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

Good day, ladies and gentlemen. And welcome to the Madrigal Pharmaceuticals' MAESTRO-NASH Topline Results Conference Call. [Operator Instructions] As a reminder, this call may be recorded. I would now like to introduce your host for today's conference, Dr. Paul Friedman, Chairman and Chief Executive Officer of Madrigal. Dr. Friedman, you may begin.

Thank you. Good morning, and thanks for joining us on this call to review our Phase III MAESTRO-NASH study data. With me today are Becky Taub, President of R&D and Chief Medical Officer; and Dr. Stephen Harrison, the Lead Investigator for the MAESTRO studies; Remy Sukhija, Chief Commercial Officer; and Alex Howarth, Chief Financial Officer, are also on the line for Q&A. Next slide. This is a fine print dealing with our forward-looking statements. Please see today's press release and our SEC filings for forward-looking statement and risk factor considerations associated with these statements and our business. Next slide. This morning, we issued a PR or press release describing markedly positive data from our ongoing MAESTRO-NASH study. While continuing to show an excellent safety profile, resmetirom at both the 80-milligram and 100-milligram arms showed highly statistically significant superiority to the placebo arm for both biopsy endpoints designated by FDA to be likely to show clinical benefit with extended dosing. Specifically, these 2 endpoints are the resolution of NASH with no worsening of fibrosis and at least a 1-stage improvement in fibrosis with no worsening of NASH. Becky will momentarily describe the top results in more detail, followed by remarks from Dr. Harrison, who we're fortunate to have with us on the call. Let me make 2 other points. First, based on these results, we intend as early in the second half of 2023 as possible to submit a subpart H NDA to FDA seeking accelerated approval for resmetirom. Second, after Becky and Stephen speak, we'll open the call for a limited Q&A session. And because MAESTRO-NASH is an ongoing 54-month blinded study, questions related to this study will have to deal directly with the topline data we've presented today. With that, let me turn the call over to Becky.

Good morning. Can I have the first slide, please? So thyroid hormone receptor beta have -- plays a key role in liver health. It's very important that, from a perspective of NASH treatment, that the thyroid hormone receptor agonists are selective for the liver with good liver uptake without any extra hepatic effects. Thyroid hormone receptor alpha, which is the predominant systemic receptor, in excess, can lead to unwanted effects especially in heart and bone. Thyroid hormone receptor beta agonists act on multiple hepatic pathways to maintain liver health by controlling aspects of fatty acid metabolism, particularly fatty acid oxidation that takes place in the liver, mitochondrial health and mitochondrial biogenesis, cholesterol metabolism and direct anti-inflammatory, anti-fibrotic effects. In human NASH, the liver has relatively low THR-beta activity, which exacerbates mitochondrial dysfunction and lipotoxicity. There's a potential with these agonists for both hepatic and cardiovascular benefits in patients with NASH and liver fibrosis. Phase III MAESTRO-NASH study design is shown on this slide. This is a registrational trial, 54 months in length, with a serial liver biopsy endpoint at week 52 and multiple other readouts -- non-exploratory readouts in the study. It was randomized 80, 100 placebo, 1:1:1, and more than 1,000 patients were enrolled for the primary week 52 analysis. The primary analysis includes more than 90% of these patients who had baseline fibrosis scores, F1B, F2 and F3. And the total enrollment of MAESTRO-NASH is up to 2,000 patients for 54 months. That enrollment is still ongoing, more than 150 centers worldwide. The key inclusion criteria of MAESTRO-NASH differ somewhat from other NASH studies. We want to emphasize that MAESTRO-NASH requires 3 metabolic risk factors in order to be eligible for screening, and the FibroScan kPa consistent with F2-F3 NASH and liver fat. There's a requirement for greater than 8% liver fat on MRI-PDFF, and if that's achieved, then the liver biopsy is conducted and the scores that were allowed are shown here. Importantly, at least 50% had F3, and a small -- only a small percentage were enrolled with various stages of F1, particularly F1B, which is a more advanced F1 stage. The primary endpoints were dual primary endpoints, resolution of NASH at week 52, and that's with a ballooning score of 0 and an inflammation score of 0 or 1 with at least a 2-point reduction in NAS with no worsening of fibrosis stage or a reduction in fibrosis stage by 1 point with no worsening of NAS. A key secondary endpoint was LDL-cholesterol lowering at week 24. The month 54 endpoint is a clinical outcome event, liver-related outcome, histologic evidence cirrhosis on biopsy or various measurements of hepatic decompensation. So I want to reemphasize what Paul said about the fact that this is an ongoing study. So patients who complete week 52 liver biopsy, continue in the study for the 54-month time period. And therefore, any data released on the 52-week endpoint, which is in support of a subpart H approval, needs to maintain that blind of -- particularly as far as individual patients in order not to impair the integrity of the ongoing MAESTRO-NASH study. So what we show here are the baseline characteristics of the population, the week 52 population that we evaluated, and this is the ITT population. And what you can see is, first of all, it's balanced across the treatment arms. But if we just look at the overall population, this is a typical NASH study. 57 years old is the average age. There's a preponderance of females; a high percentage of Latinos; high BMI of 36; very high percent diabetes, 67%; hypertension, 78%; dyslipidemia, 71%. So these are the metabolic syndrome features or metabolic risk factors that were required in the study. About 13% was hypothyroid, meaning that they were on thyroxine replacement, which is another characteristic of this population. And you can see that they had an elevated liver stiffness of mean of 13 and an elevated CAP on FibroScan. The MRI-PDFF mean was 18%. Now on baseline liver biopsy, about 84% had a NAS of 5 or higher and 5% were F1B; 33%, F2; and 62% were F3. So this was a very advanced liver biopsy population and consistent across the study groups. The eligibility biopsies were read by -- primarily by one of the central readers. And those eligibility biopsies were for patient randomization. Biopsies were entirely reread by the 2 central pathologists, and there was a very high rate of concurrence with the screening eligibility. And they were read independently by the 2 central pathologists using glass slides, which is the technology that has been used historically over the years for NASH studies. And that comprise the primary analysis read. There's a detailed biopsy review plan. And the analysis that was conducted was discussed in detail with the FDA on more than one occasion and concurrent with the statistical analysis, this plan -- the primary analysis, this plan was agreed with the FDA. The baseline biopsies rescored during the primary read is F1B, F2 or F3 by the 2 central pathologists were considered the primary population and the ITT was 966, included all patients with at least a baseline biopsy with appropriate fibrosis stage. Eligible week 52 biopsies were included if conducted before 60 weeks post randomization. Patients with biopsies after week 60 were considered missing even if they were conducted in 11 patients having a biopsy that was somewhat after week 60 due to a COVID-related reason. And consistent with the COVID guidance and consistent with feedback from the agency, they were removed from the primary analysis population. So the final population was 955. If the biopsies were rescored as F1A or C, they were considered exploratory and they will be evaluated separately. This slide shows the primary and key secondary endpoints of the study, and you can see that both doses achieved both liver biopsy primary endpoints. So at 80 milligrams, 26% had NASH resolution with at least a 2-point reduction in NAS and no worsening of fibrosis with a p-value of less than 0.0001. And at 100 milligrams, 30% had achieved the endpoint with p-value of 0.0001. Placebo rate was 10%. In terms of the fibrosis endpoint, 24% at 80 milligrams achieved the fibrosis endpoint with a p-value of 0.0002 and 26% achieved the endpoint 100-milligram arm with a p-value of 0.0001, 14% in placebo. The key secondary endpoint, which is the other endpoint that statistically controls the study, was achieved by -- at both doses and is consistent with our prior data that we've shown in MAESTRO-NAFLD or other studies, with a 12% LDL lowering at 80, 16% at 100 with p-value less than 0.0001 and an increase of 1% in the placebo. Additional liver biopsy results are presented here, and we're describing them descriptively in general at this time. So in terms of the review by the 2 central pathologists, each pathologist scored -- scores for both endpoints and both doses shows similar statistically significant and magnitude of response for both liver biopsy endpoints. So in other words, if you just took each pathologist independently, they each saw a statistically significant result for each endpoint at each dose. There was a statistical program that combined the results to generate a single treatment effect, which I showed on the previous slide, and this was discussed in detail with the agency. There are multiple additional programmatic results using these data and -- that are considered sensitivity in support of liver biopsy analyses that are not considered in this topline discussion. Biopsy endpoints were achieved independent of baseline fibrosis stage or diabetes status, including similar statistical significance and magnitude of response at both doses in separate subgroups of F2, F3 and combined F2/F3 biopsies. Other secondary liver biopsy endpoints that were achieved but -- achieving statistical significance at both doses based on not the alpha control, but they would be considered highly statistically significant, what would be called nominal p-values, included 2-point reduction -- at least a 2-point reduction in NAS with no worsening of fibrosis, at least a 2-point reduction in NAS with at least a 1-stage improvement in fibrosis, reduction in fibrosis, NASH resolution with a 2-point reduction in NAS with a 1-stage improvement in fibrosis, and the 2-stage reduction in fibrosis without worsening of NAS. Exploratory and supportive biopsy reviews include reviews of digitized slide images by the central pathologists and 2 artificial intelligence methodologies. The data for the study has been collected and is locked down. So even though we don't have output for the 2 artificial intelligence methodologies, the slides have been analyzed, and the data has been submitted to Madrigal. And this took some time to collect, but we believe that it was very important that these -- all these data sets and the statistical analysis be locked down and not conducted in a post hoc manner. In terms of safety, resmetirom was safe and well tolerated, both the 80 and 100-milligram dose. The frequency of serious adverse events was similar across treatment arms, and I also say that Grade 3 or higher events, so those are severe adverse events, was similar across the arms and if anything, it was slightly higher than placebo group. The rate of study discontinuation for adverse events was low, and that's shown below. There was about a 4% increase in the 100-milligram arm relative to placebo and about a 1% lower rate of a study discontinuation in the 80-milligram arm. And if we compare that to what we saw in MAESTRO-NAFLD, there was about a 2% greater AE rate in the 100-milligram arm. And we looked at those AEs in MAESTRO-NAFLD and they were generally mild and mild GI-related AEs in which the patient decided to discontinue. We are -- personnel at Madrigal are blinded to -- at this point to the idea of the patients and the listing for AEs. So we don't know what the AEs are at this point. The study, as you might imagine, firewalled in terms of the blinding and unblinded personnel are at a statistical [ CRO study ] to analyze, and limited personnel at Madrigal will be unblinded to individual patient data such that we can use that data to prepare the NDA, et cetera, and understand the data from the study. The incidence of diarrhea was consistent with the Phase II and Phase III data and greater frequency with the resmetirom group versus placebo, generally mild transient diarrhea, and generally mild nausea at the beginning of the therapy and the incidence is shown here. Additional analyses are an important part of MAESTRO-NAFLD and MAESTRO-NASH. MAESTRO-NAFLD is an entirely exploratory study for efficacy in the sense that we used noninvasive imaging and biomarkers to follow the drug response. In MAESTRO-NASH, this is coupled with a serial liver biopsy. And we have achieved multiple secondary endpoints, including statistically significant reduction from baseline in liver enzymes and resmetirom treatment groups compared to placebo. Reductions in atherogenic lipids and lipoproteins, such as triglycerides, Lp(a), ApoB, fibrosis biomarkers such as ELF and imaging tests such as MRI-PDFF, CAP and liver stiffness measures, were observed in resmetirom treatment arms as compared with placebo, but those details will be presented in a publication and a later date in meetings. These multiple biomarker and imaging assessments may be used in real-world clinical practice to identify appropriate patients for treatment and monitor response to resmetirom if approved. I think that they're also critically important in support of analyses, and they show a consistency of response in resmetirom treatment as compared to the placebo-treated patients in this study. In terms of the overall MAESTRO clinical development program, it's a very comprehensive Phase III clinical program. We've talked about MAESTRO-NASH and the subpart H population of the population. It is reasonably likely to predict clinical benefit based on liver biopsy endpoints has been completed at this point, and the month -- the 54-month study is ongoing. I will comment also, to be clear, that the safety in -- that we're presenting from MAESTRO-NASH is the safety at the time of the database lock. So patients who had continued in the study for 2 years or longer are -- their updated safety and AEs are part of the safety from that study. MAESTRO-NAFLD, which we read out at the beginning of this year, was a large safety study using the same dose as 80 and 100 milligrams as MAESTRO-NASH, an earlier population in general, and in more than 1,200 patients. And it showed very similar results as far as the noninvasive measures that we have so far from MAESTRO-NASH. There's an ongoing open-label extension study to MAESTRO-NAFLD and more than 700 patients. This study data is being collected as part of the NDA and does have a portion of the study in which the patients were -- it was randomized to either 80 or 100 milligrams blinded when patients entered MAESTRO-NAFLD-OLE. And then after 12 weeks, they were all switched to 100 milligrams. So we'll have a lot of data on both doses of 80 and 100. I'll just comment that we did show data from MAESTRO-NAFLD. Given that we have these 2 results now in both 80 and 100 milligrams that are pretty important, where we showed that in certain populations such as females, the 80-milligram dose achieved similar results, very similar in both the LDL and MRI-PDFF as the 100-milligram dose. So it's a very -- a great value to have a highly statistical -- significant result on the liver biopsy at the 2 doses of 80 and 100 milligrams. The males in the study, 100 milligrams was clearly better. We believe this is not a sex difference, however, it's a size difference between females and males. The other study that we started that helps -- gives us a second Phase III pivotal trial with MAESTRO-NASH-OUTCOMES, which is an event-driven clinical outcome study in patients with well-compensated NASH cirrhosis. That study has just been started in the last few months and will give us a second option that we believe will occur more quickly than the MAESTRO-NASH-OUTCOMES to show -- to help support the full approval of MAESTRO-NASH, MAESTRO studies and support the full approval of resmetirom. So overall, MAESTRO Phase III trials provide a comprehensive safety data set, more than 1,500 patients at the top dose of 100 milligrams and more than 2,000 patients on at least 80 milligrams to support accelerated approval of resmetirom for the treatment of NASH with liver fibrosis. And we'll stop there. And Stephen, if you have any comments, we'll let Stephen comment before we move to questions.

Yes. Thanks, Becky. So it's good to be here this morning. This is Dr. Harrison. So first of all, let me say, this is a breath of fresh air for the field, and that might be the understatement of the year, maybe a watershed moment. It's great news for our patients that are suffering from NASH and advanced liver disease, where it's estimated that over 20 million to 25 million Americans are battling this disease today. When we embarked on this process clinically back in 2015, I think this is exactly what we dreamed of for a therapy for this disease. It's oral. It's well tolerated. It is potentially foundational. It targets the drivers of NASH, particularly toxic fatty acids. The results demonstrate a positive impact on both endpoints as well as numerous noninvasive assessments. The additional benefits of LDL-cholesterol lowering and other lipoproteins suggest a potential impact on cardiovascular risk. Several thousand patients have now been exposed to drug without any important safety signals. A plethora of noninvasive assessments have been obtained, which really will allow for a rich data set to continue to assess these endpoints in these noninvasive tests that we have in our clinical practice that will be targeting both treatment response and potentially outcome measures. Additionally, a significant amount of artificial intelligence data, like you mentioned, Becky, have been obtained. And as we move forward, this will be evaluated in the coming days and months that could augment what we have here today. So overall, I'm very happy for this day to finally get here and to show the results that it did. I think this is just incredibly important for the field and for our patients. So maybe I'll stop there.

Thank you very much, Stephen. [Operator Instructions] Operator, would you please open the call for questions?

[Operator Instructions] Our first question comes from Liisa Bayko with Evercore ISI.

Congratulations, guys. Just super thrilled for all of you and for the field. Just the best news possible. A couple of questions. Just maybe you could comment on how you got a very nice and low placebo rate, which is very helpful. Anything in your trial design or the way you looked at the biopsies -- maybe patient selection was notably lower particularly on fibrosis than some of the other trials we've seen. And then also, I just wanted to ask about F1Bs. I noticed you didn't talk about any statistical significance there in that group but curious if you've seen anything there. I know it's a smaller -- small group of representatives. So I'm not sure if that would even be feasible, but I just wanted to ask about that.

Yes. I mean, I think the fibrosis endpoint, given this high percentage of F3, we didn't see and we don't know a lot of the details here. But the F3 rate of fibrosis in placebo was nonexistent in our Phase II, and it's definitely lower here as well. I think that's probably it. And most of the variability in the fibrosis sampling or placebo response, whatever the [ recent ] rate is based on lower fibrosis stages, so that would be the reason. Now we did also have very good biopsies, we had good samples. Dr. Harrison insisted on a 16-gauge needle. And when you have better samples, it has been shown that you can get a more reliable fibrosis stage for that. On the F1B population, yes, you're absolutely correct, very small number. And we'd just comment that we included them because the NASH CRN had determined that F1B was a progressed form. It was very difficult sometimes to tell from an F2, meaning that it has a lot of fibrosis in the central vein region. And it's a matter of whether you can see a few fibers in the portal where you suspect to see, which distinguishes an F1B from an F2 patient. And if you recall some of our earlier baseline characteristics, the F1B population that we recruited is very similar to F2. The MRE test was 3. The FibroScan was the same. The F1B population was actually more diabetic, which means they're at higher risk than the F2, and a more recent paper from the NASH CRN showed that -- in 2019 by Kleiner showed that progression to F3 over several years in the F1B population was very similar to the F2. So those are the reasons that we included them, not because they would give us a better statistical result. They did give a very similar result that approached statistical significance in terms of both the fibrosis response, meaning they had to go to F0 and the NASH resolution response.

Great. Congratulations.

Thank you.

Our next question comes from Andrea Tan with Goldman Sachs.

Congratulations on the data. One for the team as well as Dr. Harrison. Just wondering here, did the data change your view on which patient population resmetirom should be used in? And I ask this because our KOL checks have suggested that if both endpoints were met, that there would be interest in actually using the drug across all patients, including F0 through F4. So just curious your thoughts on that. Congratulations again.

I think -- I wasn't sure if you were asking Dr. Harrison or me, but if you're asking -- and he can comment as well. But this -- the approval for resmetirom, the indication is treatment of NASH with fibrosis -- with liver fibrosis. The patient population will be as treated eventually, so this is non-cirrhotic NASH. We don't have data in -- we have open-label data. We have safety data in cirrhotic NASH. But the pivotal trial in cirrhotic NASH, which is F4 has just begun. So we don't have data to support an indication in cirrhotic NASH, which is a separate indication. In terms of how patients are diagnosed in the real world, liver biopsies are really infrequently used, and this is a hot topic for discussion among many people. And Stephen, of course, can comment on this. But typically, the agency doesn't specify how physicians diagnose patients for a particular disease including, in this case, NASH with significant liver fibrosis. So Stephen, do you want to just comment on that last point about how patients will be selected if resmetirom is approved?

Yes, absolutely. I mean, I think it's great that we can have these kind of conversations about where we're potentially going to use this drug. I mean I think, look, clearly, the data suggests that patients with moderate to advanced fibrosis are at increased risk of disease progression. It's also important to remember that we want to treat the person, the patient, not just the liver. And where we're able to get extra hepatic benefits of therapy, I think that will be very helpful. And as the field evolves and as we learn more about what this drug can do, I think that'll help frame the population. But for sure, when we begin to look at how to diagnose these patients in real-world settings, there's a lot of guidance documents coming out now. The AASLD presented data on their guidance. It's forthcoming and should be out soon. The AGA has guidance. The ACE, the endocrinology societies have guidance, and all of them want to keep this relatively simple. We don't want to do liver biopsies. We don't necessarily want to do studies that aren't widely available. So while we're not at a precision medicine point yet where we can pinpoint exactly who needs to be treated, we do have a broad paint brush for identifying these people, and that's the noninvasive tests that are really outlined in those guidance documents. And I think if you hit those thresholds, those, in my opinion, would be what I would push for, for treatment. So I think it's still a work in progress, but that's kind of broad strokes where I would go.

Our next question comes from Thomas Smith with SVB Securities.

Congrats on the strong data. It looks like really pretty robust results here. Just a quick follow-up on the safety. I believe there were independent committees that were set up to review and adjudicate certain event types. Can you just remind us, I guess, how often these committees met, how much data you have from these reviews, and how comfortable you feel with the safety profile based on these reviews? And then if you could just comment on how you're thinking about the proposed dosing here? Do you anticipate seeking approval for both the 80 and the 100 milligrams, or are there still analyses ongoing that would inform that?

Yes. So our safety committee structure was that we had blinded adjudication committees for both any liver events, which could include when a -- if a patient had a hepatic complication, which really wasn't going to happen in this study at this early stage, or any potential DILI type episode. There was no incidence of an issue there. And we've talked about that previously with MAESTRO-NAFLD. And in terms of the cardiac adjudication committee, this is -- the focus is what's called 3-point MACE, which is -- or hard MACE, which is an adjudication for cardiovascular mortality, a heart attack or stroke. And then there are secondary CV type of events, which are unstable angina, emergent stenting and hospitalization for heart failure. And nothing there either related or unrelated, I mean, just very few events and no disproportion in those. So in terms of the dosing, I think we're in a great position with these data. We've always considered that both doses would achieve both endpoints. And now we've shown that. And we think that both doses will be used. And based on the data we have right now, we would seek approval for both doses. However, that's -- we will be looking very, very closely at dosing in the various subgroups in this study to come up and also, as I mentioned, analyzing the direct comparison we did in the blinded portion of our -- what's called open-label study, but it was actually blinded where we compared the doses to see if there will be a way of determining the dose that people will start on 100 or people will start on 80 or people will just use 1 or the other dose. So I think the data will drive that decision as to how to use the doses.

Our final question comes from Yasmeen Rahimi with Piper Sandler.

Congrats. I cannot -- I almost started crying this morning when I saw the press release. I mean enormous amount of work went into this, and wanted to say congrats and you deserve this win for the space as well as for yourself given all the hard work. The question that I had for you was, in the past, you guys have done a really nice job highlighting the patients with MRI-PDFF. Greater than 30% fat reduction tend to have a strong pathological response. So with that in mind, could you comment on if you saw that also in the MAESTRO-NASH that if patients are [ paused ] up on MRI-PDFF scale, so we'll have a strong response on biopsy? Any commentary there would be really helpful, and congrats again.

Thanks, Yas, for your comments. We -- these are top line data, and so we still fully intend to do the analyses that you suggest with the MRI-PDFF, of course, and the subgroups that you mentioned and also other non-invasive imaging and biomarkers to come up with, potentially, predictors of response. And we just simply -- this is topline. And so those analyses will be conducted within the next several weeks, et cetera.

Okay. So operator, go ahead. Sorry, go ahead.

That concludes the Q&A session. I will turn the call back over to Dr. Friedman for closing remarks.

So thank you all again for joining the webcast. We're tight on time because we have a lot of commitments today, which is why we couldn't extend the Q&A. We'll have a chance to talk to many of you going forward, and we can answer questions then. Very happy to have had Dr. Harrison on the line with us. Really appreciate that. Obviously, we're excited by these results, and we're going to work as quickly as we possibly can to submit the NDA for resmetirom, a well-tolerated, once-a-day oral agent. I'd be remiss before ending the call to not acknowledge the superb efforts of our Madrigal colleagues, our external collaborators and, of course, the patients who are participating in MAESTRO-NASH. With that, I ask the operator to end the call.

Ladies and gentlemen, this does conclude the program. You may now disconnect. Everyone, have a great day.