Maze Therapeutics Inc Stock

Yes. Welcome to Day 2 of Citi's Back to School Biopharma Conference. I'm Eric Joseph, Senior Biotech Analyst with the firm. And our session now is with Maze Therapeutics. It's my pleasure to have with us CFO, Misbah Tahir, to talk to us about the company. Misbah, thanks for joining us today. Maybe just for those less familiar with the company on the webcast, it would be useful to have you kind of provide us with a brief snapshot of the company and what some of Maze's key objectives are over the next 12 to 18 months.

Eric, it's great to see you again, and I'm excited to represent Maze this afternoon. Just some background on Maze Therapeutics. We're a clinical-stage biopharmaceutical company focused on developing precision medicines for kidney and metabolic diseases. We have 3 programs in clinical development. Our lead program is MZE829. That's an APOL1 inhibitor for the treatment of patients who suffer from APOL1-mediated kidney disease, or AMKD. We're currently in a Phase II trial testing MZE829 in AMKD patients, and we've guided to data from this study, which we refer to as HORIZON, by the end of this year or early next year. We've also committed to initiating a pivotal program with MZE829 in the first half of 2027.

Our second program is MZE782. That's an inhibitor of SLC6A19, a transporter of neutral amino acids. And we just announced last month that we initiated our Phase II study of MZE782 for the treatment of patients with PKU, and we've guided to top-line data from that program in 2027. Our third program is also MZE782, the same drug. We're also testing that drug in patients who suffer from chronic kidney disease, or CKD. We've guided to initiating a Phase II study in CKD in the first half of next year.

All 3 of those programs, those drugs are small molecules. They're all wholly owned by Maze, and they're all built with the support of our Compass platform, our discovery platform in which we are a discovery engine in which we harness the power of human genetics to develop precision medicines.

Okay. Great. So let's start with MZE829 for APOL1-mediated kidney disease, or AMKD. And I want to start by just thinking about the market opportunity here. The epidemiology studies give us a good understanding of who's at risk for developing AMKD and suggest that patients who have -- who are homozygous for the 2 variant alleles, right, or have 2 of the variant alleles are at substantially higher risk for progression to end-stage kidney disease. That said, how routinely is AMKD diagnosed today? And sort of at what stage of kidney dysfunction are patients typically identified?

Yes. I think the short answer to your question is there's a lot more work to be done here. And let's start with the genetics as you were starting to do. If you have 2 copies of the APOL1 high-risk variants, you're at risk of developing a more aggressive, a more accelerated form of chronic kidney disease. You're potentially going to develop CKD at an earlier age, a younger age, potentially before you turn 50. And you're likely or you're at risk of reaching dialysis earlier, potentially as much as 10 years earlier. So it's a more aggressive, more accelerated form of CKD.

Now the APOL1 high-risk variants were first described in literature only just back in 2010. And there are no approved treatments today for AMKD. And I also want to add that this is a disease that predominantly impacts people of West African descent. So we're starting from a very low base of awareness, and that's a key lever for testing, for enrollment in trials, et cetera. But that's starting to change. With sponsors like Maze in the field, there's certainly growing awareness of the disease. There's growing awareness of the need for testing. And with us and other sponsors, we're helping to drive that awareness.

So I suspect that over the next several years, the next several years of drug development and enrollment and testing will look very different than the last several years because the need is certainly there. And I think the clinical data that we all collectively are generating in this field and the potential for new treatments to treat AMKD is sort of going to drive further testing.

Has the understanding of the pathophysiology evolved a little bit over the past couple of years, particularly as it relates to other risk factors or other known contributors for kidney dysfunction with folks who have 2 copies of the variant alleles? I'm kind of thinking about this double-hit hypothesis that comes up a little bit in thinking about the indication.

Eric, I think this is an area of active exploration in the field. We estimate that there are roughly 6 million Americans who have 2 copies of the APOL1 high-risk variants. Roughly 15% to 20% of that 6 million will actually develop CKD and eventually AMKD. And so it suggests that you need a second hit in order to develop or reach AMKD. And so there's a lot of work going on, on what factors may represent that second hit, so to speak, whether it's genetic or more environmental. Some of the work we've seen and are doing suggests it might be more environmental, but there's a lot more work that needs to be done.

But we estimate currently, of the 6 million, roughly 1 million or so have CKD. And of that number, roughly 250,000 patients in the U.S. could benefit from an APOL1 inhibitor drug such as MZE829. So that to us is the current addressable patient opportunity.

Excellent. Excellent. All right. So you're running a Phase II study now, HORIZON, in patients with different subtypes of AMKD. Can you maybe be -- from a high-level standpoint, it would be good to have you walk us through the clinical activity that you're observing to date and sort of what -- if there's a target product profile you're striving for here as the Phase II program evolves and you think about advancing the candidate into pivotal development.

Sure. Let me start with some background on HORIZON. HORIZON is our Phase II open-label basket study in which we're testing MZE829 in proteinuric CKD patients, adults who have 2 copies of the APOL1 high-risk variants. And we're focusing on 3 segments within AMKD: The first is patients with FSGS; second, patients with diabetes; and then we have a third segment, patients without diabetes and without FSGS. So we're testing across all 3 of those segments, a broad spectrum of AMKD patients.

And in March of this past year, March of 2026, we released some initial data from that HORIZON study, which is still ongoing. And the results were really encouraging and positive to us at Maze. We had 12 evaluable patients. We saw responses, responses defined as achieving a 30% or greater uACR reduction. We saw responses in each of those 3 subtypes. And across all 12 evaluable patients, we saw a 36% average uACR reduction. Half of those patients received -- achieved a response of 30% or more. Within FSGS patients specifically, we had 4, we saw average uACR reduction of 62% with 3 out of those 4 patients achieving a response. So these were highly encouraging results for us.

And since we released that data in March, we've been focused on 2 priorities related to 829. The first is completing enrollment in HORIZON, and we've guided to releasing 10 to 15 patients' worth of data in each of those 3 segments by the end of this year, early next year. That 10 to 15 patient figure that I cited is inclusive of the patients we've already shown. The second priority coming out of the data release is to get a pivotal trial with 829 up and running as quickly as possible, and we've guided to initiating a pivotal study in the first half of 2027. The design and scope of that study is a topic of active discussion within the company. It's certainly going to be informed by our final HORIZON results as well as our discussions with the FDA. So a lot more to come on that.

Okay. You're not alone in the space, of course, right? We know of Vertex's inaxaplin, right, another APOL1 inhibitor. And we're expecting some data readouts there over the next 2 to 3 quarters. What comparison should investors draw between Vertex's inaxaplin and your compound?

Well, I'll certainly leave it to Vertex to speak to their drug. I think, with respect to MZE829, this is a drug that could be potent given the dual-mechanism nature of the drug. What APOL1 does is it punches holes in -- when you have the high-risk variants, it punches holes in the podocytes of the kidney. And MZE829 not only blocks those pores or holes that form, it also disrupts the assemblies of those pores in the first place. So in that way, prevents the leakage of protein.

And so you have this dual MOA that preclinically, we think could certainly be potent. And then in the initial data release from HORIZON, we saw in our FSGS patients, the potential potency of the drug in that segment and across all the evaluable patients. So we think there's potential differentiation there, and we're looking forward to the final results from HORIZON.

Just coming back to sort of efficacy expectations in HORIZON and the different -- given that you're looking at distinct cohorts, right? You've got non-FSGS, nondiabetic AMKD patients in 1 cohort, a second in diabetic AMKD and another with FSGS patients. Like is there perhaps -- are there differing thresholds of activity that you want to see perhaps -- I guess, is the threshold for what you think would be a profile that you'd want to bring forward into clinical development the same across each of the 3 buckets?

Yes. It's a question that comes up often in our discussions. 30% is the threshold that we can point to that comes up consistently in discussions with KOLs, practitioners. In the literature, there's a linkage between achieving a 30% proteinuria reduction and improved long-term renal outcomes in adjacent kidney areas. So it's the one hard number we can point to and seems to resonate with the nephrologist community.

But certainly, it's been suggested to us that in a segment such as diabetes, which is arguably much more complex set of patients, unclear what may be driving the CKD in a particular patient, that a lower threshold might be more appropriate. And if we were, for example, to achieve a 20% uACR reduction, that could still be clinically meaningful to nephrologists, to patients, because these patients have no other treatment options. And in the absence of a new treatment, they're going to continue to progress in an accelerated fashion with their AMKD.

In our HORIZON -- initial HORIZON data in diabetes, we saw 2 out of 5 patients achieve responses, one with a 35% proteinuria uACR reduction and another with a 47% uACR reduction. And that was incredibly encouraging to the KOLs to whom we showed the data because these were patients who are already on a number of background meds such as GLP-1s, such as SGLT2s, and the effect that they saw with 829 was on top of those treatments. And so that could be really important for the treatment landscape going forward.

But it also suggests, just sticking with diabetes for a moment, that if we continue to see a pattern where there are responders, but then clear nonresponders, perhaps we need to think about how we identify potential responders better going forward and enrich for that in future studies. So that's something that's certainly on our minds. We're going to wait and see what the final HORIZON data look like before we make any decisions, but we're actively exploring that whole concept.

And I suspect there might be something to learn from Vertex's programs here as well, right? And I think we're expecting a readout from the Phase II AMPLIFIED study with inaxaplin in fourth quarter or probably in [indiscernible], right? I guess how are you -- what insights are you sort of anticipating perhaps from the readout of those data? Or what are you looking for perhaps investors should be -- how should investors kind of prepare themselves when thinking about potential read-throughs?

Yes. We're going to design our pivotal program and think about our next steps for 829, mostly using what we learned from our HORIZON study and feedback from regulatory authorities. There's also an important initiative called the PARASOL initiative that may have some recommendations in the coming months. And that's a public-private initiative with a focus on accelerating drug development, clinical advancement and approvals in kidney diseases, and they have a project specifically in AMKD, and they're looking to potentially define surrogate endpoints that could accelerate drug development and approvals in AMKD.

So HORIZON, our discussions with the FDA and potential recommendations coming out of PARASOL, that's what we'll really lean on in terms of developing our next steps for 829. But for sure, data releases from other sponsors, it's going to be informative to us, informative to the whole field. Over the last 4.5 years, we've probably seen 25 patients' worth of clinical data come out from ourselves and other sponsors. And that's going to change dramatically in the next 12 to 16 months as we read out, as other sponsors read out. So it's an exciting time to be following AMKD, and we're going to learn quite a bit in the coming months.

Great. Great. Maybe we can shift gears a little bit to your second pipeline program, MZE782, in particular, the program in PKU. This is, I think, an indication that investors are becoming much more familiar with, specifically with a new market entrant we saw it come to market last year. Maybe talk a little bit about how 782, its mechanism and how it controls toxic phenylalanine accumulation in a different manner relative to the existing treatment landscape.

Yes. We're excited to move forward with MZE782 in PKU. As I mentioned at the outset, we announced last month the initiation of our Phase II study in patients with PKU and we've guided to top-line data from that in 2027. MZE782 represents a novel MOA, mechanism of action, in the treatment of PKU. And currently, when you look across the landscape, we estimate there are roughly 60,000 PKU patients in key geographies. Roughly 2/3 of the patients are on an onerous medical diet or are receiving no treatment for their PKU. And that plus some of the excitement we've seen around the new entrants suggest there's definitely a desire for new treatment options for PKU and a novel -- a drug with a novel MOA could be really helpful to the field.

MZE782 works in some different ways. It's an inhibitor of SLC6A19, which is a transporter of neutral amino acids. And recall that PKU really results from the toxic accumulation of phenylalanine in the bloodstream. And normal individuals have a functioning enzyme called PAH, which helps to break down and clear that phenylalanine. But when that enzyme is deficient, defective or not present, you see that toxic accumulation of phenylalanine, or Phe.

What MZE782 does is by targeting the transporter of Phe, it reduces the amount of Phe being absorbed or reabsorbed back into the bloodstream. It also helps to clear that phenylalanine, or Phe, through the urine. So it works in a way that is independent of existing treatments. It doesn't require the presence of PAH, and it's not looking to interact or boost the PAH enzyme. It works independently of those. So that could be a really important tool in the arsenal to treat PKU going forward.

Yes, let's -- so just kind of digging into -- or before we kind of dig a little bit into the Phase II CIPheR study, now underway. Maybe let's just talk about the learning -- the Phase I experience in healthy volunteers and what was learned there and sort of how the finding has informed go-forward dose selection and activity -- I guess, how those data sort of derisk activity in the PKU population going forward. Can you just sort of speak to what data points actually kind of drove -- have you arrived at the set of doses that you're evaluating in Phase II? And just generally, how predictive looking at Phe reduction in healthy volunteers in plasma -- or sorry, in the urine relates to -- translates to expected Phe reduction in the plasma in PKU patients?

Great. So about a year ago, in September of 2025, we disclosed results from our Phase I study of 782 in healthy volunteers. And this was a randomized double-blind placebo-controlled study of roughly over 100 healthy volunteers, a SAD/MAD study, testing a range of doses. And in addition to safety and tolerability, we were also looking to see what level of urinary Phe excretion we could achieve with 782. And the threshold or benchmark we set was informed by another sponsor who's developing an inhibitor of SLC6A19. And in their Phase I study, they had achieved a roughly 10-fold increase in urinary Phe excretion in healthy volunteers.

So with our results, what we disclosed is that not only was 782 well tolerated, but we were able to achieve an up to or over 40-fold increase in urinary Phe excretion. At our 240 mg BID dose, we actually achieved a 42-fold increase in urinary Phe excretion. So those results exceeded our expectations and certainly resulted in us moving as quickly as possible to get 782 into a Phase II study, which we've since initiated.

And what do you think is driving that sort of higher level of Phe excretion? Is it just kind of tighter affinity for the transporter, or is it a function of pharmacokinetics? Maybe you can just kind of speak to the difference in either the pharmacokinetic profile, I guess, both PK and PD profile of 782 perhaps versus the other sponsor's compound?

Yes. So I think rather than getting specifics, I'll give you the generic CFO answer, is we have a better designed molecule. And I think at 240 BID, we achieved that result regardless of being fed or fasted and so no food effect, and that certainly informed bringing that dose forward into the Phase II. And I think what we're looking to see in the Phase II now is how well does that urinary Phe excretion translate into actual plasma Phe reduction.

And you were starting to get at this with one of your other questions, but the nice thing of having another sponsor go ahead of us is that with a similar MOA is that we could see how their results in Phase I, their urinary Phe excretion, translated into plasma Phe reduction. And we saw a nice translation with their molecule. We exceeded their urinary Phe excretion in Phase I. We'll see what we achieve in Phase II with plasma Phe reduction. But that, in addition to safety and tolerability, is what we'll be primarily looking at in the Phase II.

Right. Okay. Yes, I think it probably, until your data readout, it's still going to be a bit of a question as to whether that relationship going from urinary Phe change in healthy volunteers, is that a linear relationship to the plasma Phe change in patients because we're talking about different reservoirs that we're looking at, number 1. And then secondly, probably -- not probably -- very different starting Phe levels, right? And so I guess we understand what it looks like from 1 compound here, and I guess we'll learn whether that relationship is linear or is it asymptotic perhaps with 782, which should be pretty interesting.

I guess the other thing to think about here, just is in thinking about 782 and the indication is on tolerability. So far, tolerability in healthy volunteers is excellent. You are having -- you do have a seemingly higher clearance of Phe. But along with that, you probably also have a higher clearance of other neutral amino acids. And so I think it sort of begs the question whether there are potential any downsides to increased elevated excretion of neutral amino acids via this mechanism. I guess, how does the Phase I experience perhaps or data elsewhere sort of derisk any potential consequences from that potential phenomenon, right, seeing kind of depletion of other neutral amino acids? I guess, in other words, is there a concern around a drug or drug-related Hartnup-like syndrome emerging from an approach like this?

Yes. The short answer is that no concerns have emerged so far. And so as I mentioned, in the Phase I study with healthy volunteers, the drug was well tolerated. We didn't know -- we didn't observe any issues or adverse events related to the excretion of neutral amino acids. And in the other sponsor with a similar mechanism of action, as best as we can tell from the publicly disclosed data, no such concerns have emerged from their Phase I or Phase II studies.

And then taking the case of Hartnup syndrome patients, that's where a patient has a complete loss in SLC6A19 function. And those patients seem to be managed very well with assuming a normal, standard Western diet. So no real concerns have emerged from that patient group that could impact our program at this point and nothing that we've seen so far in our clinical studies.

Excellent. All right. So just on the -- thinking about the mechanics of the CIPheR trial, this is also going to proceed through a dose escalation scheme. So maybe just kind of walk through operationally, how you're proceeding from 1 cohort to the next and the extent to which -- I don't know if you actually -- have you disclosed the doses that you're...

We have.

You have?

Yes.

Maybe just speak to those and whether any of the cohorts that you're looking at are also evaluating a once-daily dose regimen.

No, happy to. So we're testing 3 cohorts. And so we've got -- the first cohort is 120 milligrams BID. The second cohort is 240 milligrams BID. And the third cohort is 240 milligrams BID in combination with any BH4 agent. Now we certainly think there's potential for MZE782 to be potent as a monotherapy. But there may be patients when we start thinking about treatment in the context of diet liberalization, and that will become more important with the Phase III. There may be some patients who need that extra potency in order to get below the Phe thresholds that really allow one to liberalize their diet. So those are the 3 cohorts we're testing, and we'll start with 120 BID and then move forward accordingly.

Okay. All right. In the BH4 combination cohort, any concern regarding drug-drug interactions there? This the first time that you'll be kind of interrogating that.

Yes. No, it's something that -- something we do what we can preclinically and whatnot to derisk and remove concerns. And so we feel comfortable moving forward.

Okay. Okay. Great. This is a very forward-looking question, I suppose, but I wonder whether co-formulation of 782 with BH4 is something that you'd be interested in looking -- interested in pursuing down the road.

Certainly, I don't want to take that off the table, but it's not an area of focus for us currently. We're really focused on testing 782 as a monotherapy. And then as I mentioned, we have this 1 cohort in combination, and we'll see what we get there. But once we get to the Phase II, we'll look at what makes sense going forward to further develop the drug.

In that combination cohort, I presume patients are going to be coming in kind of on a steady experience of BH4 treatment, right, so they probably have lower Phe levels than the other 2 -- the other monotherapy cohorts. Having said that, what would be considered a meaningful reduction in Phe levels in that patient population, do you think?

Yes. So the threshold for -- the eligibility threshold for Cohorts 1 and 2 as monotherapy is 600. And then for the combination therapy, it comes down to 360. And so I think the question that we're looking to answer is, can we -- how far below 360 can we get because that's really a key delineation in terms of starting to liberalize diet.

And we haven't talked a lot about that, but it is fairly onerous for these PKU patients. They have to cap the amount of Phe intake they have in any given day. It's the equivalent of about 2 eggs. So it's really tough for folks, and we were certainly hearing a desire to get off of that diet with an available therapy. And if folks have an oral therapy with a relatively benign safety profile, that could certainly be an attractive option for them.

Okay. All right. Great. I know the study is just sort of -- it's still in its early phases, and we're -- there's also a stepwise fashion in terms of dose escalating. But I guess, is there a sense of time frame within 2027, you think you might be reading out from that study?

Yes. I think our official guidance is 2027. We'll have more to add as we get further into the study and see how enrollment is going. But so far, we're off to a great start.

Excellent. All right. Awesome. Well, I'm going to kind of throw in a thematic question here at the end. Our colleagues covering tech also ran a conference this week, and so I'm partially asking this on their behalf, right? But the topic of AI and the impact of AI tools is something that comes up a lot, affecting all of our line of work and all of our lines of work. And so, to you, can you tell us a little bit about how Maze currently is using AI-enabled tools across the organization and how their impact is measured?

Yes. No, we're getting that question a lot from a range of different folks and audiences. It's early days for us to be sure, but an area of increasing focus for us. We're starting to look at how we can increase adoption within the company of AI tools. We're starting to look at how we can deploy AI in various forms to improve productivity and ultimately to look at how can we shorten the drug development cycle, which would be the holy grail for us. So a lot more to come, but I would basically characterize it as early days, but an area of increasingly high focus for the company.

Excellent. All right. Great. Well, thanks so much for your time this afternoon, Misbah. We really appreciate it. Thanks for tuning in.

Thank you, Eric.

Good morning, and thank you for joining us today. My name is Elizabeth Webster. I'm on the biotechnology equity research team here at Goldman Sachs. And today, we have with us Jason Coloma, the CEO of Maze Therapeutics. Thank you, Jason, for being with us.

I guess to start, can you provide an introduction to the company and walk us through where Maze stands today with your lead assets, 828 in AMKD and 782 in PKD and CKD?

Yes. Well, thanks for having me today. Yes. So we are, if you will, a small molecule precision medicines company really focused on diseases that really haven't seen precision before like kidney disease. There's over 800 million people in the world that suffer from kidney disease. It's the ninth leading cause of death globally.

In some countries, there's a 4-year waiting list for, if you will, a transplant. And unfortunately, there hasn't been a lot of innovation in kidney disease. And that's exactly where we fit in. We develop small molecules, really focused on being able to provide precision for those patients.

Starting with 829, which is for APOL1-mediated kidney disease. There's about 1 million people in the U.S. alone who have suffered from the disease. And we've taken an approach where we've been able to demonstrate proof of concept in a broad set of patients earlier this year.

And our second program, 782, which is being developed for 2 particular indications, one for a rare metabolic disease called PKU and also equally excited about what we can do in kidney disease.

Awesome. And you relatively recently presented data from MZE-828 in the broad AMKD population. Can you just kind of go through that data and summarize what the learnings were there?

Yes. So it's 829 but...

I'm sorry, 829, yes. 829.

Thanks, okay. Yes. No. So what we did earlier this year is we presented 829 data from our HORIZON study, which is a global Phase II open-label study. A couple of things that were accomplished there. One is that we showed that it continues to build on its safety as well as tolerability profile.

The second thing that we're able to do is from across the different cohorts. We have FSGS, those without diabetes and those with diabetes. The third is we showed evidence, at least early and promising that we have a competitive profile relative to other molecules that have been described in the literature, in particular in FSGS patients.

And the third thing is that we had the ability to show efficacy in more moderate forms of proteinuria, which hadn't been demonstrated before. The totality of that particular data allows us to really accelerate what we're doing in the sense of being able to advance the program into pivotal planning, and we plan to start a pivotal study using 829 next year.

Great. And kind of walking through the different subgroups, can you kind of talk about the signal in the diabetic patients versus the nondiabetic patients? And then how you think about the data in FSGS versus the non-FSGS patients? And just kind of walk us through those subgroups.

Yes. FSGS, what we were able to demonstrate for the first time was really a group of patients that have been treated on various medications, including SGLT2, which hasn't been described in the literature before. So despite being heavily premedicated with other forms of kidney disease agents, we showed a 62% reduction in UACR, which had really exceeded anything that had been described previously.

So demonstrating our ability to have at least this concept of being best-in-class. We were also able to demonstrate for the first time the ability to show efficacy, albeit early and promising, the idea that you can go in more moderate patients, both with or without diabetes. The previous literature had really not described a lot of evidence that they would be able to see efficacy at all in diabetic patients. We saw early promising signal in some of the patients, about 40% of the patients showed efficacy even despite having been previously treated with GLP-1s as well as SGLT2s.

Great. And when we think about the change in uACR, I think you've mentioned this kind of 30-milliliter bar. Talk about kind of what's clinically meaningful in these populations from change from baseline?

Yes. If you take a look at sort of the overall clinical development of other approved agents in kidney disease, 30% reduction in proteinuria translates well to overall, if you will, benefit in looking at other measurements, including what's called eGFR or filtration rate of the kidneys, which is one of the approvable endpoints for some of the indications in kidney disease.

So the fact that you can see a 30% reduction has demonstrated in previous studies to be, if you will, translatable to other endpoints. The other thing to keep in mind is that it really is used in the KDIGO clinical guidelines, at least in the U.S. a physician and a nephrologist will look at the ability to try to reduce proteinuria by 30% with existing agents. If they're not able to do that either with dose adjustments or adding other agents, they keep going.

So they'll start a patient on, say, what's called an ACEi/ARB. They'll try to add then see if they can adjust the dose if the patient is not getting to that 30% reduction. And they'll keep trying to add agents or switch to different agents in order to try to do that. Unfortunately, patients with APOL1 kidney disease, standard of care is not efficacious in these patients.

And unfortunately, despite all that going through multiple treatments, including things like ACEi/ARBs, immunosuppressants, SGLT2s, they're not seeing that 30% reduction straight away. And so an agent like 829 shows that real promise because we're able to do that across a number of different patients in different types of patients, including FSGS and non-FSGS patients.

Great. And you're kind of targeting some near-term updates on the development path here, but where do you see clinical development for 829 proceeding? And what's your plan there?

Yes. So to take a little bit of a step back, the data that we had in March was an administrative analysis that allowed us to really be able to, #1, identify whether or not we had an active compound, which I think we did. The second thing we're able to do is really describe are there any particular cohorts that we think about accelerating and move into a pivotal study faster, which we were able to do. The third thing is to look for initial signal in more moderate patients, including those with diabetes.

Now with that in hand, we're pushing forward, in particular in the nondiabetic patient population to advance into pivotal studies. So we accomplished that. And I think what we want to do in diabetic patients is to continue to collect more data. It's early and promising as we talk about that with nephrologists in the field and the way that they might potentially treat their patients that have diabetes and have APOL1 kidney disease, they find it incredibly encouraging, primarily because they're not responding necessarily to other agents like GLP-1s or SGLT2s.

So what we'll continue to do is collect more data, see if we can tease out the signal. We did see response in some patients, as I described. What would be interesting as well is to identify are there particular biomarkers that might help us enrich to better understand the signal, right? If you look at the literature in DKD, 40% to 50% response rates are typical in DKD patients. And you can see in the literature, even the last approved product for kidney disease, which was KERENDIA in their Phase II study, they had about 20% to 25% reduction in UACR, yet that now is an approved product, right, for kidney disease.

And so if we can take that type of learning and better understand, one, can we see UACR reduction in diabetic patients as well as, #2, figure out are there particular markers that might help us enrich, that will allow us to think about the development plan for, I would say, the diabetic patient population.

Okay. Great. And if you could put a time line on those, would you say...

So we have -- we'll have the full readout for Horizon, the end of '26 and '27. So we will have the 3 cohorts of FSGS nondiabetic patients as well as diabetic patients, 10 to 15 patients per cohort that will allow us to really understand and tease out the signal. In parallel, what we're doing is advancing the nondiabetic patient cohort for pivotal planning, which we expect to start in the first half of 2027.

Okay. Great. And then just thinking about the competitive landscape, there are other kind of assets in AMKD in development. Where do you think 829 is most differentiated? And how do you think about the competitive landscape here?

Yes. I think there's -- unfortunately, there are new therapies in development. There are no approved therapies for APOL1 kidney disease today and the current standard of care, unfortunately, is not proven to really improve the patients' lives and many of them unfortunately transition not only to end-stage renal disease, but unfortunately, to dialysis.

And as you know, their outcomes are horrible when you think about what they have to shift into dialysis. And so the idea is, could you develop a particular more precision approach that allows us to really be able to help these patients. And hopefully, they'll never have to go into dialysis. I think the opportunity here from our preclinical data, we showed that we had differentiation, especially on potency, which should translate into efficacy.

We've actually seen that now clinically early and promising, in particular in the FSGS patient population, which is probably the closest one could think about in terms of apples-to-apples if you wanted to compare across trials. always dangerous to do so, but that is the closest one could do. And you can see even when we were -- our patients were treated with many co-meds, including SGLT2 inhibitors, they still had high levels of proteinuria, and we were able to reduce the proteinuria significantly.

In the FSGS patients at 62% UACR reduction and the nondiabetic patient population, nearly 50% reduction in UACR. So we think that, that is not only clinically meaningful, it gives us a differentiated profile relative to others out there that will allow us to participate and really be able to be one of the, hopefully, cornerstones for these patients.

I think the other thing to think about, too, is also in cardiorenal, I think we've all seen this. First is not always best in the sense of commercialization, order of entry is important. But I think what we can learn from others in front of us is important. We can see that in other areas, smaller [Technical Difficulty] and the board in terms of cardiorenal where companies that are coming behind with differentiated medicines, given the unmet need and the size of the population will have a role.

Great. And just mechanistically, in the preclinical data, I believe you show kind of a dual mechanism for blocking the channel here. Kind of can you speak to that because I think that's an important aspect of driving the potency?

Yes, of course. Yes. So I think probably just to take a little bit step back in terms of mechanism. What we learned over the years is that APOL1, the way that it causes disease, particularly in the podocytes or what they're called kidney cells, really under certain conditions, you get this overexpression of APOL1, usually an inflammatory response.

That overexpression caused it to be expressed in the -- if you will, in the cell membranes of the podocytes, basically punching holes in them because these are ungated channels. And then you have an influx of particular, if you will, bad actors that ultimately cause nephrotoxicity. So you have this overexpression, this ungated channel and basically having this nephrotoxicity.

The idea with an APOL1 inhibitor is minimally you want to block the pore from allowing the particular agents entering into the cell. But what 829 does relative to other molecules described in the literature is it not only blocks the pore, but it disrupts the assembly of the pore from forming to begin with. We think that's important for a few reasons. But one is that APOL1 highly turns over in the podocyte every hour, sometimes as quick as 42 minutes in the literature.

So the idea that you want to not only think about blocking the pore, but working upstream such that the pore is never formed to begin with, it makes sense that you would be able to, if you will, change sort of the proteinuria levels in these patients. And so compared to other molecules that have been described, they're only able to block the pore. They're not able to disrupt the assembly, right?

And so that's why it helped, I think, for us, partially explain the [ potency ] translating into the clinic. Again, looking at the -- saw a 62% reduction in that UACR. Previously, in terms of UACR, that was in the high 30s. And our patients were treated on the background of multiple meds, including SGLT2 versus what's been described in literature at the high-30s, the patients were not treated on SGLT2.

Got it. And thinking about what a Phase III trial could look like here, kind of walk through how Maze is thinking about that? And then your point about the background therapies, is that kind of similar to how you would think about patient enrollment in such a study?

Yes. I think there's -- I think what's nice, there's an academic working group called PARASOL that was started at the University of Michigan that was really helpful for other companies working at FSGS. There was a recent approval, which was great for FSGS patients. And that particular academic working group that they were working with the FDA. They're now working on APOL1 kidney disease.

So the good thing is we now know for FSGS patients alone from the previous PARASOL study is that proteinuria reduction is the approvable endpoint, not a surrogate or accelerated approval that is the full approval endpoint. So I think it's pretty clear in terms of regulatory path for FSGS patients. What we might be able to learn from the PARASOL study that's supposed to read out for APOL1 kidney disease this year is how that applies for FSGS patients that have APOL1 variants, right?

So it may be the same guidance or it may be different. We'll have to see what the working group comes up with. But I think that will be very important to help clarify the regulatory path. For non-FSGS patients, I think you can look at other studies that are there where they're combining both proteinuria reduction as well as eGFR slope as a basis for accelerated approval.

That will be highly informative if that's going to be possible, I guess, end of this year or early next that will help clarify at least for our program as well, the ability to kind of think about the non-FSGS patients. So I think in summary, I think FSGS is pretty clear, regulatory path in terms of proteinuria reduction. And I think we're going to learn a lot from other programs as well as the PARASOL initiative that will inform us how do we think about the non-FSGS patients.

Got it. And just kind of double-clicking on FSGS. Could you kind of speak to how that population is kind of diagnosed and just like the term of FSGS, kind of what that means clinically and how those patients might be different if they are than the non-FSGS population?

Yes. I think it's -- I think there's been a lot of education as you're inferring about what FSGS is. And it's not a clinical diagnosis, it's a histological diagnosis by looking at a biopsy. And historically, kidney patients weren't biopsied that often. It wasn't something that nephrologists had done. But I think to the credit of other companies in front of us and now approved therapies for those patients, we're learning many more patients are being biopsied once they get referred to the nephrologists.

In particular, that's helping them direct them to better care. right? So I think that's important to note. But typically, APOL1 kidney disease patients, they are disproportionately affected in the black community. So a younger individual will go to a primary care physician. They'll have elevated blood pressure typically. And for unknown reasons, they get the typical workup that URI would get if the physician saw that type of blood pressure. They'll look at their liver, their kidneys. And if they come back, they'll see even if it trace amounts proteinuria, they'll ideally refer them to a nephrologist.

Now what a nephrologist will do at that point is try to figure out if they are from the African-American community. Right now, not everyone is being genotyped, but that's being worked on to date with additional efforts in terms of, I would just say, advocacy in the community. But we're seeing more and more people being genotyped and then more importantly, getting biopsy that's allowing them to direct them to the -- to the right care.

Great. And maybe we can move to PKU and MZE782 and I think it would be helpful just framing the data that you've shared to date and kind of contextualizing it for us in this disease?

Yes. So last fall, we described 782 and published our Phase I data, which was over 100 healthy volunteers. And what we were able to do there is, again, show that it was safe, tolerable, PK was linear and predicted well and a half-life of about 11 hours that allows us to kind of think about either once or twice a day dosing. But more importantly, what we're able to do is show pharmacodynamic effect, in particular biomarkers that translate well to PKU.

Now for people that are not familiar with PKU, it's a rare metabolic disease which is left untreated. Unfortunately, you have this toxic accumulation of an amino acid phenylalanine in the blood. And what that does is, unfortunately, some of it makes its way to the brain, causing some neurodevelopmental neuropsychiatric impact. And so right now, the standard of care is to either take what's called a BH4 cofactor or an enzyme substitution therapy, which basically tries to break down that phenylalanine accumulating in the blood.

Unfortunately, not everyone can tolerate these or it's not efficacious enough to really be able to reduce. And so the magic number that you hear right now is that you would want to try to reduce phenylalanine levels below 360 micromolar in the blood, maybe even 120 micromolar in the blood. -- okay? So clinically, that's what you're trying to do.

And so unfortunately, because not everyone is doing well on these therapies, 60% of the people out there have to stay on this really onerous medical diet, where they can have very little protein. And what I mean by that is most of them can have maybe up to 10 grams of protein a day. It's basically 2 eggs. You and I will probably have 8, 10x that much in a given day. So unfortunately, people have to be on this onerous medical diet and then they can't come off of it, right? And they're on it forever.

So a new therapy like 782 provides an opportunity to address not only the patients today, who don't respond to current standard of care, but also the 60% of people who have to stay on this medical diet and ultimately try to get them off that because as we talk to patients, parents of those patients, they just want to get off this diet. And if they can do that, that will transform the way that they live because they don't have to be restricted so much and live hopefully more healthier lives.

So the data that we published, there was another company who had demonstrated proof of concept of inhibiting SLC6A19. SLC6A19 is a solid transporter expressed in the gut and the kidney. And what we're doing is basically hitting the target, getting rid of that toxic accumulation of phenylalanine in the blood by excreting it out in the urine. It's that simple of a mechanism.

And the good thing about that is you can measure even in healthy volunteers, how much phenylalanine is excreted out into urine. And we know that, that translates very well to the plasma Phe reduction eventually in patients. So there was another group that had published a few years ago that they had in healthy volunteers shown a tenfold increase in urinary Phe relative to baseline at one of their doses, right, showing that proof of mechanism. And eventually, in PKU patients, they showed a 40% reduction in plasma Phe. And the good thing about plasma Phe reduction, that's the approvable endpoint. That's not a surrogate or accelerate, that's the approvable endpoint.

So we knew what the bar was. If in healthy volunteers, we can beat a tenfold increase of urinary Phe relative to the baseline, we would have evidence that we could be best-in-class. And what we did last September is we showed at 90 doses, we're able to beat that up to 40x. So over 4x greater urinary Phe excretion than what has been described in the literature to date, which should translate into best-in-class plasma Phe reduction in patients when we run that study, which we're running right now.

Great. And in terms of kind of the mechanism and the structure of 782 help us understand points of differentiation. I think there's another oral substrate reducer in development from another company. And so how do you see that kind of biologically and mechanistically differentiated?

Yes, it's an important point. So at the end of the day, people want to just get off this medical diet. And that means getting their plasma Phe as low as possible, ideally below 360, even below 120. So if we can do that, we can get them off the medical diet in the long term.

So preclinically, we showed that we were 3 to 4x more potent, and we had talked about that before. And in the clinic, even in the healthy volunteer study, we reinforced that. They showed at one of their doses, a tenfold increase of urinary Phe relative to baseline. We showed over 40x increase of urinary Phe excretion. So reinforcing that 3 to 4x potency relative to anything that's been described in the literature to date. And what that should translate into is better plasma Phe reduction in patients, ultimately resulting in a number of people who can get below the 360 and the 120 numbers that are important clinically and get them off more importantly, this medical diet.

So what we know about the agent that you described earlier in their Phase II study, what we know is that at their low dose, they had about a 40% reduction of plasma Phe. At their high dose, they had a 60% reduction of plasma Phe. So we have a good mapping of where they are and what the bar is set. And they also had -- of the 19 patients that they reported, they had 3 that went below 360, and they only had 1 that went below 120.

So clearly, proof of concept was demonstrated, but there's room to go there. If you think about those numbers, most patients that have PKU are called severe or classical. And what that means is that they have a greater than 1,200 micromolar of plasma Phe. And so if you want to reduce that down to 360, if you just do the math, that's 70%, right?

I said at their top dose, they were 60. So they don't -- they won't be able to serve all of the patients that way. And so with our potency, our early clinical data, we have reasons to believe that we can beat those numbers.

Great. And could you kind of talk about the prevalence numbers and the addressable population sizing? And then when you speak to that 60% of patients that are on that medical diet, how big is that population in terms of...

Yes. So from a global perspective, there's about 60,000 people that have PKU. About 60% of those are still just on the medical diet. Less than 10% are on an enzyme substitution therapy. And unfortunately, not everyone responds to that because they develop antibodies over time. It also has some issues in terms of safety. So it has less than 10% usage even though it is the most efficacious for the patients.

And so most patients in that 20% to 30% are using what's called the BH4 cofactor, which is basically just a co-factor to allow the enzyme that's there that breaks down the phenylalanine to work better. That's what was recently approved with another molecule in Phase III. That's what Kuvan is.

And so our agent works in a different type of mechanism, which should, in theory, work across the entire spectrum of disease because we don't rely on the enzyme that's present to be efficacious because we just get rid of the toxic substrate to begin with. And if we can do that safely, we should be able to serve the entire spectrum of the 60,000 patients.

Great. And then before moving to chronic kidney, maybe just finishing up on PKU, remind us of your next data catalysts and potential trial initiation time lines?

Yes. So we're starting a trial that's a Phase II proof-of-concept study. We have a couple of doses that we're doing plus placebo. So this is not an open-label study. This will be placebo-controlled. So not only are we testing these in a monotherapy fashion, but we also have a third cohort, which is going to allow us to look at this in combination with a BH4, just given the fact that we have complementary mechanisms.

As we talk to investigators and some of the patients, there's a group of patients that do pretty well on the BH4s, but they can't get below 360 and they can't get below 120 -- so the idea that we can do clinically is, well, maybe we can serve those patients in combination, get them below the 120 eventually, and they might be able to get off to medical diet in the long run.

A lot of investigators encouraged us to study those patients because then if you think about from a development plan, both the low dose to high dose in combo, we potentially can work across the entire spectrum of disease. And based off of our early data, reasons to believe that we can do better than what's been seen before by hitting this target and reducing the plasma Phe, which again, at the low dose was about 40% reduction of plasma Phe.

So we're running that study right now. The data will be in '27. So we have a few catalysts coming up, as you pointed out, 829 end of the year, early next, where we'll have the full Horizon readout across the 3 different cohorts, FSGS, diabetic, nondiabetic. And then we'll also have the PKU data in '27 as well.

Great. And maybe switching over to chronic kidney disease. You've kind of talked about a potential protective mechanism here. And maybe to start, kind of what does this mean exactly clinically and mechanistically, what could be driving that?

Yes. So SLC6A19 we were the first group to describe the genetics that there's individuals out there in a heterozygous fashion, who actually were protected or had renal protection or had healthier kidneys than those that didn't have the variant. That encouraged us to think about, well, could we actually phenocopy what we call the genetics here with a small molecule such that by inhibiting the target, people in that kidney disease might also have that protection.

So we had not only been able to look at that from the genetics and published, we showed in vivo proof-of-concept in an animal model. We were the first group to do that. And then contextualizing that with one of the current cornerstones of standard of care, SGLT2. And so what we showed was that not only we were able to reduce proteinuria to demonstrate that proof of concept, we actually, in that model, did it better than the SGLT2.

And importantly, in combination, we basically got those -- the levels back to normal. So we published that, and that was encouraging. So we had the genetics, we had the preclinical data. And then last year, what we did in terms of the clinical data. Now we knew that all approved kidney disease agents had this phenomena where they showed an initial -- what's called an initial eGFR dip, which is a little bit counterintuitive because you actually want your eGFR to improve over time.

But all agents that have been approved for kidney disease, including ACEi/ARBs, SGLT2, even KERENDIA, they show this initial eGFR dip, but it plateaus and flattens out relative to the placebo such that you actually have renal protection over time. So that was interesting. And it was in the literature that you could see this even in healthy volunteers.

So we knew that. And if we had this hypothesis of genetics as well as some of the preclinical data, we said, let's look for that eGFR dip in the healthy volunteers. And we saw that. So we published that last year, which was not only do we see that in a dose response manner, we also saw that when we pulled off the patients on treatment, their eGFR bounce back, inferring that the effect is due to the treatment, not just by chance.

And the other thing that we saw is it's in the range of the eGFR dip. That's in the same range as SGLT2 that's been previously republished. So that all was very notable and interesting to us. And so that helped us better understand we could have a hemodynamic effect. And so our current understanding of the mechanism of, I would say, SLC in the context of kidney disease, one is that it could have a hemodynamic effect, which we've seen now in the clinic.

The second thing is that we know that it's a solute transporter and not just the neutroamino acids like phenylalanine, which we talked about for PKU. But it also manages these toxic metabolites, which in excess can be damaging in the kidney. So the fact if you can inhibit it, you might be able to just get rid of these toxic metabolites by just excreting it out into the urine, similar to what we do in PKU.

So not only could you have the benefit of an SGLT2-like mechanism with the hemodynamic effect, but you might also be able to just have a complementary or second type of mechanism by just getting rid of these toxic metabolites. So the idea here is that in the clinic, what we might be able to do in kidney disease patients who are not necessarily responding to standard of care in a proof-of-concept study, we would just look for UACR reduction in those patients that haven't been responding to things like an ACEi or maybe an SGLT2.

And if we see, again, a 30% reduction in UACR that would be proof of concept because not only now that we have one, the genetics, the in vivo proof of concept, the proof of mechanism in healthy volunteers, but if we can show UACR reduction in patients who haven't been responding to standard of care, that will be the first time anyone has demonstrated this with SLC6A19.

Great. Yes. In the last few minutes here, just on the financial aspect, just remind us of your cash position and your cash runway?

So yes, we just reported with a recent raise, we had a little over $528 million. That funds multiple catalysts that have been described today. So not only does it complete the HORIZON study, which we will have the data later this year, early next, the PKU proof-of-concept study, the CKD proof-of-concept study and allows us to even initiate the pivotal study for 829 in kidney disease.

Great. And just in CKD, when can we expect that next data set?

Well, we said we would start the study first half of '27. We haven't announced when the catalysts would come from that. But you can see multiple catalysts again from our APOL1 program as well as our PKU program. And the cash runway guidance that we have through that, that I just gave you is into 2029.

Okay. Great. Well, thank you so much, Jason, for being with us. Really appreciate it.

Yes. Thanks for having us.

Good morning, everyone. I'm Jason Coloma. I'm the CEO of Maze Therapeutics. And first of all, I'd like to thank the Jefferies team for inviting us to this year's conference. I'd also like to acknowledge our forward-looking statements.

At Maze, our mission is simple, yet ambitious, to harness the power of human genetics to transform the lives of patients. An emphasis for us is chronic kidney disease. Kidney disease is a silent killer. Over 800 million individuals globally suffer from this disease. It's the 9th leading cause of death to date. In some countries, there are -- there is a 4-year waiting list for kidney transplants. And there has been little innovation and no precision medicine approaches for kidney disease. And that is where Maze fits in. We develop small molecules using our genetics approach to really be able to focus on kidney disease.

Now our small molecules start with MZE829. That's for APOL1-mediated kidney disease. There are at least 1 million individuals in the U.S. alone who can benefit from such an approach. We are enrolling our Phase II right now and have -- and intend to have clinical proof-of-concept data in Q1 2026, where we could be the very first company to describe clinical proof of concept in broad AMKD patients with or without diabetes.

Our second program, 782, is focused on SLC6A19, also a small molecule approach. Now we are developing the therapy potentially in PKU, where we showed clinical, if you will, proof of mechanism for this particular approach in September. And we were also the very first group to show clinical proof of mechanism in chronic kidney disease in September, highlighted by our recent events and presentations at the large American Society of Nephrology just a couple of weeks ago. Based off the strength of our data last September, we intend to start 2 Phase IIs, one in PKU, the rare metabolic disease, and a second program in chronic kidney disease also in 2026. So 2026 will be a big year for Maze Therapeutics, where we intend to show, again, clinical proof of concept in broad AMKD in Q1 of 2026. And with our second program intend to start 2 Phase II programs, one in PKU and one in kidney disease in 2026.

Now I'd like to turn our attention to MZE829 in APOL1-mediated kidney disease. Now the genetics of APOL1 kidney disease was known for a very long time, over a decade. But unfortunately, no one really understood the function or mechanism by which APOL1 was causing disease. Our group was one of the first groups to identify the relationship and understanding of the mechanism on how APOL1 was really causing kidney disease in these patients. And moreover, we have developed what could be a best-in-class approach, could be one of the very first disease-modifying therapies for these patients.

Now APOL1 kidney disease, for investors, 4 things to highlight. Number one, that we know at least 1 million individuals have the variants in kidney disease and at least 250,000 individuals could benefit from a therapeutic approach. We have developed a small molecule that could be once-a-day dosing, best-in-class in that we have a dual mechanistic approach, which has not been described by other compounds in development. We recently, if you will, have highlighted our Phase I data. We presented that data last year's ASN event, which is the large kidney conference that gives us the confidence as we continue to enroll our Phase II and plan to unveil data that could be the first time clinical proof of concept is demonstrated in that broad AMKD patient population in Q1 '26. And the evolving, if you will, regulatory landscape of understanding how kidney disease can have a path to registration and a pivotal study, we plan to enact planning on that on the heels of that data next year.

Now we know a lot more about the patients who have APOL1 kidney disease. It disproportionately affects the black community. We know that they're younger. So those that have the variants that cause disease, they're younger than those that have -- do not have the variant. About -- what they do is they clinically present about, if you will, below the age of 50. Kidney disease is often associated with an elderly population. These patients are younger. We know that they have, if you will, more aggressive disease in the sense that they progress to end-stage renal disease as well as into dialysis about 10 years earlier, the non-APOL1 kidney disease patients. And we know, unfortunately, that the current therapies are not effective in this patient population.

So they're younger, they progress faster, and they're not responding to standard of care, and there are no approved therapies for these patients. So us and other sponsors hope to develop what could be the very first disease-modifying therapies for these patients by treating the underlying cause itself, which is to target APOL1 gene directly.

Now at least 250,000 individuals have enough proteinuric disease where they can benefit from a therapy. What we know now is about 60% of those do not have diabetes and about 40% have diabetes. That's important to highlight in the sense that we intend to be the first group that can show clinical proof of concept in patients that don't have diabetes as well as those that do not, really being able to, if you will, address the majority of patients who can benefit from this approach. Fortunately, for us, clinical proof of concept has been demonstrated on APOL1, albeit in a smaller, more defined population called FSGS, which is great because the target itself has been derisked in terms of the biology. What we intend to show is a broader set of patients who might benefit from that therapeutic approach in Q1 2026.

Now diving into the mechanism itself. We know what APOL1 is doing in the sense that it is -- there are 2 variants, G1 and G2, there's only 2 that can cause the disease and causes a toxic gain of function that ultimately results in APOL1 itself acting as a channel or a pore in the podocyte. Now the podocytes are found in the glomerulus, which is the primary filtering unit of the kidney. And what you can see here in the right panel is the fact that you have an overexpression of those pores, so you have many more of them. And ultimately, you have them ungated in a sense that they are actually just creating holes in those particular podocytes, resulting eventually into that nephrotoxicity or the toxicity and killing and damaging those cells.

So at minimum, you want to be able to block those ungated channels from acting in that way, resulting into that toxicity. What we learned about 829 is we also have a dual mechanism, which differentiates itself from all other compounds that are in development in the sense that we're not just blocking the pore, we are also disrupting the assembly of those pores from forming to begin with. And that's important because the turnover of that particular, if you will, protein is high. It's under an hour, and it highly turns over in the podocyte such that you not only want to block it, you want to be able to disrupt the assembly of it. And what we know now is the fact that no other compound that's been described to date in the literature has the same type of effect, and that allows us to not only differentiate in terms of the molecule, but address the broader population, including those that may have diabetes.

Now our Phase I results, the key takeaways are the fact that we have linear dose proportionality. You can see the 2 panels here that show our SAD and MAD data. We have the half-life of about 15 hours. This will be a once-a-day drug. We also had high confidence in the dose that we were selecting to bring that into our Phase II that basically describes that we can at least get to an EC90. That's important because not only do you want to hit this target as hard as possible, you want to be able to disrupt the assembly of that pour. And that particular data set allows us to give us the confidence exactly what we wanted to see in order to be able to start our Phase II.

Now our Phase II study is called HORIZON, ongoing enrollment, on track to deliver the data in Q1 2026. What we are doing in this case is we are screening for the genotype. So you have to have one of the G1 or the G2 variants that cause this disease. We treat for 3 months and then we follow for a month. And what we're trying to do is, again, be able to try to expand the population to a broader set of patients that allows us to the first time potentially show that clinical proof of concept.

Now what we want to do in terms of the minimum data in terms of the threshold that we are saying could be clinically significant and would demonstrate clinical proof of concept in this patient population is a 30% reduction of UACR relative to the baseline. This would be significant in the first time being able to show that type of clinical effect. And if we're able to do that, and we see the data that we would like to see in Q1 2026, we can start and initiate our planning for Phase IIb, if you will, which would be more of a registrational study next year as well.

Now why -- now in terms of our next program, MZE782. Now what we're really focused on here is being able to develop this particular opportunity for PKU patients. And so what we know in terms of PKU to date is that, untreated, basically, patients resolve an ability to have the multiple, if you will, issues, including these neurocognitive, neuropsychiatric detriments that really affects individuals across the board. What we know today is that the therapies that have been approved even to date are not able to address the entire spectrum of disease. We know at least 60% of the patients are not on therapy to date because they're inadequate. And our approach, which could be the first oral substrate reduction therapy approach could work across the entire spectrum of the PKU patients.

Now in terms of PKU itself, we know that this could have broad impact. We just had data in September to demonstrate that we could have proof of concept, if you will, a best-in-class molecule in being able to have substrate reduction in PKU. We know there's at least 58,000 individuals across the world who could benefit from this type of approach. We showed best-in-class principles relative to any other molecule that had been described in the literature to date, and we plan to initiate Phase II in 2026.

Now based off of what we knew from the existing literature, we knew that what you want to be able to do clinically is at minimum be able to reduce plasma Phe in the blood, which is the, if you will, the neutral amino acid that causes disease as that is the regulatory endpoint for approval for PKU patients. Now in healthy volunteers, what you want to be able to do is be able to show an increase in excretion of urinary Phe as it translates very well in terms of the plasma Phe reduction in patients.

So what we're aiming to do is be able to show that effect in the healthy volunteers and compare that to previous studies that have been described in clinical development to show that we could be potentially best-in-class. And if we can do that, we would be able to show that we have the ability to work across the entire spectrum of disease.

Now our data in Phase I supported advancement into Phase II. And so the key takeaways is that we had a larger healthy volunteer study. We again showed very strong, if you will, PK, PD response. We had a safety tolerability profile that really supported the ability to forward this into Phase II. And we showed, if you will, best-in-class principles, in particular in the urinary Phe biomarker relative to anything that had been described to date.

Now to highlight that, a previous study had shown at a particular dose that they had a tenfold increase in urinary Phe relative to the baseline. That's the dash line that you see here on the slide. So that was where the original benchmark was in terms of where you wanted to show some level of pharmacodynamic effect.

What we had shown here, of course, is that we had exceeded that tremendously in the ability to do that, demonstrating that we have best-in-class opportunity here, and the ability to maximize, if you will, the ability to reduce the plasma Phe such that we can show clinical benefit eventually in PKU patients. So they had shown a tenfold increase in urinary Phe previously. We showed, on average, a 40-fold increase in urinary Phe, demonstrating the ability that we can maximize the opportunity across the entire spectrum of disease.

Now we're equally excited about this concept of SLC6A19 in chronic kidney disease. We were the first group to identify the genetic relationship of looking at this particular gene in the context of chronic kidney disease. And what we had shown is that certain individuals were really protected from developing end-stage renal disease and even transitioning into dialysis if they had a particular variant of SLC6A19. And we used that information to really be able to address what could be, if you will, the next anchor in terms of the way that we can develop chronic kidney disease medicines going forward.

Now in terms of what we were able to do with chronic kidney disease, we see this opportunity as potentially being the next SGLT2, which has become an important anchor in the way that kidney disease is treated today. We have demonstrated in vivo proof of concept, demonstrating that we can be superior in particular models relative to SGLT2. We have been able to demonstrate genetic analyses that really support our work. And ultimately, in our proof of mechanism study and our Phase I that we presented in September, we were the first group to show clinical proof of mechanism that this potentially could have a renal protective effect, which is indicative of all approved chronic kidney disease medicines to date. So based off of the totality of that data, we are progressing in 2026, starting with a clinical proof-of-concept study in chronic kidney disease.

Now this data here was an important milestone for the program in the sense that it demonstrated for the very first time in vivo proof of concept in a kidney model and put that into context with SGLT2. And what you can see here in terms of the graphs, one is the ability to lower proteinuria in that particular model. And you can see in the particular bars in green, our agent. And in the orange, you can see that compared to an SGLT2. And even at an EC50 relative to what we were doing in terms of exposure, we were able to not only reduce proteinuria better than SGLT2, we were also able to reduce, if you will, injury markers in the kidney called KIM-1 and others that demonstrate that we could have, if you will, even a more superior efficacy in these particular models.

Moreover, what's interesting is that there's a potential for combining with SGLT2s. And you can see that in the black bars on both of the graphs as well, where you not only see this additive effect in the proximal tubule injury markers like KIM-1, but you're also able to see that basically get to a normal state in the proteinuria when you combine SLC inhibition with SGLT2 inhibition. Really remarkable in the sense that we might be able to really have an impact and ultimately improving kidney health.

Now what was interesting to us based off the genetic analysis that we had been sharing with nephrologists as well as some of the in vivo proof-of-concept data that we had shown to particular nephrologists, what they encouraged us before we started our Phase I was to be able to collect the biomarker serum creatinine such that you can calculate eGFR, which is a measurement of kidney function. And why did they tell us to do that? Well, they told us to do that because what they knew across all medicines that have been approved for chronic kidney disease, whether it be a RAS, an ARB, SGLT2, even NMAs, which is a new modality, if you will, for kidney disease, all of them had shown this phenomenon of having an initial eGFR dip, which is represented here in the graph, which was indicative of long-term renal protection, especially when you compare that to placebo.

So it's a little bit counterintuitive to see an eGFR dip initially, but what that did is really be able to slower the rate of decline following treatment over time, which was really indicative of the renal protection. And they saw that -- nephrologists knew that they had seen that across multiple different classes of agents, including SGLT2. So if we were able to see that in our Phase I data even in healthy volunteers on top of our genetics, our in vivo proof-of-concept data, this would really be able to reinforce what we had seen in that particular analysis.

So the first time, we are the first company to demonstrate the ability to show this eGFR dip, indicating that this may have renal protective effects in patients when we bring this into the clinical proof-of-concept study.

Now what you see here in terms of the graph is we see that there is a dose response relationship here on the right graph, really being able to see that type of drug effect. And to ensure that we see that drug effect, you see that on the left graph, which is when we pull the healthy volunteers off of drug, you can see that the eGFR bounces back up, indicative of the fact that this truly is due to the therapy itself. And the last thing is what you see in that orange bar is we compare that and contextualize it to SGLT2 represented by Empa here. And you can see that the range of what we're doing in terms of eGFR dip is quite in line with SGLT2. So one, not only do we see a dose response; two, we know that it is due to the treatment itself based off of the fact that we pulled the individuals off of the treatment and you still see the effect; and lastly, we see that in about the same magnitude as SGLT2, again, reinforcing the fact that this could have renal protective effects as we bring this particular therapy into our proof-of-concept study.

Now as you think about what we're planning for 2026, this will be a big year for us. We have our initial clinical proof-of-concept data in APOL1-mediated kidney disease. We could be the very first company to demonstrate clinical proof of concept in a broad set of patients with or without diabetes. The second thing to highlight is we've put the bar in terms of success at 30% reduction of proteinuria relative to baseline. And if we're able to do that, we could be initiating planning for a registrational Phase IIb/III study next year.

Now in terms of some of the things to highlight for investors, one is that we do have the product candidates really be able to transform the way that kidney disease is treated today, starting with MZE829 and the ability to show clinical proof of concept in that broad patient population, trying to show that we can show that 30% reduction in uACR. We also are initiating off of 782 and the strength of the data in September the ability to start 2 Phase IIs, one in PKU, the second in CKD. And the last is the fact that we've been able to demonstrate, if you will, the ability to not only be one of the few IPOs this year, we also had a financing in September that enables us to be well capitalized to deliver on all of our particular data catalysts that I've been highlighting throughout the presentation.

So our vision at Maze Therapeutics is to become one of the next-generation precision medicine companies that allows us to develop small molecules in kidney and metabolic diseases. We have the product candidates, we have the team, and we are well capitalized to deliver on all of our particular milestones that have been highlighted to date. But not only do we have the ability to develop those 2 molecules, we are just getting started. Our scientists are hard at work back in San Francisco being able to develop additional small molecules in both kidney and metabolic diseases that allow us to provide not only additional pipeline that we may develop ourselves, just like the 2 programs I described today, but could be leveraged, if you will, for partnerships as we've been able to demonstrate even with our Pompe program with Shionogi, demonstrating that we can not only be able to advance the programs ourselves, but also through partnership.

Now with that said, I'd like to thank you all for attending today, I'd like to thank our team back in San Francisco, and thank you to the Jefferies team for having us at this year's conference.

Good afternoon, everyone, and welcome to TD Cowen's I&I Summit. My name is Tyler Van Buren, senior biotech analyst at TD Cowen. For this next session, we are very excited to have a fireside chat with Maze Therapeutics. And from Maze, we have Jason Coloma, the Chief Executive Officer. Jason, thank you very much for joining me. It's a pleasure to have you.

Yes. Thanks, Tyler. Thanks for having me.

And for those of you in the audience, if you have questions, you can submit them through the web portal, and we'll do our best to get them asked. But we'll go ahead and get right into it since we have 20-plus minutes.

Jason, I thought it would be good with ASN just passing, maybe discussing some brief highlights or takeaways from the conference or observations that you guys noted?

Yes, absolutely. So for those not familiar with our story, we develop small molecule precision medicines, primarily focused on kidney disease. And so last week was the American Society of Nephrology, which is the largest conference dedicated to kidney therapies as well as for overall kidney research. And so we had 7 abstracts, Tyler, they are accepted, which was great, a great showing for our team and our scientists and had presentations as well.

But probably 3 things to highlight in terms of what we were able to kind of highlight, especially for APOL1-mediated kidney disease. The first thing is that we had published on a genotyping study. And one of the key takeaways there was it really did reinforce what we were seeing in terms of the literature on the percentage of people that had the variants. And so what we reported was the previous literature had described in a general population, around 13% of the individuals that had been genotype had the pathogenic variants that caused APOL1 kidney disease. In our study, because it was enriched for kidney disease patients, we had about 18%. So a little bit higher than the literature, but mainly explained by the fact that we had enriched for kidney disease. So I think suffice to say, 13% to 18% in that range having the variants that end up causing disease, just reinforcing the literature.

The other thing that was important from that study is that sort of the breakdown. We've always talked about 1 million individuals in the U.S. have the variants, at least 250,000 individuals could benefit from a therapy and the split between those that have diabetes and don't have diabetes. We reinforced the literature in the sense that about 40% of the individuals do have diabetes. So it's about nondiabetes, 60%; diabetics, 40% so also reinforcing the literature.

The second thing that we did is there's been some discussion, I think, in the sense that should we -- in a broad AMKD patient population with or without diabetes, should there be an effect in diabetes because there's been some previous literature talking about the, let's say, decreased incidence of disease with those with diabetes. But we and other sponsors as well have been starting to educate the field and investors is it's not about incidents. It's about progression to disease. And what we published on was the fact that those that have progressive disease, it is aggressive and progresses at about the same rate with or without diabetes. So conceptually, if a therapy should have an impact whether or not an individual has diabetes or not.

And then the other thing that we had previously published on related to that was we had found a protective variant, N264K. And just to remind folks, that protective variant protected individuals whether they had diabetes or not, sort of illustrating if we truly are, us as well as some of the other sponsors, mimicking the protective variant, which is the hypothesis, we should see an effect in either population. So what's nice to see at ASN last week was another group, Natera, reinforced that finding in terms of N264K, which really demonstrated, again, this concept that it should work in both patient populations.

And then the last thing that's probably important is, I think prevailing sort of mechanistic understanding of the hypothesis of how APOL1 causes disease. So just to first off say, the genetics were known for a very long time. No one really understood function, us and other sponsors worked really hard to better understand that the pathogenic variants that cause disease cause an aberrant, if you will, function of this poor channel in the podocyte, ultimately leading to nephrotoxicity. And we want to be able to, if you will, block that pore, if not, if you will, disrupt the assembly of that pore ever to be formed to begin with. And we published some additional data to demonstrate and reinforce that the mechanism by which APOL1 is causing disease is truly that it is acting as a poring channel. And so that an inhibitor of that function should continue to see benefit in terms of the clinic. And that's important because there were dozens of different hypotheses on how APOL1 could end up deciding disease. But now we have a better understanding of that.

So it's just really nice to see everyone that's working in APOL1, not just us, sort of the common understanding about at least 250,000 individuals could benefit. You should expect to see a signal whether you have diabetes or not. And then the prevailing hypothesis that it's acting as an aberrant pore channel seems to be sort of coalescing around that overall scientific hypothesis. So it's really great to see. And obviously, we're really happy to see all of those sort of get reinforced and accepted and being able to talk about the larger community at ASN.

That's great. Very interesting. Thank you very much for recapping that for us. Maybe we'll get straight into the MZE829 program, your APOL1 inhibitor in development for AMKD. Can you just give a high-level overview of that compound and how it compares to others, obviously, in particular, inaxaplin in terms of its properties?

Yes. So we are developing a small molecule. We have a dual mechanism, APOL1 inhibitor, and we're advancing it for the potential treatment for what's called APOL1-mediated kidney disease. Again, the idea is not only the fact that you might be able to inhibit the pore itself, but also what we call a dual mechanism is that we also have, if you will, differentiated biology relative to other compounds in the literature and the fact that we not only block the pore, but we also disrupt the assembly of that pore to begin with.

And that ultimately, in terms of differentiation, leads us to at least in in vivo proof-of-concept models in kidney models, we showed that we had a significant potency advantage relative to any molecule that had been described in the literature before and then further reinforced by other work that we had done. So the idea, of course, is that previous compounds have showed clinical proof of concept in FSGS patients previously. That's great because we know that the mechanism can demonstrate clinical benefit.

Our goal with the Q1 '26 readout, Tyler, is, number one, what we want to be able to do is be one of the first companies or if not the first company to demonstrate clinical proof of concept in broad AMKD with or without diabetes. We've set that minimum threshold in terms of seeing a signal at least 30% reduction of proteinuria relative to baseline. Happy to describe how we landed on that number. And then third, what we've been communicating to folks at ASN and this week as well is if we see, if you will, a signal and show clinical proof of concept in that patient population in Q1 '26, we could be in a position to potentially start planning for a Phase IIb/III study, much similar to obviously, what Vertex had done with their data a couple of years ago.

That's great. And just to be clear, with the Q1 update, so we're going to get data from both the diabetes and the nondiabetes AKD (sic) [ AMKD ] cohorts?

Yes. When we set up on HORIZON, our goal was because it had shown clinical proof of concept in a rare population, we'd like to show for the first time and hopefully, broad AMKD with or without diabetes. So that was the intent. And if we see that signal at 30% reduction, to us, that signifies that we have demonstrated clinical proof of concept as we've talked about that with nephrologists. And I think based off of the strength of that data, should we get there, the idea is then we can start initiating planning for what a, if you will, IIb/III design could look like.

Got it. And have you said approximately how many patients we should expect per cohort?

Yes. I think for us, in terms of the readout, people should expect an order of magnitude of what Vertex described in the first study. So that's why we describe it as an initial set of data that should allow us to -- if we see that signal in that percentage of patients, we would hope to, again, think about that 30% number, look at the totality of the data. And if we see that signal, then we can ideally push as fast as possible to something that could look more like a registrational study.

Got it. Is it possible that we see FSGS patients in Q1? Or do you think we need to wait for a pure FSGS update later?

Yes. I think our goal for HORIZON in the first readout is to show clinical PoC in that broad AMKD patient population. We are enrolling patients that have FSGS. And of course, if we have that. But just to set expectations, our goal is to focus really the investors on we could be the first company to show PoC in that broad patient population. And if we do, that gets us from a rare population of FSGS to a broader set of patients who can benefit.

Got it. And when you were recapping some of the ASN highlights, you talked about the variant being protective both in diabetes and in the broader AKD (sic) [ AMKD ] population. So just as we think about proteinuria reductions, is that variant equally protective in diabetes versus nondiabetic patients? Or is there still somewhat of a difference that might manifest in the proteinuria reductions?

Yes. I think what's interesting from the genetics, it progresses at the same rate and equally as aggressive. So that's interesting that whether you have diabetes or not. So conceptually, again, if you could intervene with a medicine, you should see an effect in both. And then the N264K, specifically the protective the variant that you just mentioned, yes, it does at the same rate, protect those with or without diabetes. So we know that's nice that we have the natural, if you will, genetic experiment showing that.

And that's why we -- for all of our programs, Tyler, we do look at not just pathogenic variants, we look for protective variants as it might allow us to better understand function and mechanism. And then as we design our small molecules as we did with this program as well as for SLC6A19, if we can mimic that protective effect, that should allow us to reap that benefit you see clinically. And so that is the goal here. And I think what's nice is we'll hopefully try to demonstrate that in Q1 '26.

Okay. And can you just elaborate on how you picked that 30% threshold, especially with the 48% with inaxaplin out there in that very small FSGS data set and your confidence in potentially being able to achieve significant proteinuria reductions?

Yes. So the previous literature described a rare patient population. This is broad AMKD, so more heterogeneous in disease, both with or without diabetes. And the 30%, probably 4 things to highlight for investors. 30% is a number that if you look at some of the analyses, it correlates very well to eGFR slope, which is most likely, just like in IgAN, the approvable pathway for particular proteinuric diseases.

The second is in the guidelines, so KDIGO, 30% is being clinical significance. uACR specifically is highlighted as what you should be measuring for this more moderate disease.

The third is if you just look historically, things like other classes like SGLT2, they had 30% as their bar eventually sort of translating to, if you will, outcome data and eGFR benefit.

And the last is, which I'm not sure if you saw this, Tyler, but PARASOL, which is a group that -- at a University of Michigan, which was instrumental in working with the agency to help define for FSGS, that rare patient population that proteinuria alone should be the approvable endpoint. The PARASOL group has taken on 2 kidney indications, one of them PMN and the other APOL1-mediated kidney disease. And in that press release, they did describe a couple of papers that they even referenced uACRs the proteinuric measurement that people should be using and the 30% number as a good minimal threshold to see clinical benefit for patients.

So those are the 4 things we highlight as the rationale for 30%. And I think that allows us to at least see where the bar should be. And if we exceed that, of course, that would be great.

Great. And preclinically, what have you seen with 829 that gives you confidence that you could have significant reductions in proteinuria in patients?

Yes. So we previously published, especially in particular, the BAC transgenic mouse model, which has high levels of proteinuria. We've done that both in the chronic setting as well as preventive and shown that we can get to doses that get to in the [indiscernible] that effectively can get into that range of at least a 30% reduction. So I think in terms of translation, we've seen that in in vitro models. We've seen that in multiple models. But more importantly, we saw that in the BAC transgenic mouse data.

And we know why that might be translatable is because we did -- we were able to synthesize inaxaplin, bring that into that model based off exposure of where their clinical dose was, and we saw a proteinuria reduction. So we already know that they showed clinical proof of concept in patients. So very good confidence in the sense of this being a translatable model where we should see at least some level of proteinuria reduction as you move into the clinical trial.

That's great. As we see this initial data and as we think about the development plan moving forward, how do you think about baseline proteinuria? In other words, does baseline proteinuria have a significant impact on the percent reduction that you can achieve in patients? If people have higher baseline proteinuria, do you have a greater potential to have a more significant reduction? Or is it independent of baseline proteinuria? Curious to get your latest thoughts there.

Yes. I think that -- I mean, at least for APOL1-mediated kidney disease, we saw that in highly proteinuric patients, which was published in the New England paper is that you had that significant drop, right? But those are highly proteinuric patients. We know that a group hasn't described at least publicly clinical proof of concept and being able to show what the 30% could be significant. But as we talk to investigators and we talk to nephrologists, setting the bar there across diabetes and nondiabetic patients, to them, they would see that as clinical proof of concept and allow us to think about justifying the progression into a more registrational study.

Got it. And because the broader patient population will have obviously lower baseline proteinuria than FSGS, it's reasonable to assume that you'll have maybe lower proteinuria reductions in that population than FSGS?

Correct. Yes.

Okay. All right. Okay. Yes. So I guess that's going to be somewhat challenging, right? Because if you look at the inaxaplin data, it's going to be in a different patient population than what you guys are showing, right? So we won't be able to necessarily compare it head-to-head, but obviously, clinical proof of concept to have the first data in the broader patient population will be really special, right, to...

Yes, I think it will be incredibly impactful because then you get closer to that at least with 250,000 instead of the, say, 10,000 patients that have FSGS with APOL1 variants, right? So I think for us, I think that was always the goal because if we -- they've already shown clinical proof of concept in the rare, we'll form our sort of like data. And they will also have data. They've also guided, Tyler, of course, that they should have data in a broad AMKD patient population by the end of '26. So I think that also will help inform that.

And I think as we think about the overall opportunity, what we always guide investors is that at least 250,000 individuals, this is at least the same size as the IgAN population, if not bigger. And then we've also described, of course, that in terms of how to think about this, there's different -- multiple agents would probably be necessary in order to benefit this patient population. And we've seen that at least in other analogs that we described to investors like ATTR-CM, where you saw diagnostic testing being set, multiple participants being able to ultimately make an impact for patients. And we're even seeing that in kidney, of course. IgAN has multiple participants that allow us to kind of think about differentiating not just on a molecule level, but also the patient populations they can address.

Great. And I have a follow-up question on the market in a second. But just -- so just to be clear, the mechanistic difference from 829 to inaxaplin, it's twofold. It's the increased potency and then also mechanistically the fact that you're inhibiting pore assembly. And is that in...

In addition to blocking the pore?

Yes.

Yes.

And the pore assembly inhibition, which is obviously a novel addition to your mechanism of action, is that -- you observed that in vitro and you guys are confident that that's happening in vivo?

Yes. And so the idea there, of course, is that something that's probably important to point out, APOL1 in the podocyte rapidly turns over. It's less than an hour. I think some people report it like for every 42 minutes it's turning over. So the fact that you want to be able to intervene at least in vivo in the sense of like not just inhibiting the pore because it's a little bit of whack-a-mole. You also want to just make sure and ideally that you're disrupting the assembly such that they never insert into the membrane of the podocytes, right? So conceptually, that makes sense to us. And I think partially helps at least to us and others explain why we are more potent in particular, in vivo systems.

Great. And 250,000 patients across the broader AKD (sic) [ AMKD ] landscape with diabetics, nondiabetics, FSGS, I mean that's clearly a very sizable population, I guess, technically almost exiting the orphan drug range. But you mentioned the 13% to 18% earlier. Was that -- that's elevated, right? I think you previously said it was more like 11% incidence of AKD (sic) [ AMKD ].

No, no. If you look at our previous -- 13% of the general population in the black community have the variants that could cause disease. That was a sort of overall number. Our genotyping study and the epidemiology around it was a little bit higher, 18%, but partially explained, I think, by the fact that we were looking specifically at those that had kidney disease. It wasn't a general population. So I think in that range, it is in there. And I think how do we get to 250,000? It's those that have proteinuric disease at a certain level that could be best, if you will, served with a potential therapy.

Got it. All right. We're up on time here, but I have to ask about 782. You guys had great data with 782, SLC6A19, as I like to call it. And just can you touch on both PKU, I mean, what you saw there, clearly, pretty striking relative to benchmarks. And then also CKD, which was a pretty interesting surprise with eGFR dip and just your conviction that mechanistically eGFR dip could lead to real efficacy in CKD or the broader almost SGLT2-like indications?

Yes. I think for the 782, which targets a SLC6A19, what we -- in terms of PKU, this is a potential therapy that can work across the entire spectrum of disease, in particular, the severe classical form which is the majority of the patients. Over 60% of the patients have severe classical form of disease, clinically defined as greater than 1,200 micromolar of Phe in their plasma. And the idea is that we don't have a therapy that relies on, if you will, the older agents that have been approved, residual enzyme or, if you will, co-factors to allow that enzyme to work. The concept very straightforwardly was, hey, why don't we just get rid of the toxic substrate itself? And if we can safely do that, we can potentially affect all of the patients, including those that aren't served by -- that are defined as severe classical.

Because what we want to do, Tyler, clinically is that 1,200 number, you want to try to reduce them below 360 micromolars. And why is that important is that, that's when you can start to remove people off of their onerous medical diet. And for some of the children, that means that you reduce the probability of seeing these neuropsychiatric or neurodevelopmental issues that happen when the plasma Phe gets into the brain, right? So we had an agent. What we reported is that there was another investigator who had shown clinical proof of concept of inhibiting the target and having an impact to plasma Phe. And they also showed that a noninvasive urinary biomarker plasma Phe, if you will, excretion map very closely to that plasma Phe reduction.

And why is plasma Phe important? That's the approvable endpoint for PKU patients and regulatory path, right? And what we showed in the study was that we had clearly best-in-class properties and the ability to excrete that urinary Phe. We knew that the other investigator had shown at one dose tenfold increase, if you will, that urinary Phe biomarker that translated in the patients on average, 44% reduction of plasma Phe. And we significantly cleared that over 42x in terms of our urinary Phe excretion.

And that gives us the confidence, of course, to now run our patient study in PKU, where we can exhibit best-in-class in the ability to reduce plasma Phe and ideally convert more patients below that 360 number. Because if we look at their data that they reported, of the 19 patients that they reported, only 6 of them got below the 360 number. So if we can do better than that, that would be excellent.

On CKD, what we did, we were -- we generated the first genetic association to the target and being able to have an improvement on kidney health. So we're equally excited about that concept in kidney disease. So we not only demonstrated genetics, we validated some academic work in terms of target validation. We also showed for the first time in vivo proof of concept in the kidney model that allowed us to see and put contextualize it to SGLT2s, which are an important anchor in kidney health. And we show that we can not only work with it in an additive sense, but we also can do a little bit better in the sense of lowering proteinuria in that model.

So we had genetics, we had target validation, we had in vivo PoC. And so what we did based off of that nephrologists gave us some feedback before we started the Phase I, which was what you should do is collect serum creatinine so that you can calculate eGFR as an exploratory biomarker because what we know, and this is important for the investors to understand, is that all approved CKD agents or therapies show this -- what's called this eGFR dip as it's indicative of renal protection in the long term. So we would like to actually see that in your -- even in your healthy volunteer study because that reinforces all of your great genetic preclinical data and gives us even more confidence to do the PoC study.

So suffice to say, we did show that. We showed eGFR dip. We showed that it was dose -- there was a dose relationship. We showed that it was true. Secondarily, we showed it was true to the agent because when we pulled them off the agent, the eGFR bounced back up. And then we also showed that it was in line where SGLT2s were in terms of seeing that eGFR dip as well.

So now with that clinical proof of mechanism data plus the work that I described earlier, I think there's a lot of excitement in the nephrology community, can we try to show clinical proof of concept in patients. And the way we've positioned that from a target product profile is there's a lot of people who aren't responding to current standard of care, including SGLT2s, and we know up to 25% of individuals are not responding adequately to SGLT2s. And we know that the 1-year discontinuation rates on SGLT2s is very high. Tyler, it's like 30%, 40%, partly due to the complications like hypoglycemia and urinary tract infections. So minimally, as a [ TPP ], if we can be just as efficacious as SGLT2, but not have the liabilities, that could be very interesting. And then moreover, if we can actually just exceed that in terms of efficacy, you're talking about something that potentially can change the way that kidney disease patients are treated today.

Yes. It could be a mega blockbuster, but that's...

That's your words.

Wonderful. Well, we're way over time, so we better wrap up. But Jason, thank you so much for the fantastic discussion, and thanks to everyone for logging in.

Yes. Thank you, Tyler. Thank you, everyone.

Thank you for joining Guggenheim's 2025 Healthcare Innovations Conference. I am Debjit, one of the therapeutic analysts and are my privilege to host our next presenting company, Maze Therapeutics. And joining us from Maze is Jason Coloma, the CEO. I don't think people are completely up to speed with Maze as yet, given your 20 -- class; of 2025 IPO. A couple of minutes on the platform.

Yes. Well, first of all, thank you for having us this year, Debjit, and thank you for the Guggenheim team and all their support. At Maze Therapeutics, we are using advanced genetic techniques to develop small molecule precision medicines for both kidney as well as metabolic diseases.

We're mid-stage in the sense that we have multiple Phase II programs. And we've been working very hard, as you know, to be one of the first companies this year to go public. We had a recent financing in September, which Guggenheim was supportive of. And that's allowing us to develop multiple small molecule approaches as we have even data coming in Q1 2026 and our lead program for APOL1-mediated kidney disease.

So let's talk about the 829 program. Obviously, there is proof of concept, proof of biology from your competitor. What would be a good outcome in AMKD and diabetic AMKD because that's the one segment that we haven't seen.

Yes. So for those that aren't aware, first off, APOL1-mediated kidney disease, there's about 1 million individuals in the U.S. that have a variant that can cause toxic gain of function or eventually toxicity in the kidney. Unfortunately, there are no approved medicines for this disease. There's about 250,000 individuals who could benefit from a therapy and us as well as a few other companies are developing what hopefully could be the first disease-modifying therapy for those patients.

And so for us, what defines success for Q1 '26, 3 things that I'd like to highlight. Number one is that we intend or hope to be the first company to describe clinical proof of concept in broad NKDs for a broader patient population. A previous, if you will, sponsor had been approved a clinical proof of concept in a rare form of disease called FSGS in the kidney. What we hope to do is be able to be the first company to describe clinical proof of concept in a broader patient population, which gets closer to that 250,000 number that I mentioned earlier.

The second thing in terms of defining success around that particular outcome that we define that as in terms of clinical significance and clinical proof of concept to be 30% reduction of proteinuria or protein relative to baseline. That would be not only clinical significant, but show for the first time clinical proof of concept in this patient population.

And then the third thing to highlight is that if we are successful in the sense of generating that data, that will allow us, if we see the signal of a significant number of patients to allow us to initiate planning what could be a Phase IIb/III registrational study and try to move as aggressively as possible to the patients.

So the first quarter data is just going to be AMKD in -- sorry, APOL1 -- diabetic AMKD and AMKD, not FSGS.

Right. Our goal from the very beginning with Horizon was to clinical proof of concept has been demonstrated in FSGS. Our goal as we designed that study was really to -- can we broaden this out to allow for even more patients with or without diabetes because our genetics as well as other work that has been done to really support that we should see activity in those particular patients, and we want to be able to try to prove that out in the clinic.

Got it. And the FSGS data, would that -- should be expected sometime middle of the year, later in the year? And are you sort of thinking about any expedited path to market in FSGS given the findings from the PARASOL group?

Yes. We haven't guided specifically on FSGS yet because we want to be able to really focus investors on the Q1 '26 readout. As we be able to generate that data and have plans to, as I said, be able to think about advancing that potentially in a registrational study, we'll give further guidance on FSGS.

Got it. Let's talk about the disease. The rate of progression from diagnosis to end-stage renal disease is very, very quick. So if you -- is 30% a target which can stop the progression of the disease, 30% reduction in proteinuria? Or do you need a deeper reduction? And sort of that can you outline how big that -- the 2 segments are, the diabetic AMKD versus the AMKD?

Yes. So I think one of the things that we were able to -- last week was the, if you will, the really important kidney conference, American Society of Nephrology was just last week. We had 7 posters and presentations accepted sort of highlighting what we're trying to do in kidney disease. And one of the posters we had in terms of the genetics, we further sort of described the ability to look at the disease progression with those and without diabetes.

And what we found, Debjit, was that people that end up developing kidney disease that have a APOL1 variant as well as start to progress, they progress aggressively and equally, whether they have diabetes or not. And why is that important is that a therapy, as you say, could potentially intervene if we can catch it early enough such that it could be truly disease modified. That's the goal of us as well as other sponsors working in APOL1 kidney disease.

So the idea for us is that we could potentially address patients that are really progressing aggressively as you say. In terms of the numbers, right, so I mentioned 250,000 individuals could benefit from a therapy. And we also had a poster at ASN that really kind of described and reinforce the epidemiology in the literature. And one of those things that we did is to look at, well, what percentage roughly have -- of people that have kidney disease have the variant. And two, of those people that have it, what percentage of them have diabetes.

So the first number is that we reinforced the literature in the sense that it's been previously described 13% of all African-Americans had the variants that cause disease. In our study, it was enriched because they had kidney disease, and they -- it was about 18% that had the variants that caused the disease, which in that range is probably safe to say 13% to 18% of the population would have the variant.

And then we also reinforced this number that about 60% of the 250,000 don't have diabetes and about 40% do, okay? So I think it's really a better understanding. What this highlights is a better understanding of the addressable patient population. Number two, that -- the split between 60% nondiabetic and 40% diabetic. And then thirdly, you have this ability that we showed that there should be an effect in both diabetic as well as nondiabetic patients.

Got it. Now these patients have to be genotyped.

Correct.

Have you had challenges enrolling the study? Or given Vertex has been in the market evangelizing the target, a large number of patients have now been genotyped.

I think we're learning a lot in the way that one has to engage with the community, in particular, genotyping and serving the community at large. Awareness is increasing, as you said, Debjit, by Vertex in front of us as well as now AstraZeneca.

So I think it's becoming more prevalent sort of the knowledge base. And that's allowing us to at least be able to engage the community. I go back to ASN, right? So I think just awareness since we just had the kidney conference, which was -- if you went to the kidney conference 5 years ago, there were like maybe 2 or 3 poster presentations on APOL1-mediated kidney disease.

This year, there was 39, right? So I think the awareness is there, and that's allowing us and other sponsors to be able to engage the community and then be able to enroll these studies. So I think that's allowing us to still be on track for the Q1 '26 time line that we said in order to show clinical proof of concept in this broad AMKD patient population.

Got it. And so the driver of the disease is really whether if you carry the scales, G1G2, G1G1 and G2G2, right? Whether you have diabetes or not, the effect size shouldn't be any different.

Yes. So to go back to what causes this. So the genetics, interestingly enough, have been known for a very long time in terms of this relationship of those that had the variants and had kidney disease. So that was understood for over 10, 15 years. No one had really in the scientific community had really understood how are these genes and these variants actually causing disease.

So our work plus other sponsors really elucidated that really this toxic gain of function is APOL1 acting as this variant channel or pore that's really causing eventually this influx of cadium causing the nephrotoxicity. That is the prevailing hypothesis. So all of the molecules being developed are let's just inhibit the pore, let's slam the door shut on this pore, and that should result in better kidney health, right?

So that was the idea is generally like that if we can do that, then we basically can ameliorate disease. Now further, there was sort of a question -- to your specific question is based off of that, we know now that progression of disease, whether or not you have diabetes is aggressive and equally progressive.

And so in theory, if you do -- if this is the underlying mechanism, if you inhibit the core, then you should, in theory, see an effect in both diabetic and nondiabetic. And then what we've defined, as I described earlier, is this 30% number allows us to see clinical significance in both of those patient populations.

Got it. So let's fast forward to middle of next year. And the Phase b first quarter data looks great. How are you thinking about your Phase II/III study? Would it mimic what Vertex is doing, trying to have a dual co-primary endpoint? Or is there a different strategy there?

Yes. I think if we see what we want to see in terms of the signal, so if we see it in broad in KD, 30%, as you mentioned, we would start to think about initiating planning for a IIb/III. I think it's safe to say what we would try to look at is a study that looks very much in that sort of framework, which is maybe a IIb/III type of design that allows us to look at additional doses to do proper dose ranging because if we take a little bit of a step back, what we wanted to do in HORIZON is pick a dose in our Phase II study to pick a dose that we can optimize to look for a signal, it's signal seeking in this broad NKD.

And if we see that signal, we can push forward, right? And so I think a IIb/III design looking at multiple doses. And then maybe just to highlight, one, we'll learn a lot about other sponsors' studies themselves and the regulatory path that allows potentially for accelerated approval. And the second thing that's important, Debjit, to highlight, there's an academic group called PARASOL, which has been out of University of Michigan, which has been very instrumental in helping define endpoints in kidney disease, in particular, for more rare forms like FSGS.

Other companies that are working specifically in FSGS are benefiting from that. And APOL1-mediated kidney disease as well as PMN has been selected by that group to work with other sponsors as well as other organizations, including the agency to look at what the appropriate regulatory path and endpoints are for APOL1-mediated kidney disease.

So why that's important is not only our involvement with that and seeing pathways for it. I think by the time we get to your midpoint '26, a lot of this information from the environment in general will allow us to inform how we think about our particular study. So a much more clearer path of how we think about sort of regulatory and interaction with the agency and other organizations globally.

Got it. And there's been a lot of discussion back and forth on 829 as it relates to inaxaplin. Is 829 more potent than inaxaplin, et cetera? Any thoughts on that?

Yes. I think suffice to say, if we go back to -- so yes, preclinically, we had the benefit of generating data relative to inaxaplin because they had disclosed the structure. But I think more importantly, if we take a step back, an opportunity like this where there are no approved therapies for the indication, there's over at least 250,000 individuals who could benefit from a therapy.

I think that obviously is an area. If we look at comparable markets or analogs like IgAN, other -- if you look at ATTR-CM, if you look at cardio, multiple opportunities will be there because there'll be different ways that you can treat these patients. And so I think the idea that you can have multiple molecules. Clearly, I think the unmet need is high enough for this is going to be a scenario just like what you saw with IgAN as well as for ATTR-CM.

Got it. So let's switch to the second program, 782. You had some interesting healthy volunteer data in September. Help us understand the path you're likely to take because the program can go either in CKD or in PKU makes you sort of a -- puts you in a unique bucket there. Thoughts on that.

Yes. So for those that are not aware, what we did in September is we highlighted data on our program called MZE782, which was designed to target a particular sol transporter that's located in the gut in the kidney called SLC6A19. And what we did is we were able to be able to use some existing data in the literature that had described this target showing clinical proof of concept by a different sponsor in a rare metabolic disease, as you pointed out, in PKU.

Now that's great because they had published so much data and information. We did -- we were able to benchmark exactly where they were to help us describe what data we might be able to generate in Phase I to show that we had best-in-class properties. We had shown that in preclinically relative to them in the sense of being more potent and being able to potentially go to QD dosing, and that played out exactly what we saw in the Phase I.

So with PKU and the opportunity there, what this opportunity could be is that it is an approach that doesn't rely on the residual enzyme present in order to be able to work in particular patients. So what this means is you take a step back is this approach could work across the entire spectrum of PKU patients and doesn't rely on what has been previously done in terms of relying on enzyme to be present.

Now that can change the way that PKU patients are treated. And so because we had other data out there, what we knew is that one of their doses, they had a particular approach that showed clinical proof of concept by being able to demonstrate a reduction in plasma Phe. That is the approvable endpoint for PKU. So that's number one, that's what they did.

Number two is what they did is they validated a noninvasive biomarker, urinary Phe excretion that correlates very well to the approvable endpoint. So it's great that they did that for the field, but we were able to use that and be able to benchmark where we are relative to them.

Now what they showed in their Phase I/II study was that they had a tenfold increase in urinary Phe. That ultimately translated into the patients in the Phase II at a 44% reduction in plasma Phe. So great they showed it. What we're able to do in our Phase I, which I think investors got excited about as well as us and the PKU physicians and the patients is that we showed a fold over 40x fold increase in urinary Phe where theirs was 10, right, demonstrating that we have the ability to potentially be best-in-class.

And we had showed a range of different doses, including QD and BID, where we can get this type of effect. So a lot of flexibility, combined with the tolerability data that we had, really demonstrates for the first time, things we saw preclinically, which we were more potent. We could potentially go to QD and we can more effectively, if you will, have an impact on PKU patients. So that's very clear.

The other thing that where we were really excited about is the concept of inhibiting SLC6A19 in chronic kidney disease.

So different concept, I know. But we were the first group to identify the genetic relationship of inhibiting this target with -- in the context of chronic kidney disease, which was great. We also had done some additional, if you will, preclinical work to validate this, target validation, knockout data. There had been some of the literature from a group at UC San Diego that we validated.

We also were the first group to show in vivo proof of concept on inhibiting the target in a kidney model and compare that to SGLT2, which is an important anchor in kidney treatment today. We did it better than the SGLT2 in that model. And there -- so you had genetics, you had target validation and you had in vivo proof of concept, very encouraging data to say that we might have an impact in kidney health.

And as we started the Phase I, we were interacting with nephrologists with this data, and they suggested to us, this is very promising, but would further reinforce the hypothesis is what you should do is collect serum creatinine, which is a biomarker as a, if you will, exploratory biomarker so that you can calculate eGFR, which is a measurement of kidney function. And if you see what's called an eGFR dip, to us, that would reinforce the entire hypothesis because we all -- what we know is that every chronic kidney disease program that's -- therapy that's been approved shows that, this concept of eGFR dip, whether it be a RAS, an ARB, SGLT2, Bayer's Kerendia, they all show this eGFR dip. So that to us would really reinforce what you're seeing preclinically in the genetics.

And we -- if you see that, we would really like to be part of that proof-of-concept study for the Phase II in patients. So we did that. And what was really nice is that we really showed that for the first time, on-target mechanism of being able to inhibit SLC6A19. We collected that serum creatinine calculated eGFR. And what you see is that the specific phenomenon of eGFR dip that is -- should be indicative of renal protection.

And I think this reinforces all the hard work that our scientists were able to do. And then what this means for the program is that what we're able to do in 2026 is be able to have not just only a PKU Phase II study, but also a CKD study where we could really be able to think about this approach, transforming 2 different diseases, which we're obviously very excited about.

So on the CKD program, this is going to be complementary to SGLT2. What does this bring from both hemodynamic and tubular detox to patients? And what's the best way to think about what a Phase II could look like?

Yes. So for those that are not aware about SGLT2, so SGLT2s were originally developed for diabetes patients. And then eventually, it made its way into chronic kidney disease. It's an important anchor in how kidney disease is being treated today. But unfortunately, there's some liabilities. Not all patients respond, up to 25% of them don't adequately have better kidney health based off of treatment with SGLT2s.

And in addition to that, there are some complications, which force particular patients to be taken off the drug. We pulled some real-world data. And what we can see is that the number of people that have to drop off the drug after 1 year is up to 30% to 40%, which is really high. And then it's partly because of the fact that they have some of these complications, including hypoglycemia and urinary tract infections, which force certain individuals off of the therapy.

Now there lies an opportunity, right? Because clearly, it has the ability to be effective in patients, but has some liabilities. So minimally, if you think about what we could do with our therapy because we are complementary to SGLT2 based off our data, you could think about equivalent efficacy with an SGLT2, but not have the liabilities of the complications of an SGLT2. Minimally, that would be an excellent target product profile.

Maximally, if we have even better efficacy with the safety that I described, one could imagine that this could really change fundamentally the way that patients are treated with chronic kidney disease, okay? So the idea that we could potentially be an important anchor starts to sort of like get really -- not only us excited, but the whole nephrology and kidney community. Now what would a clinical proof-of-concept study look like?

So to us, again, as we think about those that are not responding adequately to standard of care, including SGLT2s, if we see them sort of stabilizing at a certain high level of protein in the urine or proteinuria, one could imagine that we can introduce our agent. And if we see a certain percentage of reduction in proteinuria, say, 30%, that would be clinical proof of concept for the very first time on this mechanism, which would be tremendous.

So you guys have a lot on your plate right now, right? You've got potentially a very large market opportunity in CKD. You've got one rare disease program in PKU. You're competing with some big players in AMKD. What would be -- what would you like to keep yourselves? And what would you like to have a partner help you out with?

Yes. So 2026 is clearly a big year for us. We have a Q1 2026 data in AMKD or APOL1-mediated kidney disease. Again, we'd like to be the first company to show, if you will, clinical proof of concept in that broad AMKD patient population. 30% is our goal in terms of reduction. If we could do that, we'll be in a registrational study.

And I think the other concept in terms of what's really important for 782 is we could have 2 Phase IIs ongoing next year, which end up being -- starting in both PKU and CKD. Now to your specific question, our goal is to really be able to show clinical proof of concept next year. I think if we're able to do that, I think we'll be able to have a lot of different kind of conversations on what's the best way to advance these to patients as fast as possible.

And on the heels of that data, we'll kind of consider what makes sense. But right now, we're focused on clinical proof of concept and execution for all of these data sets. And fortunately, with the IPO early this year as well as our recent financing in September, we have the data and we have the -- I'm sorry, the capital to be able to get through all of these different milestones by ourselves.

Well, looking forward to a terrific 2026. Thank you so much for your time, Jason, and apologies for the travel hassles, but...

So thanks for having me, Debjit.

Hope you guys have fun in London. Thank you so much. Appreciate the time.

Yes. Thank you. All right.