Amicus Therapeutics, Inc. (FOLD) Earnings Call Transcript & Summary

Good afternoon, ladies and gentlemen, and welcome to the Amicus Therapeutics' conference call and webcast. [Operator Instructions] As a reminder, this conference call is being recorded. I would now like to turn the conference over to your host, Mr. Andrew Faughnan, Head of Investor Relations. You may begin.

Thanks, Gigi. Good afternoon. Thank you for joining our conference call to discuss top line results from the Phase III PROPEL study of AT-GAA, Amicus' novel therapy for the treatment of late onset Pompe disease. Speaking on today's call, we have John Crowley, Chairman and Chief Executive Officer; Dr. Jeff Castelli, Chief Development Officer; and Dr. Mitch Goldman, Senior Vice President and Head of Clinical Research. Also joining for Q&A from Amicus are Bradley Campbell, President and Chief Operating Officer; Daphne Quimi, Chief Financial Officer; and Dr. Hung do, Chief Science Officer. As referenced on Slide 2, we may make forward-looking statements within the meaning of the Private Securities Litigation Reform Act of 1995 relating to our business as well as our plans and prospects. Our forward-looking statements should not be regarded as representation by us that any of our plans will be achieved. Any or all the forward-looking statements made on this call may turn out to be wrong. It can be affected by inaccurate assumptions we might make or by known or unknown risks and uncertainties. You are cautioned not to place undue reliance on any forward-looking statements, which speak only to the date hereof. All forward-looking statements are qualified in their entirety by this cautionary statement, and we undertake no obligation to revise or update this presentation and conference call to reflect events or circumstances after the date hereof. For a full discussion of such forward-looking statements and the risks and uncertainties that may impact them, we further refer you to the forward-looking statements and Risk Factors section of our annual report on Form 10-K for the year ended December 31, 2019 and the quarterly report on Form 10-Q for the quarter ended September 30, 2020 filed with the Securities and Exchange Commission. At this time, it is my pleasure to turn the call over to John Crowley, Chairman and Chief Executive Officer. John?

Great. Thank you, Andrew, and good afternoon, everyone, and welcome to the Amicus conference call on the Phase III PROPEL study of AT-GAA. This is an exciting day for all people living with Pompe disease and their families especially those who, for nearly 15 years, have had only one option for the treatment of their disease. It is an exciting day as well for Amicus as we move one important step closer to having a second global medicine for a rare disease approved. The science for AT-GAA is entirely home-grown. It was based on the belief and skills of our Chief Science Officer, developed over 20 years working with this enzyme, that he and our Amicus team could discover and characterize a next-generation enzyme replacement therapy that would have the necessary amount of carbohydrate structures essential for optimal uptake into a Pompe patients' target muscles. And the further novel invention that we could combine this next-generation enzyme replacement therapy with a small molecule enzyme stabilizer to protect its confirmational state while in a patient's bloodstream before entering a muscle cell and delivering more active enzyme to the lysosome. We needed a therapy with the enhanced potential to reach all key muscles of disease, especially both skeletal and respiratory muscles, and it needed to be manufacturable at a very large scale, while maintaining the tight precision of its carbohydrate structures. This was an immense undertaking and one where many others have tried and failed before, but in the creation of this technology and in its development, we have now succeeded. We didn't want to discover and bring forward something of just incremental benefit. While for approval, we only needed to show that AT-GAA was not inferior to the currently approved therapy, we wanted to shoot much higher. The community needed a new medicine with the potential to be a paradigm shift. Highly successful Phase II studies for this new investigational medicine, which we refer to as AT-GAA led to breakthrough therapy designation for a full regulatory approval in the United States and globally, we needed then to demonstrate that this new drug paradigm for Pompe, AT-GAA, worked as well or better than the long-ago approved and currently solely approved enzyme replacement therapy, alglucosidase alfa, known commercially as Myozyme or Lumizyme. We recognize the great need for new treatment options for patients on the current medicine, and we know that many patients on the currently approved therapy after any initial improvements, continue to experience decline in both musculoskeletal and respiratory function. We saw the great opportunity in the near future would be to benefit these patients who may switch to a new drug therapy. The PROPEL study was designed principally to test a hypothesis that AT-GAA may be advantaged over the currently approved ERT in this switch population who comprise the vast majority of known people living with Pompe in the world today, more than 3,000 ERT-treated patients. Following regulatory feedback, we also included a small number of treatment-naive patients as well. They ask that we include 25 to 30. While we had waitlists of ERT-switch patients requesting to enroll in the PROPEL study, and eventually, we did enroll 95 of them, we looked far and wide for these treatment-naive patients who would qualify for therapy and meet enrollment criteria. It was not easy, but with the commitment of our clinical study partners, 28 volunteers were identified and enrolled in PROPEL worldwide. Obviously, these are 2 very different patient populations. And so with that, we conducted the largest ever controlled clinical study in a lysosomal disorder, enrolling 123 Pompe patients at 62 sites in 24 countries on 5 continents. 117 patients completed this study and all 117 have enrolled voluntarily in the extension study. The results of this study have profound significance for all people living with Pompe and offer the hope for treatment options. Professor Benedikt Schoser, principal investigator of the PROPEL study, will present these results virtually tomorrow, February 12, as part of a platform presentation at the 17th Annual World Symposium. And in a moment, I'll turn the call over to Dr. Jeff Castelli and Dr. Mitch Goldman to take us through these exciting data. But before I do that, let me summarize 5 important key takeaways from this study. First, we intend to move rapidly to complete the BLA submission in the United States and expect multiple global regulatory submissions for approval throughout 2021. Time is of the essence for people living with Pompe. Second, this study was immensely positive and exceeded our expectations with respect to the breadth, magnitude and consistency of responses in patients switching from the currently approved ERT. In this population, AT-GAA treated patients, on average, walked 17 meters further than standard of care on the 6-minute walk distance test, and they showed stabilization of respiratory function on FVC versus a continued decline for those patients on average, remaining on the currently approved ERT. And all key secondary endpoints and biomarkers favored AT-GAA over standard of care. Third, in the combined population of ERT-switch and naive patients, while we showed a 14-meter improvement over standard of care, we just missed statistical significance on the primary endpoint of mean change in 6-minute walk distance. Had the difference been just over 16 meters, we would have hit statistical significance. So just 2 meters more and this primary endpoint on superiority would have been demonstrated, but there were consistent and strong trends here in favor of AT-GAA. In fact, 8 of 8 key secondary and biomarker endpoints favored AT-GAA over standard of care in this combined switch and naive patient population. Fourth, and this is very important, the statistical analysis plan, which was reviewed and approved in advance by regulators, specified percent-predicted forced vital capacity, or FVC, as the first key secondary endpoint for the study. FVC is an important endpoint from a regulatory perspective, as improvement in FVC was the basis for approval of Lumizyme. And from a clinical perspective, FVC is the key pulmonary endpoint in Pompe as loss of pulmonary function is the leading cause of mortality. In the combined population of ERT-switch and ERT-naive patients, AT-GAA showed a nominally statistically significant and clinically meaningful difference for superiority on FVC, showing a 3% positive difference and a p-value of 0.023. This is remarkable. And again, it's driven by the profound results on this endpoint in the ERT switch population, where over a 4% difference was seen. In rare disease studies, it's difficult enough to show a difference versus placebo. Here, we showed a profound and clinically meaningful improvement over standard of care. Fifth and finally, we believe that AT-GAA would also be a benefit in the treatment-naive population. Although we did not achieve statistical significance in this very small sample size, in the 20-naive patients on AT-GAA, we did show a 33-meter improvement in these patients in this study. And a majority of key secondary endpoints to these patients also improved. Our submission will seek to include the treatment-naive late onset Pompe disease population as well. We look forward to upcoming regulatory submissions, which is the next step in advancing this much needed potential treatment to get it as quickly as possible to patients in need. So with that, I'm happy to turn the call over to Dr. Castelli and Dr. Goldman. Jeff?

Thank you, John, and good afternoon, everyone. We're really excited to share our top line data from the PROPEL Phase III study of AT-GAA. Starting on Slide 4. I'll review the executive summary and then go over study design before turning it over to Mitch to walk us through the detailed study results, and then we'll go back to John for next steps in closing remarks. First, as John noted, based on these results, we're moving as fast as possible towards completion of our rolling BLA here in the second quarter and expect other global submissions throughout the rest of the year. As John also noted, this was the largest controlled study ever conducted in the lysosomal diseases, and there's a wealth of data to assess. Today, we're presenting the top line data, and we'll continue to assess additional endpoints and analyses over the coming days of weeks. On the primary endpoint of 6-minute walk, we saw that ERT-experienced patients stations switching from alglucosidase alfa to AT-GAA walked approximately 17 meters further than those remaining on alglucosidase alfa with a p-value of 0.046. Of note, these switch patients had been on alglucosidase alfa, on average, over 7 years before switching to AT-GAA. Also in these patients switching from alglucosidase alfa to AT-GAA, we saw an improvement in percent-predicted forced vital capacity, or FVC, compared to a significant decline in patients remaining on alglucosidase alfa, with a difference of 4.1% and a p-value of 0.006. As John mentioned, this is a really impactful finding as FVC was the main endpoint upon which alglucosidase alfa was approved. And even more importantly, progressive loss of pulmonary function is the leading cause of mortality in Pompe. And FVC was the first specified key secondary endpoint of study. We saw a similar improvement in FVC in the overall population, including both the ERT-experienced and ERT-naive patients, where AT-GAA also showed a nominally statistically significant and clinically meaningful difference for superiority versus alglucosidase alfa with a difference of 3% and a p-value of 0.023. Also, consistent with the 6-minute walk results we saw in the ERT-experienced patients, patients on AT-GAA walked 14 meters further than patients in alglucosidase alfa in the overall population of ERT-experienced and ERT-naive patients. This difference, however, did not reach statistical significance for superiority over alglucosidase alfa with a p-value of 0.072. We also observed improvements in 2 important biomarkers of Pompe disease. HEX-4 and CK, which both significantly favored AT-GAA over alglucosidase alfa in the overall population. And across all the primary key secondary and biomarker endpoints, we saw consistent results favoring AT-GAA in both the overall and ERT-experienced populations. Before we review the PROPEL study design, and Mitch goes over the detailed results, I want to first take the time to highlight Pompe disease more broadly here on Slide 5. Pompe is an inherited lysosomal disorder caused by a deficiency of the enzyme asset alfa glucosidase or GAA. In people living with Pompe, the reduced or absent levels of GAA lead to the accumulation of glycogen in the lysosomes of muscles and other tissues. Disease severity ranges on a spectrum, but predominant manifestations with the disease are skeletal muscle weakness and progressive respiratory involvement. Often, we see decline leading to organ failure and early or premature death. Pompe can be diagnosed at various stages in life, that is estimated to affect anywhere between 5,000 to 10,000 individuals globally, although recent newborn screening studies suggest there could be significant underdiagnosis. The current improved therapy for Pompe is alglucosidase alfa for Lumizyme, which is the first generation enzyme replacement therapy designed to replace this missing or deficient GAA enzyme. Despite the availability of alglucosidase alfa, significant unmet need remains, in particular, for the many patients experiencing continued progressive decline in function after their first few years of treatment. With that as background and turning now to Slide 6, I want to remind everyone of our highly differentiated Pompe therapy AT-GAA and its unique mechanism of action. AT-GAA is our novel next-generation therapy consisting of cipaglucosidase alfa, an investigational human recombinant GAA enzyme administered into the body via IV infusion, and it's designed to target muscle cells throughout the body. And this ERT is combined with miglustat, an orally administered enzyme stabilizer. Miglustat is administered shortly before the infusion of cipaglucosidase alfa begins, and it's intended to bind and stabilize the enzyme in the circulation, allowing more of the active enzyme to be taken up in the cells and ultimately delivered to lysosomes. The combination of ERT and enzyme stabilizer has won major distinction from the standard of care and from any treatment currently in development for Pompe. The other, and we believe, more impactful advancement is the unique carbohydrate profile of the enzyme itself. Dr. Hung Do, our Chief Science Officer and his team, have been working over the past decade to develop a Pompe ERT with improved binding to target receptors for efficient uptake into cells while importantly, retaining the ability of the ERT to be processed by cells after it's taken up into the more mature active form of GAA. With that in mind, let me now direct you to Slide 7 for the PROPEL study design. PROPEL is our global Phase III clinical study of AT-GAA and late onset Pompe disease. It's a double-blind randomized study, assessing the efficacy and safety of AT-GAA in adult treatment-naive and ERT-experienced participants against the currently approved therapy. The design of this trial was based on assumptions from our Phase I/II study, the medical literature, and the feedback we've received from experts and regulatory agencies. The study was randomized 2:1. So for every 2 patients on AT-GAA, there was 1 patient randomized to Lumizyme. And we stratified enrollment to ensure we had similar proportions of naive- and ERT-experienced patients in each arm of the study as well as be stratified by baseline 6-minute walk distance. Moving to Slide 8. The primary and secondary objectives of this study were to assess the efficacy of AT-GAA compared with alglucosidase alfa on 3 particular areas: first, ambulatory function, as measured by 6-minute walk; the second was pulmonary function, as measured by percent-predicted FVC; and the third was muscle strength, health-related PROs and other motor function assessments. The primary endpoint of 6-minute walk was to be analyzed using an MMRM, repeated measures analysis, and all key secondary endpoints, including FVC or analyzed by ANCOVA. Moving on to Slide 9. We have a summary of the study endpoints. As I mentioned, the primary endpoint was, change from baseline in 6-minute walk distance, the first key secondary endpoint was change from baseline and FVC. And those other key secondary endpoints with a: change from baseline and lower manual muscle test; the PROMIS, physical function domain; the PROMIS-Fatigue domain; and change in GSGC or the Gait, Stairs Gowers, Chair endpoint, which is a composite test of motor function performance. Additionally, we assess other secondary endpoints and key biomarkers of muscle damage, as I mentioned, CK and HEX-4. Now I'd like to turn over the presentation to Dr. Mitch Goldman, our Senior Vice President and Head of Clinical Research, to walk us through the key results from PROPEL. Mitch?

Thank you, Jeff. Good afternoon, everyone. On Slide 10, you can see the patient disposition for the study. Of the 123 patients involved, 85 were stratified to the AT-GAA arm, 38 to the alglucosidase alfa arm. As you can see, the large majority of patients over 3/4 were using enzyme replacement therapy, 65 patients in the AT-GAA arm, 30 in the alglucosidase alfa. The smaller ERT-naive group had 20 patients on AT-GAA, 8 on the comparator. Of those 122 patients that are randomized, 117 completed the trial, all of whom opted into the open-label extension wanting to either continue on or switch on to AT-GAA. There is a very low dropout rate, 6 patients with [ 2 ] in total, 5 in the AT-GAA group, 1 in the alglucosidase alfa group. The dropouts in the AT-GAA arm were 2 infusion associated reactions, those being treatment-related. The others were from COVID-related pneumonia; and 2 withdrawals of consent and they no longer wish to travel to the sites for infusions. There was 1 withdrawal from the alglucosidase alfa, that being a stroke unrelated to study drug. Move to the Slide 11, please. On the next couple of slides, I'll go through the baseline demographics for the trial. A couple of things to highlight. There's balance between the 2 treatment groups across age, gender and all the other metrics. Over 2/3 of the patients in the ERT-experienced cohort has been on enzyme replacement therapy for more than 5 years prior to starting the study with average time on alglucosidase alfa of 7 years. Turning to the next slide. You can see the demographics for the 6-minute walk test and FVC. Again, they're generally balanced between the 2 treatment groups. And you can see that the data has faced validity that is the 6-minute walk and FVC are lower as expected in patients of ERT. Those measures are higher in patients naive to enzyme replacement therapy. Remember that for the inclusion-type criteria, there was no cap on FVC, no upper limits into the trial. And I think you see that reflected in the nearly 80% predicted FVC value in the naive patients. Next slide, please. So moving on to the results. In the overall population in switch and naive patients, patients on AT-GAA at the end of 52 weeks walked 14 meters further than patients treated with alglucosidase alfa. This primary endpoint, while numerically superior in favor of AT-GAA, did not achieve statistical significance over the comparator with a p-value of 0.072. As noted earlier, the first key secondary endpoint in the study with the change from baseline in percent-predicted forced vital capacity at end of study, here in this overall study population of ERT-switch and ERT-naive patients, FVC favored AT-GAA by 3%, a difference that was both nominally statistically significant at a p-value of 0.023 and clinically meaningful for superiority both when compared to alglucosidase alfa. We are very excited by this result as the decline in respiratory function and related complications are the leading cause of death in Pompe. It's also notable to call out here that FVC is an important endpoint in this disease and served as the basis of approval for alglucosidase alfa. Before leaving this slide, I want to mention that 1 ERT-naive patient in the alglucosidase alfa arm was excluded from the study analysis shown here due to use of an investigational anabolic-like steroid that impacted its baseline performance. Next slide, please. In the ERT-experienced patient population, which included the majority of the patients enrolled, 95 of the 123 total participants, here we saw a clinically significant improvements in both 6-minute walk test and FVC in favor of AT-GAA over standard of care alglucosidase alfa. Patients switching to AT-GAA walked almost 17 meters further than alglucosidase alfa, the difference being -- yielding a p-value of 0.046. Turning to FVC. AT-GAA patients lung function remained stable, increasing 0.1% during the course of the study, while the patients on alglucosidase alfa declined by 4%, and the difference between the treatment groups getting a p-value of 0.006. Next slide, please. The charts on this slide really highlight these differences in the data across the type of the study. AT-GAA is captured in blue, alglucosidase alfa shown in red. You can see there's clear separation at the first measurement, week 12, that is sustained and grows over the remainder of the study. It's important to point out here that those on alglucosidase alfa had no change in their 6-minute walk test compared to their baseline. And again, their FVC declined from baseline throughout the year on treatment period. We are very excited by these results that show that in these ERT-experienced patients ,[indiscernible] remember had been on alglucosidase alfa for an average of 7 years before coming into this trial, that AT-GAA was able to change the course of their disease, improve their walk distance and maintain their lung function. Next slide, please. Moving on to the ERT-naive population. We see at 52 weeks, the 20 AT-GAA treated participants walked 35 meters further than their baseline, which was encouraging and largely in line with expectation. Unexpectedly, however, the 7 alglucosidase alfa treated patients walked 38 meters further than their baseline. Nonetheless, the difference between the 2 groups was not statistically significant. We believe this result likely due to the heterogeneity of the disease in patients and a relatively small sample size study here. As for FVC, both AT-GAA and alglucosidase alfa treated patients showed similar declines in this measure at 52 weeks. Again, that difference was not statistically significant between the groups. There's not currently a clear explanation for the FVC decline in both treatment groups. It's possible at the relatively high baseline levels of that 80% predicted in time to enrollment may have played a role as well again, heterogenetic disease as well as sample size. You may recall that in our Phase I/II study, we did see improvements in FVC over the longer time points. So we look forward to following these patients in the long-term extension study to see how they do. I mentioned this before when discussing the overall results, but I want to reiterate here the 1 ERT-naive patient in the alglucosidase arm was excluded from the analysis shown here due to use of an investigational anabolic steroid. Next slide, please. So moving to some of the secondaries. We had prespecified secondary endpoints, just as Jeff spoke to earlier in the presentation, lower MMT, GSGC and a couple of patient-reported outcomes. So let's look at those now, and we'll focus on the overall population and the ERT-experienced patients. The next few slides follow the same format. Overall population on the left, ERT-experienced patients on the right, AT-GAA patients is in blue, alglucosidase alfa in red. Here on the lower MMT, a measure of muscle strength to lower extremities, we see that AT-GAA treated patients improved more than alglucosidase alfa patients over the course of the 52 weeks in both the overall patient population as well as the ERT-experienced patients. Next slide, please. During the GSGC, Gait, Stairs, Gowers and Chair measure, is an important and commonly used endpoint in Pompe disease that capture strength, coordination, mobility. AT-GAA patients demonstrated statistically significant improvements on the scores in this assessment compared to a worsening in scores for alglucosidase alfa, both in the overall patient population and the ERT-experienced patients. Next slide, please. Moving to the patient-reported outcomes, the PROMIS physical function. In the overall population and ERT-experienced populations, the PROMIS physical function score numerically favored AT-GAA over the approved therapy. Next slide. We see a similar result on the PROMIS-Fatigue in which AT-GAA was favored in both patient populations. So moving on to the biomarker data, creatine kinase, CK. The enzyme that leaks out of damaged muscle cells is elevated in Pompe patients. Reductions in CK may be correlated with improved healthier muscle. At 52 weeks, AT-GAA patients showed substantial improvements on TK with a 20% -- with over a 22% reduction in CK compared to an increase or worsening in the alglucosidase alfa treated patients of 16%. On this next slide, you see the urine Hex-4 measures. Glycogen is a substrate that accumulates in the lysosomes of muscles of Pompe patients. Urine Hex-4 is a measure that speaks to the degree of skeletal glycogen clearance in these patients receiving enzyme replacement therapy. Patients receiving AT-GAA showed substantial improvements with a mean reduction of Hex-4 of over 31% after 52 weeks compared to an increase, or again, that's a worsening of about 11% in the alglucosidase alfa treated patients. This next slide of a heat map presents a summary of our top line efficacy and biomarker data that you've seen throughout this presentation. Grouping the domain, the data in the domains of motor function, muscle strength, pulmonary functioning, patient-reported outcomes and biomarkers, you can see that the totality of data favor AT-GAA alglucosidase alfa in both the overall and ERT-experienced patient populations. Specifically, the 6-minute walk, the FVC, can you go back to that slide, please? The endpoint of 6-meter walk, FVC and 4 other predefined protocol endpoints, 6 in all as well as the important biomarkers of CK and urinary Hex-4 all favored AT-GAA in both the combined and the ERT-experienced patients. Next slide, please. So lastly, let's turn to the safety data from the study. Safety, inclusive of adverse events, serious adverse events, infusion associated reactions were similar between the 2 treatment groups. I'll point out a couple of things on this slide. For serious adverse events, there were 8 in the AT-GAA group, only 1 of those related to study drug, that being in infusion associated reaction, the others were unrelated. Those being, for example, a fall that resulted in contusions or lacerations, [indiscernible], removal of hardware, [indiscernible] fracture repair and so on. Nothing related to the drug. The one serious adverse event in alglucosidase alfa arm was a stroke, and again, that stroke was unrelated to treatment. As for the adverse leading -- events leading to study withdrawal, I covered those earlier in the presentation on the study disposition slide. But just to remind you, there are 2 IARs here in the AT-GAA arm, 1 serious and 1 nonserious. And again, the 1 event associated to withdrawal from the alglucosidase arm was the stroke patient. With that, I'll turn it back over to John to discuss next steps.

Great. Thank you, Mitch. Moving now to Slide 25. I will end the call here with a brief summary of the next steps surrounding our development, regulatory and manufacturing strategy for the AT-GAA program. First, Dr. Benedikt Schoser will be presenting the clinical results from the PROPEL study tomorrow at the World Symposium virtual conference as part of the 9:30 a.m. late-breaking science session. Two, our rolling BLA submission, which was initiated with the FDA in the fourth quarter of 2020, remains on track with the submission of the final module to be completed in the second quarter of 2021. Additionally, our MAA submission with the EU is expected to be completed in the second half of 2021. We also have the potential for early approval under the EAMS framework with Priority Innovative Medicines Designation in the United Kingdom, and we expect multiple additional global filings to follow. We will look to expand our ongoing open-label adolescent study in the 12- to 18-year olds, living with Pompe disease. And our clinical study for Pompe patients with infantile onset disease is expected to begin this year. And finally, in response to the many requests for compassionate use or expanded access that we have received for children with infantile onset Pompe disease, our expanded access program is for Pompe infantile patients and adolescent onset patient -- I'm sorry, adult onset patients are open and have enrolled multiple patients. Further expanded access for all Pompe patients is being considered. So in closing, on behalf of all of Amicus, I want to express thanks to all of our Pompe clinical study participants and their families, to all investigators and their site staff into our cross-functional Amicus Pompe team for their collective, unwavering commitment and supportive efforts during this unprecedented time. I'm proud of these results, of our team and the potential this therapy can provide to so many, and that we and our collaborators and these patients, conducted this study around the world and produce these high-quality results in the face of the global COVID pandemic is just remarkable. It is a testament to the need and motivation for new treatment options in Pompe. And with AT-GAA, we think a treatment option that will potentially become to do standard of care in the management of this devastating disease. For all of us, this continues to be a labor of love. Operator, we will be happy to open the call to questions.

[Operator Instructions] Your first question comes from the line of Anupam Rama from JPMorgan.

I realized on the call, you guys said as it relates to the Lumizyme-naive population that it's small number heterogeneity as the population is worth considering, but what are your -- any other thoughts on the Lumizyme outperformance in naive patients? I mean this is better than what we saw a lot, much better than we saw in the COMET. I realized that there's maybe a 25- to 30-meter difference at baseline. Could that have played a role? So just maybe how do we think about that?

Yes, that -- absolutely what made the difference was the unexpected results in that very small number, again, that's -- Anupam, the evaluable population there, that was 7 patients. And when we talked to the key opinion leaders, the experts, they had a very strong group view that, that's way too small to tell us anything other than results in those individual patients. These patients have not -- they have been diagnosed very early in the course of their disease. So we think, as Mitch highlighted, it's the heterogeneity of the disease and just a very small n in that control group.

Our next question comes from the line of Ritu Baral from Cowen.

John, have you spoken to the FDA or EMA since opening the envelope on the fileability of this application as is? I've noticed that you don't have any specific noninferiority staff in here. And also, I just got an inbound from a client, if you could just clarify the footnote on Slide 14 and 15. The thing about 6-minute walk, not normally distributed, and having an ANCOVA. Like does that factor into your past and potentially future FDA discussions before filing?

Sure, of course, Ritu. So I figured you were able to get a couple of parts in there, which is great. So on the first, we're not going to comment on discussions we have with FDA. I'll just remind everybody of the breakthrough therapy designation. We have shared these data with the FDA. We began the rolling submission in the fourth quarter. Our plan is to complete the submission in the second quarter. Of course, the standard for approval is noninferiority of this drug. We think here, we significantly exceeded that. And I'll remind everybody, too, that the next step down after the 6-minute walk on a combined population was to look at forced vital capacity in the combined population for superiority. And we achieved that. And you see we achieved that pretty significantly. Again, very much driven by the switch population. I think that's where we're very excited about the pulmonary functions that really exceeded our expectations in terms of the breadth, the magnitude, the consistency of the data and certainly the clinical relevance of that data. So we think we have a very complete package now to complete the BLA submission in the second quarter and then to move on to complete submissions for the MAA. And again, the potential for early approval with EAMS. I'll let -- Jeff, maybe you want to take, and Mitch, any color on the statistical analysis plan?

Yes. Ritu, it's Jeff. So there were a number of prespecified statistical analysis for the 6-minute walk data in the statistical analysis plan. One of the things that we also prespecified was testing the raw data for whether it was normally distributed because that data was not normally distributed, there were prespecified nonparametric and COVID tests that were the most appropriate ones to assess the data, and that's the ones that we reported in the table.

Our next question comes from the line of Joseph Schwartz from SVB Leerink.

Just to continue along some of this questioning. I was wondering if you could just expand upon or give us a sense of how consistent with a 6-minute walk changes across patients in the AT-GAA and Lumizyme arms. Did you see outliers like there were in lots in either your drug arm or the control?

Yes. We saw a very, very consistent responses. Maybe, Jeff, if you want to begin and focus -- I know there was the 1 slide that has the switch population and that showed the nice separation of the curves there in particular. So maybe comment on that and then we could add the color in the naives as well?

Yes. Joe, to your question in terms of outliers, there was the 1 outlier that we mentioned that was impacted at baseline. Other than that, there weren't specific extreme outliers like it's been seen in other studies. Overall, the data was still not normally distributed even without that outlier. But the results were very consistent when you look at sort of that number of patients improving on 6-minute walk, for example, there was significantly more improving on AT-GAA. So that overall data supported what we saw in sort of the mean numbers and what were in the analyses.

When you say not normally distributed, can you describe the type of SKU that there is?

You're pushing my limits of statistical knowledge, but there was both kurtosis and skewness, and there was not equal variances.

So it was definitely out of more -- adequate for a nonparametric.

And I think importantly, too, that was an objective test that was prespecified as well.

Yes. And what about the SKU? Was it more patients performing relatively better on Lumizyme than on AT-GAA or vice versa? Can you help us out with that?

It was mostly there were more improvements than declines. There really were no big outlier declines or sort of large declines. It was mostly some skewness in terms of improvement more than you would expect in normal distribution.

Our next question comes from the line of Tazeen Ahmad from Bank of America.

John, as it relates to being statistically significant on the switch population in particular, how much of the market is switch versus naive? And then, you are very confident about completing your application, of course, but to those out there who might question in the overall efficacy of the study, what is your response to that?

Sure. Maybe, Bradley, do you want to address, first, the patient population and the switch population? Brad, you're muted, I think, on your phone.

Yes, I am. Thanks. Somebody has to do it. I'm sorry about that, guys. So the current Pompe population, as I'm sure folks know, just passed, I think, $1 billion in reported sales, and that represents, we think, around 3,000-or-so treated Pompe patients. We estimate that, that's higher penetration in terms of treatment than what we've seen in Fabry as an example. So the vast majority of the market today is -- of the late onset market today is a switch market. And that's where we focused initially with Galafold with great success, and we'll do the same here, SMA and approval. And we think the data here with -- as I think Mitch mentioned, with superiority and forced vital capacity across the entire population, but then also the great data and 6-minute walk and forced vital capacity and switch population really positions us well for our submission and then ultimately, our target private profile.

Do you have an understanding with the FDA that you could apply without achieving the primary endpoint?

Based on what we know today, our plan continues to be to complete the submission in the second quarter of this year, again, the rolling BLA that began in the fourth quarter.

Our next question comes from the line of Mohit Bansal from Citi.

One question. So given the statistical hierarchy here, wouldn't FVC endpoint be considered exploration because it failed primary endpoint? And then if you look at the benefit, 40 meters of benefit in the baseline and same is the case for FVC, the benefit appears to be 5% to 6% from the baseline. And for clinicians, do you think is it enough for them to switch patients from Lumizyme to AT-GAA?

We think, yes, very much that it's clinically meaningful. Mitch, I'm going to ask you to comment on -- with all the key opinion leaders that we've just heard recently that it -- this data with, in your view, in terms of the clinical meaningfulness of this data.

Sure. So a few things. I'll get my thoughts first and then tell you some of the key opinion leaders' expressions. But I think that -- let's leave the naive aside for a moment. In the ERT-experienced patients, you saw that they're on 70% were on ERT for 5 years. A large -- on average for 7 years in discussing this externally is to feel that by then, most of those patients are declining on alglucosidase alfa. You can see in the trial that we ran ourselves that the compare arm was flat, whereas, we improved. You could see that FVC, their lung function is declining over time at a fairly rapid clip of 4%. If that pace carries on after a few years, it puts them at risk of respiratory compromise, whereas on AT-GAA, the walk test improved, the FVC lung function stabilized. So we think that the results are really compelling. And there's really strong efficacy there. And the naive, again, I'll go back that patients improved -- the majority of patients in both treatment groups improved. I mean, yes, the difference is with some of the outliers, in some of the small patient populations. It really is heterogeneity across the board and a few outliers. But in general, I think the strength of the data is seen from the elements I just described across the board and talking to several external experts in the field. And they're [ actually ] very much the same thing. I think that to a person, they were impressed by the 6-minute walk in these patients, impressed by the lung function, maintaining lung function.