Akero Therapeutics, Inc. (AKRO) Earnings Call Transcript & Summary

Thanks, everyone, for joining us. I'm Andrea Newkirk, one of the biotech analysts here at Goldman Sachs, and I'm really pleased to be joined by Andrew Cheng, President and CEO of Akero. Thanks so much, Andrew.

Thanks for having me.

Maybe at a high level, let's start with where the NASH field is now. We've obviously seen a lot of development, both on the clinical development side with the first commercially approved drug last year. What makes you so excited about the forward path in NASH? And where -- how do you see Akero is best positioned to take advantage of that?

That's a great question. What we see is that -- and this is hats off to the Madrigal team and their CEO, Bill Sibold. We got the first drug approved last March or end of '24. And it really speaks to the medical need and how well that drug has done. And so as a result, there are many challenges that were, I would say, investors push back on the NASH field, but have really been addressed by that. And what's so exciting is that we'll talk about this a little later. But in January, we really had some F4 data that was really a first of its kind. We were lucky to have a simultaneous publication in the New England Journal of Medicine last month, along with the podium presentation, a late-breaker at the European Association for Liver Disease. And so we're really excited about the opportunity to not only address the pre-cirrhotic population with past data, but the new data in the cirrhotic population, which has really been difficult to treat to date. Yes.

Yes. About FGF21. How does efruxifermin differentiate from pegozafermin? We also saw Boston Pharmaceuticals, efimosfermin, which is now GSKs but how is efruxifermin differentiated?

So efruxifermin is a fusion protein, Fc fusion protein. So it's fused to the Fc portion of IgG. And that is similar to the Boston -- former Boston former Novartis now GSK compound following the acquisition in May. And yet on the other side, it has -- it's a dimer. So there -- it's a divalent or bivalent compound. So there are 2 FGF21 attached. And what's different about it is that unlike the other 2 agents, there have been important modifications at the C-terminus. And what I mean by that is that FGF21 in order to be active requires a trimeric interaction. That is that there's a binding from the C-terminus to this co-receptor called beta-klotho. It's that binding that regulates the activity. And there's a natural endopeptidase that cleaves that residue 171. And we've made -- or Amgen because we're an ex-Amgen compound made modifications to that to enhance the binding. And therefore, the longer it's tethered the co-receptor, it's better able to bind to the N-terminus domain to the FGF receptor. So I know that was a long answer. But the short answer is that the pegozafermin is a pegylated single chain FGF21, and it doesn't have -- we're the only one with that modification and some of the other modifications at the C-terminus. That's the short answer.

Okay. And what -- maybe what is the -- what do those modifications really translate to as you start to think about the efficacy profile?

Sure. So they translate to better kinetics at the binding, mostly in the dissociation context. We've done some in vitro work. But really, what they translate to is, quite honestly, some of the best clinical results we've seen in the field, both in the pre-cirrhotic and in the cirrhotic population that we've touched on, the cirrhotic population I mentioned earlier, we're the only company, FGF21 or any other mechanism to demonstrate 1-stage improvement of fibrosis in cirrhotics ever. Unfortunately, the patients have very few options there. And for us to be able to demonstrate is an important milestone for patients with cirrhosis. And similarly, we've seen in terms of 1-stage improvement in cirrhosis -- excuse me, in the pre-cirrhotic population, we have about a 75% 1-stage improvement in fibrosis with roughly a 50% effect size over placebo. And again, that dwarfs the competition in the field.

Let's start with the cirrhotic data that you just presented in January. It's first very large, first large data set in this patient population. Maybe talk to us about what you saw between week 36, week 96, what that could imply if patients were to stay on even longer, what kind of effect size could you see?

So I think that's a great question. And what just to recap the data is that we showed at week 36, we showed roughly a 10% difference between active and placebo, which was not statistically significant. But at week 96, 60 weeks later, we saw roughly a 25% difference, roughly 39% versus 14% for a 25% delta. And what's important about that is the principle, as you just highlighted, that longer dosing matters. And we saw this principle in the pre-cirrhotic population, too. So it's very internally consistent. We saw roughly about a 20% effect size in pre-cirrhotics at week 24. Add additional 72 weeks, then we saw a 75% 1-stage improvement of fibrosis or 50% effect size. So the effect size more than doubled from 20% to 52%. And then in cirrhotics, it went from 10% to 25%, again, more than doubling. So we're very excited. And when you think about long-term, unfortunately, even though we're the first of its kind to be statistically significant and a 39% improvement, unfortunately, it means that 61% of the patients didn't see an improvement by 2 years, but we believe with longer dosing, they can see that benefit.

Do you expect that a plateau will be reached at some point? Or could you continue to see improvement over longer durations?

Very difficult to say for sure. But based on what we've seen from both the pre-cirrhotic and cirrhotic populations, it's not clear that there will be a plateau primarily because the compound is very effective at shutting down new collagen synthesis. And remember, a fibrotic burden or collagen is a balance between the rate of new collagen deposition and degradation. And because we've sort of shut off degradation -- excuse me, synthesis and degradation is ongoing in the background, one could see over time, even the greatest fibrotic burden potentially could get reduced.

Do you have a sense or an understanding of why some patients may take longer to respond here? Because you are seeing a deepening of that response over time, which implies new patients, and I think you've shown some data to that effect. But what distinguishes a patient who's kind of an early responder versus one that takes a little bit more time?

We're still digging into that. But on -- what it appears to be is that there's a range of collagen burden. So cirrhotics, unfortunately, some of them have -- may have as little as 6% collagen burden in their liver. Some may have had as high as 35%. So we think that the people who respond the soonest are those with the smallest collagen burden.

And as you think about your Phase III, I mean, that's a great segue. But as you think about your Phase III and the patients that you're enrolling, are you -- talk to us about your inclusion criteria to that effect?

You can imagine that as we go to Phase III, we're trying to leverage the learnings that we have from our Phase IIb study. So when we look at the people who responded, those 39% of people at the 50-milligram dose, we're really sort of reverse engineering and looking at what they call predictors of success. What are those characteristics for those patients and how can we translate that in the inclusion criteria into Phase III, and that's what we're actively doing. Also keep in mind that there are 2 cohorts. All patients will, in Phase III, be followed for outcomes. But there's a subset of patients that will get a second biopsy. So in terms of who should be in Cohort 1 and be followed for outcomes or who should just be followed for outcomes, we're trying to redirect and enhance those populations for success.

Got it. That's interesting. And maybe just one question on VCTE, which was also another measure that you reported here. How should we think about the magnitude of benefit you saw there relative to, say, Madrigal, who also reported a VCTE improvement?

So you're referring to some data that over the -- in that -- on the 50-milligram dose in SYMMETRY, what we saw is roughly a 7.3 kilopascal decline, average decline in VCTE. And what we saw compared to with the Madrigal data is roughly 6.7. Of course, whenever I make this kind of comparison, you always say comparisons about [ crosscopy ] comparisons are fraud because not exactly the patient baselines are exactly the same. At the same time, they're in the same ZIP code. And so what we would say is that we're no worse. And when we look at some of the other metrics that Madrigal put out, they commented that 52% of those patients had a 25% reduction. If you look at that as a threshold. In our comparison, the number of patients that had a 25% reduction was 70%. Again, numerically greater, but probably not outside the ZIP code.

And what is more meaningful as you try to assess the benefit of a drug? Is it BCTE? Is it the 1-stage improvement per biopsy? How do we think about those 2 endpoints?

So I think when we look at it, the biopsy is, of course, the issue because the medical expression, the issues, the tissue. So you're absolutely certain what's going on. But at the same time, recognizing that there are only so many portal tracks in a small biopsy out of a 16-gauge needle and that the liver is the largest solid organ in the body. So the question is VCTE is also very helpful because it gives you a totality of the data. So the best circumstance is no one metric or measure is ideal. And you look at the totality of the data and seeing the changes that you see, but also that's corroborated by noninvasive helps really marry those 2 data points together to give physicians and patients confidence that the drug is doing what it should be.

Got it. So multiple endpoints to get a better idea.

Correct.

As you think about the safety profile here, you did have 3 hepatic decompensation events that happen to be in the treatment arm. How should we interpret that finding?

Yes. So I think a lot of investors have asked me about that at this conference, and some of it has to do with what they're trying to understand because Phase III has an outcomes basis for liver events. And so the liver events are -- so we're clear for your audience, are ascites or esophageal varices or hepatic encephalopathy or all-cause mortality. So when one takes that lens and sort of looks backward into Phase II, what we saw is that the actual balance was that we had, unfortunately, a patient died of pneumonia on the placebo arm. And then we had 2 patients with ascites, 1 in 28, 1 in 50. And then we had 1 patient on 50 having hepatic encephalopathy. So the breakdown, if you were using the Phase III lens would et cetera be 1, 1 and 2. So is there an imbalance? Sure. There's an imbalance, 2 is greater than 1. I'm not debating that. But what we would say is the imbalance is not very large.

Got it. Is that surprising to you, though, that the annual -- or I guess, maybe the decompensation event rate was essentially, I guess, maybe 0 in your arm in placebo.

Yes. We would say that we've always expected that the decompensation rate would not be linear in how we model Phase III, we don't imagine that it's going to be an equal amount each year. We think that the numbers will grow over time. And the reason we know that is that the 5-year mortality for these patients is unfortunately about 50% and it's -- so people don't get -- it's -- they don't get diagnosed all at the same time. And so over time, we think the events will accrue.

Maybe one other question here on safety. Just the observation of bone marrow density loss. Walk us through what you did see between the 36 and the 96 weeks. I'd be curious to understand whether you measured it at the femoral neck at week 96 and what those findings were?

Right. So we measured BMD at baseline and the FDA asked all FGF21s to look at that because it was a mouse data. So the BMS pegbelfermin drug, which was discontinued, would ask to look at BMD as well. And I believe pegozafermin as well as the Boston compound have. So we looked at it at week provided data at baseline and looked at it at week 48 and week 96. And the difference from those 2 time points is from -- you asked about the change between week 48 to week 96, we saw roughly about a 2.5% decline, placebo adjusted in both the lumbar spine and femoral neck in between the first year and the second year. And when we think about it, unfortunately, the patients in our study, they had multiple medical problems. Roughly 40% of the patients were osteopenic at baseline. And even though they were osteopenic, some of them on the verge of osteoporosis, only 3% of the patients in the active arm were receiving treatment for that. And these are drugs that, frankly, were approved in the previous century and are generic and it can be given as a once-yearly injectable. What it really says is that we didn't push our physicians hard enough. We are mostly liver doctors to take care of an area that they don't have a lot of insight into, which is the endocrine metabolism and bone. And when we go forward in Phase III, we'll be highlighting these issues and to emphasize to them to -- these are their patients -- the whole patients and refer them out.

Do you expect that the 40% osteopenic rate that you just mentioned, is that reflective of the broader population? Or is that more just what you saw in the specific trial?

Yes. So I think it's clearly what we saw in this trial. But at the same time, I think when one thinks of NASH/MASH in general, especially with cirrhotics, it's a more advanced disease. As a reminder, in the baseline characteristics, the baseline patient in our study, the average patient was roughly a 66-year-old postmenopausal women. And the study was roughly 70% women and most of them, given the age range were postmenopausal. And so the bone mineral density profile for these patients is worse in general, as we saw from the baseline than most average patients. So I think when we go forward, we just acknowledge that patients with NASH/MASH, they do have complicated medical histories.

Are you surprised that we haven't seen the same profile emerge for some of your other FGF21 agents?

Yes. So what we'd say is that some of the other FGF21 agents, no one else has had run a cirrhotic study, so they don't know because they haven't conducted the study. The data for which there are available is in the pre-cirrhotic patient, pegozafermin has run a 48-week BMD in which they saw no difference in placebo versus their compound after 48 weeks. Similarly, when we ran our pre-cirrhotic population like the pegozafermin where it's been studied, we saw no difference between active and placebo at 48 weeks either. So it's really with longer-term dosing, the 2-year point in the pre-cirrhotic population plus the cirrhotic population where we've seen this difference.

Got it. Okay. So maybe too early to tell with some of these other agents. Okay. Let's move to your Phase III here. If you can just provide an update on where you stand with these trials and remind us of the expected timelines?

Perfect. So our Phase III program is just that. It's a program. There are 3 studies that comprise it. The first of which is what we call SYNCHRONY. All the studies are in the SYNCHRONY program. It's not very creative. But the first one is called SYNCHRONY real world. So these are patients who are noninvasively followed. They all have confirmed NASH/MASH phase F1 through 4, and they're being randomized 2:1 to being treated with efruxifermin 50 or placebo, and they're being followed for 1 year. There's not a second biopsy. So this is designed to mimic what happens in the clinic. No one releases a biopsy for nonstudy patients. And so we're following them primarily for safety and tolerability, but we'll also have VCTE ELF score that we'll be following on the efficacy side. We enrolled that study a little ahead of schedule, and we announced that in January of this year. So it was a 52-week study. We expect to have results in the first half of 2026. We also -- the second study is called what we call SYNCHRONY Histology. Those are F2, F3 patients who are randomized to either 28, 50 or placebo, and we're following them for 52 weeks and the primary endpoint is 1-stage improvement in fibrosis and NASH resolution is a composite. And the study is still ongoing in terms of its enrollment, and we expect results in the first half of '27. And lastly, the third study is what we call SYNCHRONY Outcomes. SYNCHRONY Outcomes being an F4 study, which we talked about a little bit earlier. We just enrolled the first patient in September of 2024, and we haven't quite given guidance yet, but we expect to later in the year.

And how are you thinking about the selection of the composite endpoint in the histology study?

Yes. So in the United States, the FDA allows as a primary endpoint, either 1-stage improvement in fibrosis without worsening of NASH or MASH resolution without worsening of fibrosis. However, in Europe, the EMA requires the and, so you need to have both. And so in that way, we just chose the composite endpoint, recognizing that it's an and, and recognizing that the building blocks are the 2 separate components, so the FDA will have what they need to have. And the FDA is aware of our primary endpoint.

And what are the implications of having that composite as you think about the study design and the powering?

Yes. So great point. We know from our 2-year HARMONY data that the composite endpoint because it's an and allowed for a very small placebo response rate because you have to achieve both. While at the same time, there's roughly about a 5x difference when it comes to those patients, roughly 35% versus 7% who achieved that endpoint. So for powering, having a small placebo is very helpful for us. And so that helped with the powering of the overall profile.

Great. And as you think about the Phase III study in F4, one of the big debates here is on the selection of the endpoint and how the regulatory agencies may accept a biopsy endpoint. What is the latest that you -- what is your latest understanding of the path forward here?

I think when one thinks about that, one has to think about what the guidance in 2019 says, which is that the first sentence of the guidance says since no study has ever demonstrated that there's an improvement in one-stage improvement in fibrosis, we have to conclude that the best way is to look at outcomes. But given the data we've already talked about from our SYMMETRY study, I think that sentence might need to be revised. So in terms of how we're dealing with the agency, we're very comfortable given the language we see from the Europeans that they accept 1-stage improvement in fibrosis in cirrhotics, and we're working with the division to get there. And I know some other companies in the space are very clear that they have a pathway, and we feel that if they have one, we are likely to have one too.

Great. Makes sense. Anything you can share? We've touched a little bit on powering and how the choice of that endpoint has improved the powering of the study for your histology study. But what can you share as it relates to your cirrhotic study?

Yes. So I think the cirrhotic study has 2 components: Cohort 1, which is powered for biopsy and the second component, which is really the outcomes. I think one thinks about the biopsy, which is a little bit easier to understand, we were initially powered for -- based on the 36-week results. But in light of the, frankly, unexpected and yet gratifying results that we saw in January when we had the 96-week data, we're reexamining the powering because the effect size was greater and the number of patients we may need in that cohort may decrease somewhat. We haven't reached a conclusion on how we're going to modify that given the nature of this cohort 1 is still enrolling. We still have a lot of runway, and we'll probably discuss that with the agency. And so we're not quite there yet.

For your Phase IIb, I guess, maybe for that, for that study there, is there the potential that the 96-week data could support approval? Or do you really need this Phase III study?

I think we really need the Phase III study. As a reminder, as good as the results are, there really were 46 patients at the 96-week time point on 50 milligrams. And there -- we believe there are roughly 1.5 million to 2 million Americans with NASH cirrhosis. So hard to see how that's going to translate in light of such a big population.

Understood. That's a nice segue to where you see the commercial market evolving beyond just Rezdiffra. Rezdiffra potentially sema getting their label expansion. And then you could be on the market as the second liver-directed agent for NASH. How do you see the commercial market?

We see as nicely developed by Rezdiffra and the Madrigal team, we see the market is quite large and quite a bit of demand for a drug that can improve and reverse fibrosis. So we're excited for the F2, F3 space, and we're excited for the F4 space.

Where do you see efruxifermin fitting in?

We think given the chronology of approvals and the idea that, a, Rezdiffra has been -- by the time we get approved, we will have been on the market for several years and the fact that when the semaglutide product gets approved later this year, and remember, it's a line extension, having obviously had a diabetes indication for more than 10 years. We're not going to be first line. And unfortunately, for most patients on these drugs, the effect size is not the largest. So there will be some patients who unfortunately were not served by either of those 2 agents. And hopefully, we can help them.

Got it. So that's for the F2, F3.

Right.

And then presumably for F4, that's where you would really find.

Correct. For F4, since we likely are the only drug to have a 1-stage improvement of fibrosis demonstrated to date, we would probably be the drug of choice for those patients since there really isn't anything else.

If Madrigal's outcome study is successful and they show benefit, albeit not with a biopsy endpoint, does that change your calculus as you think about where efruxifermin could be used in the F4 patient population?

Not really because I think the -- it all -- it's a little bit like 1-stage improvement of fibrosis in pre-cirrhotics. So it works. And I expect that Madrigal's drug is doing so well. But at the same time, as we move forward, drugs may come along that have a larger effect size. And so even if they're approved and are successful in their outcome study, drugs that as long as we are able to have a bigger effect size may have the chance to benefit patients.

As you think about how FGF21s will fit into this market space, there may be 3 of you coming on to the market in pretty short order. How do you think about the relative split between efruxifermin, pegozafermin, efimosfermin? Are there enough differentiation to?

I think it's incumbent on us as a company and 89bio and now GSK to create that differentiation between us so that physicians and patients have enough data from which to choose. And we're happy that when I think about it, that there are multiple choices because that's how they are -- we think about diabetes. Think about statins, think about PD-1s. There are multiple choices within the space. And I think the net of this is it all benefits physicians and patients.

What has your market research suggested to you on that point about which FGF21 will be the agent of choice?

Well, of course, our market research is -- we have some ideas and concepts that may not be borne out. But I would say that given the data that we do have, and we're the only compound with data, it obviously looks on our data a little more. We await the results from the other 2 companies, and we'll have a better answer for you once we see those results.

All else equal on efficacy and safety, how important is the dosing -- the dosing frequency?

So I think the dosing frequency, obviously, less frequent dosing is better, assuming that there's no difference in the safety profile. Or I think in some cases, there are in other classes, there are drugs that may be administered, let's say, monthly versus weekly, but it's multiple injections given on the same day, which may not distinguish themselves favorably compared to other drugs. So I think it all depends if it's one injection once a month with the same volume, I think that is very favorable. I'm not sure that that's the situation we currently have in the FGF21 space.

Got it. And what is your understanding right now of how payers might approach in FGF21 being on the market, particularly as it comes behind some other agents that will be potentially cheaper, more entrenched in the marketplace.

I think it's one where it's incumbent for drugs that always come later to demonstrate value. And especially if they want to charge a premium to it, then that's -- they really have to make the case that they deserve that payment. So that has to be demonstrated clinically. They have to prove -- I mean, of course, you can do that in a head-to-head study. But in the absence of that, you really have to make a very clear case that your drug merits that differentiation. And so that's incumbent on us to win that battle.

And if you're able to make that argument that there is differentiation do you think payers would forgo any type of like step-through requirements? Or is that still probably on the table?

Payers have a lot of tools to advantage or disadvantage your product, some of it based on the clinical profile, but some of it based on the economics. And so how you're able to negotiate or navigate that pathway, I think, is a little bit of a TBD because the TPP target product profile still is established in Phase III. And so we don't know what that will be. And also, we can't understand exactly how the market will be when we -- as we move forward. And a lot of it will have to do with price and has to do with rebates. And so there are a number of things that can influence whether how step edits are used and how products are advantaged or disadvantaged.

What are your thoughts on combination use?

I think combination use is here to stay. It already is here. So recall that even in our SYMMETRY study , 20% of the patients receiving GLP-1. So combination use is there. Was it combination use for MASH, was it combination used for diabetes, combination use for any other indications, obesity that were being used. I think when one thinks about a drug like a GLP-1 that has multiple indications, it's likely to be -- play a role given the patient population and likely to be here for quite some time.

Mechanistically, do you see rationale for combining efruxifermin with another liver-directed agent?

We don't have a lot of data, so we don't have any data with something like a Madrigal's compound because it was just approved last year and most of the data we have today preceded that. I think one can see drugs that can work for patients. And I think there's a lot of flexibility. We're not precluding any one of those things. But I think until the data are available, a little tougher to assess accurately for all liver-directed drugs or some, we'll have to parse that more carefully.

Maybe just on that point, as you are enrolling your Phase III study, Rezdiffra on the market now, how are you handling those patients who may have been exposed to the drug?

So I think right now, for Phase III, we're allowing those patients because it's an approved agent, but it's one of those things that we stratify by.

Got it. Okay. Interesting. And then maybe in the final couple of minutes that we have here, given the FGF21 mechanism, are you interested or do you see utility in exploring other adjacent indications outside of NASH?

We do. So I mean, I think that's one of the questions we often get asked by investors is what are you going to do about your pipeline? And when we think about that, so most commonly that's done for many companies, let's look at a novel new molecular agent. And so what would be that entity that we'd be studying. So that is a path where we could change. There are some things that are interesting to us. At the same time, given the result we have in F4 that we can reverse cirrhosis, that's particularly interesting to us because cirrhosis has been challenging for decades that it's been irreversible and really been a very challenging position for patients and physicians. So as we look at how to address that, we think we have a valuable sort of beachhead as it were, and we've demonstrated cirrhosis, which is a common response of the liver to a variety of insults. In our case, it's due to NASH/MASH, but there are other insults that result in cirrhosis. And it may be more a better use of our resources and of the agent to capitalize on the investments we've made in terms of manufacturing, toxicology, CAR-T studies, all the preclinical work that's gone in and perhaps explore other causes of cirrhosis as something we may want to follow up on.

Interesting. Any interest in -- I guess, because you do have such a potent antifibrotic agent, any interest expanding even beyond cardiometabolic into maybe like pulmonary fibrosis indications?

So we feel the fibrotic pathway is pretty common. And so the balance between degradation and synthesis of new collagen is applicable to a lot of different organs. I would say that to do that, let's say, in the pulmonary, let's say, we have to look at other fibrotic diseases of the lung. That's a little bit of a farther put, as I say. And so we probably need to do a Phase IIb study to really have proof of principle. At the same time, within the liver, I think we're less likely to do that. We potentially could go straight into Phase III. Of course, that requires consultation with the agency. So I think staying within the liver is a little bit easier, a little more -- less of a stretch, but it doesn't preclude us from going outside the liver.

Got it. Maybe the final question here, just remind us of your cash runway. Where does that get you to?

Yes. So we're fortunate to have investor support that we have $1.128 billion in our last report of our Q1 earnings. And so we get -- that gets us into 2028.

Got it. Into commercialization?

We haven't given that -- we haven't been as granular as how that gets us into 2028.

Okay. Well, with that, thank you, Andrew, so much.

Thanks again for having me.

Thank you, everyone.