Good evening, thanks for joining Madrigal's EASL Investor Event and Webcast. With me today in Vienna are Dr. Rebecca Taub, President of R&D and Chief Medical Officer, and Dr. Stephen Harrison, Lead Investigator of our phase III MAESTRO studies of resmetirom. Fine print. These are our forward-looking statements. I'd ask you to see Thursday's press release and our recent SEC filings for risk factor considerations associated with these statements and our business. We'd like to express our thanks to EASL for selecting the pivotal phase III MAESTRO NASH trial of resmetirom as the first abstract to open Thursday's general session of the Congress. The presentation signaled an important milestone for Madrigal and the NASH field. Based on the positive feedback and words of encouragement we received from thought leaders, community clinicians, and our other stakeholders after the presentation, I believe that the MAESTRO-NASH data set the tone for the meeting. Before Becky and Stephen Harrison review the detailed data, I'd like to share some high-level thoughts and perspective. First, the results confirm achievement of the primary endpoints across multiple patient subgroups with very consistent results. Second, non-invasive imaging and biomarker data support the findings on liver biopsy and demonstrate broad treatment and response to resmetirom. Lastly, further analyses of the results reinforce the strong safety and tolerability profile we've seen with resmetirom across multiple studies. I think that last point is really very important, the tolerability and strong safety profile. Let me turn the meeting over now to Becky and Dr. Harrison to review the MAESTRO-NASH data presented, at EASL. Hear me? Thanks very much, Paul. I'm my plan is not to speak, but to let Dr. Stephen Harrison go over the data that we presented here at EASL. The only thing that I want to remind you of because we didn't put that resmetirom overview slide in there, is that we have four phase III studies that make up our safety database for resmetirom beginning with MAESTRO-NAFLD-1, which read out about a year and a half ago. That was based on identifying essentially earlier NASH patients who were not participating in MAESTRO-NASH, using non-invasive technologies. With that study, we had a very large safety database that almost looks like a real-world or a real-life NASH study, using the same basic protocol design and the same doses as we used in MAESTRO-NASH, the pivotal phase III study that you'll hear about from Stephen. We also had an extension to that study, which is a 52-week extension study. We will, at some point, be talking about data from that study, but that data is very confirmatory of the dosing strategy and the safety of resmetirom. Along the way, we studied about 200 patients with NASH cirrhosis and saw very good safety at the same doses we were using in patients without cirrhosis who had not NASH. That was part of the MAESTRO-NAFLD-1, one of the cohorts in the MAESTRO- NAFLD cohorts. As a result, we were able to start an outcome study, a clinical study in patients with confirmed NASH cirrhosis that has started this year, called MAESTRO NASH Outcomes. That study will be enrolling patients with well-compensated NASH cirrhosis to look for clinical endpoints, such as decompensation, ascites, variceal hemorrhage, and traditional clinical outcome, and would support approval of both in NASH cirrhosis and support the long-term outcomes in patients with NASH, non-cirrhotic NASH, as well as MAESTRO NASH having a long-term outcome portion, as you will hear about from Dr. Harrison. With that, we believe that we have a very robust safety database, putting this all together, and a plus, great data from MAESTRO-NASH, which Dr. Harrison will tell you about now. All right. Thank you, Becky and Paul for the introduction. You know, I'm looking at this room thinking how awesome this room looks, and it's about like the data I'm gonna show you. All right. Basically, we're just gonna go over the lecture again that I gave, and I'm gonna try to take more than three breaths, so maybe we'll take a little more than 12 minutes to get through it. Happy to, you know, go through Q&A. In fact, I think I'm gonna step down, so I can... You can hear me fine, right? Okay. Resmetirom, an oral, once daily, liver-targeted THRβ selective agonist in development for NASH. You guys know that. Just a little bit about the background, we don't go into that a whole lot, but in the setting of lipotoxicity, thyroid hormone action in the liver is mediated through THRβ. That's disrupted in the setting of lipotoxic stress, and that impacts lipoprotein metabolism and beta-oxidation of fatty acids. That's really critical to really understanding what causes damage in the liver. In this setting, we've used THRβ agonism in a phase II trial. You guys know that data. You've looked through it 1,000 times. It was a 36-week trial with the primary endpoint being liver fat reduction, secondary endpoints being, you know, histopathologic. In that study, we showed benefits on NASH. When we did AI digital pathology, we actually showed pretty amazing results on fibrosis, particularly in the subset of patients who had F3 disease. That gave us great promise that this would be effective in a very large phase III trial. In addition to that, liver chemistry test, inflammatory markers, fibrosis biomarkers, all moved in the same direction, and it appeared to be safe and well tolerated in the trial, and so that led us to the MAESTRO-NASH trial. You know, here to this week, we report 52-week data of an ongoing phase III trial, when patients with NASH, with F1B, F2, F3 disease, and this will continue on for 54 months. This is the trial design, and you see the scheme here of placebo, resmetirom 100 milligrams or 80 milligrams in a equal randomization scheme, 1:1:1. To get into the trial, you had to have a biopsy. A biopsy had to show NAS of 4 or more with F1, B, 2 or 3, and at least 8% liver fat on MRI-PDFF. Biopsy was done at week 52. That was the primary endpoint. That endpoint was either NASH resolution, as you see here, with no worsening of fibrosis, or greater than or equal to 1 stage improvement in fibrosis with no worsening of NAS. I think we've beat up what that NAS is versus NASH a ton, but happy to have another dialogue about that if we need to. The baseline characteristics in this population are very characteristic of a robust, advanced, or significant fibrotic population. You see the mean age of around 58, the majority were female. Interestingly, you know, the FDA is big on diversity. We have about 20% of the patients that are Hispanic, which kind of fits that model well. They're overweight or obese. The mean BMIs are well over 30, and the majority are type 2 diabetic. Remember, in NASH, the more fibrosis you have, the older you are, the more female you are, and the more diabetic you are. That's all consistent here. In addition, about 13% of patients were hypothyroid, and about the same number were on Synthroid or some sort of thyroid replacement. FibroScans were consistent at around 13 with more advanced disease, and PDFFs showed a robust fatty liver population. That's predicted by the fact that you had to have at least 8% liver fat. You get high teens with 8%. If it's more than 10% at baseline, you get into the low twenties. That's just something you guys can throw in the back of your head. More importantly, or just as importantly, this baseline medication of GLP-1, we get asked all the time about, "You know, GLPs are here, so is there any reason to develop drugs in NASH?" Here you have around 12%-17% of patients on baseline GLPs. What we've learned from the FGF21 trials and other trials is that being on a baseline statin, Being on a baseline GLP with diabetic doses does not really do a ton to liver fat content. From cohort D of the Akero trial, where 2/3 of the patients were on a GLP-1 for a year, baseline liver fat was 11%-16%, and 89bio's data confirmed that today. Knowing that, here we have around 13%-17% on a baseline statin. This had to be very stable. They had a liver biopsy, and they still had significant liver fat and NASH and F2, F3 disease, despite being on a statin or despite being on a GLP-1. Half of them were on statins, and you see that at least 60% of the patients were F3, and that F3 number comes from the data we derived from the phase II trial, empowering this study to make sure we had enriched for F3. Importantly, there's a discussion about the methodology, and this takes you through that. All the baseline biopsies were reread by 2 different pathologists using glass slides. The ITT includes all patients with at least 1 biopsy with appropriate fibrosis stage. Okay, 1 biopsy makes ITT. 2 biopsies makes mITT. If you didn't have a biopsy at the end of the trial, it counts as a negative. Biopsies were included if conducted before week 60, and patients with biopsies after week 60 were considered non-responders, as were patients who only had a baseline biopsy. 11 patients who had their biopsy conducted after week 60, for example, COVID, you know, closures or whatnot, were included in the primary ITT, consistent with FDA guidance. All these biopsies at baseline and week 52, as I mentioned, were read independently by 2 different pathologists using the gold standard of glass slides. In addition to that, these results were combined statistically using a statistical analytical package called CMH to help generate a single treatment effect. Importantly, each pathologist scores, like each 1 individually, showed a significant magnitude of response on both endpoints at both doses. Baseline biopsies that were F1A or 1C were exploratory and were not included. Here's the primary result. I think you guys have probably seen this 100 times, but the take-home message is a very clean, very nice dose-response relationship to NASH resolution, hitting statistical significance, not just minimally, but significantly, like less than 0.001 for both doses for NASH resolution. For fibrosis improvement, again, very, very similar dose response curve, with the high dose, you know, having a 26% response with a 12% treatment effect delta, and even the 80 milligram dose having a 10% treatment effect delta, both highly significant. I think important for a phase III trial, if you have a key secondary endpoint, that's important from a statistical assessment because it does take some of the statistical power to apply toward that. That was done here looking at LDL cholesterol reduction, which we think is a critical component to this drug and differentiates us from others that are being developed in the field. Here it's significant at both doses. A lot of questions have been raised about the primary endpoint assessment, the methodology that was used. I wanted to take this slide to dive into all the other analyses that were done to support that primary analysis. Sensitivity analysis you see here, this is consensus read, as other companies have reported, the consensus methodology, and the numbers are just slightly different, but the treatment effect delta and the p values are exactly the same. Multiple other pre-specified analyses and supportive analyses were conducted, everything from multiple imputation to tipping point, other statistical models, were done. Any which way you sliced it, diced it, looked at it came up significant. You might ask, "Well, tell me about more data relative to NASH resolution or fibrosis improvement," and I'm glad you asked, because here are, like, six different analyses looking at a deeper dive into fibrosis and NASH. Now, a little more time. I didn't have time to go into this in the presentation, but each one of these we have listed here have some meaningful context if you want to apply this, to other drugs in development. A 2-point reduction in the NAS, and here you had to have a 1 point in ballooning or inflammation with no worsening of fibrosis. The response rate relative to placebo is around a 20%-22% treatment effect delta, highly significant. 2-stage improvement in fibrosis. If you look at the phase II data, people said, "Well, this drug's not gonna work on fibrosis." Not only worked on a 1-stage improvement, but it statistically significant for a 2-stage improvement, with about a 3-fold increased response rate with drug over placebo. More importantly, if you look at the late breaker poster that we had with HistoIndex, which is an AI-assisted qFibrosis assessment, that magnitude of effect change is even more enhanced for a two-stage improvement in fibrosis. What about NASH resolution and one-stage improvement? More stringent, right? What happens when you're more stringent? Placebo response goes down, but the treatment effect deltas go down as well, and so it's harder to achieve significance. That happened here despite that. Yeah, placebo response was only 5%. Treatment effect delta of around 15%. Again, a threefold increased rate of having both endpoints hit. Now, the next two sections are not ITT. These are patients that had a biopsy at the beginning and a biopsy at the end. This is more real-world, right? This is like, I wanna know how well this drug really works. If I do a biopsy at the end, I do it at the beginning and at the end. For NASH resolution, the response rate's 32% and 39% versus 11% for placebo. Really, in real-world life, you're looking at a close to 40% of patients on drug having a response. When we look at fibrosis improvement, 33% versus 16%. More than a doubling over placebo when you look at comparative biopsies. Finally, if you say, "Well, what about either or?" You hit NASH, you hit fibrosis, it's 1 in 2 almost for the high dose and 42% for the low dose versus 20% or so for the placebo. In total, it really doesn't matter how you look at it. The p-values are all highly significant relative to placebo for not one dose, but both doses. We get asked a lot about, what about prevention of progression of fibrosis? It's important to note, different than another phase III trial that's been talking about their progression data, the FDA was very strict in saying, "You can't report your F3 data progressing to F4, because that's an outcome." This is an ongoing 54-month study. You can't unblind it and show what their percentage is now, but we can for the F1b/2 patients that progress to F3, and that's listed here. What I said in the talk was, as placebo, you're twice as likely to worsen and half as likely to improve relative to both doses of drug, and that's illustrated by the black looking at worsening and the red part looking at improvement. The next question we get asked is: What about the NAS components, ballooning, inflammation, steatosis, combined NAFLD activity score? That's on the right. Ballooning, inflammation, steatosis, NAS, improvement is in red. You see the two different doses in placebo. For every one of these, there's a dose-response relationship. Not quite as good for inflammation. I think I was asked the question about that during the presentation. Pathologists never take inflammation scores to zero. It just doesn't happen. They'll always find a, you know, a white cell somewhere. You know, there's a limitation there. If you're ballooning or if your inflammation score is one, if you have an NAS of four, two for steatosis, one for ballooning, and one for inflammation, they're not gonna take the inflammation to zero. It just doesn't happen. We're a little bit, you know, hamstrung on that result. The other thing is, when you look at kappa statistics, balloon inflammation is really the toughest one for pathologists to agree on. I didn't spend a lot of time on this slide, this forest plot. The next two slides are forest plots, and these are really looking at key subgroups. The first one is on primary endpoint, the second one is on different drugs that are used. NASH resolution on the left, fibrosis improvement on the right, the direction of change for each of these favors resmetirom in every subgroup assessed. It doesn't really matter, you know, what's going on here. There's one here with age greater than 65 and for fibrosis improvement. You know, this drug is affected by body weight, older patients that tended to be lighter weight probably do better on a lower dose of drug. What about weight loss or concomitant drug therapy? I showed you at the beginning that about 14% of patients were on GLPs, 13% on thyroxines, and about half of them were on statins. Whether you look at NASH resolution or fibrosis improvement, unless you lose a lot of weight, you know, the direction of change is still. Even with losing weight, the direction of change is in favor of resmetirom. The confidence intervals cross the zero line for patients losing a lot of weight because we know losing a lot of weight does have some benefit. What about LDL and atherogenic lipids? I mean, you can see the lines going down relative to placebo in blue, that's either flatter or getting worse. We've known this for a long time, that the THR-β class has an impact on atherogenic lipids, and this is the first big trial to show definitively that these atherogenic lipids that are listed across the top are significantly lowered. Just to put context around one of them, ApoB, you see 22% change from baseline for the high dose. For every 10% drop in ApoB, you get roughly a 10% CV risk reduction. That's from the CV trials. We haven't obviously looked at that outcome here, but it's positive nonetheless. Liver chemistry test, and I have an asterisk here at the top, because this is just looking at the ALTs greater than 30 at baseline. You see a very nice response for both doses on ALT that is-- I call it the tortoise approach, because it's very steady and very consistent from beginning of treatment all the way to the end. You know, you'll see some therapy, some mechanisms action, where the drop in ALT is rapid and sustained. Here, it's more of steady and consistent drop over time, and even out to week 48, we haven't plateaued yet. What does that mean going forward? We're not sure, but we have the MAESTRO-NAFLD-1 trial that's extended to 2 years, so we can continue to follow some of this. AST drop is very nice. I got asked about what percentage of people normalized ALT, and I didn't have the answer, but because you came tonight, we'll give you the answer, 'cause I know somebody out there is gonna wanna know. About 25% had ALTs below 30 at baseline. At the end of treatment, it's about 60% had normalized ALT, meaning below 30. Gamma GT is a marker of oxidative stress, and that's one of the key components when we look at how this drug works in the liver. You know, it works to kinda clean house on all the dying or sick or you know, senescent mitochondria that just are thinking they're going off into their golden years. They don't just go to their golden years; they kinda become bad actors. This drug helps with mitophagy, the clearance of these kind of ill mitochondria. It also helps with the generation of new mitochondria. We call it mitochondrial biogenesis. As that happens, we're able to clear oxidative stress and not generate new oxidative stress, and one of the markers of that is gamma GT. Then sex hormone binding globulin, you can see at 12 weeks, it's almost maximal upregulation of SHBG, and you see it's a dose response, and it's also a predictor of response to therapy as well. So this is quite interesting. The MRI-PDFF drops in a manner that is consistent with the phase II data, reaching a maximum of around 51% for high dose at week 52 and 42% for low dose. If you provide a steatosis or liver volume correction to steatosis at the high dose, you increase that to around 60%. You know, one of the things we didn't jump into because I didn't get asked this question, but I know, again, somebody will ask me here: What about the PDFF response, you know, predicting response histopathologically? A greater than 30% drop in PDFF predicts response to not only NASH resolution, but also fibrosis improvement. Very, very nice response there. FibroScan CAP, which I don't like as much as PDFF, however, it still drops significantly relative to placebo. There's a whole story coming out with liver volume. I, you know, here we're seeing about a 26 drop in liver volume at week 52. Other drugs that have shown that, started with Mounjaro, actually, showed liver volume reduction. 89bio's shown it. Akero hasn't shown that data yet, to my knowledge. Altimmune has shown it. Interestingly, if you look at Altimmune, I think, you know, you were telling me this earlier, the reduction that I presented with Altimmune's data is around 20% reduction in liver volume, but their liver fat content reduction was more than what you see here. The liver volume change with resmetirom is even greater. What does that mean? Well, what it means to me, at least superficially, is it's not all due to liver fat content reduction. There's something else going on here. If you think back to the MAESTRO-NAFLD trial we presented in, what, AASLD, right? Last year. In the cirrhosis arm, where we had, believe it or not, the fatty liver in cirrhosis, you have bigger livers. In viral hepatitis, you get shrunken nodular livers in cirrhosis. Cirrhosis in fatty liver, you actually have an enlarged liver, so normal's around 1,500-1,600 ccs. That, if I remember right, cirrhosis was around 2,100. You still saw a reduction in liver volume in cirrhosis, but you also saw a reduction in spleen volume, and that correlated inversely to platelet count. Is there some hemodynamic change that's happening that's reflective in this liver volume? Is there some sort of glycogen reduction, water weight reduction, inflammation's getting better? There's probably a combination of a lot of different things happening, but it's not just liver fat content. There is data that shows a big liver is predictive of overall mortality, even more so than FIB-4. We don't know what reduction of liver volume means necessarily. We're gonna have to work on that. One of the things I'm interested in, I don't think I talked to you about this, is: Can we use liver volume to predict histopathologic benefit? I don't know if we could look back and see, you know, changes in liver volume correlating with NASH resolution or fibrosis improvement. The spleen volume data is very interesting as well, and you might say, "Well, why is that important?" Well, a normal spleen size is around nine to 10 centimeters, but we don't really call splenomegaly in the setting of cirrhosis and portal hypertension until it's more than 13 centimeters. There's actually an enlarging of the spleen that happens before you technically get to portal hypertensive cirrhosis, and we're seeing reductions in that. What that means, I'm not sure, but the two of these together, I can almost guarantee you, in the coming years, there's gonna be a lot of data coming out about how that is suggestive of what's really happening to these patients physiologically. What about VCTE? That tends to be the non-invasive imaging tool that we talk about at meetings a lot, and there's around 2,000 of them throughout the U.S. I don't know how many are in Europe, but, you know, it's not on every street corner, but it's probably the most common non-invasive test we have from an imaging perspective outside of a ultrasound. It is helpful in some regards. At a, at a kind of a global level of when we look at different stages of fibrosis, we see drops that are significant relative to placebo for both doses, whether it's F-one, B-two, or three. Well, I like the way this was done. This responder analysis, you know, we say, "Look, what's the coefficient of variance?" If you do one FibroScan on each of you right now, and you don't eat, you don't drink, and six hours later, I repeat it again, or tomorrow I repeat it again, what's the probability that it'll be... You know, what's the variance? It's somewhere 25% or higher. What we said here is we wanna account for at least 25% variance. It's gotta be higher than a 25% change. Then we looked at 80 milligram, 100 placebo with the different stages of disease and then combined. If you just look from here over, this is drug, and the gray is improvement. Just 30,000 feet, just looking at this, I mean, it's significantly better than what you see here, and black is worsening by 25% or more. You know, just looking at it, there's some meaningful change that's occurring, and this is at this, you know, the group level, not at the individual level. There was also important changes in MR elastography as well as ELF. Fibrosis biomarkers are all moving in the same direction as a liver biopsy, which is very encouraging. As Paul mentioned, as we know from the AdCom with Intercept, you know, the FDA doesn't just look at efficacy and say, "You hit the endpoint, you're good." There's something called the therapeutic index of a drug, which is efficacy and safety, we know the FDA is focused on that first drug being approved, having incredible safety. What I'm really excited about, equally as much as the efficacy, is the safety of this drug. I've personally been dosing resmetirom patients for a long time, since 2016, 2017 timeframe, and probably over 2 or 300 patients I've treated for a long time. I was one of the first to enroll in MAESTRO and NAFLD 1, many of those people are now in open label extension, first to enroll with cirrhotics in that trial as well. It's incredibly well-tolerated, and this data corroborates that. No difference in treatment-emergent adverse events, both mild to moderate. I commented on treatment-emergent events leading to study discontinuation. You see no difference between the 80 and placebo. There is a numerical higher and a percentage higher amount here in the 100 milligram dose, what we said there was this was really GI related and tended to occur in the 1st 12 weeks. I'll talk a minute about what those GI events were, but that's really it. I mean, if we look at fatal treatment-emergent adverse events, similar across 3-point adjudicated MACE, similar other cardiovascular events, more in placebo than in the drug. There were no DILI events, no drug-induced liver injury noted at all in the study. What about the emergent events more than 5%? It's what you would think, you know, diarrhea, 33.7, 27.6 versus 15. Even placebo, I mean, everybody gets diarrhea, right? In the EDC, the Electronic Data Capture System, more than 50% of these patients had worsening of persistent diarrhea or intermittent loose stool. You know, in EDC, diarrhea, loose stool is diarrhea, increased frequency of stool is diarrhea. It all kind of gets lumped into one heading, but there were nobody that discontinued for worse, like, for severe diarrhea. It was mild to moderate. You know, my patients generally in the study, if it happens, it happens, like, when you start dosing. Duration is around 2 weeks, and it doesn't matter what dose you're on. Same thing with nausea, that's here. The only difference in nausea is it tended to occur more commonly in women. In conclusion, I mean, you guys, we can probably add about 10 lines to this. I didn't have time to do that. Just to say, it's the first treatment to achieve meaningful effects on both endpoints and that are reasonably likely to predict benefit. Both doses were effective, so there's optionality for patients. It doesn't matter how you slice the data, how you look at it, the way we did as a primary analysis using the statistical model. If you look at individual pathologist scores, if you look at consensus scores, tipping point, multiple imputations, it all was significant. Safety is very, very good. Limitation is we don't have outcomes data with the drug yet. We feel like, and hopefully, you do as well, that really this trial and the three other phase IIIs give us enough positive data, enough safety data to get approval for the drug. We included all of the investigators, so that's great. There was a whole bunch of them, including the patients, the coordinators, the staff that have been dedicated to this trial for a long, long time. They're still dedicated because some of these guys still have to, you know, be checked out by the FDA, because with an NDA submission, a lot of these high-enrolling sites are gonna be checked out. There's still a lot of work to do to make sure everything's tucked in tight for this, for this compound. With that, I think I thought I was done. I guess I'm not done? Yeah. Just, I think there's one slide on this. We mentioned this earlier when I said there was a late-breaker poster with HistoIndex. This is the analysis from that. Not every slide went. I think it was 760 something. We don't have that listed, do we? No. I think it was well over 700 paired slides went to HistoIndex. As you guys know, that is one of the AI digital pathology companies. You send them one unstained slide. It's looked at by a laser, and the cool thing there is you get 5 different categories. fibrosis is looked at at 5 different zones in the liver, from portal, periportal, sinusoidal, perivenular, and central vein, or pericentral vein area. It's very granular in the analysis. They also look at string length, string width, cross-linking of collagen fibers. There's 168 different parameters that are looked at by these guys. Just at a top level, not boring you anymore with that, but when you look at one stage improvement or at least one stage improvement, placebo went up to 34%, but the drug effect went up to 58, or almost a doubling of the treatment effect compared to ordinal scoring. The same thing with a two-stage improvement here. Even importantly here, remember I showed you that on worsening one B, two going to three, there were twice as many placebos worsening, 34 versus around 15, 16. Here, it's three times as many. You just get a little bit... There's two things I like about this: it corroborates what our primary endpoint shows, and it takes a deeper dive into what really is happening in these patients, and this actually shows even better results. In addition, there were significant correlations between reduction in qFibrosis and reductions in PDFF, ALT, AST, and ELF score. Now I think the floor is open for discussion. Okay. Thanks. We'll begin to take questions. We also have people online, we'll be going back and forth a little bit from between people in the room and people online. First question goes to Ritu from Cowen. Thanks, Becky. Ritu from Cowen. I just wanted to ask, we were discussing earlier today on the concordance of the reads. There was something brought up in the Intercept AdCom, where the agency pushed back on their low concordance of reads. You mentioned the CMH methodology that you used, and then the sensitivity analysis that used the, like, a secondary methodology. Can you give us any more detail on CMH and compare and contrast what's done with CMH versus the consensus digitized method? Like, how different are they, really? Yeah. I can handle this one. First of all, I think what happened with Intercept was completely different. Intercept, in their primary analysis, they used a very unusual way of reading the slides, which was they read the slides as pairs. You know, like an individual patient would have slide 1 and slide 2. They read them as pairs, which had never been done by anyone, by the NASH CRN or anyone. They also, they had 2 central readers, but they didn't read the same slides. Each central reader read half the slides. You didn't have the advantage that we had, where each central reader read the entire study, that was the approved method with FDA. The FDA method that was approved for our study was two central readers with the CMH analysis. In addition, there was what we call a consensus read, which was also something we agreed with FDA, which was that we would take the slides. You know, there's 2,000 slides in the study, a baseline and a week 52. There's actually 4,000 because there's two slides for each time point. There's what's called a trichrome stain is for the fibrosis and a H&E slide for the other components. We would do a consensus read on the situations where they didn't agree on that there was a response. If they agreed that there was a response, that it was either, depending on what the response was, either a fibrosis response or a NASH resolution response, or they disagreed, or they agreed that there was no response. That if they agreed there's no response, there's no reason to, you know, look at it any further. We had them do a very blinded consensus read on the cases where they didn't have agreement, and that's what this consensus read is. What you see is that the statistical methodology of combining their original independent reads was exactly the same as after they did the consensus together on those. Yeah. What? Consensus of digital. Yeah. Somehow, that's not working. Sorry. Was that the statistical methodology agreed with the output of the consensus read of the digital methods? When they did the consensus, they used the digitized slides, right? Originally, they had the glass slides, but you can't- We were able to send the glass slides back and forth once, but you can't keep sending glass slides back and forth. I think this is why this consensus thing switched media from glass to digitized, because digitized can be shared between pathologists. One of there are issues with it, and I think you've seen this in some of the outcomes from these studies, is that not every pathologist can read-- Every pathologist can read glass slides, and that all outcomes in NASH studies are based on glass slide reads. Not every pathologist can read digitized slides. I think what you're seeing with some of these odd placebo numbers, that have come out in some of these recent studies, is certain pathologists not able to read, digitized images very well and seeing different, very low levels of improvement in certain components. At least that's my opinion. I don't know if, Stephen, you have another opinion on that. Yeah, I mean, it's pretty well known in the pathology community that, you know, looking at digitized slides, a lot of pathologists are used to glass, where they can turn up or down, and they can look in a little bit further, almost a 3D view of what's happening. You can't do that with a digitized image. There's a bit of a training you have to have to switch from glass to digital. Then some, believe it or not, the viewers aren't necessarily standardized or validated, so, you know, not all viewers are created equal. There's the viewer issue, there's the computer pixel issue, there's the fact that you can't toggle up and down through a digitized image like you can a glass slide. There's some people that can do it very well, and others that, you know, almost feel like they need to go back to fellowship to learn how to do it again. Yeah. Yeah. So anyway, the bottom line is that it didn't matter that, you know, we used the digitized images, and they came up with a consensus that looked virtually identical. So we had 2 independent pathologists, each showing that the drug was highly effective at 2 doses on 2 endpoints. Then we had, you know, the combined consensus read, which also confirmed the same thing. In addition to all the other sensitivity and supportive analyses that were done in the study. You know, statisticians will tell you that having independent reads rather than a collusive-type read of data is to them, more convincing of the result. Lisa? Hi, there. Hi, Lisa Baker from Cowen. Cowen. Sorry. I don't think I'm going to Cowen anyway. Let me know if you hear something different. Okay, it was really great to see that there was no MACE. You also talked about adjudicated DILI, no cases there. I've gotten a couple questions on DEXA scans, if you did those, any changes in bone. Can you comment at all on that? Yeah. We have commented. We did serial DEXA scans in MAESTRO-NASH, MAESTRO-NAFLD-1 extension, where that one's particularly interesting because we have baseline week 52 and then week 104. You can see over time, there is no shift in fracture risk index, as determined by BMD, with resmetirom in the high-risk population, which is the post-menopausal women. There's also, you know, We looked at exposure markers, you know, people with higher exposure. We did it every which way. We'll be, at some point, showing, you know, more data from that, but it's a very detailed analysis of what we've done on bone mineral density and DEXA scanning in a very large number of patients. The reason we did it is because, hypothetically, you know, if it was spironolactone, it might have an effect on bone. With resmetirom, if anything, we've already always seen, like in the animal tox type studies, we've always seen markers of actually more bone formation, if anything. It's been, THR-β is very different from thyroid alpha, and, you know, none of this is significant, you know, so we don't really talk about it in that regard. When we looked at, you know, sort of baseline week 52 and week 104, those data aren't complete yet, but it gives you the coefficient of variation across, you know, 3 different measurements, which is very, very helpful, given that these patients are obese and it's hard to do DEXA in them. That's where we are. We've seen no fractures. You know, they've been very balanced across all of the groups in all of our studies. Yeah, I don't know where Kia is. Kia, are you? Did you wanna do one online? Sure. We've been receiving a lot of questions from folks on the webcast, essentially starting from the introduction slide that you put up. I'll try to go through a few of them. First question from Andrea Tan, from Goldman Sachs, asking Dr. Harrison that, "Can you speak to the profile of the patients that you will put on resmetirom, and how you envision resmetirom being used in real world? Okay. Let me just make sure I have the question right. The idea is, how would I see this being used in my clinical practice? Okay. Yeah, well, I mean, look, we've been waiting. I've been a hepatologist. I finished fellowship in 2003, so 20 years, I've been seeing patients as a board-certified gastroenterologist, and before that, as an internist, since 1995. We don't have any FDA-approved therapy, and we haven't had, ever. You know, having that first drug, I don't know how many patients I've told, "Just wait, hang tight, you know, we'll give you vitamin E, we'll give you a GLP-1, we'll do whatever we can to kind of placate you and get you to the time we have our first approved drug." You know, I think for me, there's a lot of patients that are just waiting for drug approval, that aren't really responding to lifestyle modification. That's all I've been able to hammer into them, you know, for the long, long time, and they're maybe not responding to vitamin E or pioglitazone or a GLP-1 or SGLT2 inhibitor. I mean, literally, we've thrown the kitchen sink at a lot of these patients. You know, I think there's a lot that are just gonna be ready to go the minute this drug is approved. I think if you step back and say, "Okay, where would this drug really be used?" I mean, if you're a diabetic that doesn't have controlled diabetes, I probably would focus on drugs that have impacts on glycemic control first, see what benefit I can get, and then add in resmetirom. If they're not a diabetic, you know, that makes a perfect opportunity for resmetirom to be used right off the bat, even if they're not on a GLP-1. Maybe I think that's where I would start. We see a lot of people, as I mentioned at the beginning, on baseline. By the time they get to me as a hepatologist, you know, I'm seeing overweight, diabetic, hypertensive, hyperlipidemic patients. You know, they're already on a GLP-1. Many are already on an SGLT2. A lot of them are on a statin. As a hepatologist, there's a lot of people that I would immediately put on resmetirom. As a tertiary care provider, I mean, we're probably gonna jump to that right off the bat. If you're primary care, I still think you know, there's a little bit of an algorithm you would go through to get to resmetirom. I think, you know, that we would target, you know, what their primary issue is. Is it obesity? Is it diabetes? Well, how are those treated? Are they well controlled? Okay. Well, let's go there. I mean, you think about it in reverse, too. Resmetirom, where is its limitations? It doesn't induce weight loss, and it doesn't have glycemic control. If that's not your primary concern and the liver is the primary concern, this becomes the primary drug to treat with. At a tertiary care level, I let the endocrinologist and the primary care manage those other issues, and most have already done that by the time they get to me. You know, if you look at the percentage of patients that have fatty liver that I or my colleagues would put on resmetirom, it's a lot higher than it's probably GI next, then endocrine next, then primary care next. The pool of patients is bigger at the primary care level than it is at my level. It's kind of a long-winded answer to the question. Ellie? Yeah. Thank you guys for taking the question. Maybe just as it relates to the spleen volume decreases, if you could elaborate on some of the implications and learnings from that, and then just sort of any of the learnings from these results as you think about outcomes and, you know, implications from some of the data here. Then, another question in terms of the regulatory filings, how you're thinking about the range of outcomes and potential labels and sort of, you know, the different scenarios that that could be. Then lastly, just on the. Oh, my God. Okay, fine. I'll stop. Can you start with number one again? I'm done. Okay, what was the first? It was spleen volume, liver volume. Oh, okay. -reductions. Got it. Implications for- Yeah ... you know, outcomes. I mean, this is literally the genesis of a new biomarker, if you will. I don't even know how this came up. How did we decide to measure liver volume? What made you think of that? I didn't think of that. I don't know. It just started to appear a little bit, and we had, you know, so many PDFFs, you know, that we've done thousands of them. you know, it was something that our radiologist, at Duke, Mustafa Bashir, you know, is very interested in technology and also setting things up in a very automatic way. Mm-hmm. Uh, so he set up how to measure liver volume and also spleen volume, you know, in a rigorous way that was highly reproducible, which radiologists like to do. Right ... as opposed to people using FibroScan, right? Yeah. That's how we got started with it, and we saw a huge effect right away, so we kept going. Yeah. Yeah. Others have, you know, obviously, the best form of flattery is somebody else reproducing, doing the same thing you're doing. Now we see every company that's, you know, is in this game, that does MRIs, is looking at liver volume. There's not a ton of data, but there is the Naim paper from Liver International 2022, that shows a link with liver volume and overall mortality. That's even greater on a hazard ratio than FIB-4. In fact, it's 1 of the highest predictors of overall mortality. It's interesting, I mentioned this before, that when you look at viral hepatitis, I was always taught as an internal medicine doc, as a GI doc, as a liver doc, that when you develop cirrhosis, your livers became shrunken and nodular. I remember doing liver transplants on PBC patients, hep C patients, hep B patients, alcoholic hep patients, and these livers are hard as a brick, and, you know, they're small and nodular. I mean, they just look ugly. Fatty liver cirrhosis, yeah, you get nodules, but the livers are bigger, and that is turned out to be predictive of a negative outcome. Reducing volume is something that needs more work to correlate to fibrosis improvement, to NASH resolution, and ultimately to a benefit on outcome. One of the benefits, like, we talk all the time about how we started the cirrhosis, you know, cohort in MAESTRO-NAFLD-1. I don't know if you guys know, but we know we had patients that were begging for resmetirom, that I would see in clinic that were cirrhotic. I remember calling Becky one time saying, "Can I get compassionate use resmetirom for these patients? They're just begging for it." She said, "No." "Why don't we do a study? Why don't we add a cohort onto the MAESTRO-NAFLD-1 trial, open label. We'll put 20 patients in." There was so much interest that we finally capped it at 180 patients, and that has been a wealth of knowledge. It's open label, so there's no placebo control, but just following those people out to 1 year and looking at all of their dynamics, including liver volume and spleen volume, we've learned a ton. We've learned that, yes, cirrhosis and NASH, they have much bigger liver volumes than healthies, and spleen volumes get better as liver volumes get better, and that inversely correlates to platelet count. What does that mean? Well, we didn't do HVPG. We didn't do the gold standard for measuring portal hypertension, which we know is linked to outcomes. We don't ever wanna do HVPG again as a research entity. It's just so hard. It's so variable. This is almost a surrogate of that. If you can improve platelet count or... and you can shrink spleen volume, but we still need work to do to really understand what that means, right? It's more exploratory than anything else, but it gives me something to work on for the next year. It's trying to understand, what does this really mean? It's not just for resmetirom, it's for any mechanism of action where we're seeing this reduction, because this could be a surrogate marker of response to therapy, right? I mean, we just need to kinda do a little more digging on that. Yeah. That's a great question. You had another one, though. She wasn't done. Oh, yeah, yeah. She- Exactly. She had four. I had two. The second question, can you just describe what you see as a range of potential outcomes in terms of the label? On? The, FDA label. For resmetirom, like in terms of the, you know, Not outcomes like an outcome study. Oh, sorry. Yeah. Yeah. Yes. I mean, yeah, there's nothing really to say. You know, we have a very strong case for resmetirom, of a very highly clinically meaningful result. We believe a very great risk-benefit profile for resmetirom. What I'll say to that is. Yeah. The FDA doesn't approve drugs. They approve claims. You know, it all comes down to what has ended up in the filing, right? Right. I think that is what will be looked at and mulled over and discussed, and... Yeah. Our indication is the treatment of NASH with liver fibrosis. You know, that's what we obtained breakthrough status on recently with our phase III data. I guess that's the claim. That's the claim. Right. Kia, did you wanna take another question? Sure. Yeah. Thank you. Question from Jonathan Wolleben of JMP, and I know others have a similar question in mind, and they are Dr. Harrison, they're taking you up on your offer to walk everyone through again, NASH versus NAS. Oh. I know it's something that keeps coming. As a follow-up to it, or more specifically on inflammation, the difference that you're seeing in between. Okay. patients that have... All right. -inflammations, one versus higher, and how that improves, because you make the note that the ones that have one barely go to zero. Right. If you can comment on that. Yeah. Before we get started, because NASH is measured by NAS, in using the NASH CRN scoring system, I think there are some odd interpretations. NASH is not one component, what's more important is that we've had the same endpoint since we started our phase III clinical trials, the same endpoint in our phase II Lancet paper. It's approved by the agency as our endpoint that's reasonably likely to predict clinical benefit. I think, Stephen, you can go ahead and go into how someone is interpreting worsening of NASH, which in and of itself is not a measurement, right? Right. How you measure it, how you measure NASH? Yeah. I think, you know. Yeah. I remember having questions asked by the audience at NASH TAG. The first weekend in January of this year, same exact question. I thought I'd put it to bed then. It keeps coming back. There's a lot to talk about here, we're talking about a letter, N-A-S versus N-A-S-H, right? That's the conundrum, everybody says, "Well, Intercept did it this way. Madrigal is doing it this way." At the end of the day, what we've agreed as key opinion leaders in this field is we need to get rid of greater than or equal to one stage improvement of fibrosis without worsening of either of those two. We just need to stop at an endpoint being greater than or equal to one stage improvement of fibrosis, period. When we sat down with the FDA more than a decade ago and came up with these surrogate endpoints, we didn't have the knowledge that we have today, and we said: "You know, if you have one, you probably don't want the other to get worse." I mean, that made sense 10 years ago. Today, what we've learned from the elafibranor data, God bless its soul, negative study, still hasn't been published. I got it. We presented the data from Dave Kleiner, published in JAMA, from Our phase II, published in The Lancet, is when fibrosis gets better, inflammation and ballooning get better. They don't get worse, right? We never saw a case in phase II, where fibrosis improved and inflammation or ballooning got worse. If you look at the JAMA paper, very tight correlation between fibrosis improvement, ballooning and inflammation improvement. Fibrosis worsening, ballooning and inflammation worsening. If you look at the elafibranor phase III data, same exact thing. Abstract form, I got it, but it's there. You have 3 different studies showing very similar results. I mean, we're to that point now that this discussion around the nuance of 1 letter needs to just go away. Greater than or equal to 1 stage fibrosis improvement, stop, hard stop. There's nothing that needs to come after that. Having said that, you know, I think it is what it is. You see the P value, it has 3 zeros in front of it for fibrosis. Doesn't matter how you slice it or how you look at it, my guess is it's gonna come up still positive. Not still positive, exactly the same. Yeah. I mean, cause it's a very minor thing, and it's not impactful. Why anyone would think it's impactful when you- Yeah. have that type of NASH resolution score and that type of component response is beyond me. Why not the other way? Why not have a discussion with the obeticholic acid about why you didn't look at it the way Madrigal looked at it? Yeah. Is it always a discussion about? Yeah. Why didn't we look at it the same way Intercept looked at it? It doesn't really matter at the end of the day, if the meeting with the FDA is what it is, and there's agreement on the endpoint, cold stop, you're done. I don't think there's any really more discussion. There's a lot of cool things we can talk about with drug development and NASH, and that's not one of them. Okay. All right. Ed. Ed? Can you hear me? Ed Arce with H.C. Wainwright. You showed the data in LDL, as we know, works really well and obviously in TG as well. I wanted to ask your thoughts on the other atherogenic lipids, ApoB in particular. The degree to which we have data that correlates to longer-term outcomes for NASH patients, how well-defined is that? How does the agency view that, in particular for, you know, purposes of predicting, you know, full approval, perhaps in a few years? Just thoughts on that. Well, I'll give you my thoughts, and maybe Becky can speak directly to resmetirom. You know, I... What I've come to realize as an internist, first and foremost, trained generally in 24 subspecialties of medicine before focusing on GI, and then more specifically on liver, and then more specifically on NASH, is that we have to think about the whole patient. In so doing that, it's the number 1 reason for death in several studies, is cardiovascular disease, then we need to begin to think about what kind of impact can we leave in our patients that aren't just focused on the liver. Atherogenic lipids makes complete sense because atherogenic lipids lead to heart disease. That's how the whole statin field was developed and the PCSK9 field after that. The division of hepatology nutrition is focused on a liver endpoint, right? One of the things we have done in the past one and a half years is start a whole new conference called Mosaic. I started NASH TAG, and now I started Mosaic, and the purpose of Mosaic is to get after combined endpoint assessment. Having the division of the cardiorenal division, having hepatology nutrition division present, maybe metabolic division, to really break down stovepipes and begin to look at drug development from a holistic perspective, not just an organ perspective. To me, you know, I don't have any preconceived ideas that the FDA is gonna look at a MACE event in the setting of a liver disease study today. That doesn't have anything to do, in my opinion, with the label of the drug. It's more to do with, that's a nice to have. It's a really nice to have, right? We're not worsening any other organ system when we treat with resmetirom. We have a positive benefit that likely will impact CV risk down the road if enough patients are treated. you know, I don't think to me, that's gonna impact any label. Yeah. Exactly. It is a separate indication to show a positive benefit on lipids, and it's viewed in the context of an important safety feature with resmetirom. I think you were probably aware of some cases of drugs that did start to show some cardiovascular benefit that may not, like SGLT2s, that showed some heart failure benefit, where now that made sense to look at that a little more closely. In the case of resmetirom, we believe that this population of NASH is really the right target patient population, not only for the fact of it targeting the liver, but it's also a great population for the atherogenic lipid-lowering benefits of resmetirom because of the triglycerides and the ApoB. They are very the NASH patients are very much in that category of benefiting from the lipid effects of resmetirom, but it is not a separate indication. I think it will be viewed as a very positive safety feature at this point for resmetirom. Thank you. Question from Jay Olson of Oppenheimer. Was there any reduction in the portal hypertension in MAESTRO NASH? If so, could that be pursued in a potential indication? Then my own add up, if so, do I start to model for it now? Yeah. Well, we didn't measure portal hypertension in the trial. The short answer is we don't know. I mean, as we've alluded to in the conversation, there's a lot of data that we're generating that suggests as we progress in disease, and we begin to have hemodynamic changes, that maybe some of those are potentially mitigated with drug therapy. You know, those formal studies have not been done. We do have the MAESTRO-NASH OUTCOMES trial that's being done, and while no direct measurements of portal hypertension are being done, we do have a lot of non-invasive tests that are, you know, correlated to portal hypertensive changes. That'll be something to follow very closely because I do think we're gonna have a positive impact in that group of patients. Okay. Tom? Awesome. Thomas Smith, SVB Securities. Thanks for taking the questions. Just on dosing, I know you've talked about filing for both the 80 and 100 milligram doses. Can you just walk us through your approach to labeling with respect to starting dose and possible titration, either up or down? Then, for Dr. Harrison, if you do have both doses available in the real world, how do you plan to implement that? Is this the case where you've made some comments around PK variability, I guess, in patients with some differences in weight? Is this the case where you see starting everybody at a certain dose, or are there certain populations where the 80 milligram dose makes more sense? Thanks. Yeah. I'll let Becky take the first stab at it. Yeah. I mean, I think what we wanted to really highlight was that having two doses with very good effects gives us a lot of potential for treating patients in a way that is effective for that patient. What the one thing we found to be a factor in terms of the dose for a patient or what influences the exposure response parameter, is that if patients are weigh more, they will tend to need a higher dose. If they weigh less, they tend to need a lower dose. Body weight is the only influencer of, in this population, in the NASH population, of dose. There is one slide, which is the... that Stephen pointed to a little bit, but just to give an example, what would be the 65 or older age group, and if you look at both the NASH resolution response and the fibrosis response in that age group, the 80 milligram dose looked better in that age group. We looked into that carefully to see if it was just the small sample size. It's not a huge group. There are about 70 and 80 patients that are greater than or equal to 65 at 80 and 100, then the bulk of the patients are less than 65 in both. If you look at the black squares, if you could see that, you'll see that that's the one where you see a significant shift, where 80 appears to be outperforming 100 on both endpoints. Why would that be? Is there any reason for that? We looked into it, and the older age patients do tend to be smaller, have a lower body weight, and there may be some tolerability that's somewhat different. That might be, you know, in terms of choosing a dose, we think the majority of patients are gonna start with a 100 milligram dose, but there might be some subgroups like that that will be better with an 80 milligram dose. The only thing I can add to that is, you know, we've shown that SHBG is a predictor of response to therapy. You know, maybe there's a role at some point to use that to dose titrate up, even if potentially we're not getting the bump in SHBG that we would like to see. What else? I mean, I think the important thing you see is this is different from percent difference from placebo. Even in the older population, the center of that circle is to the right, so there's still benefit over placebo in all of those patients, even though the confidence interval does cross that line. These are relative nuances of how this is gonna be used, and I think if you ask me next, well, next year, yeah, EASL next year, where I probably will have dosed many patients, we'll have a better idea of, you know, is our hypothesis right in the way we do that? You got a question? Another question from webcast. Yaz Rahimi from Piper is asking, since she wasn't able to attend EASL this year, "Dr. Harrison, what is the feedback that you're receiving from hepatology community in response to our MAESTRO-NASH data presentation? Yeah. Well, I would tell you, Yaz, if you were here, so sorry you didn't come. No, I'm kidding. She, I'm sure, is home with her kid. I mean, the response has been overwhelming. I mean, I told somebody earlier, what I'd like to say at the end of the presentation is: We did it, right? Yippee, yay! 'Cause I know there's thousands, millions of patients that are super excited to hear the data that came out of this meeting. You can't do that, right? You're academic. You gotta be stoic and just kind of present the data. The response when you walk off the stage is dramatically different than when, you know, you're on the stage. The response is, you know, this is a watershed moment. It's a landmark moment for the field. It's something I've been looking for 20 years or more. Many of my colleagues have even been searching for this for longer. I think, you know, it speaks to us, our understanding of the pathogenesis of disease, learning from the failures that have come before. It's just been a, it's an incredible event for me at EASL 2023. I've been coming to liver meetings for 20 years. This is, you know, outside of hepatitis C cure, to me, this is the next best thing. Okay. I think, William. Thank you. Thank you for taking our questions. I'm William Wood from B. Riley Securities. Just going off of what you were talking about earlier, Dr. Harrison, as far as patients having on GLP one incretin therapies, I was curious if there's plans in the works for a head-to-head resmetirom versus a semaglutide 2.4 mig trial, to actually put the differential efficacy sort of question to rest. For a 2.4 milligram? Yeah, using obesity. Yeah. Doses of increased- I mean, we know from this data, these data in MAESTRO-NASH are confirmatory. We already showed with MAESTRO- NAFL that the diabetes level doses of GLPs, the patients were coming in with essentially the same liver fat content, and they were on them chronically, right? They were coming in with essentially the same liver fat content, as someone who wasn't on a diabetes level dose of GLPs. I think even, you know, GLPs, in general, it's very slow-moving. You know, it doesn't really work in the liver as far as we understand. It's working indirectly through mainly through body weight loss, and we know that weight loss certainly enhances effect of resmetirom. If you look at that weight loss graph, it's a small number of patients. There is a more weight loss patients, especially at the 100 mg dose of resmetirom. A little bit more, but it's not a weight loss drug. The separation from placebo is maintained very well in the face of this GLP. If you just look at where is the dot, it looks essentially the same. The% difference from placebo is the same. It's really not. It's not doing anything in addition in patients who lose weight. I meant at the top, the weight loss. If they lose 5%, they're not on GLPs because they're stable. If you did make them lose weight, it'd be giving you the same sort of delta relative to the mean of the study. If they didn't lose weight, you can see that resmetirom is actually even working better because this is relative to placebo, and we know very well that when there is a placebo response, it's usually because that placebo has lost a little bit of weight. Now there is no weight loss, and you can see that resmetirom is gonna work relatively better. We know that weight loss plus resmetirom is gonna work better than just resmetirom alone. We know, and we've had been in lots of GLP patients, there's no tolerability or any other issue with using resmetirom and GLP. I'm sure it would be a very good combination, but it's sort of being done for us in effect, in our, in our clinical trials. We know what the data would show. Just to add to that, I mean, you're talking about a 2.4 milligram Wegovy head-to-head comparison with, you know, a 100 milligram of resmetirom. You know, that's the heavyweight bout, you know, of the world, right? Who's gonna be the winner? 15 milligram of Mounjaro versus... I mean, I don't know if that study will ever be done. No, 'cause we need to approve first, so. Well, it might be done, it might be done like, after approval. Yeah. 'cause you could get Mounjaro or Wegovy, you know, off the rack. There might be some, in fact, I'm sure there will be investigator-initiated trials that'll be done comparing that. What I like to see is this: if you're on a GLP, look at what happens. Consider placebo as just a GLP, right? They've been on a stable dose of a GLP for a long time, had to be 6 months or more for the trial. If you look at Akero's data, again, that meant that the majority of people were on a GLP for at least 1 year. Two-thirds of them were on, when you have to set this stability of 6 months. We don't know that number. I don't know that number off the top of my head for this trial, but suffice it to say that they had to be on a stable dose for a while. Now, these are diabetic doses, the vast majority. This trial was enrolled, you know, Mounjaro and 2.4 of Wegovy weren't around at the time this was launched, but the lower doses of semaglutide were. Despite that, this looks at essentially combination therapy. They're on a GLP, they get thrown on resmetirom. In both doses, you're seeing a response. For NASH resolution, and over here, at least for the 80 milligram, you know, this is probably just a variance of small numbers of patients. Even regardless, you know, you're looking at a 10% improvement with high dose and a 20% improvement almost with low dose over a baseline GLP-1. As I mentioned before, whether you look at the 89bio data or the Akero data, baseline GLPs on diabetic doses for a stable period of time are still between 11% and 16% liver fat content. There's still a lot of room for improvement. Combination therapy, I mean, as I mentioned before, these people are coming to me on a GLP-1, and they're still sick. You know, the combination makes sense to me, but as far as head-to-head, I don't know when that'll be done. Could I say one thing about that? If you look at persistence on Wegovy or on Mounjaro, after a year, almost three-quarters of the patients are off the drug. If you did a head-to-head and it was an ITT analysis, I think we'd look pretty good at the end of that year. Yeah. Thank you. One more, if I may. In terms of you've shown really great correlation between your data, as well as a good correlation with your steatosis and ballooning, improvements in health, et cetera. I was curious, what gives you confidence going into your F4 outcomes program? Also specifically with having 700 patients in it, just kind of curious, you know, that's a little bit smaller than what we've typically seen. You know, maybe we could get your ideas on treatment of placebo and progression on some of the... Right. Yeah, I mean, I think we actually saw here at this meeting some pretty good data on outcomes in well-compensated NASH cirrhosis. I noticed that we had shown some data in our NASH cirrhosis study, that if we segmented these patients according to if they had 5% or lower liver fat versus higher or greater than 5% liver fat, these are NASH cirrhotics now, and as they progress, their liver fat goes down. We showed that they were more progressed if they had this 5% or lower liver fat than they were if they had higher liver fat. That was a presentation you gave. Right. you know, I think it was, it might have been AASLD. They did the same segmentation, and they showed quite a significant decompensation rate in that. Beyond that. population over the timeframe that we're thinking about. This was a longitudinal real-world type study, where they, the NASH cirrhosis was segmented by, you know, how much liver fat they had or how progressed they were. I think the question always with outcome studies is your event rate estimation correct? In order to protect against that... We started with 700, based on what's published as the rate of decompensation. However, the study N can be adjusted if it turns out we're not seeing that rate of decompensation. Yeah, I don't know if you saw this, there was a presentation by a pathologist from Edinburgh, Scotland, last night, looking at his SteatoSITE data. I don't remember all the details, but this was a repository of all the patients in Scotland, that, you know, many of them had advanced liver disease and cirrhosis, and they actually had a liver biopsy slide, unstained slide on each one of those patients. They also did RNA-Seq on those patients. What they did was they did the HistoIndex staining on those patients and showed that if you had perivenular fibrosis or central vein fibrosis, that predicted a negative outcome. In addition, they looked at SHBG in the cirrhotic population. If it was high, at least one case they showed didn't progress. If it was low, that patient did progress to decompensation. On RNA-Seq, they showed that gene expression of THR-β is downregulated in patients that progress to disease and/or progress to decompensation versus those that didn't. When you looked at the HistoIndex phase II data with resmetirom, and you correct for liver fat reduction and volume reduction, where do we see the fibrosis benefit? It's in the perivenular and perisinusoidal region, which matches up perfectly as a drug therapy for those patients that progress to decompensation in Scotland. I mean, that just doesn't really help us in powering, but it gives us confidence that hitting this, these patients with a THR-β agonist, like resmetirom, is gonna have a, you know, a positive impact down the road, so. Where are we, like, timing? What, how long? Yeah, I think so. Maybe we can take a couple of more questions. Richie, you have a follow-on question? Yes. Thanks. Question for Dr. Harrison. Imagine a different world where today we had, after the 22nd, we had an approved therapy for NASH. What would be the standard non-invasive testing sequence or algorithm that hepatologists would be using after they went back home from EASL to diagnose patients and to manage patients on treatment? What's the favorite algorithm right now? I'm glad you asked that question. It's different than primary care. Primary care, I mean, we've established the guidance document, and EASL has their guidance document. ACG has their guidance, AGA does. They're all the same, FIB-4, followed by FibroScan or ELF. That just gets you to me, right? When you're at me, I don't care about FIB-4. I don't look at FIB-4. I used to do BARD. I don't do BARD anymore either, 'cause that kind of fell out. You invented it, right? I invented the BARD. Unfortunately, that went the way of the dodo bird. We don't use the BARD score. We don't use an ELF fibrosis score. We use clinical, you know, acumen. We look at metabolic risk factors. If you're a diabetic, you're menopausal, you're female, you're Hispanic, your AST to ALT ratio is greater than or equal to 0.8, meaning the higher the AST, the more likely you have advanced liver disease. We look at FibroScan, we look at CAP, we look at MR elastography, all that. It's a gestalt as to who has advanced liver disease. We don't take one particular study and say, "This is it. It's definitive. This is what you have." Literally, we don't have a test that has high positive predictive value. We have lots of tests that have great negative predictive value. Many of them have two different cut points for sensitivity and specificity optimized with a high indeterminate rate in between. The more tests you do, you compress the indeterminate zone, and you have a higher likelihood that this patient has advanced disease. What I can tell you is, none of us do liver biopsies anymore unless it's a diagnostic dilemma. I used to do five a day. Now I do five a year, maybe. That's if I think the patient has two liver diseases at the same time, like PBC and fatty liver, autoimmune and fatty liver. I need to know, is it autoimmune? Do I really need to treat this fatty liver patient with steroids? I'm probably gonna make the other one worse. Let's do a liver biopsy and just see if there's interface hepatitis with a lymphoplasmocytic infiltrate, which is really the hallmark histopathologic feature of autoimmune. That's nothing like what we see with NASH. That's why we do liver biopsies. Everything else is non-invasive. I see patients, you know, every other Wednesday in clinic. These are routinely sent to me. Nobody does FIB-4, by the way. I don't ever get a patient that's sent to me, "FIB-4 1.64. What do you do, Dr. Harrison?" It's elevated liver chemistry test and fatty liver in a diabetic patient, or it's, you know, any variation on that. It's a very quick evaluation for me because I do a FibroScan. I look at their clinical history, their family history. I look at the AST to ALT ratio. You know, for me, the AST to ALT ratio is a tremendous help. AST is greater than 40. We've looked at this in our SteatoSITE database with over 5,000 biopsy-proven patients. AST is greater than 40 are highly predictive of F-two or greater disease. If I have an AST above 40, I have a FibroScan above 10, you know, to me, that's treatable disease. You wouldn't need a biopsy. I never- put pen to paper for- Never do biopsy. Do you think that there will be a written algorithm for the community gastro or, you know, the more, the non-academic hepatologist in place a year from now? Probably. I mean, that's a great point. We have guidance documents really geared toward the primary care provider, and it says, "Refer to specialist." We don't have another separate set of rules for specialists that kind of help guide them. You gave me a great idea to go back and write a paper on. It is just kind of putting it together for GI docs and liver docs. Yeah. What we've been hearing is that if that paper is in place, it'll be a lot harder for insurance companies to go then demand a biopsy. Every payer conference that I go to, and I just went to one not too long ago in Chicago. The question that came up was, you know, "What are you gonna require to pay for the drug?" The comment was, "We'll follow the standard of care. We'll follow the key opinion leader's advice." We assume that's a liver biopsy. I made the comment, "We don't..." Just like I did here, "We don't do liver biopsies anymore." They're like: Well, then we won't require a liver biopsy. You know, we'll fall back on what is recommended by these guys for non-invasive assessment. Of course, that's all predicated in the setting of how costly the drug is gonna be, right? What we showed them was the ICER data, and $19,000 was what was quoted in that document, and they're like: PA to label, PA to label. You know, it's how do we non-invasively diagnose to label, right? That's gonna fall back on the guidance document at the primary care level and at the specialty level. We need to sort that out still. Am I on here? I think there's one week left- Different calendar, right? in the second quarter. Are you still gonna be on time with submitting your IND? I guess, just as follow-ups, are you expecting a priority review? Would you also expect an ad hoc? Yeah, I mean, we're, you know, nearing the finish line with this, and it's as far as how we file, it sort of set us up getting a breakthrough therapy when we did, because it's recent, it's based on our phase III data, which we shared with the agency, including some of the safety data, and not that that's a review, but the fact that we did. I think that, you know, they expect us to request a priority review based on that sequence and that there's a very good chance that we will get it. We're past 8:00 now. Well past 8. Are there more questions from online, Kia? Actually, I think we're at a good place to conclude. Thanking Dr. Harrison for walking us through all the information today and also thanking all of our participants, both in person here and on webcast. With that said, thank you all.