Thank you everyone for joining us on our next session. I'm very happy to have up here on the stage with us, the CEO of Akero Therapeutics, Andrew Cheng. Andrew, always an exciting and dynamic time in NASH or MASH. And at the same time, EASL is going on, so I appreciate you hanging out with us rather than at EASL. But all humor aside, obviously, you are executing on your own programs here, on a phase III study, and also executing on a large cirrhosis phase II study in MASH that we're gonna talk about. So maybe you could just give us an initial high-level overview of how you see your drug fitting in, efruxifermin, fitting into the NASH landscape. Because as of including today, people say GLP-1 is gonna be great, tirzepatide shows great data, everyone's gonna be on tirzepatide. There's glucagon data also coming in, I think. I don't know when it's scheduled, tomorrow or something like that, Friday? And Madrigal's on the market. Yeah. Many things that are also on the market. How does your drug fit in, taking into account the tirzepatide data today? So I think that's a great question, and when you look at the tirzepatide data today, so they demonstrated, at certainly at the two higher doses, there's no dose response, but roughly a 21% effect size, over placebo. And- With a p-value? With a statistically significant. I did not expect that. I think that's great. Okay. When you see that, clearly the drug is already on the market for both diabetes and weight loss. Its phase III plans are still in evolution, but your question about where we fit in as efruxifermin, I think we have to recap and look at what the phase IIb 96-week data showed, and it showed that we had a 75% one-stage improvement of fibrosis compared to just 24% on placebo, a roughly 51% effect size, and that's over two years. When one thinks about our 24-week data, we had approximately a 21% effect size, very similar to tirzepatide, albeit in twenty-four weeks versus the 52 weeks they showed. So what we- So working at basically twice as fast. Correct. At least twice as fast. We see ourselves as a rapid-acting, direct antifibrotic. For patients who may be more advanced, they're F3, at that risk of progressing to cirrhosis, it's the kind of compound where using it together with tirzepatide may- Hmm ... improve or their opportunity to not become cirrhosis because... or cirrhotic. Oh. When you think about the patients who have a 21% effect size, unfortunately, that means that roughly 80% of the patients didn't have a 1-stage improvement in fibrosis during that 1-year period. As if, if you were a physician and had something else that could augment the probability of their not progressing to cirrhosis, I think it's an opportunity to partner or pair those two drugs together. That's a good point. And- So, yeah. Number one, you work basically twice as fast. You had a 20-something% delta at six months. It expanded to 50% delta at two years. Tirzepatide today shows a 20-something% delta after one year. So you're, you know, same amount, but half as, half the time. And in addition, if you ask hepatologists, you know, liver, the people who are actually treating the disease, not just the overall endocrinologist or primary care, that if the person has MASH and may or may not be on a GLP-1, let's assume many people are on a GLP-1, then 80% of those people are not gonna have a 1-point change, and therefore, they're gonna need another agent on top of that. Correct. You actually have data. Exactly. We have existing data. In the HARMONY study I just described, with a 75% one-stage improvement of fibrosis, roughly 15% of the patients were already on a GLP-1, so we were augmenting their response. We have an entirely separate study we presented at this conference last year, where we talked about patients who were taking a GLP-1 already. The average time was greater than a year. This is an all-comer study, and we added efruxifermin 50 mg for 12 weeks. What we saw is that we improved their liver fat to normal levels in 90% of those patients in just 12 weeks. They were doing well, but still had abnormal levels of liver fat and they were normalized within just 12 weeks. Yeah. In addition, their diabetes improved, their weight loss- Yeah ... continued. So it's There was added benefit. Correct. Importantly, there was not additive safety or tolerability issues because both drugs have GI nausea as a side effect. Correct. That would have been a concern when you add your two drugs together. Correct. We didn't have an additive response when it came to GI adverse events. In fact, numerically, the combination had a lower diarrhea rate compared to this, GLP-1 alone. Why would that, why would that be? So putting the two drugs together had a lower GI adverse effect? You know, it's one of those small numbers. Yeah. It was 30% versus 19. Our takeaway is it wasn't greater. Okay. So you believe that regardless of the GLP-1s, that there is a role here because of the profile of the drug, but also on top, on combination? Absolutely. Now, the important thing is that... We're just talking about the F2-3 population. In F4, and I have no idea. Well, first of all, no one has worked in F4. Yeah. There's been no statistically significant benefits in F4, and this is the highest unmet need. I'm not actually aware of, maybe you could help me out, that any of the GLP-1s are running an F4 study currently, including tirzepatide. Do we know that? I believe there is a GLP-1 study with Ozempic- Ozempic. in combination with Gilead's ACC and their FXR that is due to read out in the first half of next year. Okay. in F4. All right, so we'll look for that. But well, we know that the ACC isn't doing much to be fair, in fair regards. You might know about that, but- Yes. Fair that the ACC is probably not adding much. Correct. So then we're asking, basically asking, does the GLP-1, Ozempic, GLP-1, not clipper, clip kipper, but GLP-1, whether that's gonna have an F4 effect over, what, 12 months? Do we know? It's a one-year study. It's a 1-year study. Okay. So the probability of that showing positive results is probably not high. Yes. I mean, they have a one-year study already with GLP-1 alone, with Ozempic, and it was presented- Yeah ... in 2022, and it was placebo-like. Yeah, in F4. In F4. In F4. Correct. And now in F2, F3 as well, without getting too complicated here, that also had no fibrosis benefits as well. Tirzepatide today would sort of mitigate that. There is positive data, whether tirzepatide is a better drug or whatever, but also that, Sema also has a result-based drug. Let's move away from that 'cause I care about the F4, which is that you are running an F4 study. The initial data last year was a disappointment for investors and for you, but I want you to review that result and tell me about the two-year data that's coming around the corner, and what should we expect? Because if the two-year data is positive, that would be a huge catalyst and a huge result, and currently, it feels like nobody expects that. So do you expect it's gonna get better? Yeah. So when we look at the 36-week results that was presented last October, we were not statistically significant for the primary endpoint, which is one stage improvement of fibrosis. The results showed a 10% effect size at the 50-milligram dose, twenty- At 24, 14. Correct. Okay. 10% effect size, and we, we didn't hit statistical significance, yet at the same time, we were statistically significant for NASH resolution, which had never been seen before. And we showed some patients had two-stage and three-stage improvement of fibrosis. So that being said, that was encouraging to us, although disappointing that on the primary endpoint. In terms of what's happening at week 96, that will read out in Q1 of next year, and while not fully powered for that, given what we saw in the 96-week data for the HARMONY study, that is, we saw responses broaden, moving from 41% to 75% for one stage improvement of fibrosis, as well as deepen. We saw responses for two-stage improvement of fibrosis move from 15% to 36%, compared to just 3% on placebo, which is also statistically significant. It is encouraging to us that longer-term dosing has a role with this drug. It makes a difference. Now, when you're asking, what do we expect for the 96-week data for the F4 cirrhotic population- Yep ... we expect similar things, that we expect an improved response compared to the 36-week period. Of course, I'm not being quantitative about that, but I would just say that we expect that the p-package, when one looks at the totality data, will be better than what we saw at week 36. So currently, the results show 24% versus 14% at 9 months. You have a full expectation, based on the data we've seen, it should get, improved because longer treatment should drive more 1-point changes. Certainly, longer treatment will continue to reduce, the collagen and the fibrosis, and in some cases, 2 points, but it should get better with longer treatment. And, therefore, the 24% has a very high chance of going higher. How much higher? I'm not sure. I think we're confident that it will be higher, just as you said. How much higher? We're not certain. Now, talk about the biology there, because you've explained that it has to do with how much collagen is there. In one case, you could just look at ELF, for example. Sure. Maybe the ELF is higher. This is a measurement for the density- Mm-hmm ... of it. But, talk about that and how you think about the amount of time it takes to reduce that amount of collagen. It's a little bit complicated, but that it's just a time issue because of the biology, that there's just more collagen. So I think, Mike, what you're nicely driving at is the fact that when one thinks about patients who are F2, F3, they may have anywhere as high as 5% liver fibrosis in the liver. But when one thinks about patients who are cirrhotic, they could be 5%-30%. And the mechanism at play here is that we inhibit new collagen synthesis, but in the background is ongoing collagen degradation. Mm-hmm ... and that's at an unchanged static rate. So when one thinks about a fibrotic burden that is being not added to, but also being replaced and degraded, it's a matter of time for which the fibrotic burden should decrease, and we saw that at week 24 to 96 in the F2, F3 patients. We expect something similar, but the challenge is that the burden could be much greater in this population, as well as there could be architectural changes... So we think that in general, it's going in the right direction. The magnitude of which is, we're a little bit uncertain. 24% could get better. The 14% placebo effect, is that going higher, lower, stay the same? Placebo effect is really a false positive when one thinks about the difficulties with liver biopsy, and one looks at it this way. When one has a birthday, all of a sudden, you're 100% a year older. But at the same time, the liver doesn't work that way. There's a gradation, and while there are portions of the liver that are F4, there still may be a portion of the liver that's F3. Now, does that mean that you're, while you're been categorized for this study as at a baseline biopsy is F4, there could be regions that are not. And so when one biopsies at week 36, it's not that the liver improved, it's that you found a portion of the liver, a loci that is F3. Now, as time progresses, one would expect the amount of, F3 parts to become more a part, like- Correct. So overall, if you think about the liver, there's different, depending on where you stick the needle, different amounts of levels of fibrosis based on where they're picking the sample. You could have picked a sample here and/or just biopsy. It's not just a biopsy or reader error, but, you know, there's... This is why we do a placebo study. Correct. The overall, over 2 years, it's unlikely that portions of the liver in a placebo are gonna get better. In general, it's gonna get worse- Correct. and therefore, the probability of having a 1-point change in placebo is low. Correct. We would think that the- Lower than the noise in just 9 months. Yes. Over two years, we'd think that the disease progresses, and the proportion of F4 should increase over time, and therefore, the false positive rate- Yes -of this improvement should decrease. Decrease. Exactly. So, there's other F4 studies to look at. So Gilead and others had run F4 studies. Huge. I don't know if it's 1,000 patients, but it's hundreds and hundreds. Intercept also ran an F4 study- Yeah -using the same consensus biopsy reads, and they had what? 10, 11%? Yeah, so around, right around 10%. Yeah. 10%-11% is accurate. So over time, could we approach those numbers? Yeah. So placebo should be flattish to down, could go to 10%-11%. Your drug goes from 24%, may go to 30%. That's a modest increase. Okay. I mean, your F2/3 went from 40 to 75. True, but I think the mountain for F2, F3 to climb in terms of fibrosis reduction is. It's a smaller hill. Okay. Okay. But still, I don't. You're going in the right direction. All right. I think I'm trying to maintain some optionality when it comes to numbers. Okay. So one, easier population per se went from 40 to 75, almost doubled. Here, if it went from 24 to 30, that's a modest improvement. Yeah. 30 minus 10 for placebo is a 20% delta. Is that clinically meaningful? In this population, anything... Remember that the bar, unfortunately for these patients, the mortality for F4 patients is, at five years, is roughly 50% short of liver transplant. So I think any improvement that is, 20% would be a very meaningful improvement. Now, shouldn't that be statistically significant, Andrew? I think it a little bit depends on the sample size, and it depends on the discontinuation rate. So those factors play a role, but, I don't have my statistician in front of me to answer that. But, you know, it's not. We'd have to look at that. Okay, at the beginning of the study, it certainly was powered, and when we say powered, we wanna put some quantification on that. You power studies at least 80%, 90% is generally the rule of thumb. Correct. So 80%-90%+, as you lose patients, the powering goes down. But powering is just an artifact of the confidence level of the strength of that p-value that the result is real and not a false positive. Correct. So when we say powered, you may not be 80% or 90% powered, but you're certainly within a degree of a gradient. Jefferies' estimate, at least 50% powered. That should be a result that would happen if the result plays out as I said. You know, I think there's a lot of variables, and I think you hit on all of them. The powering goes down as you have more discontinuations. Sure. It varies as what the placebo response rate is and what the active response. So all of those things are true, and I, I agree that even though we're not fully powered, we're not 0%. All right, well, my math, and I again, I would say that Akero has not disagreed with this, but I'm gonna ask you directly, is that if you look at the F2/3 study, which was a resounding success- Mm-hmm ... your result, 2 years ago showed 40 versus 20. That is a 20% delta. Correct. What was the p-value on that? The p-value was less than 0.05. Okay, you just said 0.05. Yeah. Less than 0.05. We did not give too many zeros. We didn't give any zeros. Less than 0.05, and that was on how many patients at the time? That was on 113 patients. At that time point. At that time. 113, a 20% delta, P < 0.05. On your F4 study, how many patients are you starting with? We're we started at baseline with 182. 182. We had 153 biopsies at week 36. Okay. Where we land is uncertain at week 96. This study's ongoing. Well, in the F2,3 study, what percent final dropout did you have after two years? We lost about a quarter of patients from week 24 to week 96. Okay. So you lose about another 25% of people over another year and a half?... So then on the F4 study, losing another 25% off of 153 people is like 125 people. That's still more than the 113 that you had in the F2-3, which you just said was p < 0.05 on a 20% delta. So if you have the 20% delta, you should still have somewhat more patients than the F2-3, which showed a statistically significant result. Is there anything wrong with that math? There's nothing wrong with that math. It just depends on what- how the numbers turn out. Okay. So when is that result coming out, and what defines that result in terms of timing? 'Cause you said first half, or what, what was the- First quarter of next year. First quarter. You know, what is going on in the study now? You're obviously people are completing their- Their week nine, their- Their week 96 biopsies. There are patients who are not finished week 96 yet, so they're still dosing, and we're waiting for them to finish their dosing. What would you announce when that's has completed? We would. Or what, what would you say? Much as we did with the week 96 HARMONY results, once our last patient has crossed week 96, we'll issue a press release- Mm. and we'll also narrow the month in the quarter, which we'll release the results. Okay. When did the 9-month study read out? It read out in October of 2023. October of 2023. Okay. If someone can do quick math, you can basically take 24 months minus 9. I I think that's 24 minus 9 is 13. So it's another- 13 plus the October puts you into around November. Yeah. Yeah, November. It's 60 additional weeks from there- Yeah ... and then there's another 4 weeks for off-drug follow-up, and then there's some time for data cleaning. November, the last one is done, December cleaning. Could be earlier, but that's Jefferies' estimate. Yeah. Okay. All right, so that was the math around that. Now, tell me about, in the last few minutes, given the fact that you are running this study, and you're also at the same time running the F2-F3 Phase III study, what's going on with that study? The study, This is a phase III. Began screening in F2, F3 last December. Uh-huh. Actually, enrolled in last December, screening earlier than that, and the screening and enrollment are going well. We're very pleased with how things are going. It doesn't hurt us that we have the 75%, frankly, the best results in the NASH field ever. And so when investigators think about the compound and think about putting patients on, they're able to fall back and refer to patients with that kind of clinical results. It's very, very helpful. Right. Meaning that when, if the stronger the efficacy, and the efficacy is pretty strong, it resonates well with enrollment. So you feel good that enrollment is progressing very nicely, albeit early, but- Yeah, it's early. We, you know, we're. It's 1,000 patients. We've put that out publicly. It's going to take some time. Also, I guess next week is a glucagon readout. Glucagon, I guess on Friday, or I guess- This week. This Friday. Yeah. The Zealand glucagon NASH readout, which is also expected to be strong fibrosis benefits. Anything to say about that compound? You know, it'll be the first NASH GLP-1 glucagon readout that I'm aware of in Phase IIb. We look forward to seeing the results. I believe that about a quarter of the patients in that study are F 1, so that- Yeah ... makes it a little bit different than today's Lilly data, which is one- Mm-hmm ... 100% F2, F3. Okay, so it's a slightly different patient population. Slightly different. Yeah. Okay. Last question for you in the last couple of minutes. So we started this conversation by saying, everybody believes that GLP-1 is going to be widely used across these patients. That is a market mostly dominated by pharma. Yep. At the same time, people would say, if you have such strong results, that your product probably deserves to be in the hands of pharma. Not only to run all these studies, but certainly from a blockbuster opportunity and all of that, that pharma is the best person to, to run that. The issue is that there's been basically zero business development from big pharma in MASH. Madrigal, a drug, that is approved and on the market, continues to remain independent, and no one has jumped on that one yet. So, how does one explain the challenges that basically pharma has exuded not much interest in this space, and that doesn't resonate well with investors? So I would say that, that what you've described is true from an M&A perspective, but when one thinks about this space, you see that large companies, we just talked about Lilly, we talked about Boehringer, we talked about Novo, are very much in the NASH, MASH space. They continue to have internal programs in this space. I think it's very difficult to know what third parties are thinking, but we, I would note that we're the only mid-cap company that has a large pharma investment. Pfizer invested in us- Mm ... in 2022, and they're one of our top 10 shareholders. So, you know, for what that's worth, they, we were investors in us, and we speak with them. Yeah. On a- That was the Pfizer Innovation Fund- Correct ... which speaks regularly to the BD team at Pfizer? Yes. Okay. I think a lot of this, you know, when we reach and get to time points that have meaningful data, I think it's things that we share with, you know, larger- Yeah ... organizations. Yeah. Fantastic. We'll look forward to execution, look forward to the F 4 data, which is right around the corner, and I know investors are gonna be zooming right in on all the details about that, just like it did for the F 2/3 results and the F 4 data. But high interest in that huge event, and obviously an opportunity, I would say, given the stock has pulled back. But there's gonna be high interest in this, and we look forward to your data. So do I. Thanks again, Mike. Thank you. Thank you, Andrew.