AN2 Therapeutics, Inc Investor Event Transcript, Nov 9, 2026
AN2 Therapeutics, Inc. (ANTX)
AN2 Therapeutics received FDA clearance for its polycythemia vera trial, plans to dose 10 patients in the sentinel cohort in December and expects milestone updates throughout 2027. The broader trial will enroll 75 patients, with initial data expected around May. AI summary
Conference Transcript - ANTX 2026-11-09
Moritz, Host
In conference, my name is Moritz. Our next presenting company is AN2 Therapeutics, and joining us from AN2 is their CEO, Eric Giesem. So, to begin with, congrats on the IND clearance. For those not familiar with the story of AN2, could you just give us a brief intro into AN2 Therapeutics, your boron chemistry platform, and the programs you currently have in development?
Eric Giesem, CEO
Yeah, and I'll be happy to. Thank you for having us here. So AN2 is a boron chemistry platform company, and we've discovered all of our molecules internally. We have three compounds in clinical... So we have our polycythemia vera, and then we have two other ones, one for Chagas disease and one for non-tuberculous mycobacteria lung disease. So we have three that are just starting phase two that we're going to have data readouts, and then we have a bunch of things coming. We just announced last week a program for hypophosphatasia that is going into development. That will also start phase one next year and then a number of other oncology programs behind that. So we have a very robust pipeline, I think, that has a lot of value creation potential and addresses a lot of high unmet needs. So we're kind of agnostic to the disease area, per se, because we're across now oncology, infectious diseases, bone disease. But we follow the biology, basically. So anyway, we're very excited about the potential of all these programs and the future set up for AN2.
Moritz, Host
Great. So let's start with PV. Your phase two PV trial is about to kick off and has three parts. You'll have a sentinel cohort followed by an open labor part one and then a randomized withdrawal part two. Can you just walk us through the rationale of this particular trial design?
Eric Giesem, CEO
Yes. So for PV, we have lots of clinical observational data showing that our drug lowers hematocrit in a very red blood cell-specific way. It's oral and so forth. So the main question on the table is, does this translate to PV patients? And we believe it has a high probability of doing that. And so the primary purpose of the study initially for the sentinel and then part one is its open label is to really see that proof of concept and translation in PV patients with this therapeutic. If it does, we're going to feel quite bullish about this. Of course, you know, we haven't dosed a PV patient, so we'll be looking at safety and the amount of efficacy, I think. And the other important part is we know what the dosing curves look like in non-PV patients. But we don't know NPV. So we'll be exploring three different doses to kind of see where we land on the dose curve. So that's kind of the part one, but that'll give us high confidence that we have the ability to control hematocrit of patients who have very poor control, you know, less than 50% is kind of the data on that. So we think we can really improve upon that. And then the part two is more of just taking drug off and showing, okay, without drug, it clearly goes back to, reverts back to baseline, their kind of baseline status, and we can pick up other PRO activity in a blinded fashion and so forth. So that's the main purpose of that. And then what's not shown there is a part three, which will be a long-term extension, which will be durability over long periods of time, which will be important to gather.
Moritz, Host
And size-wise, how many patients do you expect to include in each of the parts of your study?
Eric Giesem, CEO
Yeah, so the sentinel group is a low-dose run-in of 10 patients into a bigger part one, which is 75 total patients. So we plan to take 75 kind of through the whole.
Moritz, Host
Trial, yeah. And what's the current status of the sentinel cohort when you plan to dose the first patient?
Eric Giesem, CEO
Soon. We're excited. We just got regulatory clearance from the FDA we announced, as you mentioned earlier, and also simultaneously we announced we have Australia and India. So we plan to dose the first patient yet this year. So, I mean, we're going to get started on this. We have, you know, kind of get the show on the road and with lots of kind of data and milestone updates throughout 2027.
Moritz, Host
So when are we going to get the first data cut from this?
Eric Giesem, CEO
Yeah, always a big question. You know, all this is... We don't have experience with these sites and PV trials, so it's always how quickly can we get patients enrolled. But we think the short answer is we think the sentinel group will be in the first quarter. But that's mainly just a low dose, but mainly a safety readout to make sure. And also, is there an over-response or a more profound response, let's say, for PV patients than we saw in non-PV patients? So pick up, you know, kind of where we are on the dosing curve early, and then we'll expand that from there.
Moritz, Host
But early, Q1-ish. So you mentioned this is the first time you're testing this drug in PV, but importantly, unlike many other early stage drugs, your drug already has a very large existing safety database. Can you just give us some color on that? What data have you already collected?
Eric Giesem, CEO
Yeah, so we, you know, this was a side effect, essentially, of our non-tuberculous mycobacteria lung disease. So there we have, you know, 105 patients out to, you know, six months, some even out to 12, 15 months. So we've got a lot of data there, but we also have, you know, something like, I should get the exact number, But we've run 10 studies, so we have hundreds of patients of data where we see this kind of phenotypic profile that we have with this drug in patients and then in volunteers in phase one studies. And then we've also run a tremendous amount of... We've done a lot of preclinical studies in NHPs where we see exactly the same profile and done a lot of PKPD modeling on this exact effect because we wanted to really understand the effect on red blood cells before we went into long-term trials. So we have a tremendous amount of knowledge about the effect, which you normally wouldn't have. We also have done, you know, chronic tox studies and the NHP at super therapeutic doses, you know, more than 10x, the clinical doses we expect to go in this trial over long periods of time. So I think we've got a tremendous amount of both safety and, you know, quote-unquote efficacy data, I guess, for hematocrit across that. So that's very unusual, and I think de-risks to some extent, other than we now need to see if it works the same in PB.
Moritz, Host
Can you touch a little bit on the mechanism of your drug by which it actually lowers hematocrit? And just, you mentioned this already briefly, but does it have any effect on white cells or platelets?
Eric Giesem, CEO
Yeah, so we, as I mentioned, we noticed this effect on red blood cells early on in NHP models. So we did a lot of modeling. So we know, you know, the effect, the kinetics happen over the first 4 to 12 weeks, And then there's kind of a plateauing or a compensatory effect. We know that it occurs kind of in late-stage erythropoiesis. So it's, you know, kind of just prior to kind of reticulocytes where hemoglobin is incorporated into cells is where we know. We don't know the exact mechanism of action. We're exploring a number of those. We know that it doesn't affect bone marrow, like precursor cells or... Any of the progenitor kinds of cells, because we've done bone biopsies on NHP tox studies, both, you know, kind of long-term effects and so forth. So there's no effects kind of on that. So we've narrowed it down to where we know where it happens. We think it's in globin synthesis side of the equation, because there's not a lot of change in iron stores or levels like maybe you You get with hepcidin-type agent. They starve the cells of iron, and that's how their mechanism works. So we know it's a different mechanism of action, but we don't know. And so we're exploring a number of hypotheses around that. But we know it's, you know, very predictable, monitorable, reversible, all the things that you would want to have.
Moritz, Host
So let's talk about efficacy a little bit. So this could, if successful, could be the first oral drug in PV. So what do you think would be the minimally viable efficacy threshold?
Eric Giesem, CEO
Yeah, I mean, I think there's, fortunately, there are a number of agents out there, so most recently, Rust for Tide has kind of set a benchmark, you know, all of those. Drugs are showing the ability to control hematocrit, you know, at 60, 70, you know, pretty, pretty high levels compared to kind of the, the, the experience. So, you know, they're all injectables. So would an oral have to be equal to that? I don't think so. We don't, we haven't yet, since this program is fairly new for us, we haven't done a lot of commercial kind of target product profile assessments at this point. But I think there's some leeway to be less than, but we're quite optimistic based on the data we've seen in non-PV patients that we can achieve high levels. of efficacy, you know, even as good as or, you know, if it's great, it could be better than, but we're optimistic we can do that. But I don't think it would have to be as good as those, but we'll, we'll get back to you on that question in the future.
Moritz, Host
Great. Anything else you'd like to add on the PV side of things before we move on to some of your other programs?
Eric Giesem, CEO
Yeah, no, we're just super excited to get started and start seeing data and, you know, it's going to be a pretty quick pace. So we're going to have lots of inflections and, you know, important updates as we get on into 2027.
Moritz, Host
Maybe just one more question on that. We sort of talked about the Sentinel data as the initial readout. Could you just give some color on beyond that what sort of the cadence of readouts is going to look like? What do you think?
Eric Giesem, CEO
Yeah, I think, you know, it's back to kind of it depends on the enrollment. But, you know, the Sentinel group, we're going to start in December. So we'll start getting, part one is 28 weeks. So we'll start hitting those kind of timelines around May for the initial patients. So we'll have a pretty long set of data for some of the first patients. So, you know, our guidance basically is what have others done that have preceded us. So, for example, resveratide reported out after seven patients where, you know, a few hit 28 weeks and less. So it'll, you know, sometime it could happen around mid-year where we would start to have pretty robust sets of data out over long periods of time in PV patients. But it'll depend on, you know, kind of our statistical analysis, how many we have, and, you know, where we think. And then just kind of updates throughout the rest of the year. So pretty near term.
Moritz, Host
Great, so you've got two other programs with clinical data expected in 2027. So lots more exciting things going on besides PV. So let's start with Chagas and your phase two. What's the trial design there, and when do you expect to start dosing?
Eric Giesem, CEO
Yeah, so we're in the final throes of the regulatory approval process, and the plan is to begin the trial soon afterwards. So probably the first dose will happen in the very beginning. We're already in kind of startup mode at the beginning of 2027. But the trial design is a 60-patient, two-dose trial, and the statistics are quite high because it's either sterilized and get rid of parasites or you don't. So we've previously reported on three NHP studies with naturally acquired infections, and once you get above 10, 15 animals, the stats are quite profound. It's either because the efficacy is very high, so you get like 90% clearance or 100% clearance, or you get zero on the stat. So it's pretty easy to see that signal, proof of concept. So the study's going to start, we're doing it in Bolivia, which may seem unusual, but Bolivia has the highest rates of Chagas in the world. About 10% of Bolivians have Chagas disease, which is parasites in your heart. So I don't think anybody wants heart. I say heartworms, but they're trypanosomes. But, you know, like we treat dogs and cats and things. We're freaked out about that. So imagine if you're a human being with this. So anyway, we want to... To get rid of those, because it's a silent killer, and it shows up kind of late in the game. You have, you know, a heart failure. You drop dead of an aortic aneurysm. So it's really bad. There's, you know, there's, I'm kind of going a little different, but there's 10 million people with Chagas. So imagine 10 million. Humans with parasites, and there's no real effective drug for treating these chronic patients of that, and so that's what this program's all about. So the trial will enroll 60. There's a lot of patients in Bolivia. Back to your question, these sites can recruit 20 to 30 patients per site per month. Because there's so many patients. So we expect to enroll it quickly and read out in the second half of next year. So we're super excited about that. Great.
Moritz, Host
So obviously this disease is quite prevalent in South America, but what about the U.S.? What's the current prevalence estimate and sort of how reliable is that number?
Eric Giesem, CEO
Yeah, so the estimates were made on the U.S. blood supply from donors. So we screened the U.S. blood supply because you obviously wouldn't want, it can be transmitted through blood. You know, HIV, HCV, etc. So the estimates are 300,000 plus. You know, there's people who believe that's way underestimated because the typical patient is from Latin America, and they oftentimes don't give blood unless there's a family member that needs surgery. So it's an underrepresented. But 300,000. So, you know, we believe in the U.S. It's quite a large market opportunity for this drug.
Moritz, Host
And what gives you confidence in the success of this particular program?
Eric Giesem, CEO
Yeah, this one is about as sure. Infectious diseases translate usually from preclinical models to human efficacy. This one is even better than that because we've done three monkey NHP studies, as I mentioned, that were infected naturally in the state of Texas. So the kissing bugs that are infected exist kind of from Indiana, if you drew a latitude line south. So in the south, lots of animals who are outdoors, like dogs and so forth. So we've done three monkey studies of naturally infected, genetically diverse parasites. They're natural carriers, and we've cured all of those. So I'm about as confident as you could possibly be in this working in people. The other thing that gives us confidence is we were part of a team that discovered a drug called a coziboral, a boron compound that hits the exact same target for another trypanosome disease called African sleeping sickness, and it had 96% cure with a single oral dose treatment in sub-Saharan Africa and is now approved by the EMA for treatment of that disease. So same mechanism, trypanosome disease, You know, we're saying chemistry, so I think. It's about as sure of a bet as you can get, yeah.
Moritz, Host
And if you're indeed successful with this program, would that make you eligible for a PRV?
Eric Giesem, CEO
Yeah, so Chagas is hard-coded in the Orphan Drug Act under tropical diseases as a PRV-eligible disease for NCEs, which this would be. So, yes, it would also be eligible, in addition to the market potential in the U.S., I think, and other developed countries. It's PRB-eligible.
Moritz, Host
Great. So let's spend some time on the other clinical program, M-Obsessus. So why do you believe you can be successful with Obsessus after your Phase III miss in MAC?
Eric Giesem, CEO
Yeah, so we ran the Phase III MAC study we ran and In treatment refractory MAC, and kind of to set the historical context, this is after InnsMed's error case was approved for that population. So what we found was in our trial, we ended up getting patients who were really refractory, substantially more than the error case was in their pivotal trial. So, for example, we had patients had nine years of average. They had three years. We had, you know, two-thirds of our patients had multi-drug resistance, including amicacin. They had zero because that's how their drug works. So they had very little multi-drug resistance in the background therapy. And we had patients with lots of cavities. So I think what ended up happening was we ended up with a very, very refractory patient population with that group. And so, you know, we've done a lot of PK-PD analysis and so forth. So what we've looked at is our strongest data is in another section of NTM called Mycobacterium abscesses. We have 250 times more potent against that bacteria. Probability of hitting target attainment is higher. We're doing it in treatment-naive, like early-stage patients, and so it should work from everything we know in the animal models we look at. The other thing is Paratek has amatocyclin. They ran a successful phase two recently in the same trial design that we're using, and so our drug looks substantially better. There's a static in that model. Narcicidal. So there's just a lot of things, I think, that say this should work.
Moritz, Host
And could you just briefly touch on the OMIT need for an oral in this space?
Eric Giesem, CEO
Yeah, this is another one. By the way, for Chagas, there's no approved drug for chronic Chagas in adults. NTM for abscesses, there's no drugs FDA approved for this. So right now, they use off-label IV-only antibiotics. So if you're a patient, you're going on drugs like imapenem, amikacin for long periods of time with daily IVs. So it's really like chemotherapy, and patients really do not want to do this. So our vision of this is, petroboril is an oral drug, so having an oral regimen for patients would be a dramatic improvement over what they have now, and our drug is shown to be as efficacious. In fact, I was just with the professor at Johns Hopkins who ran the animal studies, and he was like, I couldn't believe it when I saw the data that your drug is as good as imipenem, which is the IV gold standard drug. So patients have to endure IV antibiotics for months and even years, if you can imagine, which is just, they really are in a bad spot. So it would be a big improvement for them.
Moritz, Host
Great. Let's move on to the preclinical side of things and talk oncology for a bit. So you're also developing a pan-mutant PI3K inhibitor. Can you just briefly highlight some of the differences versus some of the more advanced assets in this space?
Eric Giesem, CEO
Yeah, so we've been working on PI3K, so we bring something unique, which is more on chemistry, I think, to the table. So there's different binding and so forth. So what we have noticed, this was a space that we acquired a company that had an ongoing program, and we were really trying to understand where do we fit. So how we look at this is, you know, there's some really good compounds coming through. So like Relay, which you guys are very bullish, it's a good compound and really improves. It really works on, it has a separation over wild type for kinase mutants. And, you know, it has a 10x or so separation. But against Helicase, which makes up probably half of the mutants, There's almost no separation over wild type, which is where you get hyperglycemia. So to us, that's kind of the second generation. And the third generation are more like synovation, which Novartis just acquired for $2 billion, which are pan mutant. And so they have, you know, 50x against kinase and separation over wild type and maybe, you know, 10 to 20x over Thank you. A lot of the IP around all these programs are all over top of each other, so I think we have a real potential here to have a unique profile that's got great IP and freedom to operate.
Moritz, Host
So what is your timeline for this program, and at what stage of development are you looking for a strategic partner?
Eric Giesem, CEO
Yeah, so we're still in research. We're really trying to find a great compound because we want to make sure we feel good about it going forward. So I think, you know, we're going to need some time into next year to start developing. Our understanding from potential partners is that they don't necessarily need to have clinical data. You know, preclinical data would be, there's a lot of validation in the models here that show, you know, if you have this kind of separation and avoid all the toxicities that are known, you can show that in animal models.
Moritz, Host
So, yeah, great. Yeah, I think that's it from our end. So thank you so much for taking the time. Thank you very much.