All right. Welcome, everyone. Thanks for joining us. Next up, we have Jeff Finer who's the Co-founder and CEO of Septerna. Jeff, I'm going to turn it over to you, go through some slides, and then we'll get into it.
OK, great. Thanks, Gavin. I appreciate the introduction. My opening remarks will include some forward-looking statements, so take that into consideration. For those of you learning about Septerna for the first time, we're a company focused entirely on GPCRs, or G protein -coupled receptors, and we've discovered a new way to do drug discovery that we call our Native Complex Platform. That platform has quickly led to a portfolio that we'll spend the majority of the time talking about today with validated targets. Each has an early clinical readout, each represents a significant market opportunity, and we're well capitalized with cash runway into at least 2029. Our programs are outlined here. SEP-479 is our second-generation PTH1 receptor agonist for hyperparathyroidism. I'm sure we'll spend a lot of time talking about that. SEP-631 is an MRGPRX2 negative allosteric modulator for mast cell-driven diseases.
This one's in phase I, and we anticipate being able to share data in the first half of next year. Earlier stage program includes the thyroid-stimulating hormone receptor, TSHR, where we again have a negative allosteric modulator, or NAM, focused on Graves' disease and thyroid eye disease. Beyond that, we've got an incretin receptor agonist program where we did something quite unique here. We discovered a novel binding pocket that allows us to activate multiple incretin receptors simultaneously, and this became part of a collaboration that we announced in May with Novo Nordisk. A quick, I want to share a little bit of data just to kind of kick things off. SEP-479 is, again, for our hyperparathyroidism candidate. PTH receptor has been a historically challenging target.
Many companies have tried for many years to find small molecules and more or less struck out, but we've actually cracked this receptor a couple of times. We found two unique binding pockets that each allow us to activate the receptor with an oral small molecule. Used our platform to optimize those very quickly to drug-like candidates. Our first drug candidate was a compound called SEP-786, which unfortunately ran into an issue in phase I. We had to discontinue that trial back in February. Importantly, this compound showed us initially that a small molecule can mimic a peptide and have the activity that we would hope to have in a healthy volunteer. We ran into a case of unconjugated bilirubin increases, and then post-discontinuation, we discovered the mechanism for what that was. It's a potent inhibitor of an enzyme called UGT1A1, which is a bilirubin conjugation enzyme.
Fortunately, behind this 786, we had a follow-on program that we were quite excited about where we were aiming for a potentially longer half-life originally, and at the same time, we were going after the other binding pocket. So 479 is completely structurally unrelated to 786, has no evidence of UGT1A1 inhibition, but importantly, has significantly improved pharmaceutical properties. Here's some data looking at 479 in a rat surgical model of hyperparathyroidism. Here we surgically removed the parathyroid glands. These animals develop all the hallmarks of hyperparathyroidism with low serum calcium levels and high phosphate levels. What I'm showing you here is that in a 28-day study where we're looking at detailed time courses on four of the 28 days, we're able to normalize both calcium and phosphate. Importantly, the dose range here is actually significantly lower than comparable data for 786 that we showed earlier.
This is active at 0.15 mg per kg once a day, whereas 786 required a much higher dose. In terms of the PK properties for 479, it's looking like we're in a range with a projected human half-life in the 40-80 hour range, which we think could end up being quite good if we're able to do that and support once-daily oral dosing. Here's the data I really wanted to share, which is a monkey PK/PD study. This is a study where we're trying to simulate what a healthy volunteer study should look like. Importantly, healthy volunteers are quite different than hyperparathyroidism patients in that they actually have an intact ability to dial up and down their endogenous PTH levels. We're seeing exactly this effect in a monkey. This is a seven-day, once-a-day dosing study followed by a five-day recovery period.
If you look at the right graph, which shows endogenous PTH levels, after just a single dose, we're able to take endogenous PTH levels down about 80%. Once that's bottomed out, we can start to see increases in serum calcium. We'll talk a little bit about the levels that we're looking for in healthy volunteers, but this monkey study simulates this quite well. We're looking for increases in calcium of about 1 mg per dL, and that is based on the historical precedent of peptide therapies that were in clinical development. If you look back at the starting doses of what they looked like in phase I, they looked like they increased calcium about 0.5 to 0.7 mg per dL, and then many of these patients you have to titrate up.
We wanted to be able to ideally bracket that 1 mg per dL range in a healthy volunteer. 479 is moving along quite nicely. Wrapping up IND enabling studies, we completed studies in rats and dogs. Well tolerated. The only effect was hypercalcemia, which was completely a non-target expected effect. We decided to go the extra mile and do an extra study in cynomolgus monkeys, and that is in progress. Assuming we get through that, this compound will be entering the clinic in the first half of next year. A second program to just introduce you briefly to is our MRGPRX2 program. This is a negative allosteric modulator for mast cell diseases. In terms of our profile, we set out to build a great compound with high potency. We're in the single-digit nanomolar range to high picomolar based on the assays.
Broad inhibition, we inhibit every MRGPRX2 agonist that we throw at the receptor. We've got a very long residence time and a mechanism that we call an insurmountable negative allosteric modulator. What that means is when the compound is bound to the target, it can't be outcompeted by excess amounts of endogenous agonists. We're quite confident in a once-a-day profile, and we've got some data I'll show you here on preclinical PK and PD. This is an animal model that we think is quite relevant to urticaria. It's a knock-in mouse where we knocked in the human gene to the mouse. What we do in this study is we dose the animal with our drug, then we administer a blue dye. The blue dye tints the blood slightly blue, and then we do an intradermal skin challenge.
At the side of that intradermal skin challenge, you get extravasation of the blue dye. Ideally, if we're able to fully turn off this receptor, which we believe we can do, we will get no extravasation. That's exactly what we're seeing here. This is fold over vehicle, so one, the dashed line there is basically the zero extravasation line, and we're able to completely knock this down. The best human translational model is primary human skin mast cells. Again, when we get those cells, treat them with Substance P, which is another MRGPRX2 agonist, then look at tryptase release. We're able to knock that down as well. This program is in phase I currently. It's a standard SAD-MAD design. Once-a-day dosing. The MAD dosing is for a period of 10 days.
For each of the patients in the MAD portion of the trial, we're doing an intradermal skin challenge with Icatibant. Icatibant is an approved drug that causes a known skin reaction, and that skin reaction mechanism is clearly through X2 stimulation. That skin challenge is being done pre-dose for all the patients in the MAD, and then again post-dose as well on day nine of the 10-day dosing period. This study is well underway, recruiting well, and we're hoping to be able to share the data for this study in the first half of next year as well. With that, I think we can transition to Q&A.
All right, perfect. I'm just going to go by order of programs in the clinic, or I should say status in the clinic.
Absolutely.
It's actually starting off on the X2 side. For the phase I study, is the Icatibant challenge only for the MAD portion, or are you doing it in the SAD also?
Now, we're only doing it in the MAD portion of the trial, but we're doing it with every subject in the MAD portion of the trial.
OK.
We're going to have a lot of data there.
Perfect. What dose of Icatibant are you using?
We're using 10 mcg per mL as well as 100 mcg per mL. These were the same two doses that Evommune used, and we decided that why not use the same doses so that we can cross-compare.
I guess from the Icatibant challenge, can you test out the negative allosteric modulator thesis? Because I guess presumably you want to see no dose response between a 10 or 100 Icatibant challenge. Would that be the ideal outcome there?
Yeah, I mean, the ideal outcome is that we have a dose that we can actually completely squash the activity. With what we'll be looking at, maybe I'll just go back to the slide there for a moment here. One second. There we go. Oops. What we're looking at is both the low and the high dose. We've got a negative control, which is saline, and a positive control, which is histamine. What we're going to be looking at is sort of the difference between the baseline Icatibant challenge and saline, and trying to see what percentage of the way can we go from on an individual basis from that baseline all the way down to saline. We think that the more we can actually kind of suppress that effect, the better. We'll be looking for that at both doses and again, seeing how much inhibition we get.
Yeah. I guess what I'm trying to think through, and I don't think we really have any data on this, so it's kind of speculation at this point, but with the NAM, and you're increasing the concentrations of the ligand, you presumably, once you hit the right efficacious dose, presumably won't need a higher dose at like the 100 Icatibant challenge. Is that what you're looking to see?
Yeah, I mean, that's exactly right. The insurmountable negative allosteric modulator means that once the, and again, we have a very slow off-rate, once we've got the compound on the receptor, it can't be outcompeted by excess amounts of endogenous ligand. Exactly to your point, once we hit a dose, we should be able to theoretically completely prevent it.
Yeah, makes sense. Is next steps for development after this Icatibant challenge study? The other precedents in the X2 space have done a relatively smaller open-label SIENDO study. Are you planning to do that or go kind of right into a controlled study?
Yeah, we think the more quickly we're able to get straight into CSU, the better in a controlled study.
Yeah, that makes sense. I guess what other external data points are you looking at over the next year or so to kind of tease out if there's additional efficacy left on the table by other X2 inhibitors?
Yeah, obviously we're going to be following the Evommune space as you are as well. Really looking forward to their data in the middle of the year next year on both CSU and atopic dermatitis. I think we're hoping they're quite efficacious there, and we'll see how ours plays out in terms of just how it may stack up.
Awesome. All right, let's switch. Let's go over to the PTH1R side of things. On the PK/PD profile, you noted a 40-80 hour half-life. Are you also able to measure receptor occupancy, if that's the right word, to just kind of see what the off-rate looks like? Because theoretically, the PK/PD profile could be different than the half-life implies if you stay bound to the receptor for a longer period.
Yeah, yeah, it's a great question. We've got a lot of molecular pharmacologists at our company, and we think a lot about receptor occupancy. It's a question that we think about a lot for each of our programs. Just to compare this to the 631 program as an example, with an antagonist, you want to actually completely, you want to have the highest receptor occupancy as possible. With an agonist, surprisingly, and a lot of people don't realize this, you only need actually very low receptor occupancy to have an effect. In fact, with some agonists, a low receptor occupancy, even single-digit percentages, can be quite effective. We've done some calculations based on peptides, including the PTH peptides. Others have done these studies as well. To be effective, a PTH peptide only needs to be on 2% or 3% of the receptors, surprisingly.
I mean, this is just, again, a fundamental difference between agonist and antagonist, where what we're aiming for is just completely different between the two. We have done some preclinical studies with 479 to cross-compare it to the peptides, and we believe we're getting very similar receptor occupancy to the peptides.
Do you have like an upper ceiling on the receptor occupancy for what you're aiming for? I guess at what point do you get 80% or 90% of the effect?
Yeah, I think you end up with most of the effect that we would need to get into this 1 mg per dL range with only a couple percent.
Yeah.
Again, just as a reminder, this is a condition where every patient needs to be titrated to the right dose. It is really an empiric thing. Each individual patient is going to be titrated to the right dose of the drug. We think that is going to end up in a relatively low receptor occupancy at that level.
That makes sense. I guess 1 mg per dL we noted before. Why not higher? I guess that kind of opens the conversation into translating healthy volunteer into patient and how you think about reading across the pharmacodynamic measures.
Yeah, so again, that number comes from looking at the peptides that have been in development. If you look at all the peptides that have been in development to date and look at what their eventual starting doses were in hyperparathyroidism patients, and then extrapolate back to what they showed in healthy volunteers, it was always below 1 mg per dL. So it was around the typical range was about 0.5 to 0.7. The total dynamic range isn't that high either. If you look at the peptides in terms of where they ended up in terms of treating hyperparathyroidism patients, it's only factor two or three at the most. If you go up two or three times in dose with a peptide, you really only get up to maybe about 1.2 or 1.5 mg per dL.
In a healthy volunteer patient study, and as you can see the healthy monkey data here, we can go up higher than that. It's just that in a healthy volunteer, it's not ethical to push somebody's calcium level that high in reality. There is another nuance here, which is that in hyperparathyroidism patients, we can probably dose higher than in healthy volunteers just because the hyperparathyroidism patients actually start out naturally at a lower level. This is something we're going to have to work through as we go because you want to be able to demonstrate that all the drug levels that we will have eventually in hyperparathyroidism patients are at least safe in healthy volunteers.
I guess from the first-gen compound, what did you learn on the whole PK/PD relationship? What can you learn for the 479 program? Are you planning to put out the first-gen data at any point?
I think the main learning there, again, was that a small molecule can do what a peptide does. We learned that as expected, as we're seeing in this monkey study, the first thing you see is endogenous PTH going down. So we saw that going down. We went down low as in this study, started to see calcium increases, but we weren't quite at the level that we thought would support a starting dose and then have some headroom above that before the hyperbilirubinemia effect kicked in.
Yeah. All right, that makes sense. Let's touch on TSHR, actually. Where are you in the DC selection process? Are you optimizing more for potency? Is it about tuning the selectivity? What's the key focus right now?
Yeah, TSHR is a very tricky development path. Obviously, a number of companies have tried and had trouble there. On the selectivity side, there are a couple of very closely related receptors, FSH receptor and LH receptor. We want to have selectivity there. We've been able to get very selective compounds, so that's less of an issue. One of the tricky things that a lot of companies have run into here is optimizing both potency and pharmaceutical properties at the same time. Generally, the more hidden pockets on the receptor are ones where as you optimize potency, you tend to make pharmaceutical properties worse. We've had to discover different binding pockets, and we think we're in a binding pocket now that's going to allow us to kind of thread that needle and hit kind of a sweet spot between potency and pharmaceutical properties.
We think at this point we've got line of sight to a development candidate. We're not at the point where we want to give any guidance on timing, but I think there's a good chance this becomes part of our story in 2026.
Awesome. I guess even beyond the initial healthy volunteer study where you can tease out some of the pharmacodynamic measures, how do you think about developing this longer term? We kind of go Graves, thyroid eye, do a TED prevention, a few different options here. What are you guys thinking about?
Yeah, we think the place to start is Graves. There's a lot of Graves patients out there that are not well treated on antithyroid medications, significant unmet need there. The mechanism really should treat, as you said, it should be good for Graves, it should be good for TED. We do think the home run potentially is treating Graves patients that haven't developed any thyroid eye disease symptoms yet or even visible manifestations. If we can prevent that progression, I think that's going to be an important part for us. The idea is start with Graves and then add on from there.
Awesome. We're actually right at time, but I think that was really all the key topics. So really appreciate you joining, Jeff, and hopefully talk next year.
Yeah, thanks again for the invite.