Inventiva S.A. ($IVA)

Great. Welcome, everyone. Very excited to have Inventiva here with us today ahead of a very exciting year for you into a Phase III readout. We're getting a lot of investor questions on this. So a lot to talk about today. I'm Ellie Merle, one of the biotech analysts here at Barclays. Joining us from Inventiva is CEO, Andrew Obenshain, I'm so sorry if I butchered your name, Chief Executive Officer; and Jason, Chief Medical Officer. Thank you both so much for joining us today.

To begin with, can you give us an overview of lanifibranor and the design of your Phase III trial as we head into the data later this year?

Yes. So maybe I'll start at a high level, and I'll hand it to Jason for the Phase III design with lanifibranor. So we are a one-asset company. Lanifibranor is a pan-PPAR agonist that we're developing in a Phase III study for [ MASH ]. We -- it is a product that's been in development for quite a number of years. It's a rationally designed molecule designed to design out the liabilities of former PPARs. So learned from Actos and Avandia, for example, and designed out the liabilities by making a more moderate gamma binder. And then also it's a balanced PPAR. Another insight was that these PPARs are networks. And if you affect one, you affect them all, and therefore, you want to hit them all in a balanced way. Before we go into the Phase III, I do want to comment on the fact that we were required by the FDA to do quite an extensive toxicology work, 2 years in animals that were -- that was due to the past PPAR experience. Those came out that we press released those. Those actually had Actos and Avandia, going through those. They would not have survived. And we showed that we had a very good tolerable profile, went into the clinic. We've done 5 or 6 Phase I trials. We had a positive Phase IIb trial published in the England Journal of Medicine. So quite an extensive background before we even got into this Phase III. But maybe, Jason, you can talk about then how we've started...

That's great. As Andrew said, NATiV2 was a successful Phase IIb trial, published in the England Journal of Medicine now about 5 years ago, 2021. I think in that era, the excitement, it sort of lost 5 years later, but it was the first drug ever in NASH to not only show a meaningful effect at 6 months, but it had done so on a dual co-primary endpoint of both NASH resolution and fibrosis improvement in the same patients. So NATiV3 just builds on the strength of that study. NATiV3 is largely similar in many cases identical to NATiV2. Obviously, it's longer. It's 18 months, but it's still the same basic construct, 3 arms, 1:1:1, 2 doses of lanifibranor against placebo. The primary endpoint is somewhat unique, though, in the NASH field. It's the composite of both NASH resolution and fibrosis improvement of one stage and more in the same patient. And we think that, that reflects some of that unique biology Andrew was sort of hinting at and that we -- the ability of the drug to get both intrahepatic and extrahepatic drivers of NASH showed in NATiV2 that we can, in the same patient, resolve all of the features of NASH. And therefore, we brought that forward as our primary endpoint in Phase III.

And how should we think about the powering of the study?

Well powered. For a Phase III study, we are -- we've been guiding that we are powered to above 90% on the primary end, the co-primary endpoint of NASH resolution and fibrosis improvement. What we are not giving too much specifics about is what's the effect size, [ Jason ]. But what we can say is that we took a more conservative approach than NATiV2. Our effect size was 24% there with a placebo response rate of about 7%. What we did was we took a higher placebo response rate and therefore, a lower treatment effect and powered to that. And then we overenrolled the trial slightly, and we built in -- because it's an 18-month trial, and it was getting going during COVID, so we were a little nervous. We built in a much larger sort of dropout rate than would otherwise be expected in a Phase III up to 30% dropouts, and we would still maintain and preserve that power.

Okay. So in my view, and jump in if I'm wrong, that sounds like quite conservative powering where if you're 90% powered to show what the 18% placebo-adjusted effect size in Phase II, you probably could show meaningfully lower than that and still be statistically significant. Fair?

Fair. Yes.

Another way to say that is in 247 patients, we hit statistical significance on all 3 NASH endpoints. And now we have a 1,007 patient double-blind, randomized controlled trial, we are very well powered.

And at a much longer time point where you should see increasing effects over time.

Exactly right.

And how should we think about that? How much that could...

I think broadly speaking, I mean, we don't know. In the end, it makes -- there's a rational argument made that longer treatment is going to result in deeper effect. I think the 3 lines of evidence that support that are pretty straightforward. Other sponsors have shown it. Akero has shown it, for example, when they went from 6 months out to longer. Intercept showed it back in the day when they revealed top line data on fibrosis improvement. 18 months later, they continue to follow those patients, and they had deepening of the fibrosis effect by TE. So that's very good. And lastly, the biology of lanifibranor, the way that PPAR agonists work, the sort of metabolic reprogrammers that work at the level of DNA transcriptional activation, that just takes time. So 6 months, no doubt, we saw an effect. We're very confident in that, but it's really early for gene regulation and transcriptional activation to show themselves fully. So if all those 3 things line up, we think it's reasonable to think we're going to get a deeper effect.

How do you expect the patient baselines to compare in Phase II versus Phase III?

Yes. I'll start, it's largely similar. The big difference is we enroll more patients in the U.S. Therefore, there's more diabetes patients in the U.S. So we have 55% diabetes patients in the trial, 42% were in Phase II. And actually, the stage of disease, F3 travels with diabetes. The more metabolically dysfunctional patient, the more likely they are to have later-stage disease. And you see that with the baseline of a higher baseline of F3 patients in the trial. The other difference in the trial was that GLP-1s really weren't around during the Phase II. We had 14% of patients that started on a baseline GLP-1, not the [ MASH ] dose, the low dose for diabetes. Those came in on a stable dose. We had another about 10% of patients that dropped in during the trial, but should be balanced by arm across placebo and drug. Anything else?

Great. That's helpful. Sorry for jumping into the trial design questions. You can tell where we're getting a lot of questions. But maybe taking a step back, my favorite topic, [ MASH ]. How do you see the landscape? There's a lot of drugs in development. You have Rezdiffra, you have GLP-1s now available. And then, of course, FGF21 is coming along. Where does lanifibranor fit in the development landscape?

Yes, absolutely. So just to take a step back first, broadly, we're at the very beginning of this market, right? We've just had 2 approvals recently, more that are going to be coming. We are in the very beginning of what will eventually be multiple classes of drug and multiple drugs in those class and growing quite substantially as a market. Where we are today is we think there's about 375,000 patients in that F2, F3 space where lanifibranor will compete. In that space, we believe there's going to be a baseline use of GLP-1s. GLP-1s for us are friend, not foe. And then physicians will layer on an oral to address the fibrosis on top of that, either Rezdiffra or lanifibranor. I'll get back to that in a second and how we see that playing out. But just to comment on the FGF21s, we really view those as staying in F4 for a number of reasons. First of all, we believe the pharmaceutical companies will premium price those. And there injectables competing in an oral market. They've got some toxicities, both bone density and GI that are going to really prevent them from moving into that F3 market. So we're really seeing the staying in the F4 market. So bring us back to the F2, F3 market. So how does the physician choose between lanifibranor and Rezdiffra? So the way we segment this 375,000 patients is into 4 boxes, right? It's a 2 x 2 grid. You think about F2 and F3 and they're roughly equal, 50% in each. And you can think about nondiabetic diabetic, and it's about 40% nondiabetic, 60% diabetic. So you end up with 4 equally sized boxes relatively equal, a little bit bigger for the diabetic. And where -- what's the natural habitat with these drugs? Well, if you think about an F3 diabetic patient, right, physician, let's say, lanifibranor has just launched and a patient walks in and the doctor needs to make a decision about what world to put that patient. Let's say we duplicate our Phase II and say, okay, well, super worried about fibrosis because I don't want that patient to go to F4. I want to pick up the biggest hammer I have on fibrosis, and that's going to be lanifibranor, 18% versus 12% if we duplicate the Phase II. And if patients got diabetes, I know that's driving the disease. The metabolic function is driving this disease, and I need to address that, but probably already on diabetic medications, but why not add something that can reduce HbA1c even further. So lanifibranor becomes a very natural choice there. A patient walks in the day after lanifibranor launches, and there's an F2 nondiabetic patient. Well, that's probably a patient I'm still going to put on Rezdiffra for now, right? It's a proven drug, it's comfortable with it, doesn't have diabetes. I'm not worried about the fibrosis going to F4 overnight. I'll stick with it there. And then you can see that as the physician gains experience that a very natural place for lanifibranor to go is then to the other F3 patients without diabetes or to the diabetes patients with F2, and that's how we see the market evolving over time. And if we end up getting better F2 than 18%, maybe we double what Rezdiffra has, then I think that you migrate into that whole market.

Yes. And that 18% was at 24 weeks now, not...

That's correct. Yes. It was 18% after 6 months versus 12% after 12 months.

Great. And so -- and then in Phase III, you're looking at 18 months?

Yes, correct. And we do expect a deepening of effect and hopefully some better results.

How do you think about the breakdown between the different prescribers, especially given this effect that you're seeing in diabetic patients as well as maybe more physician familiarity from endocrinologists with the PPAR class. As you think about the segmentation, I mean, it seems like -- right now, most of the NASH patients that are being treated are being treated by hep, GIs, less so in the endocrinology segment. Where do you see this evolving in the coming years? Because I mean, correct me if I'm wrong, but I would guess that a lot of NASH patients are sitting with endocrinologists right now.

Certainly, the endocrinologist is probably keeping them in the opinion of the GI and the hepatologists probably keeping them too long. So -- and this is something we will make a decision now about how we do this. We'll definitely be monitoring it. But overall, you can think of -- we clear that we're not going to be looking to get switches in terms of patients off of Rezdiffra, on lanifibranor. We're looking for new patients. So we'll monitor closely kind of how the market is growing, where the diagnosis is happening and where those new Rxs are. I think GIs and heps are clearly going to be a call point and a starting call point for us. And whether we add endocrinologists at the time of launch remains to be seen or to what scale. But I think certainly, there's at least a subset of endocrinologists that should be in the targeting and launch.

And what proportion of patients do you expect in your Phase III to be on GLP-1s at baseline?

So baseline GLP-1 use in NATiV3 is a little less than 15%. So they needed to be stable treatment, non-NASH dosing. So you're talking, one, it's not all sema. I think all eyes and heat and light are on sema, but it's not only sema, there's liraglutide, dulaglutide as well. But whatever it is, they need to be on stable dosing and it needs to be the diabetes dose because it's for concomitant comorbidities, diabetes in this case and not for NASH. So we like that data set across both. We have 2 cohorts, the main 1,000 patient randomized double-blind and then a sort of expansion cohort for screen fails of about 400 patients. That would give us a data set of about 300 patients. And I think as Andrew mentioned earlier, GLP-1s are going to be the backbone of the therapy, particularly patients entering at that F1, F2 side. Ones that are caught later by the heps, GIs, not clear that they'll be on GLP-1s, but clearly, the earlier ones will be. The idea that we'll be able to have data that say that we work and play well with GLP-1s will be fantastic. So we think it's going to be a good data set.

And then the mix of F2, F3 patients in your trial that you expect?

It's contemporary. So we have roughly 2/3, 1/3, 2/3 F3, 1/3 F2, which is fairly similar to what [ Madrigal ] had in their Phase III program. It's a little different than our Phase II, but the world has changed. As Andrew said, we have more diabetics today that both have NASH. So that's also reflected in the diabetes status. We have about 55% diabetics in the NATiV3 trial compared to about 45% in NATiV2. So both of those really reflect the contemporary practice today.

And that's helpful because you're enriching for the patients that were better responders in Phase II.

Well, it's interesting. We actually did a responder analysis. We published it after the NEJM article in 2024. Independent of diabetes status, the drug actually works equally well. So even if you strip out the F1s from NATiV2, the drug works slightly better in F2, F3, but I'll say here that it works equally well. So one, we're not getting a free ride on the back of the sort of lesser severity patients. And second, diabetes isn't giving us the kind of tailwind. But on the other hand, in the presence of diabetes, or hemoglobin A1c reductions like in our LEGEND study approach one full point. So that has all the makings of a drug that could be valuable to endocrinologists as they're trying to manage both diseases. So although it doesn't work better on histology, it does have the glycemic effects in the diabetics.

Right. Makes sense. That's helpful. A question that we get from investors a lot is around safety, particularly the history that the PPAR classes have with safety. Can you walk us through the history of the safety of the PPAR class briefly? I know there's a lot to go through on the PPARs, but just kind of an overview of that and where you see lanifibranor falling and your confidence in the safety.

I'm going to hand that one to you.

That's so kind I'll actually pick up where Andrew left off. So lanifibranor was rationally designed by a group of chemists that were PPAR biology founders. They had created fenofibrate back in the day. And their goal was to sort of get the biology that these metabolic reprogrammers, PPARs could do without a lot of the baggage of some of the therapeutics that have been on the market. So the stepwise approach was design the molecule. Second, run the toxicology data in conjunction with FDA to say what's true for this drug and what's not true. And back in 2019, we had publicly talked about that, that toxicology program Andrew mentioned, full 2 years to 2-year rodent 1-year nonhuman primate, discharge many of the traditional PPAR risk, things like cancer, muscle damage, kidney damage sort of went away. And that it showed that as predicted from the structure of the drug, we had gamma effects, but they looked a little muted and blunted compared to like pioglitazone or rosiglitazone. So very good. Rubber meets the road in clinical trials. And what we showed in NATiV2 sort of confirmed all of that. So if you look at an adverse event table in NATiV2, it looks like a lanifibranor adverse event table. It doesn't look like other PPARs. It's its own footprint. What's absent are a lot of the PPAR alpha delta safety issues, good. But what is present are some of the gamma concerns, but they do appear to be muted compared to what you see with longer-term bigger studies with pioglitazone. We do have weight gain, for example, but it appears to be less than what you would see in pioglitazone, and it tends to plateau after about 6 months or so, good. The peripheral edema that one sees associated with that weight gain appears to be less. And the final arbiter of that is the congestive heart failure signal. Now obviously, we're running a blinded clinical trial, but based on NATiV2, we like that step off a lot, less weight gain, less edema and much, much less heart failure compared to sort of historic norms. So I think in summary, we like our safety profile, not because it's better than pioglitazone, it's our own. And it's been sort of matched with exactly what the people that designed the drug 15 years ago were hoping that they would get in the clinical environment. Stripped of a lot of the safety baggage, but still getting you the same intrahepatic and extrahepatic biology that made PPAR so interesting in the first place.

What are you seeing in terms of weight gain? And how does that compare to what's seen with pioglitazone?

So why don't you start with the weight gain and then I'll go to pio.

So I think -- the first thing to understand is about 50% of patients in our Phase II did not gain weight. 20% gained very modest 2.5% to 5%. And there was a 30% they gained 5% or more. What's interesting is the way the doctors code that, right? If you look at the adverse event table, a weight gain is only 10% because in many -- this is not perceived as an adverse event. It's really a tolerability issue and physicians actually only recorded 1 out of every 3 patients with that weight gain over 5%. So that's the facts of where it is. Maybe I can hand it to you to talk about kind of what the biology is, and then I can talk about how we think this will play in the market.

Yes. So on the one hand, the gamma effect of pioglitazone and rosiglitazone, these are incredibly strong metabolic reprogrammers. In fact, pioglitazone probably has some of the best glycemic data that exists ever. I mean they're fantastic. They even had outcomes back in the day. It was in the 1990s. I don't think people think about outcomes in that regard. But they actually had really good outcomes in diabetics that had macrovascular disease when they looked at like what we would think of as sort of a MACE endpoint today. Hazard ratios look pretty similar to what we've seen in the monitor. The problem was that it came along with this question of weight gain edema and congestive heart failure. So what we see is, generally speaking, the weight gain that looks -- it's about proportionately numerically less than what you might see with pioglitazone. More patients tended to gain more weight with pioglitazone and it lasted a longer time. So if you follow those patients for years like in the PIVENS study, the weight gain kept going up and up and up, and it reached a plateau maybe at year 3 as opposed to something like maybe month 6 or 9, which is where we're landing. And the overall magnitude of the weight gain, which is what Andrew just showed is less. And when you step that down, the peripheral edema, in some cases, looks to be fivefold lower than what we reported in some trials. But if you look at the label, for example, it might be two to threefold lower than what was reported. So we like that. We call it a sort of blunted or attenuated gamma effect, but it still is very clearly a gamma effect, which is really just salt and water retention in the end. So that's what all this comes back to is that the gamma works by modulating salt reabsorption and you get some peripheral edema and fluid retention.

But then I think just to bring it back to that market segmentation that I talked about, that it is so a chance of weight gain is going to be part of the profile. For that F3 diabetic patient population, that is a perfectly acceptable part of the profile in terms of physician choice. For the F2 nondiabetic, that's probably where it becomes a little bit more of a liability unless your fibrosis score is so good that it overcomes that, right? So it just plays into that market dynamic.

We've seen in labels across the PPARs, black box warning for congestive heart failure associated with the edema in warnings for bladder cancer risk. As we think about your profile, what are the reasons why you think you're seeing a lower risk relative to pioglitazone?

So it's a tricky question, right? Because in the end, FDA gets a big vote, too. So -- and we don't have full data yet from NATiV3. So it's all highly speculative. But that being said, the 3 elements that matter are what's your drug, your class, what are your data? And can you manage the risk in a trial environment or market environment? Number one, we have a different scaffold. We are not a TZD. We're a completely novel agent. So we're not pio that also happens to have alpha and delta. We are a completely novel agent there. Second, we'll have our own data sets, as we just talked about, they will or they will not look differently in the end than pioglitazone that will matter. And lastly, how are we doing in the clinical trial when the FDA ultimately looks at were you able to manage whatever effects you see. If they're comfortable that it's a different drug and the data look different and you're managing risk well, well, then there's no reason for a box warning. That being said, FDA has decisions of their own, and they may still say, you're going to get a box no matter what. But we actually think we're in a really good position to walk in with a straight face argument, not the same drug. We have our own data set. And we think we've done a really nice job in a trial environment of managing our risk. Therefore, this is the label that we're proposing based on our data. We'll see what we get.

That's helpful. And what are your plans in F4?

I'm sorry.

Your plans in F4?

Do you want to start?

Yes. So we are -- we will be approved under conditional approval in the U.S. for F2 and F3. So we'd have to do an outcomes trial. We are choosing to do that outcomes trial in F4. I'm willing to find that a little bit more. So we'll start that trial after we get data. We have to have substantially started by the time we file our NDA. But we're looking to choose a patient population that is sick enough that it can progress to outcomes, we can measure outcomes, but not so that we can't reverse the disease. And we've identified a patient subgroup that has portal hypertension that we believe is the right target for that trial. I'll hand that over to you to maybe to explain the biology.

Yes. I mean this is -- honestly, this is a really active area for those that are listening. It just -- it will develop over the next couple of years. But this idea that we can now identify easily at the bedside, a group of patients that are either advanced F3 or F4, but they have the physiology of cirrhosis, portal hypertension, as Andrew said, that you can identify them easily at the bedside really makes the possibility of putting them in clinical trials really attractive. So this idea that you can have a group of patients in the trial that have the portal hypertension that puts them at much higher risk of liver-related outcomes, that's fantastic when you think about a trial design perspective. But what we like the most about it is that, that group of patients appears to match the biology of lanifibranor really nicely. The intrahepatic and extrahepatic modifying elements of lani on the alpha, delta and gamma PPAR receptors, each of those pathways are driving some of that portal hypertension. So we have a fair bit of preclinical data that we published over the years, including at the liver meeting last year, AASLD in Washington sort of outlining that data. So we're excited. And we have about 75 patients roughly in our expansion cohort in our Phase III program that was a really nice view of the safety of both doses of lanifibranor. So we're excited to begin to talk to FDA about this and build that trial.

Strategically, we've seen 3 acquisitions in the [ MASH ] space in the last year. You are a fraction of [ Madrigal's ] market cap. Those are just fast. I'm not speculating. How are you thinking strategically about your plans for commercialization and what might make the most sense, whether you go it alone or seek a partner?

Well, [ Madrigal ] has really laid out the path for how a midsized biotech can commercialize successfully and drive to profitability. So we plan to follow that path. So you should expect us to -- if we have positive data to raise a lot of money to be commercializing ourselves. Now along the way, yes, you're right, there might be some strategic interest that might prevent us from realizing that mission. But we are planning to commercialize ourselves in the U.S. definitely and then secondarily in Europe.

What is your cash balance and cash runway currently?

So cash runway goes to about Q3 '27 right now, so well on the other side of data, which gives us a lot of room to make some strategic decisions post data.

Great. Just a question on the confirmatory study. How are you thinking about how you might design that and start it shortly before we might get the data for [ Madrigal ]? I can imagine that's not...

Not exactly. I think what FDA wants is very clear on this. You need to have the trial meaningfully underway at the time of approval, at the time of filing that the trial started. And then prior to that, we have a lot of opportunity to engage with FDA on exactly how they define that. That's the point of it. So it's not a get going and let us know how you're doing. It's active engagement with them at the prefiling meeting, the interim mid-cycle review meeting. But our plan is that clearly, we will wait for data to have any meaningful start on that study, but we're already laying the groundwork for that now.

Great. Awesome. Well, I think we're out of time, but I appreciate both of you joining us today and certainly an exciting year ahead, ahead of the Phase III data.

Thank you. Thanks for having us.

Thank you.