might expect for an early program where you're trying to figure out how it works, where it works. We've done multi-hundred cell line panel screen to see which cell lines had sensitivity. We've done so far a small mouse clinical trial of about 30 different rodent xenograft models. We have more cell lines and mice ongoing. We've done selectivity assays to make sure it's not hitting other targets. The takeaway is that we're seeing single-agent regressions in difficult models, for instance, triple-negative breast cancer model among other models. We're seeing really nice activity. So far, we're seeing good tolerability. We've done pilot tox studies. It looks well tolerated at exposures, multifold above the level associated with activity. Good selectivity, good oral bioavailability, single-agent regressions.
Of course, the next steps are GLP tox, GLP manufacturing, and continuing to explore the breadth of the activity across many different tumor types. We have not yet defined publicly what the clinical development plan is, partly for competitive reasons, partly because the work's ongoing. That is something we would probably look to disclose at a later date.
I see. Okay. Look forward to hearing that.
Thank you.
You guys also have a proprietary diagnostic tool called Echo. How does that work, and how does that factor into your development of these novel assets?
Because ecDNA is a relatively new area of biology and therefore a new biomarker, up until now, there's been no clinical assay to detect ecDNA. What academia does is they use these research-grade tools that rely upon whole genome sequencing. We know that clinically, most patients don't get whole genome sequencing, so that wouldn't be a standard of care. We really wanted to be able to slot into standard of care, which is typically these multi-gene onc panels like Foundation Medicine or Caris. What our team was able to do was to develop a proprietary bioinformatic algorithm, ultimately a software that uses standard multi-gene panels. It takes those data as its input. It ingests the data from a panel. It recalculates them and can determine whether these circular units of DNA, ecDNA, are present. That is what Echo is. It's a software.
It takes standard input data. It spits out, is there ecDNA? If so, how many copies? What genes are amplified on it? What it allows us to do for all of our clinical studies of all of our programs is determine the ecDNA status of each patient enrolled. So far, we have not made it a condition of requirement. We've enrolled more broadly just based on the presence of oncogene amplification. We can retrospectively ascertain ecDNA status to analyze in an unbiased way, is the ecDNA status correlating with any activity or safety or any other features we're interested in.
I see. Okay. You mentioned earlier that you guys were the first company that sort of formed to target the ecDNA space. Has the mainstream pharma sort of companies, have they been getting into the space? Maybe talk about how you think about the competition for your program.
Yeah. Yeah. The good news is they've started to. I mentioned earlier that I'm heading to this conference overnight for the next few days in London. It will be attended or well-represented by major pharma companies, folks we know. Looking forward to interacting with them there. There's another private biotech company called Econic who was also formed a little bit after us to go after ecDNA. They'll be there. I would say within pharma, they're all aware of ecDNA. They're dipping their toe in their water. They're attending these conferences. They're taking meetings with us. Probably not a lot of them have internal ecDNA-focused research groups yet. I think they're trying to get up to speed by attending the conferences and interacting with us.
In some cases, companies like AZ or Merck Ex-U.S. have been funding postdocs at academic institutions just to work on this biology. That's a very early innings of what I think will be a multi-inning game.
I see. Maybe talk to us about your cash position and then what is guided in the runway for that. What studies are included in that and what studies are not because of some changes in terms of the combination?
Yeah. We ended Q1 with $138 million on the balance sheet. As Rich just mentioned, with the operational focus that I've described today, the 355, 825 combo, bringing 940 in the clinic, we've got an efficient organization, just under 40 people. We project that that cash will last us into 2028, so approximately three years from where we sit right now. Most importantly, it allows us to drive the clinical development of both of those efforts. What I mean by that is within that cash runway timeframe, we expect to have meaningful clinical readouts from the 355, 825 combo, as well as from 940. When I say meaningful clinical readouts, what I mean is a reasonable number of patients treated in an expansion cohort at the recommended phase two dose.
That was really a critical part of the recent focused efforts was to make sure that we had sufficient runway to get two independent clinical shots on goal. They're non-overlapping mechanisms. Therefore, they're completely diversified or non-synchronized risk. They're both potentially first in class. They're both wholly owned. We should have real data from both.
I see. Okay. The stock has been trading down since 2024, since early 2024. What has been the biggest pushback that you've been hearing from investors? What do you think investors are not appreciating?
I think it's hard to find stocks in our industry that haven't been trading down since 2024. Certainly, the overall state of the biotech market has not been helpful. On top of that, being a company within a year or two of IPO, I think you guys probably have the stats. I want to say something like 90% of IPOs from the last two years are down. That doesn't help. I think more fundamental to Boundless Bio, and I appreciate this and think about it every day, is we are doing something new. It's very exciting. I think people love that we're trying to innovate and go and fulfill an unmet need that hasn't been addressed. That also means there's just the intrinsic biological risk of trying to do something new. In this environment, that means it's very much a show-me story.
The burden is on us. The onus is on us to prove that our novel approach works. We haven't done that yet in the clinic. I think this environment is one where you don't get credit for trying. You get credit for having done it. That's what my team understands and is motivated and looking forward to overcoming that burden, that onus, and demonstrating that we're on to something. I think when we do that, we'll be rewarded for that.
Right. And then based on what you said, I mean, do you think this is the right time for investors to get into the stock? Or do you see it as wait until you guys have some more meaningful data?
I think it's a very exciting time to be a part of Boundless, whether that's those of us who are employees there or those who want to participate as an investment. I mean, if you just think about the formula for what innovation in America or anywhere should be, you got to start with an unmet need. It's unquestionable that we've got that 25% all cancer. Statistically, we're survival. You got to go after it with exciting differentiated biology. We're doing that. We're going after this area of biology that's really taken off over the past decade, opening plenary at AACR. We're the leading company doing it. If one wants to discover new innovation, this is a great way to do it. We're doing it with a great team. My broader management team is extremely experienced.
They've had success both developing drugs, but also a multi-billion dollar M&A for shareholders. The team loves the science. They're focused on it. We get along well. I think that's a really nice formula. When you put that formula together and you add to it multiple first-in-class assets that are wholly owned that are going to have clinical readouts in the next 12 to 24 months, all within the existing cash runway that we have, I'd say what's not to like.
Great. One final question before I turn it to you for closing remarks. How do you see sort of your vision for Boundless? Is it going to be continue to look for these novel targets and this novel science? How do you think about sort of that vision for the next couple of years?
Yeah. In the abstract, I love the unmet need we're going after, and I love the biology that we're doing it with. I think it's a fishing well or honeypot that has not yet been fished dry. In other words, there are more targets that we've discovered that we think are worthy of pursuit. In an ideal world, we have the capital to continue identifying novel targets, identifying novel drugs, bringing them forward, demonstrating clinical concept. I recognize that that model can be constrained right now by prevailing macro conditions. As stewards of investor capital, we will always be wise and thoughtful about how to create value in a way that's appreciated. I think it would be a nice thing if we could continue to truly invest in the biology.
Great. Zach, it's been a pleasure hosting you for the first time, at least for me at Goldman Sachs. I will turn it to you for any final or concluding remarks.
Yeah. Thanks, Rich. It's been fun to sit up here with you for the first time. Great to kick off the Goldman Sachs Conference this way. Thank you, everyone, for your attention.
Right. Thanks, everyone.