Excellent. All righty.
Over here?
That's perfect. All righty. Good afternoon, everyone. I'm David Risinger. Very pleased to welcome Tectonic's leadership team on stage here. We have Alise Reicin, who's the company's CEO. It's very much my pleasure to welcome you, so thanks for being here.
Thank you for hosting us.
Of course. I thought it would be great for you to just start off with some high-level comments, and then we'll take it from there.
You know, when we started the company, our goal was to actually develop a pipeline of agonists and antagonists against GPCRs that sort of give us flexibility and the ability to make good database decisions, and I think we are just there now. We've now got two programs in the clinic in three different indications, which I think gives us the type of flexibility that we wanna have, and we're excited about both programs. Our first one is TX45. I'm sure we'll spend a lot of time talking about that. It's our long-acting relaxin that we are in the clinic in two clinical studies now, one Group 2 pulmonary hypertension, and we expect the readout from that Phase II, our APEX study, late 2026 or early 2027.
We just opened our first site for our PH-ILD study, Pulmonary Hypertension Associated with Interstitial Lung Disease. Just starting screening there. Our TX2100 program, which is an APJ antagonist, it's a nanobody fused to an Fc, and we are studying that for Hereditary Hemorrhagic Telangiectasia. Are you not hearing me?
I think your mic fell. It's anywhere.
Hereditary Hemorrhagic Telangiectasia. Perfect. We're in Phase I normal healthy volunteer study. We should have data from that later this year, then hopefully starting in parallel a Phase Ia/Ib and Phase II study.
Excellent. Great. Maybe we could start with TX-45 and just looking ahead to those Phase II APEX top-line results that you mentioned are expected in late 2026 or early 2027. Investor expectations are quite low in the wake of Lilly's and AstraZeneca's negative news on their injectable relaxin candidates. Could you just comment at a high level and then let us know what you'll be focused on when we get some data revealed from AstraZeneca? We don't know when they'll share details, but would love to get your thoughts.
Well, let's start with the Lilly failure, which was, I think, driven by two things. First of all, what have they found there? They saw extra congestion and an increased rate of hospitalizations for heart failure. You don't want that in heart failure patients. I think it was driven by two things, the patient population that they chose and maybe also very high doses. I think both of those could have played a role. Patient population, they went into preserved ejection fraction heart failure patients, HFpEF, who had to have been discharged from the hospital within the last two weeks for an acute exacerbation of their heart failure. These are patients who are likely fluid overloaded at randomization 'cause they're not completely dried out after leaving the hospital, and you get treated with IV diuretics for acute heart failure, which actually makes you more sodium retentive.
These were patients who you would not want to give a drug that could increase sodium retention and relax them like other vasodilators. If you relax the vessels around the glomerulus in the kidney, the glomerulus will see decreased pressures and will wanna hold on to sodium thinking that the individual is actually dehydrated. Now, you might think, well, that, then how can that drug work in heart failure? There are other drugs like hydralazine that also do that, but they're actually approved for the treatment of heart failure. I think it ends up being the degree to which you get the fluid retention and whether the positive aspects of the drug outweigh that. I think they took the most sensitive patients who are likely to retain even more fluid because of their sodium-avid state, and then they gave them very high doses.
Our hypothesis is with lower doses and in a patient population that is stable, euvolemic at baseline, on stable meds, that the benefits will outweigh the risks. We may get a little bit of fluid retention. I think we will, but not to the degree that they saw in Lilly. I think the AZ data, they haven't shown any data, but they didn't highlight that they had excess hospitalization. Unlike the Lilly data, we know that study went to completion, and they had an interim analysis where they looked at right heart cath, and didn't stop the study for futility. For all those reasons, that and also our own data that we have in IDMC looking at the data, they know that because of the Lilly data, we wanted them to focus on fluid retention.
I think that what we're seeing feels different than what the Lilly study saw. Now, AZ their slide said they discontinued the sub-Q program. Interestingly, Sharon Barr, their head of medicine, said the study was paused. There seemed to be excitement about their ongoing relaxin, oral relaxin. They're studying a cohort with straight HFpEF and a cohort with straight HFrEF. They gave us no other information about the sub-Q program, and so it's very hard to come to any conclusions. Was it drug? Was it dose? Was it patient population? It's a very heterogeneous study. HFpEF and HFrEF, IpcPH, meaning low pulmonary vascular resistance, and CpcPH, high pulmonary vascular resistance. They had about 65 patients per treatment arm. Our study's gonna be about 60 per treatment arm.
It's just unclear how many patients in their cohort match the patients with Missy in our study. I think until we get more data, it's hard to.
Sure.
It's hard to interpret it.
Makes sense. Just remind us about the timing of the oral readout for them and-
Well-
It sounds like it's gonna be the same, sort of disclosure issue for them because they have different patient groups. Irrespective of-
They've got two cohorts.
They'll have to disclose.
I think they.
Okay.
I can't say.
Sure.
Eventually, they're gonna need to disclose it.
Eventually, yes.
The study's done. It ended on February 10th. They should have data soon. We think they must have had data in the early fall, and yet it took them until February to put anything out. Their slides say they'll have data second half. The head of their steering committee sort of sent out a little teaser on LinkedIn, where he said, "Oh, we just published the study design, and we're so excited, and stay tuned for data soon.
Okay. Very good.
I don't know if soon means six months or, you know, six weeks. So.
Excellent. That's great.
We do know for the subcu study, they have to put results on ClinicalTrials.gov in July.
That's right. Okay, excellent. Maybe you could just now turn to the differentiation of TX45 versus those two candidates. You know, just so that everybody is sort of level set on how you design TX45.
Okay. It's a relaxin that's combined with an Fc, and that's to increase the half-life. We've also spent a good year re-engineering it to decrease the isoelectric point. Relaxin itself. Did I lose it again? Relaxin itself is a high isoelectric point protein, and what can happen to high isoelectric point proteins, they're highly positively charged, and they get bound to negatively charged hyaluronic acids that line the glycocalyx of the blood vessels. In the alpha phase, you see. Like within the first 24 hours, you lose a lot of PK. By increasing. You see that in, if you look at the PK profile, for instance, of the AstraZeneca compound, which is also a relaxin attached to an Fc.
What we found was in those high isoelectric point relaxin Fcs, there was a disconnect between the in vitro and the in vivo potency. They looked much more potent in vitro than they did in vivo. By decreasing our isoelectric point, number one, the PK looked much better. I think it's why we have the longest half-life, and we could do every four-week dosing.
Mm-hmm.
In addition, our in vitro and our in vivo efficacy are just spot on. When I talk about in vivo efficacy, I'm talking about renal blood flow. Both rat and humans, everything correlates beautifully. The AZ compound, as I said, is also a relaxin bound to the Fc. It's got a higher isoelectric point, PK not quite as good, which is why they had to do their dosing is every two weeks. You know, in vitro, they look more potent than us. If you look in rat renal blood flow, they look less potent than us. It's really a little bit hard for us to say exactly how we compare to them in that regard. Then Lilly has a relaxin bound to a nanobody-albumin.
I've heard some KOLs question whether that could somehow lead to a greater increase in, fluid retention. I don't know how to put those two together, so I don't put a lot of credence in that. That is a difference. We look pretty similar to, with the way they do on renal blood flow.
Okay. Excellent. Remind us about the dosing in APEX.
Lilly was weekly dosing. If getting that shot and getting high peak levels, once weekly is important, that may have played a role. AZ is every two-week dosing, and we're studying every two-week and every four-week dosing. Our base case all along has been that our every four-week dosing is likely to be our dose.
Remind us about the number of milligrams that you're
300 mg Q2 and 300 mg Q4.
Got it.
At trough with the 300 Q2, the exposure is about 3-4x what we get at the 300 Q4.
Got it. Okay. Thank you. Sorry. With respect to the rationale for enriching for CpcPH, could you talk about that? Because we haven't seen that specifically, you know, being done by Lilly or AstraZeneca to date.
Lilly went into a straight HFpEF population. They didn't even look for pulmonary hypertension. You're right, AZ did not enrich for CpcPH. Our primary endpoint of PVR reduction is in our PVR greater than 3 Wood unit subpopulation, which is about 70%. Lilly's PVR endpoint was in the overall population, and in the IpcPH population with a normal PVR, our Phase Ib data suggests you won't decrease the PVR because you get a floor effect. If the majority of their patients had IpcPH, it could have led to them missing on their primary endpoint. Why did we do that? Here's the hypothesis. Patients with heart failure, why do they have a decrease in exercise tolerance?
Well, when they exercise, they need to bring more blood into the left ventricle, and when they do in HFpEF, it's a stiff ventricle that doesn't relax during diastole, and as more blood goes in, pressures on the left side of the heart go up, your wedge pressure goes up, and fluid goes into your lungs. In PAH, why do you have a decrease in exercise tolerance? In PAH, you have a high pulmonary vascular resistance. You've got a small tube to get all that blood through, and as you need to get more blood into the left side of the heart, you actually can't get enough blood in. We think in CpcPH, both are playing a role, both the high PVR and the left heart failure.
If you can decrease the pulmonary vascular resistance and send more blood into the left side of the heart, while at the same time enabling the left side of the heart to accommodate that increase in blood, we think you could get almost a further increase in the exercise tolerance compared to just straight HFpEF. We think there'd be efficacy in straight HFpEF alone. We just think this could be additional efficacy. Think of it as the more severe patients. Very often if you're trying to show a signal in a disease, it's easier to show a signal in the more severe patients. We'll have to see. We have some IpcPH patients in our study, and we'll be very curious to see the data from AZ as well to see if the efficacy looks just as good there.
Not on PVR, 'cause you're not gonna see something there. What do you see on wedge? What do you see on six-minute walk in those patients?
Got it. Very helpful. Could you talk a little bit about what you would view as success in this upcoming set of results from APEX?
If we see a 15%-20% reduction in PVR or pulmonary vascular resistance, I wanna see some other hemodynamic measures that suggest the left heart is working. Does cardiac output go up? Does wedge go down? Then we wanna see a numeric increase in six-minute walk. If I see anywhere above 15 meters, I'll be happy, and those likely won't be statistically significant.
Excellent. Okay, great. Well, actually, before we go to Group 3 PH-ILD, one more question. How should we think about the effect and the time course over the 24 weeks of the trial?
We think you'll get the vasodilatory effects and the lusitropic effects, its ability to relax the left ventricle within the first day or two, first few hours even. Then the antifibrotic effects take time. We think six months is sufficient. We know sotatercept really is all about the antifibrotic effects. They start to see some improvement by 12 weeks, and it looks like it continues to get better out to week 26.
Excellent. All right. Turning to PH-ILD, can you outline the mechanistic rationale, for, relaxin in the context of historical, you know, vasodilator results being disappointing in PH-ILD?
First of all, mechanism you want to know.
Yes.
All right. It's about bringing down the pulmonary vascular resistance, right? That's what you need to do. You don't have to improve their heart function. Their heart is normal, at least the left side of their heart is normal. It's all about bringing down the pulmonary artery pressure and the pulmonary vascular resistance. Our PH HFpEF data was very impressive in that regard. We said, "Okay, so are there other forms of pulmonary hypertension we could work in? PH-ILD, high unmet need." Then the antifibrotic effects, we thought potentially I wouldn't develop relaxin as a straight IPF drug, but if we had some antifibrotic effects in the lung, that could also be beneficial. Our preclinical data suggests that you could as well.
Got it.
Now, you asked why did the other drugs fail.
Yes.
treprostinil is the one drug that has been successful as an inhaled agent. Failed systemically. Why? It resulted in what's called V/Q mismatch. It's ventilation, that's the V, and the Q is perfusion. It's a very, very strong pulmonary vasodilator. Why did it fail? The feeling is that it dilates parts of the lung that aren't getting perfused, resulting in the V/Q mismatch. Then it became sort of a generally held sentiment that all systemic vasodilators do that. If you actually go back and look at the data, that's just not true. In fact, there was a study doing head-to-head treprostinil, which results in V/Q mismatch, versus a PDE5 inhibitor, I think it was done with sildenafil, in PH-ILD. The sildenafil didn't, which we know is a systemic vasodilator, didn't result in V/Q mismatch.
Oh.
In fact, if you look at some of the early studies done with sildenafil, it suggested that there could be efficacy. It's just when they went to the larger studies, they went into a straight ILD population, not necessarily a PH-ILD population. You know, guanylate cyclase also failed in PH-ILD. It brought down PVR. It didn't result in V/Q mismatch. That they didn't report. But what you looked at there was cardiac output drove the decrease in PVR, not decreases in pulmonary pressure.
Mm-hmm.
We'll be looking to see make sure that we're seeing decreases in the pulmonary pressure as well.
Excellent. If you could just talk about, you know, development there and timelines ahead.
We just initiated that study, and it's a four-month study with 20-25 patients. PVR reduction in PVR will be the primary endpoint, and we haven't really given guidance on that study. I sort of wanna see how recruitment's going. You know, I wouldn't be surprised if we've got some data in the 2027 timeframe, but that's not official guidance.
Excellent. Okay, perfect. Let's turn to the latest big news on TX2100. Congrats on the launch of that program. You obviously recently hosted a detailed webcast, but if you could just, you know, summarize. You know, I know it's a lot to cover, but if you could just summarize the key highlights, that would be great.
Yeah. It is a selective and specific anti-angiogenic agent, and we know that HHT is a disease of upregulated angiogenesis, sort of angiogenesis gone awry. We also know that it's been proven that anti-angiogenic agents work in this disease. There's data on VEGF antibodies, pomalidomide, and pazopanib. They've worked in preclinical models of the disease, and they've translated to the clinic. The issue with them is long-term tox. These were oncology drugs that were repurposed. Durability is an issue. They're not approved, so insurance is an issue, and no one really knows how to dose. Everybody chooses their own thing. Anti-angiogenesis has been proven, and now we can improve on it. Why do I think we can improve on it? Well, it's more.
APJ is mainly expressed in the endothelium, so it's not expressed all over the body. When we antagonize it, well maybe we will be more specifically antagonizing the cells that have upregulation of it. In addition, it's typically quiescent, and it's really upregulated in the disease state. VEGF, on the other hand, is much more widely expressed. It's very important in homeostatic mechanisms, both in the vessel maintenance as well as in the kidney and other places, and that's what leads to the toxicity. Preclinically, we've got very strong data in two separate preclinical models. We know where the genetic disease, you know, mutations are. They decrease the ALK1 pathway. You can decrease the ALK1 pathway in the animal models. Phenotypically, they reproduce what you see.
Having very strong data preclinically, where we showed you get a decrease in AV malformations, improvement of hemoglobin, and hematocrit, and a decrease in bleeding that looked better than the VEGFs and AKT, it gives us confidence going forward. We did 13-week non-human primate studies. We went up to 100 mg per kg, gives us about a, we estimate about a 70-fold margin, and we didn't have any safety signal.
Excellent. Can you talk a little bit more about those two animal models and?
Yeah
You know, what species and additionally.
These are rodent models and mouse models. The most widely used, and what you'll see the most, are neonatal models, where you give them an antibody to BMP9 and BMP10 and that inactivates the pathway. They're sick little pups that have chronic blood loss, and they don't live long. In that model, AV malformation was reduced similarly with anti-VEGF and an AKT inhibitor, but we looked much better on bleeding and hemoglobin restoration. Then the second one, which very few companies use, I think Hanny Al-Samkari, who's one of the big KOLs, said, "Oh, wow. You guys were willing to go into this model? This is a really, really sick model. Nobody else has tried this." It's a ALK1-inducible adult model.
The reason why we like that is the vasculature is an adult vasculature, right? Neonatal, maybe you'd say it's not quite similar. In that model again, we looked better than the VEGF. On day seven, hemoglobin looked similar between the VEGF and ours. By day 12, the VEGF didn't look as good. Again, that may be a hint of the durability issue you can sometimes see in patients.
That's great. Excellent. Maybe you could just talk about the mechanism relative to competing therapies in development.
you have VEGF. Downstream of VEGF is AKT and ERK. We think both are important in angiogenesis. If you inhibit AKT, you inhibit AKT in every cell in the body, and you are going to get the toxicity that's associated with that. There are a whole bunch of companies that are working on AKT inhibitors. VidarIs is one. They had a The New England Journal of Medicine article with their AKT inhibitor. There definitely was some efficacy, and there were also some safety issues which limited their dosing, I think. I don't think they were getting the full efficacy that you get from that. They only got to up to about an IC50. We think if you go above an IC50, you'll get even better efficacy, and they had hyperglycemia in some patients, rash and GI toxicity. Now, there's a whole
There are two other companies that are working on AKT1 selective agents. I think that should help with the hyperglycemia. I'm not sure that that's gonna get rid of the skin rash, or the GI tox, and you're not inhibiting ERK, and so that might also limit the efficacy a little. The other really elegant approach is Diagonal's approach, where they've got a bispecific antibody that results in a clustering of the receptor and reactivation of the receptor. They've got some nice preclinical data with efficacy in that in the animal models, not the ALK1 knockout. They wouldn't have worked in that because it's a double mutation of the ALK1 and their drug wouldn't work. I think the major issue with Diagonal is potentially safety. You're activating it. Do you overactivate it?
There is a theoretical issue about mosaicism. There's a literature that patients with HHT can get a second hit mutation in some of their blood vessels, and if you had that, not clear that that mechanism in those blood vessels would work as well, so could that somehow in some patients impact efficacy? Alnylam has a siRNA to plasminogen that doesn't address the disease. The disease is a disease of angiogenesis. What it does is stabilize the clot, essentially. Tranexamic acid is used in mild patients. The efficacy isn't great. The question is, with an siRNA, do you get better efficacy? It's also tranexamic, it's the same thing. It stabilizes the clot. I think we'll see. It's a different approach, but it's not disease modifying.
Got it. That's very helpful context. Thank you. We're almost out of time, but would love to have you wrap up and just comment on the platform because 100 was discovered and developed using your proprietary GEODe platform and I think you know, that's obviously relevant and you'll be advancing other pipeline candidates down the line from that platform.
We're pretty excited about the platform, and we've got a lot that's under the hood in the discovery space. Hopefully at some point we'll be able to talk about that. The platform, you know, historically, developing biologics against GPCRs has been quite difficult. They're cell membrane proteins. They're seven loops that go in and out of the cell membrane, and when you try and purify them for drug discovery, they fall apart, in which case you can't really do drug discovery. One of the key elements is our ability to purify the receptors while keeping their natural form so that you can then make antibodies. The other thing is making antibodies using mouse models, which is so common, doesn't really work for GPCRs because they're very well conserved, and so the mouse doesn't see them as being non-self.
We have proprietary yeast display libraries that we've sort of optimized for GPCRs. We look at both nanobodies and antibodies. Sometimes the structure of a nanobody that can go deep in is preferable to an antibody. Then we have a whole bunch of engineering techniques that sort of optimize. I think what you can expect is the rest of our pipeline or much of the rest of our pipeline are bispecifics.
Got it. Wonderful. Well, like I mentioned, we are over time. This has been great. Thanks so much for.
Thank you.
being here with us today.
My pleasure.