Madrigal Pharmaceuticals, Inc. (MDGL) Earnings Call Transcript & Summary

Good morning, everyone. I'm Ellie Merle. I'm one of the biotech analysts here at UBS. Thank you, everyone, for joining us for day 2 of the UBS Global Healthcare Conference in person. Very happy to have Madrigal Pharmaceuticals here with us for a fireside chat this morning. Joining us from Madrigal are Paul Friedman, Chairman and Chief Executive Officer; Alex Howarth, Chief Financial Officer; and Rebecca Taub, Founder, CMO and President of R&D; and Remy Sukhija, Chief Commercial Officer, who may be in the audience here for some Q&A. And with that, I thank you guys so much for joining us.

And maybe just to kick it off, can you tell us sort of high level some sort of background on the company? Where you are? And it's certainly been an eventful past couple of years and an eventful year to come. And sort of frame for us all the important milestones that we should be looking out for.

Yes. Well, this is a private company that on its own got through Phase I, made API and was ready to go to Phase II, but did not have the funds to do it. And it turned out fortuitously that Synta Pharmaceuticals had, had a Phase III failure for an HSP90 inhibitor for metastatic lung cancer, non-small cell. And it ended that in about 5 years ago, the companies did a reverse merger. Madrigal -- the company became Madrigal, the public entity. And the funds were available in Synta that allowed Madrigal to go ahead and do the proof-of-concept study that has propelled us to the point we are currently. And we are currently well into Phase III with key readout in the fourth quarter of this year. Becky is the founder of the company and was the CEO of the private company. And it was felt that there was going to be so much clinical activity that she needed somebody to tag along and help out. So that's kind of how we got to where we are.

Awesome. And maybe just in terms of framing, kind of a huge year for you guys with the Phase III readout. But before we get into that, kind of tell us a little bit more about resmetirom, the mechanism of action and what we've learned so far from the Phase II data.

Yes. So as Paul said, the resmetirom, which is a thyroid hormone receptor beta agonist, is in Phase III development for the treatment of NASH in patients who also have significant liver fibrosis. And the mechanism of action is that it interacts with a liver receptor for thyroid hormone called thyroid hormone receptor beta. And this receptor is responsible for the metabolic benefits of thyroid hormone in the liver such as lipid lowering, LDL cholesterol lowering, triglyceride lowering. And as we've discovered also for the treatment of NASH and resmetirom doesn't interact with the alpha thyroid hormone receptor, which is a very important receptor in many tissues of the body but also responsible for the adverse consequences of excess thyroid hormone, such as increased heart rate and bone effects. So the beta receptor, which is predominant in the liver, mediates the metabolic benefits that we're looking for. And also, as we've learned and has been known for many years, is very important in the liver for normal liver function, which NASH patients do not have. We've shown that patients with NASH or fatty liver disease and inflammation in the liver have a defective thyroid hormone beta pathway function and resmetirom in Phase II showed activity in all components of NASH, including reduction of steatosis, which is very profound and can be observed within a few weeks of starting the therapy, also reduction of inflammation, ballooning, liver injury, reduction in liver enzymes from baseline in these patients. So it's treating all aspects of NASH, including in patients who have these reduction in liver fat, reduction in information. They also observed reduction in liver fibrosis. So that was in the 36th week Phase II study, where the primary endpoint was actually reduction in steatosis as measured on non-invasively on MRI-PDFF and then serial liver biopsies, which showed a NASH resolution and trends to fibrosis reduction. We then took a very similar design study into Phase III, using somewhat higher doses than we had used in Phase II. So we have placebo, 80 and 100 milligram, orally, once a day in Phase III. It's a 52-week serial liver biopsy study called MAESTRO-NASH. And it's in up, where it's more than 1,000 patients of liver biopsy readout for MAESTRO-NASH. As Paul mentioned, it is near the end of this year in the fourth quarter, and that would be the basis for submitting for filing for subpart H or accelerated approval with the FDA. We also completed another greater than 1,000 patient MAESTRO-NAFL study. It was a noninvasive study in patients with earlier forms of -- earlier NASH that was completed and read out in January of this year. So together with these two large Phase III trials and a number of other trials that we've completed, we have abundant safety data with resmetirom that we believe will support the filing for the treatment of NASH.

Great. You also recently amended the design or at least the statistics of the Phase III trial and added a second outcome trial. Could you tell us a little bit more about the Phase III design as well as the recent announcements in terms of the Phase III and outcome study plans?

Yes. So for -- based on the FDA guidance document, the FDA is recognizing two liver biopsy endpoints, potentially to support approval for the treatment of NASH. And this is in patients who are felt to have significant liver fibrosis. They're not cirrhotic. They're called non-cirrhotic NASH. But the stage of fibrosis is at least F2, which is moderate fibrosis, and F3, which is considered to be advanced fibrosis, the precursor of F4, which is cirrhotic NASH. So the two endpoints are resolution of NASH, which means resolution of the inflammation in the liver, which is characterized by an inflammation score and what's called the ballooning score. And that was the original primary endpoint that we had for our 52-week serial liver biopsy study. However, there is a second endpoint, a 1-point reduction in fibrosis that initially, we hadn't included as an endpoint. But more recently and upwards of a year ago, we engage with the agency about adding that endpoint because what it can do is then you have two primary endpoints, which is called dual primary endpoints. You don't have to make both endpoints, you can make either endpoint and then have a successful trial. Also, the fibrosis endpoint was moved up the hierarchy from a key secondary into the primary endpoint, recently, that we announced. And so we have two doses, 80 and 100 in Phase III, and then two primary endpoints. And we also have another key secondary endpoint for LDL cholesterol on, which is very critical in these patients who die most commonly of cardiovascular disease. So what the change that is being referred to is moving that fibrosis endpoint into -- from a key secondary into the primary. So moving it up the hierarchy. And in part, that statistical plan is supported by the fact that we're adding a second outcome study. So the primary MAESTRO-NASH study has a serial liver biopsy at week 52, but the patients don't leave the study at week 52. They continue for a total of 54 months or 4.5 years to be followed for clinical benefit. So that's an outcome, a clinical outcome. And that's used to support full approval for the indication. The serial liver biopsy at the end of 52 weeks is considered to be a reasonably likely to predict clinical benefit, but it's not establishing clinical benefit, whereas the month 54 establishes clinical benefit. So the second outcome, which is something that is highly favored by the agency, having two Phase III confirmatory trials allows us to have better statistics in the first trial as well, where we're guaranteed a 0.05 outcome or alpha, as is called, in the first trial and also in the second trial. And the second trial is actually looking for improvement relative to placebo in cirrhotic NASH, and we can talk about that a bit more. But the point is, we had two outcomes demonstrating clinical benefit in this population. And then we have more alpha for the first trial. And that also supported adding that second primary endpoint.

That's helpful context, especially in thinking about the alpha and the likelihood of success. I mean, I'm paraphrasing here, but it sounds like you're more confident on statistical success with this modification.

Correct. Yes. I mean, it gives you a lot more power for the study. So yes.

Yes. And you have, I guess, to use a terrible sell-side line, two shots on goal now with the primary endpoint. Can you tell us a little bit more about this new outcome study? I think it's interesting that you don't necessarily need to use some of these biopsy measures for outcomes. And maybe elaborate how this could potentially get to a full approval faster.

Yes. So I think one of the things -- so the NASH field is probably, some of you are aware, has evolved quite a bit in the last several years prior to the first trials with obeticholic acid that I think read out some time in 2014, it really what we're -- significant serial liver biopsy studies. And this was not that large of a study. I think it had 200 participants at that time, a very large study. And there have been a couple of Phase IIIs that are even larger since then. The liver biopsy is an invasive procedure. And it's very difficult to conduct a study where you have serial liver biopsies. And Madrigal is one of the few with well over 1,000 patients in a Phase III serial liver biopsy study. But certainly, if we could show clinical benefit without having to rely on liver biopsies, that would be better, particularly in patients who are cirrhotic. It's not at all customary to conduct liver biopsies in patients, where it's already established that they have cirrhosis. So what our new outcome study, which is called MAESTRO-NASH outcomes, does is, it enrolls very well-compensated NASH cirrhosis patients. So they've never had liver decompensation. They really transition fairly recently, from a temporal perspective, from advanced non-cirrhotic NASH into cirrhotic NASH. And so the idea is to follow them and look at decompensation events that usually occur within a few years of that transition into cirrhosis. And we've already studied with resmetirom in one of our trials called MAESTRO-NAFL. More than 180 NASH cirrhotic patients. So we know that [ resmetirom ] -- and we'll be talking about some of these data at EASL, coming up in about a month. But we know that resmetirom is active in NASH-cirrhotic patients. It's patients are doing very well and that the rate of decompensation when patients with NASH cirrhosis are on resmetirom is extremely low, based on our early data. And there's a series of different activities, so improvement in liver enzymes, reduction in liver volume, reduction in liver fat, reduction in liver stiffness or presumably the fibrotic score based on FibroScan and MRE in NASH-cirrhotic patients. So the outcome study is simply comparing resmetirom-treated NASH cirrhosis patients to placebo-treated NASH cirrhosis patients and looking at the rate of decompensation is characterized by new ascites. These are markers of portal hypertension ascites, hepatic encephalopathy, bleeding varices or increasing MELD score and a couple of other measures like liver transplantation, which would occur very rarely, or overall mortality. So that study will be in parallel to MAESTRO-NASH. And we expect the number of events to support the clinical benefit to accrue in 2 to 3 years as compared with the 4.5 years for the MAESTRO-NASH outcome study, which is actually a serial liver biopsy study, where they're having another liver biopsy after 4.5 years. So noninvasive versus invasive patients who are further along on the spectrum, compared to earlier non-cirrhotic, and this is why we think it will be completed sooner. And the opportunity to support full approval is something that FDA has been clear on, using this full approval for non-cirrhotic NASH using the NASH-cirrhotic outcome study. And in addition, the potential for approval for cirrhotic NASH, which was -- is very important. There's no treatment for these patients. So that's a critical unmet need in a growing population. So this is -- NASH cirrhosis is actually becoming more and more common as this disease evolves.

That certainly makes sense. And since you alluded to the regulatory landscape and thinking about full approval, maybe we can talk a bit about accelerated approval, which hopefully would come before full approval. This has been one -- I know we've spoken about that's been a big debate in the -- at least in the investor community. What drives your confidence in kind of understanding of the pathway in NASH from an accelerated approval standpoint?

Yes. So we -- based on our Phase II data and a lot of noninvasive data that we have with [ resmetirom ], we know the drug is very active in NASH that is anti-steatotic, anti-inflammatory, anti-liver injury. And also, as we've shown, reduces fibrosis, especially in patients who have the NASH reduction. The fibrosis component is driven by the NASH, the form of hepatitis. This is the same with all liver diseases that cause hepatitis. The fibrosis follows as a consequence of the hepatitis. So resolving the NASH was demonstrated in Phase II in a much smaller study. We have 10x more patients. So the study is extremely well powered to demonstrate NASH resolution. And it's also well powered to demonstrate fibrosis reduction in patients with NASH.

So I was going to say that there is this misconception in the investor community that the goalposts have been moved since obeticholic acid and Intercept got their CRL. The FDA has gone out of its way, including a publication last year in hepatology to make it very clear that the goalposts haven't been changed at all and that they still adhere to the requirement to make one of those two endpoints to be able to put a subpart H submission in. One of the issues and one of the reasons that Becky's team designed MAESTRO-NAFL, the noninvasive study that read out in January, was to make certain that we had an adequate safety database because for chronic therapy, you really need -- you need a lot of patients on the drug that you -- the dosage you're going to market. And the -- one of the main issues with the [ obeticholic ] acid study was the fact that they had nowhere near the number of patients at the 25-milligram dose, which was the dose they wanted to market. And so we have dealt with that hurdle by running the second parallel study with like 1,200 patients. And every indication from the FDA has been solid that there have been no changes in the goalpost, that these requirements are the same as they have been for the last 5 years.

That's helpful and reassuring. Maybe, since you mentioned it, can you elaborate a bit more about what we learned from this NAFLD study, both in terms of safety as well as some of the noninvasive markers for weeks 2 activity, particularly perhaps on the metabolic front? And then also, what we can expect to learn as well at EASL when we learn a bit more about this data set?

Yes. So to summarize. MAESTRO-NAFL, as I mentioned, was a 4-arm and then became 5-arm but 4-arm, randomized, double-blind study, in which patients were identified -- fatty liver disease patients were identified non-invasively. And I also mentioned that it was an earlier population from MAESTRO-NASH. So how did we identify these patients? Well, we identify the patients in MAESTRO-NASH and MAESTRO-NAFL very similarly. First of all, NASH is associated with metabolic disease, a very high proportion of obese -- about 90%, diabetes more than 50%, dyslipidemia, hypertension. And if you have more of those metabolic risk factors, the likelihood of actual NASH, not just fat and liver, but inflammation in the liver is much higher. So both of our MAESTRO trials require 3 metabolic risk factors to start with. So in order for patients to screen, they have to have that metabolic profile. And then we use a test, a noninvasive, office-based measurement called the FibroScan to then categorize the patients in terms of their potential risk to have fibrosis. And many of the same sites were participating in both MAESTRO-NASH and MAESTRO-NAFL. And so if a patient had a FibroScan score, a stiffness of 8.5 or higher, then they needed to screen for MAESTRO-NASH. And if they had lower than 8.5 but higher than 5.5, which was consistent with an F1 stage of fibrosis, then they could screen for MAESTRO-NAFL. Or if they [ screen fell ] from MAESTRO-NASH after all of the testing but they had F1 on their liver biopsy, then they could enter MAESTRO-NAFL. So MAESTRO-NAFL itself, the noninvasive testing, was a fiber scan, as I mentioned. And then it was followed by an MRI-PDFF, where they had to have at least 8% liver fat, which meant that they, by definition, had NAFL, so NAFL presume NASH. And there are a series of other safety tests, and they entered the study. And the readouts, as we mentioned, were noninvasive. So they had serial MRI-PDFFs, serial lipid measurements, large safety analyses for 52 weeks. And we -- we also serial -- FibroScan serial MRE, which is another test for liver stiffness and other biomarkers. So to your point, it was a very large safety study, 52 weeks. And what we saw that we announced at the end of January and then another brief update in May was that the safety in the study in this very large population was consistent with our Phase II study. The only AE that was disproportionate in the resmetirom population was an early increase in loose stools, very transient or diarrhea. There was no increase after the first few weeks of therapy relative to placebo, and it didn't result into drug discontinuation. So the discontinuation rate, based on an adverse event, was only 1% or 2% in the arms of the study. I mentioned there were 4 arms. There was also a parallel, open-label 100-milligram arm that was randomized at the same time as the 3 double-blind arms. 3 double-blind arms were 180 and placebo. And we had read out the open-label arm at the end of last year. So the data that we discussed and going to be discussed in much more detail next month at EASL and then wave-breaker presentation, was that the primary endpoint with safety was achieved and that the key secondary endpoints, which were LDL, ApoB, triglyceride lowering, MRI-PDFF lowering were also achieved. And the serial measurements in the fibrosis scanning devices, we were able to show that the number of patients -- the patients on resmetirom, who had serial [ FibroScans ] or -- and/or serial MRE tests, the drug-treated groups had more responders than the placebo-treated group. And that was statistically significant for both the FibroScan and the MRE tests. So we believe that's consistent with the Phase III NASH study, where you're looking at the responder rate, the percentage of patients that have a fibrosis reduction in each arm compared -- in each treatment arm, compared to placebo.

That's helpful. And turning to the NASH Phase III, maybe for those in the room and on the line, can you remind us some of the design changes or design from the Phase II to Phase III in terms of the dose, patient population and how to think about that?

Yes. So the differences between the Phase II and the Phase III, I already mentioned. The major difference is that it's a 10x bigger study in Phase III. So it makes the powering much more appropriate for a serial liver biopsy study. The concentration of patients who are -- the advanced F3 population is a lot higher in the MAESTRO-NASH study than we did in the Phase II. But I will comment that with the fibrosis benefit we saw in the F3, the Phase II study was the most significant, based on both on biomarkers and a liver biopsy review, using a technique called second harmonic generation. So when patients have more advanced fibrosis, they're more established. There's very little placebo effect in our study, and that's where we saw the biggest drug effect. So more than half of the patients in MAESTRO-NASH are what's called the F3 stage of fibrosis. The dosing is very different in phase -- the Phase II was essentially a dose-ranging study. We were trying to arrive at the Phase III dose. So half the patients were on 60 milligrams and another half were on 80 milligrams, and there were very few at a higher dose of 100 milligrams, so 80 or 60. And we showed that 60 was not a sufficient dose for Phase III. So we ended up with a higher dose of 100 milligrams in Phase III and another 80 milligram. There's an 80-milligram arm, 100-milligram arm in the Phase III MAESTRO-NASH study. So we've got a higher dose. We showed that 100 milligrams was safe in MAESTRO-NAFL and that the magnitude of effect on the MRI-PDFF is significantly higher at 100 milligrams then -- even then 80 milligrams. And we showed that in MAESTRO-NAFL. And in the Phase II, we tied the effect on MRI-PDFF to NASH resolution, including all component improvement in NASH as well as fibrosis reduction. So more patients achieving that 50% mark or that 30% mark on the PDFF means more patients with NASH resolution and fibrosis reduction.

When we get this top line later this year, how are you thinking about what's meaningful, I guess, on NASH resolution without working in fibrosis and improvement in fibrosis, each of those endpoints, respectively? Although, I guess, they're intertwined, but what magnitude of effect size is meaningful, both statistically but also from a regulatory and commercial perspective, in your view?

Yes. So I think what we've shown is that [ resmetirom ] it has a very high responder rate. So essentially, all patients will respond to resmetirom with some benefits and LDL lowering and liver fat reduction, et cetera. But NASH resolution means that they can't have any little residual this, that or the other. It's a bit artificial. So one of the things we showed in Phase II is that going along with this PDFF reduction, more than 70% of the patients actually had improvement in their NASH, and a high percentage had resolution. Around 40% had resolution. So if we look at -- compared to placebo, let's say, the placebo rate, which is typically -- we powered the study based on very conservative numbers. But let's say, the placebo rate of NASH resolution in these populations on order of high single digits or low double digits, then we expect the differential of -- to be statistically significant for NASH resolution. But we powered it with a relatively low percent -- but we're understanding that there's a lot of patients who have also improved. So we have other endpoints that look at NASH improvement. And then with fibrosis, there's a similar kind of situation, where there's a placebo response that we believe is primarily based on sampling variability of the liver. And there's always about a 20% apparent placebo response in -- almost always. There's maybe one study where it's lower, but it's typically 20%. And that could just be sampling. F1 versus F2, slightly different area of the liver, et cetera. So you need to get above that 20% to show the drug effect. And so anything statistically significant in the 30s, above that amount, would be statistically significant. And we believe that with the risk benefit of resmetirom, that's really what we're looking for. So we're not as concerned about the absolute what is that number, as long as we have a statistically meaningful, statistically significant outcome to the study.

I just would add to that. I agree with that. I mean, there have been enough failures in this space that people should be over the moon about statistical significance for a primary endpoint. But the data is going to show that those patients, who did not quite resolve their NASH, really got a lot better. Maybe there's a balloon cell that we're still seeing in the high-power field. So that data will be published and/or will be in the label in some way. And I think our data that we've gathered with healthcare providers, who take care of significant numbers of these patients, indicates that they're looking for something that's [ been approved ]. And they have a great need and stables of patients who are either currently on nothing or on non-approved therapies.

Absolutely. And doing biopsies is no small feat, particularly amidst a pandemic or endemic COVID, I guess, however you want to [ leave that ] now these days. How are things going from a timing perspective and as well as the data analysis? And how we should think about the time frame for last patient visit and the amount of time it takes for the pathologist to read the biopsies?

Yes. So we -- the pandemic had more of a significant effect on our noninvasive trial, MAESTRO-NAFL, in terms of the data analysis and et cetera and also patient visits, whereas MAESTRO-NASH was highly protected in terms of the patients, and discontinuation rate was not higher than we've expected right from the get-go. So really, the COVID impact, except for a delay for a few months, primarily at the beginning and primarily in Europe, actually. Beginning of the pandemic has not impacted MAESTRO-NASH. In terms of the data plan and data analysis, going very well. And I think, having read out one large Phase III trial that we have the experience, we've hired a new Chief Officer of Data, who's really experienced. And so it's all really going well as far as the data.

Awesome to hear. And so Alex, NASH is a large space and presumably quite expensive to commercialize and run these large studies. Can you remind us where you guys are from a cash runway perspective and potential plans for commercialization, particularly thinking about any strategy ex U.S.?

Sure, absolutely. So just to level set from a cash perspective, we just announced our Q1 results. We ended March 31 with $220 million in cash. As part of that announcement, actually, in May, we completed a term loan financing, totaling $250 million facility. Of that facility, we've drawn down $50 million. So in addition to the $220 million at the end of March, we've drawn down another $50 million, leaving a facility of obviously $200 million left, which aligns with our operational growth, going forward, into registration and commercialization. And again, just to round it out, we do have an existing ATM facility with $160 million available to us. So in the round, we're very well positioned, going into the data readout. And I have a number of options for raising additional capital to fund ourselves, going forward. In terms of the ex U.S. commercialization, as you mentioned, we've been very public about saying that we're going to be looking for a partner ex U.S. So unfortunately, Remy's not here today, but he's clearly spearheading the effort for commercializing in the U.S. Just a quick [ slide ] there in terms of the -- you mentioned sort of the significant sums required to commercialize. We actually believe we can be very effective with a specialty care-focused salesforce of circa 150 to 200 field force. So it's not some huge primary care salesforce that requires that sort of gargantuan commercial investment. But then in terms of the ex U.S. partnership, as I say, we are looking to conclude that deal after we get the MAESTRO-NASH data readout. We've already got the funding for it. We've got obviously, under Becky's leadership, the team and the resources and the wherewithal to complete those studies. What we're looking for is a partner on the back of that data readout to then commercialize ex U.S. And obviously, that's a global infrastructure. That's not something that we're going to be doing ourselves. And we're already in discussions, as of now, with partners, lots of interest, as you can imagine, a very, very late-stage asset and a new market opportunity with, to date, a very sort of attractive profile.

Yes. It's exciting. With one minute left, let me see if there's any questions in the audience or on the webcast. Okay. Well, I'll ask one more and then I'll let you guys go. You alluded to this earlier, back to you, Becky. But how do you think about minimizing reader variability when it comes to the biopsies, particularly when the close statistics like this very much matter?

Yes. So I think with our Phase II study, we had one primary reader and then we have a second reader who read the whole study. And even though -- and they got the same result, even though they didn't agree on every patient, every biopsy, the bottom line was the same. We also had the study read by an AI [ method ] that got the same result. And we're going to show that -- those data at EASL. So we believe that if the drug is effective, that the variability doesn't have as much impact and you'll get your result that you want. And that's kind of how our plan has evolved with our same two readers.

Phase II.

Got it. Well, that's very helpful, and we're very much looking forward to seeing the data later this year. Thank you guys so much for joining us.

Thanks for having us.

Thank you.

Thanks.