Okay. Good afternoon or evening, everyone. Welcome to the UBS Healthcare Conference. I'm Eliana Merle, one of the Biotech Analyst here at UBS. Very happy to have Sagimet here with us for a fireside chat. Joining us is David Happel, Chief Executive Officer, and Eduardo Martins, Chief Medical Officer. Thank you both for joining us.
Thank you.
Maybe before we go into some of the specifics of the data, can you give us an overview of your lead program and the recent milestones?
Yeah, sure. Thanks, Ellie, and thank you for inviting us to come out. Our company was founded really on this unique target of fatty acid synthase, FASN. Our approach, therefore, begins with understanding how overactivity of FASN plays such a critical role in the development of a number of underserved conditions, including MASH, which is our primary focus, but also in acne, where we have some really compelling data, and in certain solid tumors that are highly dependent on overactive FASN for progression of disease. To solve for this overactivity of FASN, we have developed a portfolio of highly differentiated FASN inhibitors led by our lead program, denifanstat, which really seek to target the underlying or root cause of the disease and the primary culprit being fat accumulation or de novo lipogenesis.
We have a great deal of work that we're doing in these other indications beyond MASH, but the bulk of our data really is in MASH, and what we've demonstrated through preclinical and clinical studies is very consistent that the molecule independently and simultaneously really targets the three drivers of MASH, and that's fat, inflammation, and fibrosis, and it does so independently, directly, and frankly, is probably the only molecule that's under development that does it as effectively as denifanstat does. Denifanstat is the only fat synthesis inhibitor, the only real fat inhibitor that's in development for MASH. Everything else is more on the fat burning side, fat oxidizer, fat mobilizer, and that includes GLPs, FGF, and THR.
Those molecules, as you know, really rely on reducing fat to translate into downstream benefits and inflammation and fibrosis, whereas denifanstat has shown, I think particularly very clearly in the clinical data, that it really is effective at reducing inflammation and fibrosis directly and clearly is a very potent antifibrotic. The molecule has now been tested in well over 700 patients at this stage through all of our clinical studies, including MASH, acne, and oncology. It's been extremely well tolerated, and therefore we don't really see any limitations to any of the treatment populations that we're pursuing development. We're really excited. We completed our phase II-B biopsy study this past year and reported out the top line in January.
We reported further information, further data at the European liver meeting in Italy a few months ago, which we demonstrated that or showed that denifanstat had a really pronounced effect in more severe disease. Those patients that have F3 stage fibrosis, and it gives us a great deal of confidence as we head into our phase III program that we can have a very successful program and get this molecule out to patients eventually.
In terms of the mechanism, just diving into that a little bit more, maybe what's some of what's known around sort of the natural biology around FASN?
Sure. That's a very good question. We actually know quite a bit. FASN is a key enzyme, we like to say, in the Goldilocks spot in the de novo lipogenesis cascade, or DNL, as Dave mentioned, which basically is the process through which sugars are metabolized into fats. So that happens as part of the normal physiology, and it's very important in the liver, as well as, and we'll talk about that later, also in the skin, hence our second program that we are pursuing in acne. Now, FASN and DNL in particular, of course, the process is critical for fat accumulation in the hepatocytes. So you're making fat, and that fat accumulates in the hepatocytes.
As they accumulate, they balloon, eventually they die, and the debris from that death will activate inflammatory cells, chronic inflammation, stellate cells that will deposit fibrous tissue, then the disease progression all the way to cirrhosis, transplantation, death, or cancer. Now, the other interesting thing about DNL and FASN is not only it's important for fat synthesis in the hepatocytes, but also important for the activation of those inflammatory cells and activation and differentiation of the stellate cells. So it does more than just burning or mobilizing, as Dave mentioned, the fat in the hepatocyte. It actually does that very well with all its downstream benefits, but also directly hits the inflammatory cells, in this case, the Kupffer cells in the liver and the stellate cells. So it's a fascinating biology, no pun intended, and that our molecule is the only one in development for that.
Great, and maybe just a little more on, I mean, the fat synthesis makes sense around that mechanism, but just in terms of more specifically how it's working on inflammation and fibrosis?
Sure. Of course. To activate inflammatory cells, there is a structure called the NLRP3 inflammasome, which is again a cascade of events. DNL is important, and of course, with DNL goes FASN for NLRP3 activation that goes to the inflammatory cells. And similarly, a very similar structure, just with a slightly different signaling pathway, is also involved in the activation of stellate cells. So it's a different mechanism than just fat accumulation. It's the actual DNL cascade itself doing other things.
Interesting. Makes sense. And turning to your data in MASH, you reported your phase II-B data earlier this year. Can you give us an overview of that data?
Sure. The molecule did what we expected it to do. The results were very robust, as you've seen our presentations before. Basically, what we showed was, one, we hit the primary endpoints, both of them, MITT and ITT, but the most important thing, we go back again to fibrosis, again, the driver of prognosis. We showed a profound effect, not only overall and fibrosis improvement without worsening of MASH, but in particular in patients with more advanced disease, those F3s, and they need the most on the cusp of becoming cirrhotic. They need to really reverse that as fast as one can. And what we showed was in the F3s, we hit one stage improvement without worsening of MASH with a 36% delta over placebo. When we looked at an even stricter measure, two stage improvement, we hit with a 30% delta over placebo.
In addition to that, we also showed that when we looked at progression to cirrhosis, and progression to cirrhosis is one of the outcomes for full approval, 5% of the patients on Deni, and I'm using Deni for short, 5% of patients on Deni progressed versus 11% on placebo. So thinking forward several years down the line, our expectation is that this gap will probably widen even more. Now, another important thing in our analysis of not only the primary endpoints, but in particular the two secondaries that are the ones required for registration, so fibrosis improvement without worsening of MASH or MASH resolution without worsening of fibrosis, we hit MITT as well as ITT. And the ITT analysis that we used was the same that the agency requires for approval, which is missing equals failure. So that's as strict as it gets.
You don't have a second biopsy, we register this particular patient as a therapeutic failure, and that's the expectation that any drug in development has for phase III.
Yeah. Absolutely. Very strong data from the efficacy perspective. Maybe if you could just, from your view of the MASH landscape, there's obviously a lot of drugs in development, a lot of different mechanisms. Where do you see this molecule fitting, particularly in light of the recent data that you've presented?
Sure. It's the perfect first line therapy, and that goes with the efficacy. In addition to that, because of the mechanism of action, it is a perfect backbone for multiple different combinations. And I'm going to use one example, and then I'm going to mention data from the phase II-B. The combination is a hypothetical, is with resmetirom. So we are reducing production of fat. We are directly hitting the other two cell types, and resmetirom is burning the fat. So it would be a nice combination, if you will. Another one, which, yeah, let me call it a combination, which is with the GLP-1. When we looked at our data from the phase II-A, we had 10% of the patients that were on the GLP-1, stable GLP-1 prior to screening. We had 42% of those responding to denifanstat and none from placebo.
We expect many more as a proportion to be enrolled in the phase III, 30%-50% based on what our advisors tell us. So that would be another great avenue for combination. And most likely, in the real day-to-day, given the proportion of patients with type II diabetes and all that, it will already be there. So we'll be adding because you need an effect for the MASH itself, not just for the metabolism.
Yeah. And I think just to add on to that, as we look forward beyond the F2, F3s is into the F4 patient population, the lack of progression into that population that we demonstrated was statistically significant. And as the liver becomes less fatty, a mechanism that is truly antifibrotic would seem to be highly logical. And we do anticipate this molecule could work very effectively in that population. As an oral, once daily small molecule, as opposed to an injectable, I would think it would be highly attractive, particularly if the data continues to be as strong as we expect. So I would say the cirrhotic population also offers a significant opportunity.
I agree. Dave said it all. It's targeted going to the patients in the most need. And at AASLD, we'll be presenting data on artificial intelligence analysis of the liver biopsy showing that what we presented earlier on just one sliver of the data, we're going to now show the whole data set, and that goes beyond the human eye and reproducing and expanding on the good results we saw. I can't go into much more detail because the embargo is up to the 15th, so we can't discuss too much more, but encourage you to watch the presentation. The data will be, not surprisingly, very good.
Great. Looking forward to it. In terms of the safety tolerability profile, maybe just an overview, and then we can go into some of the ones of focus.
Sure. The drug is oral, once daily, very well tolerated. Importantly, especially in a chronic liver disease, no DILI signal, no bone signals, no heart signals, no muscle loss. The drug was very well tolerated overall. The one adverse event that we saw more was hair thinning, which actually, I must say, caught us by surprise. We had only seen before this study two out of 102 patients at the dose that we treat having that. We believe this particular instance had to do with the fact that the study was conducted through two massive waves of COVID, Omicron, and then the subsequent wave. Patients did have COVID, which is a cause in and of itself, COVID vaccines, which also are associated. We had patients on levothyroxine, which is also linked to hair thinning. We had one patient that accidentally took much higher doses, overdose by mistake.
And we had a few patients on Paxlovid for their COVID. The ritonavir on Paxlovid bumped the levels up to the cancer levels of drug, if you will. Importantly, in discussions with our advisors, we know how to manage this moving forward for the phase III. And contextualizing this, we saw a 6% discontinuation rate. When we look at hair thinning overall, if you look at some of the GLP data that have been published, about 7% of the patients also had hair thinning. So it is much common, much more common than we believe phenomenon, if you will. One final thing, what are we going to do in the phase III? If needed, we will use biotin, vitamin D, biotin and/or vitamin D supplementation, scalp oils. If need be, we can dose reduce. We did that in the phase II-B. And interestingly, two things happened.
One, as expected, hair regrew. Two, three patients that were on 25, two of which they had actually a histological response. The third one had a fat reduction response by MRI. And we had five patients that remained on 50, and the hair thinning stabilized. So bringing this all into context was the one adverse event that caught us by surprise. We don't expect it to happen anywhere near that in the phase III. Nevertheless, we know what to do, and everything is in place for the study to go very well and get to another set of robust data.
Just thinking from a commercial perspective and how this could play out, GLP-1s have the aesthetic benefit of weight loss. How are you thinking about how this profile could play out commercially?
I mean, I think, as Eduardo pointed out really eloquently, we do think of this molecule as both a strong candidate for monotherapy as well as a combination therapy with fat burners. I think it's unlikely that you'd want to combine two fat burners, I think, to help patients reach treatment goal largely because I think it'd be difficult for most patients to stay on them for all of the reasons that I think Eduardo pointed out. But to combine it with a fat inhibitor that is well tolerated, I think makes a lot of sense. And I think it makes a lot of sense to the scientific community. And our data, certainly in the phase II-B study, supports combination use with patients that were on GLP therapy. Those patients had an enhanced fibrosis reduction that was statistically significant despite the small number of patients.
That data actually replicated what we saw preclinically in data that we presented at AASLD a year ago, which showed that denifanstat was very effective in reducing fat and inflammation. But when you combine the two together with semaglutide, that there was an enhanced treatment effect. We took that a step further with preclinical data at EASL this last summer, in which we presented data in combination with resmetirom. It showed the same thing that the data with Sema did preclinically, and that is that denifanstat was more effective in reducing fat and inflammation and fibrosis independently. But when you combine the two molecules, there was a pronounced effect in fibrosis. That suggests that this molecule can either be a great standalone molecule or can be used in combination with other molecules to help patients reach treatment goal.
I suspect that the latter is going to be critical because patients are going to have a hard time staying on fat burners for any great length of time. That, I mean, that's just been proven by the most recent data, I think, that suggests that it's just hard to stay on a GLP, for instance, for a length of time. Having an additional therapy to help those patients, I think, is going to be extremely attractive.
Yeah. I think it's interesting that you saw this imbalance in the MASH study, but this really wasn't seen in the other studies. And so with the mitigating factors, perhaps this is not something we see as much in the pivotals.
Yeah. We don't expect. Yeah.
Maybe, I guess, turning to the phase III plans, you recently had your end of phase II meeting with the FDA. How are you thinking about your phase III design?
Sure. The Madrigal set a very good footprint. And I think we, as well as the others out there developing drugs for MASH, we more or less follow the same model, in particular because of the FDA guidance. You are much more restricted. So the plan initially in F2, F3s is to have an efficacy trial where, like in the phase II, we are, one, not enrolling F1s. We did not enroll in the 2B. We're not going to enroll in the phase III, solely F2, F3s. And two, we are going to enrich for F3s. At least 60% of the patients will have to be F3. It's a biopsy trial. Like any MASH trial, it's going to be a large study, about 1,200 patients or so. Biopsy in the beginning, biopsy at the end of one year for accelerated approval, and the patients continue on to outcomes.
As I mentioned a few minutes ago, the most likely outcome in an F2, F3 population is histological progression to cirrhosis. Now, the other thing has to the other trial, because there are two for this particular approval, filing for approval, is to complete the safety database. There is a minimum requirement for patients receiving therapies chronically for indications that impact a large proportion of the population. And to fulfill that database, what we are doing is a parallel study one year where patients that fail screening to the efficacy trial, they run into that trial. They can roll into that study. There is no biopsy in the end, but that plus the main study, when we hit one year, that's going to be our population that will be filed. We would also like, as Dave said, to explore F4 because the mechanism of action is totally suited for that.
Absolutely. And I do want to talk about cash runway, but first, I think it's important we touch on acne, which I think is a really interesting program, very different, but the mechanism totally makes sense. Can you tell us a bit about where you stand there and some of the upcoming potential data from your partner?
Yeah. Great question. Thank you. So the acne opportunity is significant. There hasn't been a novel new oral acne medication in over 40 years since isotretinoin was approved back in 1982. And the only thing that has really come to the market have largely been topical agents. And our molecule, based on the data, the phase II data that Ascletis, our partner in China, produced and released in May of 2023, showed a highly statistically significant reduction in lesion count, improvements in total score in IGA in that population in a relatively small study, which was really encouraging. We had a chance to socialize it with a number of key opinion leaders in the dermatology community, and they're very enthusiastic about this molecule in this space based on the data. Based on the strength of that data, Ascletis launched a phase III study at the end of last year.
They just announced about two days ago that they completed enrollment in the phase III study, and they expect to see top-line results in Q2 of next year, so we're all very anxious and looking forward to that, and we expect it to be very, very similar to the phase II outcome, so based on that, we have a next-gen molecule that shares a very similar PK profile to denifanstat that we are filing an IND on, and we plan to start a phase I study early next year and get that program rolling. We know that it would be of significant benefit to that population, to the community.
How does one think about the potential opportunity in acne just from a patient numbers perspective? What's currently out there? I know we've discussed Accutane and the lack of sort of novel options.
Yeah. Accutane is always going to be the drug of choice for cystic acne for the most severe forms of acne, and where we think that Deni makes the most sense is in the moderate to severe population. 42% of the population, particularly the adolescent population, suffers from moderate to severe acne, and roughly half of those are seeking treatment, so it's a very, very large population. Many of those patients seek over-the-counter alternatives largely because there isn't a prescription medication that has worked very effectively.
The drugs that you can use right now are oral antibiotics in the doxycycline, tetracycline, minocycline family, and those are not really sustainable molecules to use long-term, particularly for adolescents who like to get outside and get moving around, so we anticipate that Deni can slip right into that population very, very effectively into the broadest patient population that's seeking treatment from their dermatologist. It's a very large population. I mean.
Yeah. Certainly an interesting opportunity there.
Yes. No, it's fascinating. It's one that when I joined a couple of years ago, I always had my eye on because of the early data that I saw, and I thought it made a lot of sense. I've spent some time in dermatology in my past and worked on an isotretinoin derivative, and I thought if we could have a molecule that would be nearly as effective as isotretinoin, but a much kinder, gentler version of it, that it would be highly, highly attractive to the dermatology community, and that seems to be playing out.
Yeah. Absolutely. And as we think about the cost of development just between both MASH phase III versus an acne, and I know separate molecules, and that's an important consideration here. But yeah, how much does an acne program cost?
It's much less expensive than a MASH program, for sure. The acne programs, all studies are three months long. So that should tell you immediately that it's considerably less costly. We haven't provided guidance yet, public guidance on the cost of it, but you can sort through various databases to probably understand that. But it's considerably less expensive. For instance, based on our exit cash runway in the second quarter of $188 million, we could largely run the entire dermatology program for acne with that cash and get to the outcome. For MASH, it's an entirely different story. We need about, I think, another roughly $200 million to get to phase III data readout in order for us to be able to enroll the study completely and to run it to its finish, at least for the accelerated approval, Subpart H approval submission to the FDA.
We're working feverishly to fill that gap now.
Yeah. I guess, can you talk about what you've said around your cash runway and your current cash balance?
Sure. Our current cash will get us. I think we've communicated publicly pretty consistently through the end of 2025, and that remains. We do need to secure additional funding to get through the phase III study for MASH, and that is the goal. We don't really want to start the study until we have full access to the cash. We're prepared. Eduardo's team has done a phenomenal job in preparing and being ready to start the phase III. And we're equipped to do that by the end of the year, but we really need to solve the financing. And we're well on our way. I will say that the data on this molecule and the uniqueness of it, the differentiation of it as a fat inhibitor is highly attractive. And we have been very fortunate to draw a lot of attention.
Certainly a critical period that we're entering in terms of the strategy here. What's your philosophy when it comes to BD?
For MASH, look, we're having a number of discussions with all parties, and that includes pharma. It includes traditional equity. It includes potential and royalty arrangements, which would leave the asset unencumbered through phase III, which is highly attractive both to our internal stakeholders at our board level, but also our internal investors are becoming quite enamored with that. So we have a number of options that are in front of us, and we just need to solidify them and be prepared to move forward rather quickly.
Great. We will stay tuned on that front. As it relates to the phase III acne data, which I guess is coming in the second quarter next year from your partner, what should we be looking for there? What's good data? What's mediocre data? What are you hoping for?
I think if it's consistent with the phase II data, it'll be a blockbuster for sure, right? I mean, we have seen very, very few data sets outside of isotretinoin that even reflect anything remotely close to the reductions in lesion count that we saw on these patients. I think what we're looking for is essentially a replication of the phase II data plus or minus and also statistically significant improvements in total score in IGA, which the data set would suggest that they're equipped to get there.
As it relates to the phase III program and MASH, what's your expectation? And I know you alluded to this around sort of background use of GLP-1, tirzepatide, and then how you would treat potential background use of Rezdiffra if that would be allowed or not.
Sure. So starting with the GLPs, as I mentioned, our advisors estimate 30%-50%. Some actually go higher, but that's roughly where the majority is. Now, with Rezdiffra, again, in consultation with our advisors, the plan is to have at least a six-month washout. Because if somebody just started or started recently, you don't know if there is momentum going on or not. You need to have a clean data set, and you need to make sure that groups are well balanced and all that. So there needs to be a complete washout there.
Understood, and I guess heading into AASLD, what are you looking to see across the landscape? I know we spoke a bit about ESSENCE already, but yeah, curious your thoughts on all things.
Sure. I mean, I think the most important thing in the landscape for me are the three abstracts that we have. The artificial intelligence will be an oral presentation by Mary Rinella. There is a poster by Rohit Loomba on the difficult-to-treat population. And another poster, internal data on a mouse model of hyperlipidemia and atherosclerosis, again showing very interesting results. My perspective on the outside data, ESSENCE, of course, that's what everybody is looking at. I think there are some very interesting abstracts from Madrigal that I want to see. There are some interesting, not a lot, but some interesting data from the FGFs. And a lot of data on mechanisms of disease. And those, of course, attract me a lot. But Dave, any comments?
I think you've covered it. I'm anxious to see the fuller data set from ESSENCE to understand the full treatment effect as well as the safety database. I'd like to see a little bit more granularity on the safety database to see the tolerability profile and the discontinuation rate. 72 weeks of data.
Yeah. Yeah. It'll be interesting to see. Last question. What's your latest sense on how long enrollment will take? There's a lot of phase III enrolling.
Sure. 18 months is our expectation.
Okay.
There are lots of patients out there. And with the GLPs, more and more are coming in. And of course, again, back to our phase II data, that is something that is positive for potential patients interested in studies. It's encouraging information for them.
Great. Thank you all and thank you for everyone for joining us.
Thank you.
Thank you, Ellie. Thank you for inviting us.