Inventiva S.A. (IVA) Earnings Call Transcript

Ladies and gentlemen, thank you for standing by, and welcome to Inventiva's NATIVE Phase IIb top line results in NASH. [Operator Instructions] I must advise you the call is recorded today, Tuesday, the 16th of June, 2020. I would now like to hand over to your first speaker today, Mr. Frederic Cren. Please go ahead.

Hello, [Foreign Language]. Very pleased to welcome you on this webcast. Of course, you can understand from the tone of my voice that I'm extremely satisfied and pleased with what lanifibranor has shown in the NATIVE Phase IIb data. We really believe we have a compound that is active in NASH, and the data has clearly established that and we are in a very good position for the next step that include also a Phase III. I will be sharing the floor today with Pierre, who will go through the presentation with Marie-Paule Richard, our CMO, and also we're very pleased to have our both PI of the study, Sven Francque from the University of Antwerp and Professor Manal Abdelmalek from Duke University. It's great to have them with us, and they'll be able to answer all of your questions. And then, of course, as I just mentioned, at the end of the presentation given by Pierre, we will have a Q&A session to address all of your questions. So Pierre, with that.

Yes. Thank you, Frederic. So good morning, everybody. Yes, we are very excited to share with you the good news on the outcome of the NATIVE trial with lanifibranor in NASH. And before diving into the data, I would like to go through a few highlights. So firstly, lanifibranor has met the primary endpoint with statistically significant reduction after 6 months of treatment with the 1,200-milligram dose of the Steatosis, Activity, Fibrosis score. The SAF score, which you know combines hepatocellular inflammation and ballooning, is with no worsening of fibrosis both in the intention to treat and the per protocol populations. Lanifibranor also met key secondary endpoints including NASH resolution with no worsening of fibrosis and improvement of liver fibrosis with no worsening of NASH. And here again, in both ITT and per protocol populations. What comes out from this trial is that actually lanifibranor is the first drug candidate to achieve a statistically significant effect on the FDA and EMA primary endpoints that are relevant for seeking accelerated approval, meaning NASH resolution with no worsening of fibrosis and improvement of fibrosis with no worsening of NASH. Lanifibranor continue to show a favorable safety and tolerability profile. And this positive top line results support Inventiva's decision to move forward with the clinical development of lanifibranor and to enter into a pivotal Phase III trial. So you're all very familiar with the profile of Lanifibranor. But as a recap, it's a pan-PPAR agonist with moderate and well-balanced activity on the 3 PPAR isoforms, PPAR alpha, PPAR delta and gamma. And as you know, there is a strong scientific rationale supporting the concept that this unique mechanism of action would be very well suited for the treatment of NASH with the potential to address most, if not all, features of NASH, and particularly steatohepatitis and fibrosis. And as shown in long-term nonclinical tox studies as well as recent clinical study, lanifibranor is well tolerated and safe. Next slide. So in that context, the NATIVE trial investigated the efficacy and safety of 2 doses of lanifibranor in patients with biopsy-proven NASH and the combined score of inflammation and ballooning of at least 2 points out of a maximum of 4, as determined by using the SAF score. As recently communicated, by using this inclusion criteria, we successfully randomized the population of severe patients, with 73% of the patients having a NASH score equal or superior to 6 and 76% having fibrosis F2 or F3 score. The study randomized a total of 247 patients and was stratified on type 2 diabetes status. So I will present the data in 2 populations: the intention to treat or ITT population, which is also the safety population and defined by the 247 patients, randomized, having received at least one dose of lanifibranor or placebo. And the per protocol population, or PP, defined as patients randomized, having paired biopsies and no deviation impacting efficacy reserves. These deviations could be, for example, alcohol consumption or use of forbidden concomitant medication. So in the next slide, this is really just a reminder that the NATIVE trial is a global study with a total of 71 sites, North America, so Canada and the U.S., in Europe, in Australia as well as in Mauritius. So in terms of efficacy endpoints, the primary endpoint was the decrease from baseline to week 24 of at least 2 points of inflammation and ballooning and no worsening of fibrosis as measured by the SAF score. The secondary endpoint on which we will show you data today, our resolution of NASH and no worsening of fibrosis, improvement of fibrosis by at least one stage and no worsening of NASH, resolution of NASH and improvement of fibrosis by at least one stage and then several endpoints related to [ metabolic ] factors such as change in glucose metabolism parameters, like fasting glucose and insulin, glycated hemoglobin, change in liver function test, ALT, AST, GGT, change in main plasma lipid parameters, we will see HDL cholesterol, LDL cholesterol and triglyceride. And the other outcome measures like inflammatory and fibrosis biomarkers are still under evaluation and will be communicated later. So the -- this slide is just to remind you that the NATIVE trial uses both the SAF and the NASH scoring system. The primary endpoint, which is, again, a decrease of 2 points of SAF activity score and no worsening of fibrosis, obviously, uses the SAF score, whereas the endpoint resolution of NASH with no worsening of fibrosis uses the NASH score as did most of the recent trials with competing drugs. And fibrosis improvement was measured using the CRN fibrosis score, as did recent trials with competing drugs. So in terms of patient disposition. So 247 patients randomized and treated. We had 81 patients randomized and treated in the placebo arm, 83 in the 800-milligram and 83 in the 1,200-milligram arm. 91% of patients completed the 40-week treatment in the placebo group and 93% in each lanifibranor treated groups. The number of dropouts and dropouts due to AEs were well balanced between the 3 arms. We had 9% of dropouts in the placebo arm and 7% in each lanifibranor treated arm. There were 3 dropouts due to adverse events in the placebo group and 3 dropouts due to adverse events in each lanifibranor treated arm. So in terms of patient baseline demographics and characteristics. There was overall 58% of females, a mean weight of 93 kilos, a mean BMI of 32.9, 42% of NASH patients with type 2 diabetes and all were well-balanced between the 3 arms. We had an overall mean SAF activity score of 3.3, well balanced between the 3 arms. Overall, we had 73 of patients with a NASH score equal or above 6, with slightly more patients with a NASH score equal to or above 6 in the lanifibranor treated arms relatively to the placebo. There was overall 76% of patients with an F2 or an F3 score. And here again, with slightly more F2 or F3 patients in the lanifibranor treated arms relatively to placebo. Baseline liver enzymes were in the range of what is expected in this patient population. Triglyceride levels were borderline high and HDL cholesterol borderline low, also in line with expected levels in this patient population. And in diabetic patients, all had glycated hemoglobin of 6.5 or above and were well-balanced between the 3 arms. Okay. Next. So now the results. So in this slide, you can see that lanifibranor met the primary endpoint with a highly significant effect achieved at the dose of 1,200 milligram in both the ITT and per protocol populations after 6 months of treatment. 49% and 55% of patients fitted with the 1,200-milligram dose had a reduction of 2 points of their combined score of inflammation and ballooning with no worsening of fibrosis in the ITT and the per protocol populations, respectively. There was a positive trend with the 800-milligram dose, with 41% and 51% responders in the ITT and the per protocol populations, respectively, but this did not meet statistical significance. So in conclusion, lanifibranor met the primary endpoint in both ITT and per protocol population. Regarding resolution of NASH and no worsening of fibrosis. Lanifibranor produced a dose-dependent and significant effect on the resolution of NASH with no worsening of fibrosis at both the 800- and 1,200-milligram doses. In the ITT population, 33% and 45% of patients achieved resolution of NASH with no worsening of fibrosis after 6 months of treatment with the 800- and 1,200-milligram dose, respectively. And in the per protocol population, this endpoint was met by 40% and 49% of patients with lanifibranor 800 and 1,200 milligram, respectively. If we look now specifically to the F2 and F3 patients, which constitute the typical Phase III population, we see actually that the high level of efficacy of lanifibranor is maintained with [ both ] doses being highly significant. In the ITT and per protocol populations, if we look at the size of the effect relatively to placebo, there are nearly 4 and 5x more patients achieving resolution of NASH without worsening of fibrosis at the 800- and 1,200-milligram dose, respectively, when compared again to placebo. Now regarding improvement of fibrosis and no worsening of NASH. After 6 months of treatment, lanifibranor also improved fibrosis by at least one stage without worsening of NASH, with highly significant effect at 1,200 milligrams in the ITT and the per protocol populations. At this dose, 42% and 46% of patients improved fibrosis by at least one stage without worsening of NASH in the ITT and per protocol populations, respectively. Now very interestingly, if we look now at the number of patients that improved for both resolution of NASH and fibrosis, we see a highly significant effect of lanifibranor at 800 and 1,200 milligram with 3 and 4x more patients having NASH resolution and fibrosis improvement in the 800- and 1,200-milligram dose, respectively, compared to the placebo and the ITT population. Now regarding liver enzyme ALT, AST and GGT. All were quickly, so within a month, brought back to normal levels and was maintained to such levels all along the study time course. If we move now to lipid, we see that on the lipid profile, as previously observed in our Phase IIa, both doses of lanifibranor rapidly increased HDL cholesterol and decreased plasma triglyceride. There was no change in LDL cholesterol. So in NASH patients with type 2 diabetes, both doses produced a quick and highly significant decrease in fasting glucose as well as a decrease in insulin levels, indicating improvement of insulin sensitivity. Glycated hemoglobin was reduced by more than 0.5% in the overall diabetic population. So in terms of safety, lanifibranor continues to show a favorable safety and tolerability profile, which is consistent with observations from previous clinical trials. I'll start with the serious adverse events. There were a total of 13 serious adverse events. We had 3 SAEs in the placebo group, 2 of which were one mild cardiac failure and one moderate urticaria. There were 3 SAEs in the 800-milligram group and 7 in the 1,200-milligram group, and we will see the details of those serious adverse events in the next slide. These adverse events, AEs, were generally mild to moderate in severity. Yes. Thank you. Back to the slide. There were 5 regulated AEs leading to drug withdrawal, 2 in the placebo group; one in the 800-milligram group, which was moderate diarrhea; and 2 in the 1,200-milligram group with one mild cardiac failure and one mild diarrhea. Regarding body weight, and consistent with known insulin sanitizing pharmacology, a modest weight increase was observed with a mean of 2.4 kilos in the 800-milligram group and 2.7 kilos in the 1,200-milligram group. And regarding edema, we had a total of 14, 1-4, patients that reported peripheral edema, 2 in the placebo group, 5 in the 800-milligram dose group and 7 in the 1,200-milligram group. All of them except one were of mild intensity, all were transient. There were only 2 patients with treatment-related peripheral edema in each lanifibranor treatment arm, and that there was no treatment discontinuation due to edema. Now regarding the serious adverse events I was talking about on the previous slide. So as I mentioned, there were 13, 1-3, serious adverse events, 3 in the placebo arm, 3 in the 800-milligram per day arm and 7 in 1,200-milligram per day arm. After excluding biopsy-related serious adverse events, there were 3 SAs in the placebo group which was one risk factor, one cardiac failure, one urticaria. There were 2 SAEs in the 800-milligram group, one pancreatitis, one undifferentiated connective tissue disease. And there were 4 SAEs in the 1,200-milligram per day dose group, one angina and stable, one gastroenteritis, one pyelonephritis and one foot operation. So to put this efficacy data in perspective, we can see here that compared to other Phase IIb data, lanifibranor is the only product that induced NASH resolution with no worsening of fibrosis in more than 40% of patients, and this after only 6 months of treatment. And likewise, regarding fibrosis, lanifibranor is the only product that induced fibrosis improvement with no worsening of NASH in more than 40% of patients. So in conclusion. Lanifibranor met the primary endpoint with a statistically significant reduction after 6 months of treatment of the SAF score combining hepatocellular ballooning and inflammation, with no worsening of fibrosis in both populations. Lanifibranor also met the key secondary endpoints in the ITT and per protocol population. Lanifibranor is the only product that has shown a statistically significant effect on the 2 key endpoints of NASH resolution with no worsening of fibrosis and improvement of fibrosis with no worsening of NASH. The product continues to show a favorable safety and tolerability profile. And this result is really motivating us to move forward and enter into a pivotal Phase III trial. And the next steps are finalization of Phase III synopsis and protocol, which is ongoing. We plan to meet with FDA for an end of Phase II meeting and EMA for a scientific advice meeting in Q4 this year. We are going to finalize also the other Phase IIa study, which is running at University of Florida with Professor Cusi testing the effect of lanifibranor 800 milligram in NAFLD and diabetic patients on insulin resistance, liver-related insulin sensitivity. And of course, the plan is to launch pivotal Phase III trial in NASH. And before turning to the Q&A session, I would like to extend my deepest thank you to patients, caregivers, investigators, advisers and our team for their continued confidence, relentless dedication and strong commitment throughout this trial. Thank you very much.

Thank you, Pierre. And so operator, now we can move to the Q&A session.

[Operator Instructions] And your first question today comes from the line of Lucy Codrington.

I've got a couple, please. So firstly, with the weight gain, when did that occur? Does it kind of occur consistently or is it a very near-term effect? And would that be expected to continue to increase with longer treatment duration? Secondly, were there any drug interruptions required related to the adverse events? And then I know you were going to look at the performance between diabetics and nondiabetics. Is there any difference in the response rates between those populations, if that data is available yet?

Okay. Lucy, thanks for your question. So 3 questions. Pierre, do you want to tackle those?

Yes, the last one actually. So yes, we analyzed the data according to diabetic status, at least from the key parameters that I presented, and there is no difference. Actually, the drug is active, as efficacious in -- on the parameters in diabetic patients as in nondiabetic patients. Regarding body weight gain, when this -- we don't have the -- as of today, I don't have the precise answer to your question about when did this body weight gain occur. I think that it goes along with anyhow, improvement of insulin sensitivity. You know that this is -- this body weight gain is just due to the fact that you restore the activity of the adipocyte, which under insulin resistance status, is dysfunctional and leading to the accumulation of fat ectopically into, for example, into the liver or into a set of muscles or into the heart. And when you correct this, when you improve insulin sensitivity in the adipocyte and the adipocyte is now capable to store fat again so, the body weight gain is, I would say, good, healthy weight gain goes along with improvement of insulin sensitivity. So I think that 2 together would probably be improving at the same pace, I would say. Then your question about drug interaction potentially explaining adverse events. So we have quite a compelling package of drug-drug interaction studies performed with lanifibranor, and there is absolutely no risk of [ DDI ]. So I don't think that the serious adverse events that we are -- that we have in the study, are linked to drug-drug interaction.

It was good to know that the drug-drug interactions -- I was -- I'm sorry, I wasn't clear. I meant drug interruptions. So were there any patients that had to have their dosing paused, for instance, the patients with the peripheral edema or did they continue the treatment with no pauses?

No. No, they all continued the treatment. There was no dropouts due to peripheral edema.

Your next question comes from the line of Lenny Van Steenhuyse.

I was actually looking at the [ F2/F3 ] population, showed an interesting difference in placebo rates while maintaining high responder rates in the treatment arm for NASH resolution. So I was wondering is this effect restricted to NASH resolution or do you also see differences in fibrosis improvement responses along patients in different stages of fibrosis?

So yes, we've made the analyzed on the F2 and F3 subgroup versus the overall population for NASH resolution. And as you have seen, we have observed a very significant improvement of NASH resolution without worsening of fibrosis in this F2 population, which is the population to be used in Phase III. And we have done that also for the fibrosis improvement without worsening of NASH. And the results in terms of effect size is identical to what we see in the global population, okay. So we have not performed the statistical analysis yet, but the size of the effect regarding improvement of fibrosis and no worsening of NASH is identical in the F2, F3 population compared to the overall population, the data you've seen.

Yes, okay. That's clear. Now you have a compound that's, of course, partly an insulin sensitizer so there's a rationale here for looking at more early stages of disease. On the other hand, you're showing impressive data on fibrosis so that may raise questions on the drug's potential perhaps in an F4 setting. So based on this kind of data set, what kind of opportunities are you looking for, for lanifibranor beyond the F2, F3 segment? Is that something that you're looking into at this point?

Yes, that's a great question, actually. So you've seen the size of the effect of the drug on fibrosis improvement in this population. And as I said, the effect is the same in the F2 and F3 population, indicating that there is really a noticeable [ mechanism ] of action that actually kicks in quite rapidly because this is the first time we see such an effect after 6 months. So yes, there is -- if you -- if we add to that a number of preclinical studies we have recently completed and communicated at ASLD last year, where we've seen a major improvement of not only liver fibrosis but also portal pressure in models of cirrhosis, there is certainly an opportunity also for lanifibranor in the F4 population.

Okay. Perhaps a very last question from my end. Looking at patient recruitment, this was, of course, mainly skewed towards Europe in this case. I was wondering whether or not there are particular differences in European and U.S. NASH populations. Are these more or less heterogeneous? Or are they really indistinguishable. Just to have a feel if you have a Phase III population with perhaps more U.S. patients, could that make a difference? Or is this really a very broad effect?

Yes, no. So we have not performed fully the analysis. From what I have seen roughly, there is no major difference, no difference between Europe and U.S. responders. But again, this is part of the work that we are now doing in term of data crunching to be able to design the most appropriate Phase III study in term of endpoints, of dose, of population et cetera.

Maybe Sven and Manal, as you have had a patient in this trial, do you want to answer the question if there are, according to you, significant differences among patients between U.S. and Europe?

Well, I think that we know, of course, that some of the problems, namely of obesity is, perhaps a little bit more severe in the U.S. than in Europe, but I think that roughly the phenotypes of the patients are quite comparable.

I agree completely. This is Manal Abdelmalek. And while the underlying risk factors related to lifestyle and dietary drivers for disease may vary across different regions of the world, the primary inclusion criteria of definite NASH with fibrosis and the treatment effect noted is not determined by the risk factors as it is in effect a measurement of therapeutic effect. So we will learn more about regional, global and lifestyle variants as we embark on larger Phase III global trials.

Your next question comes from the line of Patrick Dolezal.

A few for me. So first off, do you anticipate deeper improvements on NASH disease or fibrosis with a longer dosing period? And what duration study do you imagine for the subsequent Phase III? Is that likely to be similar in length or more likely to be in the 12-month range?

Good question. I'll give it a try first, and then maybe, Pierre, Manal and Sven can complete. In the Phase III, I would say, it's a bit too early. We really need to deep dive in the data. I think one thing we have learned from this trial is that, I think it was a well-designed trial because we gave it a lot of thought, a lot of work and we look to do exactly the same for the Phase III. I think we have an amazing set of data with this NATIVE trial and that will help us design the best trial. What is the dose, first? What is the duration? We've seen a significant effect on both endpoints at 6 months. Can we build on that? What is the endpoint we should put forward and discuss? And that's why the team is already working on preparing the discussion with FDA that is planned in the fall and the scientific advice with EMA. [ Do no -- ] Pierre, Manal, Sven, anything else to add?

Fully agree.

Fully agree.

Yes. Me too.

Great. And on the safety side, can you provide a little bit of detail on the AE of mild cardiac failure and how that's defined clinically. And considering there was one case in the placebo arm and one case in the lanifibranor -- in one of the lanifibranor arms, did the presentation of this AE differ between the 2 patients? And I guess, more specifically, did it occur in the lanifibranor arm concurrently with other signs of congestive heart failure, such as edema, rapid weight gain, ISMEA, et cetera?

So we -- today, what we have, like mentioned, one case of heart failure in the placebo group, one case in the 1,200-milligram dose group and one case in the 800-milligram dose group. So we need to go back to these details and particularly look at the history of the patients. And today, I cannot give you the details that you are requesting simply because we have not done yet the full analysis of this safety event.

Okay. And I guess, for both of our physicians on the line. Do you currently treat NASH patients with pioglitazone? Why or why not? And how is edema, weight gain, risk of CHF tolerated in this patient population? And then I guess, how do you think about the prospects of using lani versus pio? Obviously, there's some mechanistic differences but would love to hear your perspective on that.

Well, from my side, in Belgium, there are strict limitations in the reimbursement of pioglitazone. So we have a few patients that are treated with pioglitazone but that's always in close collaboration with the diabetologists because they are allowed to prescribe it and I, as a hepatologist, I am not. Now overall, the side effects are rather mild. Of course, we should be careful in comparing populations because we know that -- I think the few cases we have observed in NATIVE with some fluid retention are far beyond what is globally reported for pioglitazone. But of course, we should take into account that you cannot compare straight away the population. So that's something to take into account. But globally, I think this compound appears to be quite safe compared to pioglitazone, and as you pointed out yourself, there are mechanistic differences that might help us to explain this difference. So I think that with the results we have here in the metabolic profile, I think this is a major -- potential major step forward. And I, of course -- we feel quite confident to use these kind of drugs for patients with NASH, given their particular phenotype.

In response to your excellent question, here in the U.S., I have utilized pioglitazone in the treatment of NASH for patients with diabetes. In other words, patients who otherwise had an indication for insulin sensitizing agent for the treatment of their diabetes and such practices in keeping with the guidances of the American Association for the Study of Liver Disease. With that being said, however, the utilization of pioglitazone in general and even in my own clinical practice for use in patients with NASH is steadily dropping in light of the broader landscape and armamentarium of medications that is currently available for the treatment of diabetes. So as opposed to 8 years ago or 5 years ago, I utilized pioglitazone in my clinical practice very rarely in the present day.

The next question comes from the line of [ Derrick Arquilla ].

Congrats, Frederic and Pierre, on the data. Just a couple of questions from us. And just first on edema. It looked like there was a very low rate in the study. That was great. So just wanted to get a sense of whether you'd expect any changes with a longer study of 12 to 18 months potentially in Phase III? And if not, what mechanistically gives you confidence that you might not see an increased rate of edema in a longer study? And then second question on a Phase III study design, if you can give us any color in terms of whether or not you bring in multiple doses. I mean, it looks like the 800 mg is definitely active. And just wondering, again, if the longer duration of treatment, 12 to 18 months, you can also see some good results there?

Thank you, Derrick. So on the edema, from what I remember, they tend to appear at the beginning of the study. So what would you say...

Yes, yes, yes. Certainly, yes, we have seen this rather early in the study. I think that from the experience we had also in the SSc trial lasting 12 months, the appearance of edema started quite soon. There was no increase between the 6 and the 12 months. You can also refer to studies with Pio, for example, where you can see that the frequency of edema is occurring rather soon. And to put things on perspective, nevertheless, I would like to restate the fact that the edema that we see in our trial in terms of frequency is much lower to what is reported with Pio. For example, if you look at Pio frequency of edema after 6 months of treatment, if you look at the Pio 45 milligram, you can have, according to different studies, between 14% and 23% of patients showing edema. Whereas in our 6 months trial, we have only 6% to 8% of patients. And this, despite the fact that, in contrast to Pio, we have a very strong effect on fibrosis and on NASH resolution. And also mention the fact that we -- in the study, we didn't have any dropout due to edema. So I think that we are much, much better than what has been reported for by Pio on this edema. And regarding your question on whether a longer duration of treatment with 800 could reveal an efficacy close to 1,200, this is a possibility, I think so. It's interesting to see that on the metabolic parameters, actually the 2 doses are actually efficacious. So it could be that an effect of the 800-milligram dose in fibrosis, for example, could occur between 6 and 12 months to a level that what has been achieved with the 1,200 milligram. This is a possibility. We will look at that more closely by digging into the data.

Your next question comes from the line of Ed Arce.

Congrats on this fantastic data set. Really, a landmark with the first results, even at 6 months, to show positive results on both FDA-acceptable endpoints. Around that, first question is, you saw some pretty significant results in both resolution of NASH and fibrosis. And I noted that, that was in spite of some fairly high placebo rates, 34% placebo rate in the primary endpoint. And even on the improvement of fibrosis, that was a 29% placebo rate. So the first question is around that, and your comments around it seems fairly high compared to other studies. Also, I know you've kind of touched upon this, but have you decided on a particular dose to move forward with Phase III? And could there be 2 doses tested in a pivotal study. And then one last question on the phase -- or excuse me, the fibrosis 2, fibrosis 3 baseline for the 76% of patients that had that, you noted that, that was the slight higher in the 2 lani arms versus the placebo. I'm wondering if that could have any -- had any potential impact on the fibrosis improvement that you saw.

On the placebo.

Yes. So okay, so the placebo effect, yes. Well, on the primary endpoint, this was the first study, really, where measuring the 2 points of reduction of activity of the SAF score. So yes, so it was unprecedented. So we had the placebo rate that we had, and that's the [ point ]. The NASH resolution, the placebo was within the range, I would say, of what has been reported for the semaglutide or even the elafibranor Phase III trial, 15%, 17%. The fibrosis improvement, placebo effect, was in the range of what is reported, 24%. But whatever the placebo size -- actually lanifibranor, we've managed to produce a specifically significant effect relatively to placebo, which again, is a testimony to the good efficacy of the drug.