Good day, everyone, and thank you very much for joining us at the H.C. Wainwright Liver Conference. Next up, we have a discussion with Sagimet Biosciences, and joining me from the Sagimet team is Dave Happel, CEO, Thierry Chauche , CFO, Rob D’Urso, SVP of New Products, and Eduardo Martins, CMO. Dave and team, thanks very much for joining us. Now, maybe just to start, for those newer to the story, can you start by giving us a high-level overview of Sagimet and the scientific focus that differentiates your approach in metabolic and inflammatory diseases?
Yeah, thanks, Brandon, and thanks everyone for joining. Really excited to share our progress today. Just by way of background, Sagimet is a clinical-stage biopharmaceutical company. We're developing novel fatty acid synthase inhibitors, that's FASN inhibitors, that target dysfunctional metabolic and fibrotic pathways in diseases resulting from the overproduction or overexpression of the fatty acid palmitate. FASN is a regulator of lipid synthesis and a key enzyme that's implicated in multiple diseases, such as our primary focus in MASH and acne and in certain tumors and oncology that are absolutely dependent on FASN for progression of disease. Denifanstat works by inhibiting overactive FASN, which, if left unchecked, will lead to progressive disease in these underserved disease states. Denifanstat targets fat accumulation or lipogenesis, which is a root cause that is common to all of the disease states that I mentioned and that we are pursuing development.
Therefore, this offers a highly differentiated approach to treating the diseases. Denifanstat is really the only fat synthesis inhibitor in clinical development at this stage, particularly late-stage development with late-stage data. Our FASCINATE-2 study in MASH, phase II-B clinical trial with denifanstat in liver biopsy-based primary endpoints, met our primary and secondary endpoints, particularly in the most severe MASH patients in the study. Those F3 patients where denifanstat demonstrated industry-leading data demonstrating one and two-stage improvements in fibrosis that were statistically significant. We recently presented data at EASL this past summer from the phase II-B study in patients that were digitally diagnosed as having F4 stage disease or cirrhosis of the liver, in which 11 of the 13 demonstrated a one or two-stage statistically significant improvement in fibrosis, which was again parallel.
Based on the strength of this data in more severe patients and our preclinical data that showed that the combination of our FASN inhibitor plus resmetirom demonstrated an amplified response in markers of MASH disease, including MASH resolution and fibrosis, we recently initiated, as a result of the strength of this data, a phase I PK study in the combination of denifanstat and resmetirom, and we expect top-line results in the first half of next year. Just really briefly on the dermatology front, earlier in June, Ascletis our partner in China reported impressive statistically significant phase III results of denifanstat in moderate to severe acne patients. The phase III trial was conducted in 480 patients in China, and Ascletis reported that all primary and secondary endpoints were met. We have since initiated a phase I trial of our next-gen FASN inhibitor in June.
I should also point out that Ascletis has already had pre-consultation discussions with NMPA in China, and they've been given the green light to go ahead and submit. Those are in a nutshell our progress today, and I'll turn it back over to you, Brandon.
Thanks very much. You touched on the biology behind FASN inhibition. Can you just sort of dig into it a little bit more and how this differs from the mechanisms of other metabolic drugs like GLP-1s or THR-β agonists?
Sure, Brandon. Of course, as Dave mentioned, FASN is a key enzyme in the lipid synthesis cascade, a cascade that we refer to as de novo lipogenesis. This is a key pathway that is implicated in different conditions such as MASH, but also acne and certain types of tumor. Now, specifically in MASH, FASN inhibition targets the three key drivers of MASH independently and in parallel: fat accumulation, inflammation, and fibrosis, and those in three different cell types, as we know, hepatocytes, Kupffer cells, and stellate cells. Importantly, it's a once-daily oral molecule. It blocks fibrosis directly by targeting and inhibiting the activation of stellate cells in addition to the downstream effects of defatting. This mechanism of action on stellate cells is unique, which probably and most likely is the reason why we observed the profound effects in fibrosis reversal in patients with more advanced disease.
As we know, fibrosis is the main determinant of prognosis, and the more fibrosis there is, the likelihood of poor outcomes increases. Therefore, back to, as we mentioned, the big differentiator is the direct activity in addition to the benefits from defatting upstream. Now, specifically in relation to how it differs from other drugs, GLP , THR-β , and FGFs all act indirectly on inflammation and fibrosis because they rely just on the defatting component. There are no receptors for those molecules on stellate cells. THR-β are an important mechanism in MASH because they increase the beta oxidation, or colloquially saying burning, of lipids and that defats the liver, which actually plays a role, and we can discuss this if you wish later on, on the combination therapy that we are launching because of the complementarity of mechanisms.
Now, specifically regarding GLP s, all GLPs act outside of the liver. That's where the GLP receptors are. They decrease appetite, they induce weight loss, and they improve insulin sensitivity. With that, there is less sugar coming into the liver and less substrate for fat production. In broad strokes, this is how the mechanism differs from everything else out there.
Great. I want to just turn to the actual data right now. You reported phase II-B biopsy results earlier this year. Can you walk us through some of the key findings of FASCINATE-2?
Sure. Yeah, the results that you mentioned, we announced January last year, and those were data coming from the completed phase II-B FASCINATE-2 trial. This study was in patients with biopsy-confirmed F2, F3 MASH. They were treated for 52 weeks, biopsies at screening, biopsies at the end of the study. So denifanstat met both primary endpoints and showed statistically significant improvements in fibrosis and no worsening of MASH, as well as MASH resolution without worsening of fibrosis, which are, of course, the two biopsy endpoints that the agency uses for accelerated approval. More importantly, our analysis, our ITT analysis in this study followed the strictest FDA requirement of missing equals failure. In our analysis, any patient that did not have a biopsy at the end of treatment would be counted as a therapeutic failure. This is as strict as it gets.
In addition, we presented and we continue to present AI-based digital pathology data at AASLD last year, and we will be expanding on that at AASLD this year with the same platform, as well as new data with the second platform as well. Specifically for the results, if we look at, since I mentioned digital AI, when we used the HistoIndex platform, the second harmonic generation, we saw that there were some patients that had, by digital AI, an F4 phenotype, which they call qF4, and which overlaps significantly with the human. We observed decreases of one or two Q fibrosis stages with AI pathology. This is yet another reason why we believe the impact on advanced disease makes this a very differentiated therapy.
Now, specifically back to the humans, when we looked at the patients by traditional pathology with the most advanced disease in our trial, so F3 patients, we saw that looking at a one-stage improvement in fibrosis, there was a 36% delta over placebo in response. When we went for an even stricter parameter, two-stage improvements in fibrosis, we saw a 30% delta over placebo. Taken together, these results are unique and we believe are leading in the field, which gives us great confidence moving forward with the program in more advanced disease.
Great. Why do you think the molecule performs so well in the more advanced disease?
Very good question and important for the patients. Of course, fibrosis drives prognosis. The mechanism of action is really the reason and the differentiator here. Let's remember the defatting of the liver, which is very much what other drugs do, will lead to reduction in inflammation and reduction in fibrosis, so indirect downstream effects. Yes, denifanstat does that as well as a potent defatting agent. Now, the other important part, and that's again back to the unique component, is that inhibiting FASN also inhibits activation of Kupffer cells, so activation of inflammation. Without that de novo lipogenesis cascade, without FASN, the activation of those cells is blunted quite considerably. Similarly, the mechanism also applies to stellate cells. Our data and others have shown that FASN and FASN-driven mechanisms are important for stellate cell activation and differentiation.
I think we can say there is a double whammy on the disease in response to denifanstat. Defatting, downstream benefits, and then the direct hit on all three key cell types in the dysfunctional metabolism in this disease, ultimately leading to the fibrosis benefit that we observed in the studies.
Thank you very much. I want to follow on something you mentioned earlier, right, which was the digital pathology and AI. Can you just elaborate on the AI-based quantification support or extension of the traditional biopsies and help us understand better how you are using it?
Sure, of course. I think the biggest difference, and you said that, the quantification. The human assessment, the scoring systems that we have for any liver disease, for that matter, are based on the architecture of that fibrosis, which, make no mistake, it's very important. What the difference is, the digital pathology, first and foremost, it not only shows you the architecture, but it shows how much fibrous tissue there is. Not only the shape, but also how thick or thin those scars are. In addition to that, it also allows the identification of fat deposition, or in the case in response of denifanstat, fat reabsorption in areas of the liver that, because of the subtlety of the changes, are very difficult for the human eye to identify. Those areas, improvements there, have been associated with prognosis. You put all this together, you can see the benefit of the AI.
There are some other things that, like any other machine learning AI system, we don't quite know how AI functions. That's the big thing, isn't it? The observations keep growing. The system sees something, and then it starts identifying other components and gives us that granularity, which also allowed us to look into even more detail into those patients with very advanced fibrosis. We saw that by the HistoIndex platform, there were, as I mentioned a few minutes ago, there were patients that had what they call qF4, overlaps largely with the human, and we saw the significant benefit in one or two-stage improvement in fibrosis in those patients. Put it all together, human plus AI, and I always say human plus AI, you can really see the benefits of different drugs, and in particular, denifanstat.
Thank you. That's very interesting. Maybe just staying on that kind of topic, do you view those digital pathology metrics as part of the FDA's evolving framework for MASH endpoints? How should we think about sort of.
That's an excellent question. Yes, a number of things. The field is evolving fast, and of course, the FDA, which is composed of scientists and the division of multiple hepatologists, they follow this very closely. We believe digital pathology will play a big role, in particular in F4 patients. We know that at least one of the available platforms has received positive opinions from both FDA and EMA. This is a force. Why do I say this is a force?
Recently we saw a press release from FDA stating that they have accepted a letter of intent from Echosens for the FibroScan technology as a likely robust surrogate for response to drugs, and therefore companies can propose studies for registration based on FibroScan, in other words, the vibration-controlled transient elastography or VCTE, which is effectively reflecting clinical practice into regulatory approval. This goes back to my thinking of the agency, the division always being very close to the clinic. Now, our data, going a little bit towards the non-invasives, showed in several different non-invasive parameters that the drug was statistically significantly beneficial, VCTE as well. In addition, and I consider that to be a very important marker, the VCTE-based FAST score, combining VCTE and AST, has been shown to be a very strong correlate of fibrosis, perhaps even more robust than VCTE alone.
It's just that VCTE on its own has a larger body of evidence because it's been around for a while. Put this all together, and we believe, one, digital pathology will play a big role in F4, continues to, and two, non-invasives are probably going to become the parameter for non-cirrhotic MASH, which is again something we welcome significantly.
Thank you. I do want to move on to combination therapy. I know you sort of alluded to that, we'll get there, but there's a school of thought that MASH will ultimately require combination therapy. Why is denifanstat well suited as a backbone for combination regimens?
Sure. Yes, and I agree with you, and I believe the hepatology community, I'm a humble hepatologist following the big names out there, we believe that you need one potent drug, denifanstat being a potent drug, but there is more benefit to be gained if you have a mechanism-driven, science-based approach to it. I talked about the mechanism of action of denifanstat hitting all three cells, hepatocytes, Kupffer cells, and in particular stellate cells, but there is always something to be gained. Denifanstat inhibits, so if we go all the way up to the hepatocyte, we inhibit fat formation and accumulation. You then look at resmetirom, which is a potent defatting agent through a different mechanism, as we talked about, beta oxidation or burning lipids. Those two mechanisms are complementary, and we have shown in animal data that the combination is synergistic.
In addition to that, specifically for this combination, not only are both drugs well tolerated, but also the way they are metabolized is different. They're metabolized through different CYP enzymes. We are, of course, in a phase I study right now to assess the combination, and we believe we will see that the two drugs can be combined in humans, as we have shown in animals. Finally, when you induce beta oxidation and when you induce lipolysis, there is a counter-regulatory loop increasing FASN activity. By combining again, we are blunting the upregulation of FASN induced by resmetirom. As you can see, mechanistically, there is everything to suggest that this is a good way to go.
Great. I think there's preclinical models that suggest synergy with GLP-1 as well. Maybe, you know, what have you learned from your own data and patients who are already taking GLP-1 therapy, perhaps during FASCINATE-2?
Sure. Yes, in FASCINATE-2, we saw that patients on a stable GLP, and I mentioned stable GLP because if you look at the FDA guidance for development, they need to be on a stable dose for three to six months prior to enrollment because you don't want to see changes in weight. Of course, it makes it difficult to analyze. What we saw first, we showed that in animals. In different mouse models, we saw that the combination was also synergistic, which speaks to mechanism. We had the in vivo confirmation. In the FASCINATE-2 study, we saw that in both MASH resolution without worsening of fibrosis and in fibrosis regression without worsening of MASH, about 42% of patients with either assessment responded to the combination, whereas none on placebo. The effect of GLPs had already happened in those patients, if you will. This is important.
We are seeing that GLPs are being used quite considerably in that population, and we believe the proportion of patients, the percentage of patients receiving a GLP, is only going to increase. Therefore, our results also give us confidence that in the new clinical scenario, if you will, the drug would be well positioned. It's not only from a clinical trial perspective, but this is all supported, as we do, with evidence from in vitro and human studies.
Thanks very much. Definitely a lot going on in MASH. I could carry on talking to you, Eduardo, about this for a lot longer, but in the interest of time, I'm going to just switch gears to acne now just because I think that's an incredibly interesting part of the company as well. Can you just talk about the rationale for FASN inhibitor in acne and maybe just provide an update on your acne program?
Perfect. Thanks, Brandon. Acne has four main pathogenesis. There's an increased level of sebum, hyperkeratinization, increased C. acnes, and ultimately some inflammation. In phase I studies, denifanstat has been able to demonstrate that it has a direct impact, as Dave and Eduardo were saying, on de novo lipogenesis, which ultimately has the ability to reduce sebum, which is one of the most significant drivers of acne. In phase I studies, we've been able to demonstrate that and also that denifanstat or oral FASN have had the ability to reduce inflammation. Mechanistically, the product makes a lot of sense for acne and its core pathogenesis. From a program update, we have a business partner in China, Ascletis . They've been developing denifanstat, as Dave was saying, for moderate to severe acne.
They've been able to complete a successful phase II and phase III program, both in moderate to severe acne. The phase III program, when you think about it with the U.S. lens in mind, is a very similar design to what the U.S. FDA would want. They looked at primary endpoints that were aligned of IGA success, inflammatory lesions. The study was 12 weeks and had roughly 480 patients. The study met all of its primary and secondary endpoints, and the product was generally well tolerated in patients. It's been a successful program there, and our Chinese partner, as Dave mentioned earlier, has had their interactions with the NMPA in China and are planning to submit the program in China. In parallel, Sagimet in the U.S., we've been using a sister molecule, TVB-3567, for our development of our acne program.
That program has entered into phase I, a typical MAD/TAD study design in June.
Fantastic. How would you expect to see a FASN inhibitor used to treat acne in practice, just given, one, the strong data we've seen out of the Chinese partner, and then secondly, the different options that patients often try first with limited success?
Yeah. Unfortunately, acne for the last 40 years has really been an underdeveloped market in dermatology. There's been very limited significant innovation in oral mechanisms. An oral FASN represents a significant opportunity for a new mechanism of action to address a core aspect of acne. As you think about moderate to severe acne patients, every single one of them has an increased level of sebum. An oral FASN has a direct impact and has the ability to be applicable to every single patient because they all have increased levels of sebum. As an oral product, oral denifanstat would have the ability to be used for larger body surface areas because sometimes a topical is not really the best if you're applying it for your face, your chest, your back. You want an oral option.
Lastly, because our mechanism is a unique mechanism in the oral category, it will complement a lot of products. We think that the product does have the applicability to reach a large opportunity of moderate to severe acne patients and be a real tool for dermatologists. When you think about the overall U.S. acne population, there's about 50 million Americans that have acne. Of that 50 million, roughly 20% of them are moderate to severe. You think about about 10 million moderate to severe acne patients. With most recent data, we see that only 5 million of those people are actually seeking treatment in dermatology offices. We think that there's a really nice opportunity for denifanstat, not only to help the patients that are currently seeking, but it also will expand the acne market.
I think when you think back in dermatology, and a lot of people don't know about acne, but we've seen it in dermatology in the psoriasis market and in the atopic market, where both of those markets were being managed by generic topical steroids, generic oral products. When these large molecule biologics came in, they had the ability to increase the value of every single patient by offering really new treatment options with great efficacy and great safety profiles. The value of the patients went up, and a significant market expansion happened too. I think in our minds, and when we speak to dermatologists, they feel that that's what's going to happen, and this is what denifanstat or an oral FASN has to offer in the acne market.
Thanks very much, Rob. In the interest of time, I'm going to end our discussion there today. You know, Dave, Rob, Eduardo, Thierry, thank you very much for joining us. Very interesting conversation.
Thanks, Brandon.
Thank you, Brandon.
All right.