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

Good morning, ladies and gentlemen, and welcome to Akero's 36-Week Analysis of Akero's 96-Week Phase IIb SYMMETRY Study. As a reminder, today's conference call is being recorded. I would like to turn the call over to Bill White, Akero's Chief Financial Officer and Head of Corporate Development. You may begin.

Before we begin, I ask you to please to read the disclaimers presented on Slide 2. And we'd like to remind you that various statements that we make during this call will include forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995, including expressed or implied statements regarding implications of this analysis, Phase IIb SYMMETRY Study of efruxifermin or EFX, our future plans, including our plans around the development of EFX, prospects and strategy, financial goals and guidance, product attributes on pipeline, drivers of growth, and other statements that are not historical facts. Such forward-looking statements are not guarantees of future performance, and therefore, you should not put undue reliance upon them. Management's assumptions, expectations and opinions reflected in these forward-looking statements are subject to risks and uncertainties that may cause actual results and/or performance to differ materially from any future results, performance or achievements discussed in or implied by such forward-looking statements, and the company can give no assurance that they will prove to be correct, and will not provide any further guidance or updates on our performance during the quarter, unless we do so in a public forum. Please refer to the press release we issued this morning, and risk factors included in the company's filings with the Securities and Exchange Commission for a discussion of important factors that may cause actual events or results to differ materially from those contained in our forward-looking statements. Prior to this call, we issued a press release regarding data from our Phase IIb SYMMETRY Study, which is available on our website at akerotx.com under Press Releases in the Investor Relations section. The presentation will walk through on today's call will be available immediately afterwards under Events and Presentations in the Investor Relations section of our website. I will now turn the call over to Andrew Cheng, Akero's President and CEO.

Thank you, Will. Today, I am pleased to introduce new data from our Phase IIb SYMMETRY Study of EFX, an Fc-FGF21 fusion protein, we are developing as a treatment for NASH, including patients with liver cirrhosis due to NASH. Although obviously disappointed that the week 36 primary endpoint of fibrosis improvement was missed, we are encouraged by the totality of today's data and believe it supports moving EFX forward into Phase III in the cirrhotic population. In the trial, a trend towards fibrosis improvement without worsening of NASH was observed, including 4% of patients in each of the EFX treated groups who experienced a 3- or 2-stage improvement from F4 to F1 or 2 fibrosis. The trial demonstrated statistically significant results for NASH resolution, which we believe is the first such report and statistically significant results on many other noninvasive markers of liver fibrosis. These results are consistent with previously reported EFX studies in patients with cirrhotic and pre-cirrhotic NASH, and continue to show EFX's potential to be an effective NASH therapeutic. Overall, though no head-to-head comparative studies have been conducted, we believe these are the strongest data reported to date for any product in this difficult-to-treat population. The SYMMETRY study was designed to have 2 on-treatment biopsy reads for each patient. The 36-week analysis we are describing today and an end of treatment analysis at week 96. Given the trend in fibrosis improvement and the statistically significant results in NASH resolution and markers of liver injury at week 36, we believe another 60 weeks of treatment with EFX has the potential to yield higher rates of fibrosis improvement in cirrhotic patients at the completion of SYMMETRY. Additional context helps set the stage for today's data. First, patients with cirrhosis due to NASH face a poor prognosis. As seen here on Slide 3, the overall 5-year rate of survival without a liver transplant among F4 patients is about 50%. Not only is the survival prognosis poor, but as cirrhosis progresses to decompensation and liver failure patients' quality of life is severely compromised. Put simply, cirrhosis is cancer-like in its potential morbidity and mortality impact on patients, and in the economic burden it imposes on the health care system. Unfortunately, there are no approved therapies to treat cirrhosis due to NASH. Arresting progression to liver failure or reversing cirrhosis would, therefore, represent a major medical advance. With more than 3 million Americans estimated to have cirrhosis due to NASH by 2030, the unmet burden is high. Quite simply, any therapeutic that helps reverse fibrosis would be a game changer for a population with no other approved options. Second, [indiscernible] endpoints in patients with cirrhosis due to NASH has proved elusive. Slide 4 shows the landscape of some of the leading programs that have evaluated different mechanisms in Phase IIb or Phase III placebo-controlled trials in patients with cirrhosis due to NASH. Most of these programs have been discontinued. Development of semaglutide remains active, but the placebo-corrected rate was negative 18% for fibrosis improvement without worsening of NASH in a 48-week Phase IIb trial. Slide 5 shows a similar landscape for the NASH resolution endpoint. Here, we see a 13% placebo-corrected response rate for semaglutide, with no other mechanism higher than 6%. Importantly, each of the programs on these 2 landscape sides was studied for at least 48 weeks and as long as 96 weeks. A much higher proportion of collagen accounting for 5% to 35% of the area of a biopsy from cirrhotic liver, compared to up to 5% for pre-cirrhotic liver represents a substantial challenge to address and helps account for the repeated program discontinuations in this indication. Against this backdrop, we believe the totality of today's SYMMETRY data stands out as perhaps the most promising publicly reported placebo-controlled data set to date in patients with cirrhosis due to NASH. 22% and 24% fibrosis improvement without worsening of NASH was observed for the EFX 28 and 50-milligram arms, respectively, compared with 14% for placebo. As mentioned earlier, 4% of EFX patients also experienced at least a 2-stage reversal fibrosis from F4 to F2 without worsening of NASH. There is also considerable evidence that EFX improved liver health. Notably, we observed that NASH was resolved at approximately 60% of EFX treated patients. The first drug reportedly showed a significant effect on this endpoint in patients with cirrhosis due to NASH compared to 26% for placebo. Likewise, noninvasive markers show that EFX reduced synthesis of new collagen and overall fibrosis to a similar extent seen in patients with moderate to advanced fibrosis F2 and F3 as reported last week in Lancet Gastroenterology and Hepatology. Markers of liver injury will also significantly reduce consistent with effects seen in precirrhotic patients. Crucially, these improvements were sustained over 36 weeks in contrast to the only other FGF21 analogue Pegbelfermin tested in a comparable patient population. Based on these observations, we are confident that EFX's antifibrotic effect is exerting clinical activity and cirrhosis, and believe that longer treatment EFX could potentially continue to reverse cirrhosis and improve overall liver health. The prespecified 96-week biopsy analysis in SYMMETRY scheduled for late 2024 will evaluate this. I'll now turn the presentation over to Kitty to walk through the study design and data in more detail.

Thank you, Andrew, and good morning, everybody. I'd like to begin with a review of the design of the SYMMETRY study, which is shown on Slide 6. The SYMMETRY study is a Phase IIb randomized, double-blind, placebo-controlled, multicenter dose-ranging trial. All patients had biopsy-proven compensated cirrhosis fibrosis Stage 4 due to definitive NASH or cryptogenic cirrhosis, presumed secondary to NASH. Subjects with cryptogenic cirrhosis were limited to approximately 20% of the total study population. Importantly, the design of SYMMETRY allowed for an early efficacy evaluation at week 36, as well as an end of treatment evaluation at week 96. After the 36-week analysis, patients remain on their randomized treatment out to week 96 when they will receive an end-of-treatment biopsy. The study enrolled 182 patients with approximately 60 in each treatment group, randomized to receive once-weekly subcutaneous doses of either 28 or 50 milligrams of EFX or placebo. The primary endpoint was the proportion of subjects who achieved at least a 1-stage improvement in fibrosis without worsening of NASH at week 36. Secondary measures include NASH resolution, a combination of fibrosis improvement and NASH resolution. Change from baseline and noninvasive markers of fibrosis, liver enzymes, glycemic control, lipoproteins, and body weight at 36 weeks as well as safety and tolerability. Patient disposition is shown on Slide 7, 182 patients were randomized. One patient was randomized but not dosed, 26 patients or 14% discontinued prior to week 36, 11 of the 26 discontinuations were described as being for other or administrative reasons. The remaining 15 patients discontinued due to adverse events, 2 in the placebo group, 5 in the 28 milligram group, and 8 in the 50 milligram group. Of the remaining 155 patients who completed week 36, 2 patients declined to be biopsied a second time. Thus, there were 153 patients in the liver biopsy analysis set used for the primary endpoint. Of these, 57, 46 and 50 were in the placebo 28 and 50 milligram dose groups, respectively. This study enrolled patients with advanced liver disease, including patients with either cryptogenic cirrhosis or definitive NASH. The analysis set for NASH resolution endpoints excluded those with cryptogenic cirrhosis who didn't meet definitive NASH at baseline. That is the NAFLD activity score of greater than equal to 3, with a score of at least 1 in each of the components of steatosis, ballooning and inflammation. Consequently, the analysis set for NASH resolution is compromised of 126 patients, with 46, 38 and 42 patients, respectively, in the placebo 28 and 50 milligram dose groups. Cryptogenic cirrhosis is sometimes referred to as burn-type NASH, and is associated with advanced fibrosis and a higher level of risk in terms of liver decompensation or death. On Slide 8, we review the baseline demographics, which are consistent with what has been reported in fibrosis stage IV NASH clinical studies, and were representative of the compensated cirrhotic patient population. The majority of patients were female. And this advanced patient population, 17% to 26% of the patients had cryptogenic cirrhosis across the 3 treatment groups. The noninvasive markers of fibrosis, ELF, Pro-C3, and liver stiffness levels are higher than we would have seen in our HARMONY trial, and consistent with more extensive sarcosis. AST and ALT values range from 35 to 40, with ratios approaching a 1:1. Approximately 80% of the patients had type 2 diabetes, and they were being extensively treated with approximately 1/4 of patients receiving GLP-1. Serum triglyceride levels were borderline high, but lower than seen in pre-cirrhotic patients in HARMONY. The advanced stage of disease was evident in the medical history of SYMMETRY patients, with the majority taking more than 10 concomitant medications to manage their comorbidities of NASH. Before diving into the histology results, it's important to highlight that we use the same biopsy analysis process that was used in HARMONY, and this is reviewed on Slide 9. Our consensus approach in SYMMETRY was to have all biopsies read by 2 well-trained pathologists. Each pathologist scored each individual biopsy sample independently. Our pathologists were blinded to subject and treatment assignment. Biopsies were shuffled to ensure the pathologists were also blind to visit sequence. Our protocol specified that if the bio -- as consensus couldn't be reached, a third trained pathologists would adjudicate between the 2 scores. All biopsy analyses were resolved by consensus, to a dedication was not acquired. On Slide 10, we report the results of the primary endpoint of fibrosis improvement of one or more stages without worsening of NASH, which was observed in 24% for the 50 milligram dose group, and 22% for the 28 milligram dose group, compared to 14% placebo response. These translate to treatment effect sizes of 8% and 10% in the 28 and 50 milligram doses, respectively. The table on the right highlights patients in each of the EFX treated groups who experienced a greater than or equal to 2-stage improvement in fibrosis, improving from F4 to either F1 or F2. Despite the short treatment duration, we believe these response rates are numerically higher amongst those reported to date and placebo-controlled clinical studies and patients with compensated cirrhosis. We believe the trend here may lead to a higher treatment response with longer treatment out to week 96. Slide 11 reports the results for the percentage of patients who achieved NASH resolution. The rates of NASH resolution in both EFX treatment groups were statistically significant, with 60% or more NASH resolution. These response rates are more than double the 26% placebo response. We believe that reducing steatohepatitis as indicated by NASH resolution, should ultimately in time lead to reversal of fibrosis and slow our arrest progression to liver failure. Using the intent to treat our ITT analysis set, which treats any missing data failure, both EFX groups maintained statistical significance, with 47% of the 50 milligram and 51% of the 28 milligram group showing NASH resolution compared to 24% for placebo. Slide 12 shows the percentage of patients in each group who achieved both fibrosis improvement of at least 1 stage and NASH resolution at the same time. 14% and 21% of the 50 and 28 milligram EFX group achieved both endpoints compared to a 9% placebo control. On Slide 13, current and discontinued programs reporting fibrosis improvement without portioning of NASH in patients with compensated cirrhosis have been arranged from left to right by steady treatment duration. Information relevant to interpreting each data set is provided above the corresponding histogram. Head-to-head trials comparing these various investigational drugs have not been conducted, so the differences in these trial designs and other factors make cross trial comparisons challenging. Nonetheless, the effect sizes, meaning the difference between treatment and placebo for both doses of EFX are amongst the highest reported to date. Slide 14 places the NASH resolution rates in the same landscape. As you can see here, on a low cross-trial comparisons of inherent limitation, EFX is the only compound to demonstrate statistically significant improvements in NASH resolution in this patient population. Noninvasive markers used to diagnose and monitor fibrosis and cirrhosis indicated that both doses of EFX decreased synthesis of new collagen and reduced inhibition of remodeling of fibrosis in a statistically significant manner. Slide 15 depicts change from baseline for the enhanced liver fibrosis panel. The change from baseline is shown overall in the left hand most histogram, with significant reductions in the EFX group relative to placebo after 36 weeks. The corresponding changes for each component of the panel are shown in the next 3 histograms. Moving left to right, these are P3NP, a marker of collagen synthesis associated with liver fibrosis, TIMP-1 is a marker of inhibition of collagen remodeling, and hyaluronic acid, a major component of the baseline -- the basement membrane to which liver cells attach. EFX treatment group showed statistically significant reductions in NIM levels of P3NP and TIMP-1 at both doses, while the change in hyaluronic acid was not statistically significant. The extent of production in P3NP and TIMP-1 is comparable to what we observed in the 24-week HARMONY study in F2, F3 patients, as is the limited impact on hyaluronic acid. Moving on to Slide 16, which depicts main change from baseline for 3 more noninvasive fibrosis markers. The first histogram shows Pro-C3, another marker of collagen synthesis in the liver. The second plot shows liver stiffness measured by FibroScan. And the third histogram shows FAST, a score which combines liver stiffness and the liver injury marker, AST and is widely used to assess risk of NASH progression. Both doses of EFX showed statistically significant reductions in Pro-C3 relative to placebo, consistent with the reduction in P3NP. Moving to the middle plot, EFX reductions in liver stiffness were statistically significant relative to baseline. However, the reductions were not significantly different from placebo, which declined to a similar extent. These -- this appears to be anomalous with, by reference to prior studies, with EFX in pre-cirrhotic and cirrhotic patients. In the rightmost plot, EFX showed statistically significant reductions in mean FAST score relative to placebo. Again, the extent of reduction from baseline in these markers is similar to what was observed in the HARMONY study. Moving on to Slide 17, which depicts main change from baseline for 2 markers of liver injury, ALT and AST. Baseline levels of these enzymes are lower in cirrhotic than pre-cirrhotic patients because of the loss of normally functioning hepatocytes. Nevertheless, 50 milligram of EFX significantly reduced both levels of enzymes relative to placebo, reaching a maximal effect between 12 and 16 weeks, which was then maintained and diminished through to 36 weeks of treatment. The magnitude of reduction in ALT and AST was smaller for the 28 milligram EFX group, but still significant relative to placebo at most intervals after 8 weeks. Substantial and maintained reductions in markers of liver injury have been consistently observed following treatment with EFX for periods of 16 to 36 weeks in patients with both cirrhosis and pre-cirrhotic NASH. One of the goals of treatment for patients with cirrhosis is to prevent progression of disease to portal hypertension or decompensation. To that end, platelet count significantly decreased from baseline by around 7% in both EFX dose groups compared to a slight reduction for placebo. Consistent with this trajectory of preventing -- sorry, preventing progression, other markers of disease progression, including bilirubin, INR, MELD and Child-Pugh score remained stable or showed slight improvements. Let's shift now to safety and tolerability. Slide 18 summarizes patients who discontinued from the study. Starting at the top of the table here, there was 1 death in a placebo patient who had pneumonia. 21 serious adverse events were reported, which were balanced across treatment groups and reflect the complex comorbidities of this advanced NASH population. The only event reported in more than 1 subject with angina which incurred in 1 patient on placebo, and 1 patient on EFX 50 milligram group. None of the SAEs reported were determined to be drug-related by the clinical investigators. A total of 12 patients were discontinued due to drug-related adverse events or AEs. 1 in the placebo group, 3 in the 28 milligram group and 8 in the 50 milligram group. The majority of these were due to grade 1, 2 diarrhea, with 5 patients discontinuation due to diarrhea in the 50 milligram EFX group, and 1 each for the placebo and the 28 milligram dose group of EFX. In the other category, for the 28 milligram group, 1 patient discontinued due to retching and vomiting, and the other due to palpitations and feeling jittery, which are likely symptoms of a relative hypoglycemia, which may occur with the fast-acting insulin sensitizer like EFX, particularly during the first months of treatment. In the 50 milligram group, 1 patient discontinued due to soft feces and nausea. Another patient discontinued due to a suspected allergic reaction with flushing, vomiting and fatigue that was quoted to hypersensitivity. The third patient discontinued due to a grade 2 patch of a rash at the injection site. There were 3 nondrug-related adverse events that led to treatment discontinuation. In the placebo group, 1 patient, as I mentioned before, died due to pneumonia. For the EFX 28 milligram, 1 patient with a history of cardiac disease experienced cardiac failure, and will require replacement of their preexisting cardiac pacemaker. Another patient discontinued due to drug hypersensitivity, related to 1 of their concomitant medications, diltiazem.. Slide 19 presents an overview of the drug-related treatment emergent adverse event. Consistent with previous evaluations of EFX, the most frequently reported adverse events were transient, mild to moderate, gastrointestinal grade 1 to 2 events. The majority of injection site reactions were mild grade 1. As reported in previous studies, even though we see increased appetite, which reported as an adverse event, mean body weight reductions were observed across the EFX dose groups. No clinically meaningful changes were observed for heart rate or diastolic blood pressure. However, at week 36, increases of 4 to 7 millimeters of mercury and -- sorry, in systolic blood pressure were observed in the EFX dose groups. Based on nonclinical observations on skeletal development, bone health has been monitored in all clinical studies with FGF21 analogue. It's important to note that fatty liver disease is associated with core bone health, in part because of postmenopausal women who constitute a large proportion of the patient population. Small reductions in bone mineral density were observed for the EFX dose groups in the lumber spine region of less than 1%, and in the femoral neck region in the range of 2% to 3%. The I'll now turn the presentation over to our Chief Scientific Officer, Tim Rolph.

Thank you, Kitty, and good morning, everyone. By way of context to the effects of EFX on markers of glucose metabolism in patients with cirrhosis shown on Slide 20, almost 80% of the study population were being treated for type 2 diabetes, with a high proportion on medications to boost insulin levels. This slide depicts mean change in baseline and passing levels of C-peptide, a market of insurance secretion, insulin and glycosylated hemoglobin or HbA1C. Treatment with both doses of EFX for 36 weeks showed statistically significant reductions in serum levels of C-peptide and insulin, consistent with EFX, improving insulin sensitivity. EFX treatment was associated with reduced levels of hemoglobin A1c, but not significantly more than placebo. This may and partly attributable to this population of patients generally being well controlled, particularly with patients in the 50 milligram EFX group, having a mean baseline HbA1c of 6.6%. It may also be attributable to a fourfold higher proportion of patients in the placebo group than the EFX groups, with poor glycemic control at baseline. That's a hemoglobin A1c of greater than 9%. To put the effects of EFX on serum lipids, on Slide 21 in context, pretreatment levels of triglyceride were at the high end of normal. Those of low-density cholesterol were generally close to goal of 100 mg per deciliter, reflecting the substantial proportion of patients taking statins, while those of high-density cholesterol were low, reflecting the high proportion of patients with diabetes. This slide depicts mean change in baseline. Treatment with 50 milligram EFX showed statistically significant reductions in triglyceride and non-HDL cholesterol, and statistically significant increases in HDL cholesterol, and numerically lower LDL cholesterol. The effects of treatment with 28 milligram EFX followed a similar pattern to the higher dose, but the magnitude of improvement was slightly smaller. Moving on to Slide 22, which shows mean change from baseline for serum adiponectin, a pharmacodynamic measure of agonism of FGF21's receptors -- receptor rather, FGFR1c in adipose tissue. After 36 weeks of treatment with both doses of EFX, levels of adiponectin were approximately doubled, while placebo didn't change. EFX has beneficial effects on adipose tissue, enhancing insulin sensitivity, and reducing release of fatty acids is key to establishing a healthy whole body metabolic environment in which hepatocytes are not overloaded with energy. This reduces hepatocyte stress and decreases the rate of loss of normally functioning hepatocytes. Slowing or preferably preventing further loss of normally functioning hepatocytes is essential to preservation of liver function and avoiding progression to failure in patients with compensated cirrhosis. Moving on to Slide 23, which depicts change in baseline in body weight after 36 weeks treatment. By comparison with placebo, 28 milligram of EFX appeared to be weight neutral, while for the 50 milligram dose, body weight trended lower with a significant decrease from baseline. This pattern of response is consistent with that observed in 3 prior separate studies of EFX with pre-cirrhotic patients up to 16 to 24 weeks of treatment, including on top of stable GLP-1 receptor agonist. I will now hand over to Andrew to conclude the presentation.

Thank you, Tim. In conclusion, we are encouraged by today's SYMMETRY data, and believe it supports moving EFX forward into Phase III in patients with cirrhosis due to NASH. Although no head-to-head comparative studies have been conducted, we believe the totality of these data are the most encouraging reported to date for this difficult-to-treat population. Despite the short treatment duration, we see a trend in fibrosis improvement, including 3 stage and 2 stage improvements for some patients. In addition, we have statistically significant results, both for NASH resolution and for many noninvasive markers of liver fibrosis. Based on these observations, we are confident that EFX has the potential to improve liver health and whole body metabolism in patients with cirrhosis. With $659 million in cash as of June 30, 2023, and runway into 2020 -- 2026, we are well positioned to move ahead with our Phase III program. Specifically, we are pleased to have supportive data from our 2 placebo-controlled Phase IIb studies to advance EFX into Phase III for both patients with NASH due to cirrhosis and patients with pre-cirrhotic NASH fibrosis stages F2 and F3. Taken together, today's SYMMETRY study and last year's HARMONY study, in which over 300 patients have been treated with EFX, show that EFX has the potential to treat the core facets of NASH. EFX has been observed to reduce fibrosis, resolve NASH, and improve noninvasive markers of liver injuring fibrosis, reduce liver fat, improve glycemic control and lipoprotein profile, and reduced body weight. These are the hallmarks of an investigational drug, well matched to a complex disease affecting the whole body as well as the liver. The Phase III SYNCHRONY histology and SYNCHRONY real-world studies in patients with pre-cirrhotic NASH stage F2 and 3 are well underway and remain on track to enroll the first patient by December of 2023. We look forward to discussing today's SYMMETRY results with the FDA to help finalize plans for incorporating patients with cirrhosis due to NASH into our Phase III program. And we are excited to continue developing therapies that have the potential to transform the lives of patients living with serious metabolic diseases. Thank you.

[Operator Instructions] The first question comes from Michael Yee with Jefferies.

I have 2 questions for you, Andrew. First, when you take a look at the study and step back, do you think it was a higher discontinuation rate and dropout rate that would have impacted this? And what gives you confidence on a longer treatment duration would be helpful here? And if I may ask a follow-up, there was a disclosure on the slides around bone mineral density [indiscernible] or 4 to 7 millimeter increase in systolic blood pressure, can you make a comment about that? That would be something that I think that could be more catching.

Yes. So Mike, I think the -- as we mentioned during the call, overall, we're pleased with the totality of the data. The discontinuation rate was a little higher than we had expected, but at the same time, represents the -- to some degree, some of the overall advanced disease that these patients have. In terms of -- you asked about blood pressure. Maybe, Tim, if you want to make a comment about that, that would be helpful.

Yes. The increase in systemic blood pressure appears to be associated with the reduced capacity of the liver in these patients to maintain blood glucose. So when they are exposed to a rapidly acting insulin sensitizer, the glucose levels are very quite markedly, and that's leading to activation of the regulatory hormones, glucagon, adrenaline, both of which are known to increase systolic blood pressure. We would expect them going forward into Phase III that we'll be using, adapting the doses of the antidiabetic medications to allow -- to be in better control in these patients.

And I guess, Andrew, to clarify, I was asking what gives you confidence that longer treatment duration would be the results in this study?

There were a number where I can get to all of them. So I think, when one looks about -- at the mechanism of action, I think that it's really driven by the additional collagen that these patients have. It's just so much more than what we've seen in the pre-cirrhotic population. And when we look at the response rates, especially the deep response rates in the patients with 2 and 3 stage improvements in fibrosis, it gives us the confidence that a mechanism is very active, and it just takes more time to reduce to get the collagen down from these higher percentages at baseline.

[Operator Instructions] The next question comes from Eric Joseph with JPMorgan.

I too, I'm just a little bit curious about sort of what gives you confidence in moving forward with the cirrhotic NASH Phase III program based on the data here today as opposed to perhaps getting a little bit more from the end of treatment assessment at 96 weeks. And then, this potential for cryptogenic NASH, I think, is a new variable in thinking about the context for an F4 study. I guess, what's sort of -- to the extent there are -- are there any measures that could be taking in a Phase III program to sort of reduce their participation and perhaps get a clearer signal?

So Eric, thanks for raising that. So I think, recall that the Phase III program, per the FDA guidance, is an outcomes-driven measure. So I think when one thinks about 1 stage improvement in fibrosis, that's not necessarily directly correlative with outcomes-based end points. And so I think for us on that, it's -- we think that we've seen improvements in liver health overall, and it's as well as improvements in natural solution that will also contribute to improvements in outcomes overall. In terms of cryptogenic cirrhosis, I think these patients represent a part of the cirrhotic spectrum. And they have a little more advanced NASH, and I think we've -- and in consultation with the FDA, have chosen to limit the patients to about 20% of the population. And I think that's something we may consider to do. But of course, that's pending discussions with the agency, which we haven't had.

[Operator Instructions] The next question comes from Mike Ulz with Morgan Stanley.

Maybe just a follow-up on the thinking in terms of the Phase III plan for F4 patients. Could you -- I guess, some of the scenarios around that, could you start that Phase III study near term? Or would you want to see the results from the 96-week end of treatment SYMMETRY study before you move forward there?

Mike, it's a good question. I think -- we think we have enough data to move forward in Phase III, but it's important that we haven't consulted with the agency on this. We had an end of Phase IIb meeting in March of this year, which we said that in pursuing the alternative pathway where we use the F4 population, outcomes-based approach as a confirmatory study, in principle there, we're in agreement with that. But at the same time, we -- both sides agreed to say, let's come back and readdress this once we have SYMMETRY results. So since we're announcing the results today, we haven't had a discussion with the agency. We are going to reach out to them and hope to talk to them in the near term.

We have time for one more question. That question comes from Liisa Bayko with Evercore.

Just two questions from me. First of all, was it prespecified to take out the cryptogenic NASH patients? And can you give us the data on an ITT basis? And one more, for FDA, for your Phase III study, is a histology biopsy endpoint [ shown ] on the table? Or are you thinking more along the lines of an outcome study for the cirrhotics?

So I think for the cirrhotics, this is, as I mentioned with Mike's question, we're still discussing with the agency. I think the agency at this point doesn't view histology as an approvable endpoint. However, as you and I have discussed in the past, the EMA does view it that way. So I think how we reconcile that and -- requires some discussion with the agency. So I think that, that's a little bit of a TBD. In terms of the ITT, it's on the slide, Liisa, on the endpoint slides, we put that on the data. I don't have the slide right in front me. I couldn't really tell you right off the top of my head. Maybe Kitty, if you have that, if you could just help Liisa.

Yes. So I'm just wanting to check whether it was for the primary endpoint you wanted that data?

Yes. Sorry, maybe I missed it because there was different cuts going on, but just...

Don't worry. So it was 13% for placebo, 18% for the 28 milligram and 19% for the 50 milligram.

And then just final question was on the cryptogenic cirrhotics. Was it pretty specified to exclude them from some of the analysis? Or what was the plan there?

Yes, that was all prespecified.

Thank you for participating in today's program. This does conclude the conference. You may now disconnect. Everyone, have a great day.