All right. Good morning, everyone, and thanks for joining us at the Morgan Stanley Global Health Care Conference. I'm Mike Goldsmith, one of the biotech analysts here, and it's my pleasure to introduce Tien Lee , CEO, and Tim Kiefer, CSO of Aardvark Therapeutics. Before we get started, I just need to read a quick disclaimer. For important disclosures, please see the Morgan Stanley Research Disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. With that, Tien, thanks for joining us, and Tim, you as well. Maybe to start, you can just provide some sort of introductory comments for people that might not be familiar with your story.
All right. Great. Yeah, thank you very much. I really appreciate the opportunity to be here. Aardvark is a San Diego-based biotech company. We're developing small molecule oral drugs to aggregate hunger. We have a lead program in phase III development for Prader-Willi syndrome.
Great. Maybe we'll start with Prader-Willi, and for people who aren't familiar with that disease, maybe talk a little bit about what those patients have to go through and what some of the current treatment options are.
Sure, sure. Yeah. Prader-Willi syndrome is a chromosome 15 abnormality related to a lack of paternal chromosome 15 segments. It's spontaneously occurring in about 1 out of 15,000 live births. There are about 10,000 to 20,000 patients in the U.S. and about 300,000 to 400,000 worldwide. Patients with Prader-Willi syndrome suffer from a panel of different conditions. The most well-known quality of life impairment is the unabating, unrelenting hunger that most of them experience, starting from the ages of four to six, and then it perpetuates for the rest of their life. It causes a lot of challenges for the family. Patients' families have to lock away food and so forth. There are cases where patients with unrestricted food access have eaten to the point of stomach rupture. It's a very extreme hunger condition.
Additionally, these patients also suffer from really extreme anxiety, oftentimes also systemic inflammation, and very severe constipation often.
Got you. Maybe let's move to ARD-101. That's your treatment for Prader-Willi syndrome. You mentioned in phase III, and maybe you can talk about the mechanism of action and how that's unique.
Okay, Tim?
Yeah, sure. ARD-101 activates the taste receptors that line the gastrointestinal tract. They're very concentrated on enteroendocrine cells. These are cells that secrete regulatory proteins, ones that you're probably very familiar with, like GIP and GLP-1. This drug, ARD-101, goes into the lumen of the gut, activates those taste receptors on those enteroendocrine cells, stimulates the release of these hormones, and these convey messages to the brain, feelings of fullness, dampening hunger, and ultimately that leads to a reduction in body weight and many other complementary effects, such as improving glucose homeostasis.
Can you talk a little bit about just CCK and how that's involved in these patients? You know, why is that sort of the reason or one of the issues with developing this disease?
Sure. CCK is a gut peptide, one of these hormones released from enteroendocrine cells in response to meals. It does dampen hunger, and that's been shown clinically with injections, for example, of CCK or small molecule agonists of CCK. What differentiates kind of what we're doing is that it's more than just CCK. It's activating taste receptors on the cells that release CCK, but also the other enteroendocrine cells that activate these neural pathways to the brain. CCK will activate the vagus nerve that feeds to the brain, activates these satiety centers in the hypothalamus, and induces that feeling of fullness and reduction of hunger.
Maybe you can talk about the difference between the hunger and appetite and how those complement each other, maybe.
Sure.
I'll let Tien cover that one.
Sure. The way we look at it is there's a dichotomy between appetite and hunger. Appetite is a pleasure-seeking drive, a reward-based drive, and hunger is a pain avoidance drive. For instance, if you really like a delicious food, it's appetite that's primarily driving that need to eat or desire to eat, like if you like chocolate cake or ice cream, for instance. Hunger could be represented as when, let's say, all the food available is just very low quality, like dog food, for instance. You wouldn't want to eat it voluntarily. If given enough time with enough fasting, hunger would compel you to eat that, and that's what hunger represents. There's satiety for appetite and satiety for hunger, and they're both simultaneously regulated.
We believe that GLP-1 drugs and other drugs are good at aggregating appetite, but less evident for reduction in hunger, which can be represented by some biomarkers like ghrelin. We think our drug is particularly useful in aggregating hunger as a differentiated and complementary approach.
Makes sense. Maybe just talk about how validated this mechanism is. I know in the past there's been some CCK agonists that were tested, and what were some of the learnings from those studies?
Yeah. Following up with what Tien was just describing, CCK is very well validated to affect hunger and reduce meal sizes, but it doesn't have long-term effects. I think GSK reported at 24 weeks, there wasn't a large diminishment of body weight, and I think that's why they dropped the program. There's also concerns with peripheral administration. One thing I think we need to keep in mind is that delivery of these hormones is normally physiologically regulated from the gut, and the levels that are given with injectables are pharmacological and 24/7. They're aiming for once a week or even once a month delivery, which means it's steady state, whereas these are designed to be released during meals. They have other effects. CCK, for example, is released from those gut cells in response to fat.
It stimulates the gallbladder, stimulates the exocrine pancreas, and you don't want to be doing that chronically. I think some of the limitations that were borne out in the clinical trials reflected that. A differentiating factor here is physiological release. 99% of the drug stays in the gut, and it's just stimulating the release of those normal hormones that act synergistically to regulate appetite and hunger.
Makes sense. Maybe just talk about some of the tolerability that we're seeing with these CCK agonists.
That's, again, related to the elevated levels that are achieved with injectables. I think in rodent studies, there were effects on the exocrine pancreas, for example, because this is a normal thing it does, but when it's expressed 24/7, it's overstimulating the exocrine pancreas. That was observed in rodents, inflammation, pancreatitis, for example, and overstimulation of the gallbladder as well. I don't think in humans they saw effects on the exocrine pancreas, but it was still concerning given the preclinical observations.
Gotcha. Okay. Maybe we can move to some of your phase II data with ARD-101, and you shared some very promising results. Maybe just walk us through that early data and what made it so promising.
Sure. We had 18 patients that completed dosing with ARD-101 with Prader-Willi syndrome, and we were very encouraged by that. We had only a 28-day treatment course because at the time, the company only had GLP-Tox data for coverage for 28 days of dosing. Now we have long-term GLP-Tox, and we can dose theoretically in perpetuity. For that very short 28-day period, we saw very encouraging reductions. The primary endpoint, which is the primary endpoint in our phase III, is a HQCT-9 score or hyperphagia questionnaire of clinical trial. It's the validated endpoint that measures hunger in Prader-Willi subjects. We saw reductions of about 8 points on that scale, which was very, very encouraging. In addition, we have anecdotes of reduction of anxiety and inflammation as well in multiple subjects.
I think even though without the expectation of what they were going to be feeling, I think the anecdotes were very rapid onset of feelings of diminished hunger in the majority of subjects.
Can you talk a little bit more about the HQCT-9 score and just how it's conducted and then what's a meaningful change in that scale?
Sure. It's nine questions, and then you can have a score of zero to four, and the primary caretaker for the patient is the one that's supposed to be filling out that score. It's a subjective measure. As far as what a meaningful reduction is, I think there's an acceptance that even a one or two-point change, if it didn't have a concomitant trade-off in terms of safety risk and so forth, a lot of patient families and clinicians would say that would be clinically meaningful as well, too.
Yeah. Makes sense. You have a question there that's a little bit subjective. What are some of the challenges in running studies with that type of endpoint?
Of course. I think it's important to note that this is a limitation of subjective scales, right? There's an importance to try to normalize and have an understanding of what the scoring should represent. You know it's, for better or worse, the validated endpoint for it. We are actually looking for a number of secondary endpoints as well in our phase III study, including body composition and other scores to measure anxiety and also other biomarkers as well. I think over the course of time, we can have a good reliability, even accepting a very wide standard deviation that we think we can get statistic on the phase III.
Maybe talk a little bit about the dosing you tested, and then how you selected the go-forward dose for your phase III.
Sure, sure. Initially, for the first 12 subjects in our phase 2 study, we had a 200 mg b.i.d. Right now, phase III, this is just four tablets taken twice a day. Sorry, that's the dose it is now. When we actually observed the data, there were some patients that did a little better than others, and we thought that one of the contributing factors was that if patients took their drug at mealtime, they had not as good a response on average than ones that took it between meals. We thought there could be some food dilution effect at play. Our dose is very, very far away from the theoretic toxic limit of our drug. In the second part of our phase II study, we tested a dose escalation protocol, and we saw encouraging results that gave us confidence that 800 mg b.i.d. was an optimal dose to utilize.
That still represents a dose that's far below the toxic limit of our drug. We didn't see any new AEs that caused concern at that dosing. Our go-forward dose will be a target of 800 mg twice a day.
Could you talk about why the safety profile was so clean at that higher dose?
Sure. The drug has a very unique property. It's 99% gut-restricted, so less than 1% systemic availability. It's very paradoxical because most oftentimes you're not looking for a drug with low oral availability. The target of our drug is in the lumen of the gut. The gut restriction is not an impediment, but actually a feature that really augments the on-target efficiency of the drug and reduces the systemic exposure. That lack of systemic exposure also diminishes concern for renal, oral, cardiac toxins, and so forth. I think the physical restriction of the drug in the gut is very contributory to its unique safety profile.
All right. You've seen very promising activity, very good tolerability, and now you're advancing in your phase III HERO study in Prader-Willi. Maybe just talk a little bit about the trial design there and your recent decision to include younger patients as well.
Sure. The HERO trial is a one-to-one, so the ratio of one-to-one for placebo versus active. We're going to look at the HQCT-9 score at week 12 of treatment versus the baseline, and that will be our primary endpoint. As mentioned before, we have a number of other secondary endpoints that we have embedded within the trial. I'm sorry, what was the other part of your question?
Expanding into younger patients.
Yes, yes. In our phase II, we had tested patients 17 years and up. In our current phase III protocol, we have 13 years and up. We've recently announced intent to lower the age to lower ages for two reasons. We want to also make sure we have a robust label upon the completion of the study, and also since a lot of the patient need is for younger patients as well, and also historically across the competitive space. Per our expectation, younger patients are expected to have a more robust response to treatment than older patients in general because of more neuroplasticity and so forth. There are a couple of fold reasons for doing that, and that's our intent right now.
Makes sense. Maybe just talk about enrollment so far and how that's tracking versus your expectations.
Yeah. Enrollment's going very great, very strong. Just came from an investigators' meeting for European and Asian investigators. There's a lot of enthusiasm, and I have full confidence we can fully enroll and complete the study per our guidance.
Yep. You plan to share some data there in the third quarter of next year. Maybe talk a little bit about other differences in the phase III trial versus the phase II. We talked about younger patients, but anything else notably different?
Yep, yep. Younger patients, higher dose in general than the part one, which is important. A lot more fastidiousness for controlling the environment and tracking the environmental controls because one of the issues was, you know, if a lot of Prader-Willi households, if there's a very strict control of food access, that actually can itself control HQCT-9 scores a bit. If you loosen those, it can influence the score. We have a lot more precision in tracking and doing that. Patient travel is also like an impediment for compliance. Right now, because of the number of sites that we have, most patients will have a very local clinical trial site for themselves. We also have a larger panel of biomarkers, like ghrelin, for instance, a biomarker that we did not formally assess in our phase II, but we will be doing that in the phase III. We're very optimistic.
Also, younger patients, as I mentioned before, that will be expanding the group of patients.
How should we think about those changes impacting HQCT-9 score, like the longer duration, younger patient? Should it be the same, or is there a reason to think it might be different?
Our expectation is that the score will deepen. We actually have many, many reasons. Prader-Willi subjects are also oftentimes very pattern-oriented, so some things that would convey benefit take a longer time to manifest. There's also a look-back period in the HQCT-9 questionnaire itself. Over longer periods of treatment, you would expect that score to deepen as well, too, which is one of the reasons why we're confident that having a 12-week endpoint will kind of unmask even a greater benefit than a shorter-term treatment would.
Yep. Makes sense. I guess one of the other sort of challenges with these studies is high placebo response rates. You kind of discussed a little bit about ways to manage that, but you know, just talk about the risk there and how you've mitigated that.
Sure. One of the known drivers of placebo effect is, for rational reasons, in the competitive space, the inclusion criteria usually has a minimum cutoff score for HQCT-9. We've chosen one that's actually congruent to what others have chosen, which is a score of 13 or higher to be included in the trial. Knowing this, parents of Prader-Willi children may at least have an incentive to overreport scores to qualify for trials. Once they're in the trials, with that pressure gone, that objective drop then contributes to the measured placebo effect. The way we're ameliorating that is we actually, once they're already in the trial and qualified, will ask for another baseline before they start dosing. At the time they're asked that second baseline, they're guaranteed to be in the study respectively, right?
We have the option to, for the primary assessment, assess patients that still have a score of 13 or higher at the second baseline.
What's in terms of a successful study and sort of what's the bar for success there for the study, and then more, you know, commercially as well?
Sure. Now, you know, I make an argument of what a clinically meaningful HQCT-9 reduction is, and we're powered to have it. If we assume the effect size of our drug is six points above placebo, which we actually, as a company, think it exceeds that, if that were the case and we assume a standard deviation of six, 90% confidence only requires us to have 23 patients per arm. If we thought the true effect size were five points above placebo, then we'd only require 32 per arm. At 45 patients per arm, we're pretty confident that we're there in terms of powering.
Makes sense. Maybe just talk briefly about the competitive landscape and where ARD-101 might be able to fit in.
There is only one approved drug for hyperphagia for Prader-Willi syndrome. I think the commercial uptake has been positive. I think the majority of patients still have clinically meaningful hyperphagia that still could benefit from additional treatment. I think the landscape is such that it's not only one drug for this condition. It's multimodal. The mechanism of action of our drug, however, is very differentiated, complementary. We think that our safety profile is not additive to others because of the gut restriction. For instance, one of the contraindications of use for the approved medication is renal insufficiency, which is not an impediment for patients for our trial. Given the profile, we think we address a lot of the patients that might not have an optimized solution with the available treatments.
Got it. Maybe we can switch to ARD-201 that's in your obesity program. Maybe just give us sort of a quick background there on that molecule.
Sure.
Sure. Yeah, happy to. ARD-201 is the combination of the ARD-101 we were just talking about with an oral dipeptidyl peptidase-4 inhibitor. If you're not familiar with that drug, it's a very successful class of drugs for diabetes. It works by extending the biological half-life of the gut hormones we were talking about. Therein lies kind of the concept. Given that ARD-101 stimulates the release of gut hormones, it's complementary to use a DPP-4 inhibitor to extend the biological half-life. That's kind of the concept. Two orals, DPP-4 inhibitors are very safe and effective, been used in millions of patients. The concept is to put the two together.
Makes sense. You recently shared some interesting preclinical data. Maybe just walk us through that and the key findings there.
Sure. Yeah. It was really, really exciting findings. First of all, the model is high-fat-fed mice. It's diet-induced obesity, a very gold standard model for testing therapeutics for obesity. It's been highly translational, including for the incretin class of drugs. After the animals achieve sufficient obesity, we gave them ARD-201 or placebo and also a group just getting the sitagliptin or the ARD-101 alone. In that model, we achieved 19% weight loss within 30 days. That's getting on par with the GLP-1 receptor agonist class of drugs that need to be injected daily in that mouse model. That was really quite exciting. We did another study where we asked, could we achieve weight maintenance in animals that lost weight on a GLP-1 receptor agonist? In this case, we used tirzepatide. We got the expected weight loss of around 20% with that drug in the animals.
We divided the group into some that stayed on that drug or switched to ARD-201, and the weight maintenance was the same. That's really important because it suggests that we can transition from an injectable to the oral and achieve the same weight loss. Finally, we included a group where we had a low dose of tirzepatide combined with ARD-201. By day 30, we had 30% weight loss. Really exciting because it was a one-tenth dose of tirzepatide. You know the limiting factors with the injectable GLP-1 receptor agonists are nausea and vomiting. That's born because of the high doses that are given chronically elevated pharmacological levels.
We think it's quite attractive to consider that ARD-201 combined with a very low dose of tirzepatide or a GLP-1 receptor agonist may be able to not only achieve weight maintenance, as we've shown in this model, but also continue the weight loss and achieve a greater percent weight loss.
Makes sense. Maybe talk a little bit about how that data informed your thinking in your phase II study and some of the changes you made there.
Go ahead.
Oh, okay. Yeah.
We are very, very encouraged about that. It kind of inspired us to focus on this because of the real unmet need to address a number of features. I think one of the more appreciated areas for a need for innovation is how folks can get an off-ramp. If they've achieved weight loss on GLP-1, how can they bridge off of it and maintain their weight loss? By the way, there's a paradox of increasing ghrelin, which we believe represents hunger, while people are on GLP-1. It goes to the point of appetite and hunger being differentiated. Your body compensates for decreased appetite with GLP-1 by increasing hunger. We believe that's what fuels the weight regain when you get off of the GLP-1. If we want to see benefit and protection against that weight regain if one were to have lost weight on GLP-1 and then go on ARD- 201.
Our trial, the POWER trial, our prevention of weight regain trial, is meant to explore exactly that. We are going to be taking patients that already have established substantial weight loss on a GLP-1 looking to bridge off of it and then seeing the protection from the weight regain, which we are now inspired that it's with a highly predictive model that we can demonstrate that. The POWER study should allow us to see data much sooner than we had earlier intended with that study design. The second study, the STRENGTH study, is to look at prospective weight loss with our drug and also in combination with GLP-1 to really replicate what we see in the preclinical data. That's very, very encouraging to show that it's complementary and synergistic. We are very hopeful to see many positive things.
I would also add in our preclinical data, we are also encouraged that body composition advantages may exist with our ARD-201 formulation as well. Those are all focuses of intent for our upcoming trials for the POWER and STRENGTH trials.
Can you maybe talk a little bit about the timelines around those two studies?
Sure. For the POWER trial, we'll start before the end of this year. The STRENGTH trial, we've stated the first half of 2026. The POWER study readout should come around the back half of 2026, sometime similar to the timing of the HERO study readout or a little bit after that.
Makes sense. Maybe talk just a little bit about the strategy in obesity. It's obviously a very competitive space. It sounds like maintenance is maybe where the focus might be, or is that the right way to think about it?
I think there are several niches in where Aardvark's tool want to come into play. One of them is prevention of weight regain. Another is even prospectively as a first-line treatment because as an oral drug that we wouldn't expect to cause nausea, because just as a reminder, sitagliptin is not known to really be notably causing nausea. I'm of one belief that it's not associated with a high degree of nausea as well, too. In combination, we wouldn't expect that. As a possible first line and as a complementary therapy, that could also potentially allow for a much lower dose of GLP-1 receptor agonists that would not elicit nausea. I think that's kind of the multiple tiers of prospective use. We also think that our drug ARD-201 can be very useful in inducing favorable change in a number of metabolic parameters.
We saw hints already in our clinical data from before with hyperlipidemia and hyperglycemia. We think it is likely to induce improvements in both. There's another avenue to pursue supplementary to everything I mentioned before. We'd be able to elicit that signal in the studies that we have planned.
Makes sense. Maybe just talk about your current cash position, you know, runway, and sort of what that covers in terms of your clinical plans.
Sure. Our last Q is that we had $141 million in cash, and that should get us into 2027. That's a drivable date that's after the key readouts that we have anticipated, including the HERO and the POWER trial.
Great. Maybe we can switch gears a little bit, and I can ask a couple sort of macro questions. The biotech team this year is sort of asking all their companies some of these key macro questions and would love to get your perspective on it as well. Maybe with the first one, it's, you know, with the rise in China biotech innovation, how are you thinking about Aardvark Therapeutics' competitive position here? Will this influence your sort of R&D or BD strategy?
That's a good question. I think a lot of concerns always are for, you know, drug targets. Everybody's worried about being obsoleted because of some of the competitive program in China. I think we're sufficiently late stage and we're not aware of any other program targeting tasks to our end. The unique part about this target is normally the precision and the binding affinity of a target is important because there's always a trade-off for a side effect profile and, you know, everything is a risk-benefit analysis. Given the very unique safety profile, the gut restriction, I'm not saying that there couldn't be a better agonist, but I think our drug is so well optimized that I think we're in a very good position, especially with our late stage of development. We have orphan drug status as well.
I think that makes us a little bit less concerned about that setup.
Yeah. Makes sense. Maybe another macro question. Just, you know, how are you currently leveraging AI or thinking about AI's future disruption potential?
I don't know if Tim has any thoughts on that.
Yeah. We embrace AI for sure. We use it daily. We use it a lot in our writing predictive models. For example, in preclinical studies, we've started using AI to kind of predict outcomes, draw out hypothetical results, and help us refine study designs. Once you start using it, there's no looking back.
Yeah, how good are those sort of predictions or models?
They've turned out to be quite good. The study that we kind of walked through with ARD-201, for example, we used AI to help us pick doses and strategize durations and different groups, and help us with the analysis. It turned out to be pretty darn accurate.
Do you think that's going to be the trend going forward, or is it going to depend on which type of studies you're asking?
It is only as good as the information that it relies on to make the predictions. As I mentioned, ARD-201, part of it is the dipeptidyl peptidase-4 class of drugs. We have a really good understanding of how the CaSR works and what hormones it is evoking from the gut, so it was able to come up with pretty good predictions. I think in other cases, maybe less so. Even going through the process gets you thinking about aspects that you may not have considered otherwise. It is pretty useful, I think, in that respect. Of course, it does hallucinate, though. We have found that with math and with sources of information. You have to keep challenging it, but it is just going to get better.
Great. It looks like we're just about out of time, so why don't we end it there? Thanks so much, guys. Appreciate it.
All right.
Thanks for the questions.
Yeah.
Thanks for the time.
Thank you, Mike.