Poxel S.A. (POXEL) Earnings Call Transcript & Summary

Ladies and gentlemen, welcome to Poxel PXL770 Phase IIa Trial Results Announcement. I now hand over the call to Mrs. Jonae Barnes, Senior Vice President, Investor Relations, Corporate Communications and Public Relations. Madam, please go ahead.

Good afternoon, everyone, and welcome to our conference call today to review the results from the PXL770 Phase IIa trial. I'm Jonae Barnes, Senior Vice President, Investor Relations, Corporate Communications and Public Relations. With me on the call, Thomas Kuhn, CEO; Pascale Fouqueray, Executive Vice President, Clinical Development and Regulatory Affairs; David Moller, Executive Vice President and Chief Scientific Officer; and Julie Dubourg, Medical Director, Clinical Development. They will all be available for Q&A after our call is complete. Earlier today, we issued a press release for the PXL770 Phase IIa results. For those of you who don't have a copy, you can access it in the Investor Relations section of our website, poxelpharma.com. As part of this presentation, we will be using slides for this discussion. You can access them on Poxel's website in the Investors, Company Information section under the name of PXL770 Phase IIa results. Before we begin, I would like to read our forward-looking statement. Some of the statements contained in this presentation constitute forward-looking statements, which are statements that are not historical. These statements are based on the company's current strategy, plans, objectives, assumptions, estimates and projections. Investors should therefore not place undue reliance on those statements. The company makes no representation, warranty or prediction that the results anticipated by such forward-looking statements will be achieved. And such forward-looking statements represent, in each case, only one of many possible scenarios that should not be viewed as the most likely or standard scenario. Forward-looking statements speak only as of the date that they are made, and the company does not undertake to update any forward-looking statements in light of new information or future events. In the context of COVID-19 outbreak, which was declared a pandemic by the World Health Organization on March 12, 2020, the company is regularly reviewing the impact of the outbreak on its business. As of the date of this conference call and based on publicly available information, the company has not identified the occurrence of any material negative event on its business due to the COVID-19 pandemic that remains unresolved. For today's call, Thomas Kuhn, our CEO, will provide opening remarks. Dr. David Moller, our CSO, will review the Phase IIa results. We will then move to the question-and-answer portion of the call. I will now hand the call over to Thomas.

Thank you, Jonae. Hello, everyone, and thank you for joining us today. So today, we achieved a very important milestone when we announced the encouraging results of the PXL770 Phase IIa trial for the treatment of NASH. Before I hand the call over to David, I would like to take a few minutes to provide a brief overview of the company for those of you who are not familiar with Poxel. Beginning with Slide 3. Poxel is focused on developing innovative drugs for the treatment of metabolic disease, including type 2 diabetes and NASH. We have a strong foundation in this therapeutic area dating back to the company's inception when it was spun out of Merck Serono. We have a mid- to late-stage pipeline as well as earlier-stage programs from our direct AMP kinase and deuterated TZD platforms. Please go to Slide 4. Our pipeline includes our most advanced drug candidate, Imeglimin, which is globally partnered for the treatment of type 2 diabetes. Through our partnerships with Dainippon Sumitomo Pharma, a new drug application is currently under review in Japan. Sumitomo is currently targeting a product launch in their fiscal year 2021, which is April to March. They also have developments and commercialization rights for China and 11 other Asian countries. Metavant, our partner for all other countries not covered in the Sumitomo agreement, has ongoing dialogue with the FDA to finalize Imeglimin Phase III plan for the U.S. and Europe. The Phase IIa results for PXL770 for the treatment of NASH that we announced today support the potential to advance the development of this molecule into a Phase IIb biopsy-driven trial. For our second program in NASH, PXL065, we initiated DESTINY 1 early September. It's a streamlined, dose-ranging Phase II trial for the treatment of NASH. As a reminder, PXL065 is a new chemical entity derived from pioglitazone, specifically deuterium-stabilized single stereoisomer, which has been shown to retain NASH efficacy without triggering PPAR gamma-related side effects in preclinical studies. This mechanism involves modulating non-genomic pathways such as the mitochondrial pyruvate carrier. We also have an agreement with ENYO Pharma for FXR agonist, which is currently in a Phase IIa program. For our earlier-stage preclinical programs, we are advancing work related to our direct AMP kinase and deuterated TZD platforms. Today's results give us further conviction related to AMPK and its utility for several metabolic disease. I will now hand the call over to David.

Thank you, Thomas. Good afternoon, everyone. Before diving into the results, I'd like to first briefly review the AMPK mechanism as well as earlier results for PXL770 that led to the positive outcome reported this morning. Let's turn to Slide 5. AMPK is a master regulator of several important metabolic pathways, including pathways also linked to the control of inflammation. Importantly, we know the conditions associated with overnutrition are also associated with reduced AMPK activity in key tissues like liver in the case of NASH. Thus, direct activation of AMPK is an approach to restoring normal AMPK tone and a compelling and novel target for NASH and a range of other chronic and rare metabolic diseases. Turning to Slide 6. The downstream effects of AMPK activation lead to PXL770's potential ability to address each of the key hallmarks of NASH, which we define as steatosis, inflammation, ballooning and lastly fibrosis. Unlike some other mechanisms which act primarily to reduce steatosis, several lines of evidence, including our internal data, show that AMPK activation can independently affect inflammation, fibrogenesis and hepatocellular death. In addition, root causes, including insulin resistance, can be ameliorated. Here, you can see how our data in rodent models and human cells are beginning to align with and translate into emerging clinical results in the target population. Significant effects from the Phase IIa study that we report today are highlighted in red in the right hand column. I'll review these findings in subsequent slides. Overall, the profile is very encouraging for the future, specifically anticipating the potential for further efficacy when histology data would be available. Turning to Slide 7. More specific results from multiple preclinical studies are depicted here. These studies have shown consistent and promising results with respect to each of the core elements of NASH. I'd also like to note that PXL770 may also be differentiated from other compounds in development for NASH since AMPK activation also has the potential to treat major comorbidities by specifically targeting cardio-metabolic risk factors such as hyperglycemia and insulin resistance. Moving now to Slide 8. Our preclinical data and evidence from the literature also highlights the potential of PXL770 to show even greater additive or synergistic benefits to treat the root causes of NASH when combined with other agents in development. We've studied several combinations in a classic rodent NASH model. These include PXL770 with semaglutide, a GLP-1 receptor agonist; obeticholic acid, an FXR agonist; and MGL-3196, a thyroid hormone receptor agonist. As shown in the example on this slide, co-administration of PXL770 with semaglutide resulted in further improvements in several key disease-related parameters, including hepatic steatosis and plasma ALT. Additive benefits were also observed with other agents studied to date. These studies highlight the potential for complementarity of use with a broad range of other classes in NASH. On Slide 9, we have summarized here the cumulative prior clinical experience with PXL770. In over 130 healthy subjects, we observed suitable PK properties and good safety and tolerability. In the more recent 4-week PK/PD placebo-controlled study in 16 likely NASH patients, we met our endpoint and objectives. PXL770 demonstrated a PK profile that was similar to the one observed in healthy volunteers and was also observed to be safe and well tolerated. Notably, PXL770 also induced a statistically significant suppression of de novo lipogenesis, which is linked to AMPK and known to be an important contributor to the accumulation of liver fat or steatosis that underlies the pathophysiology of NASH. Despite this also being a nondiabetic population, we also noted a greater effect on de novo lipogenesis in patients with more severe metabolic dysfunction. A statistically significant improvement in glucose tolerance during a glucose challenge test was also observed in this study. Moreover, we observed additional significant improvements in several indices of insulin sensitivity. This trial then confirmed the target and pathway engagement with PXL770 and its potential utility in other metabolic diseases. These results also represent the first human experience with a specific AMPK activator, and they support our hypothesis that this novel mechanism could be safely harnessed and have a beneficial role in controlling key pathways that lead to liver injury in NASH or utility in other indications. Turning now to Slide 10. STAMP-NAFLD, the trial's name, is a Phase IIa 12-week randomized, placebo-controlled, parallel group trial which enrolled 120 presumed NASH patients with or without type 2 diabetes. The trial was conducted at several sites in the U.S. The primary enrollment criteria was evidence of hepatic steatosis or NAFLD based on the percent liver fat mass greater than 10% as measured by MRI-PDFF. The trial evaluated 3 different dosing regimens of PXL770 versus placebo. This included 4 groups of roughly equal size. Patients who received PXL770 at 250 milligrams once daily, 250 milligrams twice daily, 500 milligrams once daily or patients who received placebo. The primary endpoint of the trial was the relative change in liver fat content based on MRI-PDFF, an imaging-based biomarker that allows for quantitation of steatosis in the entire liver. Other secondary measures on multiple disease-related parameters helped to confirm efficacy in NASH as well as to shed light on other related indications. Slide 11 depicts baseline patient characteristics. Patients were overweight or obese and 41% to 47% of each group had type 2 diabetes. Patients with diabetes were well controlled and treated with monotherapy or diet as reflected in normal mean fasting glucose levels and mean hemoglobin A1c, HbA1c values for each group, which averaged around 6% or below the 6.5% threshold for diabetes. Liver fat content was 16% to 22% at baseline and generally balanced across the groups. Importantly, mean ALT levels at baseline were also nearly normal, either at or below the upper limit of normal, which is 41 international units per liter. Moving now to the Phase IIa results on Slide 12. MRI-PDFF was used to measure effects on liver fat mass. PXL770 was observed to produce dose responsive effects with a statistically significant relative mean decrease of 18%, a p-value of 0036 from baseline at 12 weeks in the 500-milligram QD group and 14% in the 250-milligram BID group. Both placebo and 250 milligrams QD of PXL770 produced negligible effects. The proportion of patients who achieved greater than 30% relative fat mass reductions in the 500-milligram group was also substantially greater than in the placebo group. In addition, some patients were highly responsive and were observed to achieve up to 85% relative reductions in liver fat mass. As shown on Slide 13, a more pronounced effective PXL770 on liver fat mass was observed in the 41% to 47% of patients who also had a diagnosis of type 2 diabetes. In these patients, a dose-responsive decrease in relative liver fat content was observed with an effect size averaging 27% in the 500-milligram QD group. This key subgroup represents approximately 47% of all patients with NASH and is known to be at higher risk of progression to more advanced stages of fibrosis and end-stage liver disease. It's also known that endogenous AMPK tone is lower in tissues from animals and humans with type 2 diabetes. Slide 14 displays the effects on ALT levels. Although baseline ALT levels in each group were within the high normal range, we did observe a significant 6.3 IU per liter change in the mean absolute ALT levels at the top dose of 500 milligrams QD. In addition, the pattern of changes in ALT suggest that we have not reached a plateau at 12 weeks. We are performing further analysis of this data to assess these findings and to examine effects on other related parameters. Slide 15 shows effects on hemoglobin A1c in each group, including both patients with and without diabetes. Here, we observed a dose responsive and significant decrease. Not shown here was a parallel strong trend towards reduced mean fasting glucose that was also observed. As a reminder, baseline hemoglobin A1c values for each group were within the normal nondiabetic range, consistent with the presence of mild, well-controlled diabetes in those patients with a diagnosis. The placebo-adjusted effect of negative 0.3% at 500 milligrams QD is therefore clinically meaningful. By comparison, the landmark diabetes prevention program also studied nondiabetic glucose-intolerant patients. After 6 months of treatment with metformin, a mean placebo-adjusted decrease in HbA1c of approximately 0.15% was observed in 850 patients. Slide 16 summarizes the safety findings from the STAMP-NAFLD trial. Treatment with PXL770 was observed to be very well tolerated in all 3 treatment groups. The number of patients in each group who experienced treatment-emergent adverse events was very similar to the number in the placebo group. The most common AEs included mild to moderate GI symptoms, infections and headaches. There were 3 patients with serious adverse events. Two had AEs classified as unrelated to study drug. In one patient, increases in ALT were classified as a serious AE. In this specific patient, the increase began before randomization and was persistent with substantial elevations in ALT that were also observed several weeks after discontinuation of therapy. This case is, therefore, unlikely to be related to treatment with PXL770. Overall, the safety profile is consistent with that observed in greater than 140 subjects studied in the Phase I program, both health subjects and NAFLD patients. Let's turn now to Slide 17. In summary, results from the STAMP-NAFLD Phase IIa trial lead us to conclude that PXL770 has a good safety and tolerability profile, as I described on the preceding slide. This oral first-in-class agent was observed to produce significant efficacy with respect to reduction of liver fat mass and to improve other key measures, which included reductions in ALT and hemoglobin A1c. The greater response observed in patients with type 2 diabetes is noteworthy since this is a high-risk subpopulation accounting for nearly half of all patients with NASH. Along with preclinical data showing independent effects on inflammation and fibrogenesis, including in human cells, the results of this study support the potential to advance PXL770 for the treatment of NASH. As a key next step, we are considering a Phase IIb biopsy-based NASH trial, which could confirm safety and further evaluate important histologic endpoints, including inflammation and fibrosis, which were not assessed in this Phase IIa trial, and to confirm its potential in the high-risk subpopulation with diabetes. Based on the clinical results obtained to date, its distinct mechanism and preclinical combination results, we believe that either monotherapy or co-administration of PXL770 with other agents could provide broad treatment for patients with NASH. The results also support potential for PXL770 and the AMPK platform in other chronic and rare metabolic diseases. Finally, Slide 18, I'd like to briefly mention our efforts to examine other indications and potentially leverage other molecules from the platform. The results announced today for PXL770 pave the way for potential pipeline expansion. We envision that PXL770 or other AMPK activators could be harnessed to treat a range of disorders, including rare metabolic diseases such as adrenoleukodystrophy or mitochondrial disorders, rare and common forms of chronic kidney disease as well as additional endocrine disorders. We have strong preclinical efficacy data for several of these conditions, including diabetic nephropathy, adrenoleukodystrophy and polycystic kidney disease. Thanks for your attention and interest in our results. Operator, you may now open the line for questions.

[Operator Instructions] The first question comes from Jason Butler from JMP Securities.

First off, based on the data you have today and the prior Phase I data and preclinical data, is there any rationale to consider higher doses or different administration frequency? And then you talked about the fact that some patients had really good responders, up to 85%. Beyond the type 2 diabetes population, were there any other attributes that describe this patient group? And I guess, can you give us a sense of how many patients were in that higher responder group?

Yes. Maybe I can start, David, and then please add anything there. So with regards to the first question, I think you're right. We factor the analysis that we do, so we will have to look at various options there. One of the key goal we had from this trial was really to assess whether we want to go on a twice daily or on a once-daily regimen. And as you've seen from the data and as you've heard from David, the once-daily regimen seems to bring really a more significant efficacy with a good safety there, which is good. We'll then now have really to look at the pharmacokinetic profile. We'll then have to correlate that with the pharmacodynamic profile there. And so I mean, we will need to wait for the remaining data to consider this. But all options are really currently on the table there. David, then, do you want to take the second question on this highly responsive patients there?

Yes, I can start and maybe Pascale or Julie would want to add. Yes. It's an interesting question. Obviously, we just received this data recently. We're analyzing it as we speak, and we'll potentially have some additional parameters to look at that may or may not be able to help us identify the more responsive patients. But we do think that what we have seen so far, and it's consistent with what I mentioned from the PK/PD trial, is that patients with more severe metabolic dysfunction, in this case, patients with type 2 diabetes, seem to be clearly more responsive in general. And that to us is kind of interesting and encouraging results. Anything to add, Pascale, Julie?

No, I think what you say, David, is what we have as data today. But well, we will wait to have the full data to look deeper into difference between type 2 diabetes and nondiabetic patients.

Okay. That's really helpful. And then just 1 more for me. As you think about the Phase IIb trial design, obviously, the type 2 diabetes population is a relevant consideration. But other things that you're thinking about? And maybe give us the context given your Phase IIb trial design for the 065 program?

Pascale, do you want to answer this one?

I'm sorry, I didn't get the question.

It's on the Phase IIb. So maybe I can rephrase and start to answer. So Jason wanted to have more detail about what we're planning for the -- what we're planning really for the Phase IIb. Is it going to be in diabetic patients there? Or any other detail that basically we believe could be important for the Phase IIb.

I think this study, this proof-of-concept study, was really important for us to get a sense of the size of the effect of PXL770 and to confirm the previous data that we have observed in the PK/PD related to the better effect even in more severe insulin-resistant patients, which seems to be the case here. I think it also gave us a lot of information of the dose regimen and the dose with BID or once-daily and dose range. So I think we really have to first get the complete view of all the data before moving forward with the Phase IIb design. But well, honestly, the design might be very classical with biopsy and -- but I think we need to get all the data before...

Maybe you...

Getting too much on that. Yes.

Maybe one point of clarification, Jason. I think in contrast, as you know, for PXL065, we can benefit from the 505(b)(2) pathway coming from pioglitazone. So that's why we designed this streamline Phase II program, still with the aim to assess the efficacy, however, with the primary endpoint to be the liver fat mass reduction and to be correlated as a secondary endpoint with the histology data. Here, for 770, as you have understood, we believe that there is much more than the liver fat mass that we've seen here during this Phase IIa, given that the drug hit really the target. And it's been demonstrated that by activating AMP kinase, we could also benefit from additional efficacy independently from the metabolic benefit. We can benefit from inflammation benefit and fibrogenesis benefit. And so that's why I believe that histology data will be really pivotal in this Phase IIb. And as you have understood as well, maybe treating longer, could be also very important to have the long-term efficacy data. So as Pascale said, it would be more like a classical Phase IIb in biopsy-proven NASH patients treated for 9 months, 12 months or more. And then, of course, around this, there could be a multiple option. But really, this is going to be the core component of this trial.

And next question comes from Lucy Codrington from Jefferies.

I've got a couple. Firstly, am I right in understanding that these were per protocol responses? And if so, did the ITT population differ at all? In terms of the safety, were there any discontinuations due to the treatment-emergent adverse events? And then lastly, I do appreciate you still need to analyze the full data set and think about the Phase IIb, but is the plan to include a broad NASH population initially in the Phase IIb with some subgroup analysis or could you actually consider focusing on subpopulations sooner rather than later? And then actually, I have one more. I just noticed that there was quite a much higher proportion of female patients in the 2 higher dose groups. I wonder if there are any implications as a result of that.

Thank you very much, Lucy. Pascale, maybe you want to take these questions, starting with the per protocol versus the ITT population there?

Yes. This is the per protocol population, but we also have data in the ITT population, which are very close to the data observed in per protocol. We believe that this per protocol was more accurately representing the data for this study.

For the second question, Pascale, on the safety, any discontinuation due to adverse event?

Yes. There was some discontinuation, but 1 or 2 discontinuation per group and without any specific, I would say, reason in terms of safety.

For the third question, I think, Lucy, yes, I think we definitely have different option on the table because this broader benefit that we've seen on type 2 diabetes is not just coming from this trial but also from the PK/PD trial. And we know, and David described this well, we can expect really a broad benefit targeting NAFL in diabetic patient. So do we want to really solely focus on this population for the Phase IIb or the rest of development or include a significant number of this? I mean, that remains to be fully analyzed there. But certainly, it would be a key area of focus. And maybe I don't know whether David or Pascale, you want to elaborate on this?

Yes. I think it's a very interesting and attractive concept, and it's something that we're actively discussing with our expert advisers and also continuing to look more closely at the data. So we don't have any definitive answer, but that's definitely on the table.

And so for the last question...

And last question would be for females, the imbalance of males and females, maybe Pascale, you could?

Pascale, on this last question on the imbalance between -- on the 2 top dose for female and male there?

Yes. We don't believe that this made an impact on the result because, honestly, in terms of PK, there is no evidence of any difference between male and female. So there is no currently reason to believe that this may influence the result.

Next question comes from Andy Hsieh from William Blair.

Congratulations on the encouraging results and finishing the trial. So my question has to do with the trend line. So I think, Thomas, you mentioned a little bit about the ALT trend, which is that it has not plateaued after 12 weeks. I'm just wondering if the same observations could be said about the liver fat reduction and also potentially glucose sensitivity, HbA1c, just curious if you can comment on that.

Yes. Thank you, Andy, for this question. Pascale, do you want to answer this one?

So yes, well, we have only 1 MRI in this study as it was a very short duration study of 12 weeks. So we don't know whether all the effect has been captured after 12 weeks or if we can expect more deeper decrease with time. What we have observed is that the decrease in ALT is not something immediate, but it's a little bit delayed. So this may also get deeper with a longer duration of treatment. As far as the glucose control and A1c, well, the patients were at very low A1c baseline to start with. So well, I think that the decrease that was observed in this study for these kind of patients and this glucose control, I think it's already a very meaningful decrease.

Maybe I could just add to Pascale, a couple of comments on these questions, Andy. So as you know, HbA1c is an integrated measure of glucose. It takes time, no matter what the mechanism, to reach a steady state. And with most glucose lowering agents, you're nearly there but not all the way there by 12 weeks, and it usually takes 24 weeks to reach a steady state. So we would expect potentially some further effect on glucose. But as Pascale said and as I said earlier, what's remarkable here is that the baseline HbA1c was really not in the diabetic range, and yet we still see significant reductions. So that's very encouraging to us. And then as far as the ALT is concerned, again, as Pascale said it and I mentioned it, we don't think we've hit a plateau there yet either. And many of the other NASH studies similarly show that, in some cases, depending on the mechanism, you don't reach a steady state until you get out to 24 weeks or beyond in ALT.

Got it. Got it. Yes. So I also have a PK question, if you don't mind. So just curious if there's kind of an explanation in terms of the BID dosing and the QD dosing. So 250 BID, 500 QD, there seems to be a difference in terms of potency for those 2 dosing schemes. Is there some sort of scientific explanation or PK explanation for that?