Capricor Therapeutics, Inc. (CAPR) Earnings Call Transcript & Summary

All right. It looks like a good number of folks are connected online. So we're going to get started. So welcome, everyone. Thank you for joining us today for this webinar with Capricor Therapeutics, where they're going to be providing an overview of some of the new HOPE-3, 2-year open-label extension data and then having an update ahead of the November PDUFA date. Just some general housekeeping. This meeting or webinar is being recorded. So if you have to step away if you have someone that this would be relevant. Two, the recording should be up on the website in about a week's time. And then at the end of today, we will have an open Q&A. So during the presentation, if you have questions, there's a Q& box at the bottom of your screen. Please questions there. We will try to get through those and the ones that were submitted ahead of time as we can. So we're joined by a few members of the Capricor team as well as Dr. Craig McDonald and Dr. Chet Villa today. So we'll hopefully be able to answer all of your questions. And I think with that, I'm going to hand it over to Linda Marbán to get us started.

Good morning or good afternoon, depending upon where you are. I want to thank you for joining today's call. I also want to thank PPMD for making this presentation possible and look forward to your questions at the end. Today, we will present an update on deramiocel for the treatment of Duchenne muscular dystrophy, including our updates on our most recent data presented at the World's Muscle Society at Hiroshima Japan last week. Next slide, please. Next slide, please. This is Capricor standard forward-looking statements that we presented as a publicly traded company. To remind you, our filings are up-to-date and available on the SEC website. Next slide. Well, if you just heard from Eric, joining me on today's call is Dr. Nathaniel Hogan and Dr. Adam Calvin, 2 Capricor's lead bio-esthetician as well as Dr. Craig McDonald, our national PI who hails from the University of California Davis as well as Dr. Chet Villa, a pediatric cardiologist from Cincinnati Children's Hospital Medical Center, obviously, in Cincinnati, Ohio. We are looking forward to updating the DMD community on the recent developments around deramiocel, including what has been submitted to the FDA. We will not take questions today regarding our interactions with FDA as they are sensitive, and it is important to maintain that confidentiality. Thank you for respecting that guidance, and we look forward to providing updates as they become available on our BLA applications. Now let's talk about deramiocel and Duchenne muscular dystrophy, the important topic of today. Next slide, please. While we were disappointed in the vote by the advisory committee, we were pleased that the FDA recognized and hear the voices of the patients, primarily in the open public hearing and decided to extend the PDUFA date to November 22 in order to review the additional data from the open-label extension and the supportive evidence that HOPE-3 did, in fact, demonstrate a clinical as well as statistically significant effect on treating both upper limb dysfunction in a primarily nonambulant DMD population and also in treating the cardiomyopathy associated with GMP. We believe exciting OOE data presented today will be important in moving that program forward. To remind you, the HOPE-3 data showed a 4.55% improvement in performance of the upper limb 2.0 and those treated with deramiocel at 12 months compared to a placebo group based on the statistical analysis plan we submitted, which was statistically significant, showing the trial met its primary efficacy endpoint of performance of the upper limb 2.0. We use percent change from baseline as a primary efficacy endpoint because as data continue to emerge, it appears that the best way to assess the performance of the upper limb and also the clinical meaningfulness to a person with DMD primarily based on the work of Aviva Gillman and [indiscernible] published in 2025. However, we saw a statistically significant and clinically meaningful improvement and absolute change from baseline in PUL 2.0. So no matter how you look at it, the data shows a benefit in sustaining upper in function in those with advanced DMD. I don't have to reiterate this to you that HOPE-3 represents one of the largest DMD data sets ever collected. And therefore, there is true strength in the data as you will see in the presentation from Dr. Nathan Hogan in a few minutes. FDA granted us a refinement of the indication from cardiomyopathy to one more focused on what HOPE-3 was powered to detect which was a change in upper lymph function. The new indication, therefore, is for treating those with upper limb impairment at ages greater than 9, which encompasses a large proportion of those with DMD. Next slide. Now as I just mentioned, the new refined indication is treatment of upper limb impairment and those older than age 9. Now how did we get there? Let me provide some context, please? Most of you know we filed in PLA based on the HOPE-2 and the HOPE-2 OLE results based on recommendation from FDA that we do so. Upon their review, they decided the data set was too small and issued a CRL in July of 2025, asking for a new trial with DMD cardiomyopathy as the primary efficacy endpoint. However, since we had a complete data set in HOPE-3 and 106 of you had enrolled in that study, we offer them HOPE-3 as a way to tier the CRO. They agreed and requested we lead the indication of DMD cardiomyopathy and maintain the primary efficacy endpoint at PUL 2.0. They stated in writing regulatory flexibility when they reviewed the data. As you know, based on their briefing book for the Advisory Committee meeting, the interpretation of the HOPE-3 data was very different from ours, primarily based on the fact that they use the statistical analysis plan version 1.1 that was never meant to be applied to the full data set. In addition, the key secondary endpoint of HOPE-3 was left vetricular ejection fraction in the ITT population. ITT is intent to treat, and that means everybody that enters into the trial has to be analyzed. 1/3 of those in the intent to treat population had healthy heart. So while the data was strongly suggestive of clinical benefit, the strictest interpretation of that data did not show left ventricular ejection fraction as statistically significant. However, and I want to reiterate that it's essentially our numeric interpretation because you can't make a healthy car healthier. This made interpretation of the voting question by the Advisory Committee regarding cardiomyopathy as an approvable endpoint more challenging. Today, Dr. McDonald will show data the dose of DMD cardiomyopathy has sustained and statistically significant improvement in heart function as measured by left ventricular ejection fraction which is most important -- which is the most important group to assess the benefits of deramiocel in treating the cardiac dysfunction associated with DMD. You will also see that the upper limb data continues to be strong and that there is the durability of effect. We are delighted that based on the strong support from the DMD community on the major unmet medical need, FDA allowed us to submit a major amendment and the data submitted to FDA and those data from the open-label extension recently presented at the World Muscle Society will be shown here today for you. Next slide, please. To remind you and for your review, the results from HOPE-3 were recently published in the Lancet, 1 of the top medical journals in the field and so we will not focus on that data today except to provide the context I mentioned earlier. We are proud of the results from HOPE-3 and believe that the Lancet reviewers carefully assess the data prior to accepting the manuscript. Please feel free to reach out to us at Capricor or any of the authors of the paper for further context around the HOPE-3 data should you have any questions after today's presentation. Next slide, please. To my view, we have been looking at treating DMD for a long time now. And what we take great confidence is the consistency of the data across multiple clinical trials as well as over 1,000 infusions over 8 years. Each clinical trial from HOPE-2 onwards show clinical relevance with at least a 1-point change in the performance of the upper Loom 2.0 and those treated with deramiocel compared to the placebo group for the natural history cohort. In a few minutes, Dr. McDonald will show you what a one-point change looks like and many of you on this call can tell us what a 1 point change feels like. Importantly, 1 key element of alignment with the FDA has been that a 1 point change in the full 2.0 is clinically relevant. I would like to take you -- thank you for taking time out of your day to join us and hear about the new data from the open label extension and also why we believe in the strength and the lack of fragility of the HOPE-3 data. Now I will ask Dr. McDonald's to present the new 2-year open-label extension data from the HOPE-3 clinical trial. Thank you. Craig?

Great. Thank you, Linda. And I'm really pleased to be able to present these updated data that really focus on longer-term treatment effects in the HOPE-3 trial participants and I'll be discussing today an overview of the HOPE-3 open-label extension study where patients were followed out for an additional year after the original randomized placebo-controlled clinical trial. We'll discuss some HOPE-3 free open label extension skeletal upper limb functional results as well as the cardiac -- some cardiac MRI results, and we'll also provide some update with regard to the open-label extension safety data as well. Next slide. And before I get into the data, what I'd really like to emphasize is that a 1 point change in the PUL 2.0. There's no question has been determined to be clinically meaningful. The FDA has actually provided feedback to Capricor on a number of occasions, written feedback that they believe a 1 point change in the PUL 2.0 was clinically meaningful and clinically relevant. In addition, I was actually quite pleased to see that the FDA actually in their slide presentation actually stated in their slides that they believe that a 1 point change in the PUL 2.0 was clinically meaningful and clinically relevant. And just to illustrate this, this actually demonstrates on one of the functions of the PUL to bringing hand-to-mouth function, which has been deemed by the DMD community to be 1 of the most important functional issues related to quality of life in a non-ambulatory population, the ability to bring a hand to the mouth and feed oneself independently. On the left, if we can click this first image here. This patient actually is bringing their hand the cuff fully to the mouth without compensation, they actually have a 200-gram weight in that cup and they're easily able to bring that cup to the mouth without any compensation. This will be scored as a score of 2 on the bringing hand to mouth function. If we go to the next slide, this is a patient with more advanced DMD. This patient actually is still able to bring the hand to the mouth the cut to the mouth, but they do so with compensation. And so -- as you can see here in this image, they have to lean forward quite a bit. They use their other extremity to help bring the arm up to the mouth. So they would actually be scored as a 1-point score of the bring-hand-to-mouth function. So the difference between the first image the first video and the second video is essentially a 1 point change in the PUL 2.0 assessment going from a score of 2 normal ability to score 1 with difficulty in compensation. I think most of you will agree that the differences there are certainly clinically meaningful to a patient with Duchenne. If we can go to the next slide, I'd like to, again, just bring people up to date with regard to the Phase III open-label extension study design. So year 1 compared patients who were treated with deramiocel with placebo in a double-blind fashion. So essentially, these patients were followed for 12 months, they remain blinded to the treatment effect, the clinical investigators, the clinical evaluators all remained blinded to that treatment effect. And then the placebo patients then essentially had a delayed start in beginning deramiocel. So after 12 months, they actually transition to open-label extension treatment where they were actually treated with deramiocel. And then the other patients that actually were originally treated with deramiocel continued deramiocel, so we can actually see the treatment effects over 2 years in the deramiocel-treated patients in the original placebo patients of the delayed SARP patients we can actually essentially look at the treatment effect with each individual patient, the impact of treatment for that patient transitioning from placebo to deramiocel. And so there were actually 98 patients, 49 per group that actually entered the second year of the study. 82 of those patients completed a full 2 years of study, 40 of the early treated patients and 42 delayed treated patients. There were some patients that actually right around this time, the there was an FDA approval with accelerated approval for instance, for gene therapy. And some patients opted to transition to commercially available treatment, such as gene therapy. But 82 patients actually completed a full 2 years of study. In addition, 85 of the patients actually had core obtained at month 12, 40 early treated patients, 45 of the late treatment patients. And then 69 of those patients went on to have MRI scans at month 24. 31 early treatment and 38 of the delayed treatment patients had cardiac MRI studies at month 24. If we could advance to the next slide. So I think an important consideration is what is the balance of the patients that actually with early start patients versus the delayed SAR patients. The patients on average were 15 years of age at randomization. 84% of the patients were non-ambulatory in both groups are not walking independently. And if you look in the early start treatment patients versus the delayed start treatment patients, they're actually quite well balanced on a randomization, 15.3 years versus 14.4 years. The agent DMD diagnosis was very similar. The percentages of patients who are non-ambulatory, 83.7% of deramiocel early start patients versus 3.7% of the placebo delayed start patients. Their PUL entry score, the proportion of patients that had lenses we're quite well balanced between the 2 groups and also the proportion of patients that had PUL entry scores of 4 to 6. Those scores would be patients that were able to reach up to the shoulder, to the scalp or fully overhead. The proportions of the patients who were very similar across the 2 groups. And the -- in fact, the total PUL 2.0 score was well balanced across the 2 groups. Importantly, there was a group of patients that had genetic mutations that were actually predicted to the slow progressors, just 1 of the patients in the deramiocel early start group were in the slow progressor Group. And 5 of the delayed start patients that were originally in the placebo were actually in that slow progressor group. So that distribution of slow progress, it actually favored the placebo, at least in the first year of the trial. In terms of proportions of patients on standard of care systemic oral corticosteroid treatment 100% of the deramiocel treated patients, 98% of the placebo-treated patients were on standard of care steroids. Next slide. So just to remind you, the HOPE-3 clinical trial actually did meet its primary prespecified endpoint, which was a change in the PUL 2.0 total score. This study originally was actually when we expanded the cohort to a full 106 patients -- the study was actually powered statistically powered to show a 1.1 point change difference in the PUL 2.0 score. So the original powering of the study was actually based on the ability of the study to show a 1.1 point treatment effect on the PUL 2.0. There was some confusion, I think, in the ADCOM over a 1.5 point change. But actually, the study was actually statistically powered to show a 1.1 point change. And in fact, in the prespecified primary analysis, which was percent change from baseline, there was less decline in the treated patients versus placebo by 4.55%, a p-value that was statistically significant using the prespecified primary endpoint analysis as a prespecified sensitivity analysis that was submitted prior to the unblinding of the study, there was actually a sensitivity analysis based on the absolute change in the PUL 2.0. And in fact, that absolute change actually was 1.1 points. So there was 1.1 point less decline in the treated patients versus the placebo patients. That p-value also man statistical significance, nominal statistical significance of 0.05. And I think it's -- I think, rather remarkable that the study was powered to show a 1 point change on the poll. That was the original hypothesis of the study. And indeed, that's exactly what we showed in this Phase III randomized placebo-controlled trial was 1.1 point change on the PUL 2.0, which again, has been deemed by the FDA as well as investigators in the field to be clinically meaningful. -- if we go to the next slide. And I want to just -- before we advance, I want to just -- if we go back, but I want to just make the comment that even though both groups here appear to be declining, there's about slowing of the rate of decline in the durmiosel-treated patients, over slowing of disease progression in the germiosil-treated patients. And when you think about that, when you actually treat patients over many years, they're when you have a slowing in the rate of progression, there tends to be a cumulative effect over time. So the difference between the treated patients and the untreated natural history controls or the -- what would be expected in the Duchenne population. -- the actual magnitude of that treatment effect becomes greater as you follow patients out to 2 years and 3 years of time. If we can go to the next slide. This actually shows the hope 3 open-label extension results really showing that the benefit really is associated with the treatment with and not just simply the passage of time. So if we look at the data first on the left, these are patients actually the delayed start patients that were placebo in the first year. So the dotted line is that patient group's disease progression over the first year of the trial. And the red line is that same group's treatment progression over the second year of the trial after they've started on deramiocel. And the difference we actually see over a 12-month period of time, is actually a 1.56 point difference. So they actually do much better. They have much less slowing progression in the second year of treatment, which is when they start deramiocel as shown in red as opposed to the placebo year which is the dotted data shown there on the left. And interestingly, when you look at the patients treated with deramiocel over the 2-year period of time, really comparing their disease progression in their first year of their treatment with deramiocel versus their disease progression in their second year treated with deramiocel, we see very similar rates slowing of rates of disease progression, comparing their first year of treatment on deramiocel, they are treated as shown in the dotted line versus their second year term. So the difference here 12 months is really quite negligible. There's really no difference in their first year disease progression versus their second year disease progression. I want to point out here that the axis, the scales of these 2 graphs are actually different. And there really is slowing of progression they are seen in the deramiocel-treated patients. We see very similar data looking at the mid-level PUL data with essentially a 0.9 point treatment difference and a 1% change value almost 7% comparing their first year of treatment versus the second year of treatment. So we're not only seeing a change in the open-label extension with deramiocel treatment on the total PUL, but also on the mid-level PUL as well, which has actually been chosen to be a primary end point in many emerging new Duchenne muscular dystrophy clinical trials looking at the non-ambulatory Duchenne population. Please can go to the next slide, please. And importantly, are these results really quite robust in terms of is this just a function of some peculiar features of the of the placebo-treated patients. And what happens when we actually compare the long-term treatment effect of deramiocel, both on the delayed start patients shown in the solid red here, and the deramiocel-treated patients shown in the solid blue with natural history data, polished natural history data that has been following patients with Duchenne over a full a 3-year period of time. And that data is actually shown in the black dotted line there. This was actually data published by Karate fairly recently in 2025. the predicted chain over both 2 years period of time in 3 years period of time, it shows almost an 8-point decline in the in the performance of upper limb over 36 months at a greater than 5-point decline in the performance of upper limb over the 2-year period of time. In addition, there has been a recent prognostic modeling data that has been published and actually has been put together by the analysis group. A very sophisticated group of statisticians that have been doing a lot of work in Duchenne dystrophy, in addition, working with data from if next in Europe where they've actually developed a prognostic model showing where we -- based on the initial baseline characteristics of each individual patient, what would their predicted 12-month and 24-month trajectory be with the performance of upper limb. And that's the data shown in the dashed line there -- the red dash line and the blue dashed line. And so essentially, if you take each patient's individual baseline characteristics, how severe they are, what their clinical characteristics are a baseline. And so in this case, we compare the solid blue line which are the 2-year deramiocel-treated patients with the prognostic model for that group over 2 years. You get a minus 3.4 point decline in the deramiocel treatment, whereas you get a minus 5.88 point decline over 2 years' period of time in the prognostic model patient. Similarly, in the delayed start patients, the solid red line there, I think, which is really quite striking is those placebo patients actually followed the published and the predicted trajectories during the first year of treatment. Those lines almost completely are superimposed or overlapped. And then as soon as the delayed start treated patients start deramiocel, you can see in the red solid line, there's almost a hockey stick phenomenon there where the patients all of a sudden start to progress more slowly over time. And you can really see the deramiocel-treated treatment effect there as those patients transition from placebo treatment to their second year of treatment. And so again, those patients compared with their prognostic model. It's not quite as robust treatment effect as we see with the 2-year deramiocel-treated patients. But nonetheless, they have a minus 3.7 point decline versus the prognostic model of minus 5.3 points. And that's with only 1 year of deramiocel treatment with the delayed start. So I think the key takeaways here is these early start patients show up minus 3.4 point decline versus a 5.8 point decline 2 years of deramiocel-treated patients. The delayed start patients, again, are showing a minus 3.76 point decline versus the prognostic model where there's a 5-point decline after switching a month 12. The decline slows once placebo patients start on deramiocel and so these open-label model-based comparisons as well as the comparisons with the published natural history data are actually quite supportive of the randomized Hope 3 clinical trial results. If we go to the next slide, I'd like to now turn our attention to the really important cardiac MRI data. And so while the BLA amendment has really switched the focus to the upper land because there's really more robust upper room data. I think it's still important to point out that we're still quite optimistic and quite excited about the treatment effects we're seeing with cardiomyopathy. And just to remind you, in the clinical trial, if we actually look at the patients that actually have impaired cardiac function when they entered the study. And there were 64 patients out of the 106 patients in the clinical trial that actually had impaired cardiac MRIs and have evidence of cardiomyopathy. If you look at the deramiocel-treated patients from the clinical trial, they had essentially complete stabilization of the progression of their cardiomyopathy over 12 months, whereas the placebo-treated patients declined as anticipated. And that treatment effect is about 2.8% and data that has been published by cardiology colleagues, Dr. Sasan and his group and collaboration with Dr. Bill and others. It's shown that about a 3% change in cardiac MRI is actually predictive of a significant improvement or a difference in long-term survival. So this 2.8% change, although it seems modest is actually clinically significant in terms of being predictive of survival. And statistically, this was statistically significant in the clinical trial. When you look at the overall population of all 106 treated patients that looked at patients who actually had normal heart at entry into the study. And you wouldn't have ever designed a study necessarily to show a cardiac treatment effect by including patients that already had normal hearts with a short-term trial over, say, just a 12-month period of time. So it wasn't surprising that we see a more robust treatment effect with the cardiomyopathy population. If we go to the next slide. And again, when we look at the open-label extension cardiac MRI data in both the ITT population, these are the patients, again, that included patients that did not have cardiomyopathy. We see a 1.8% predicted change percent change in cardiac left ejection fraction. And at month 12 there, there's -- month 12 was a 2.4 point treatment effect. We're still seeing a 1.8% treatment effect there. If we go to the cardiomyopathy population, we're still seeing, again, a 2.7-point change treatment effect on left vertical ejection fraction over 12 months and a nearly 2% change over 24 months. So again, I think these data really point to continued preservation long term after 24 months of cardiac MRI left ventricular ejection fraction. It stayed near the baseline with patients on deramiocel, whereas the mass patients from the national history -- external natural industry controls actually continue to show more significant decline over that 24-month period of time. Next slide. Again, to summarize the safety results. What this infusion actually involves because of concerns about hypersensitivity reactions. We treat patients with a single dose of oral prednisone and antihistamines, essentially, they take either 1 or 2 just single administrations just prior to the infusion. And then H1 and H2 blockers to help prevent reaction, they usually get their deramiocel infusion over 60 minutes' time and then they're observed for about a 2-hour period of time by the care team following the infusion. And the long-term on label extension data shows that the side effects were well balanced between groups, mostly mild, short-lived and manageable. The hypersensitivity reactions were generally mild to moderate reactions, premedication with the high-dose oral steroids and H1/H2 blockers largely reduce the severity in the occurrence of these hypersensitivity reactions. And as we follow patients now in the HOPE-2 and HOPE-3 open-label extension trials, out to a period as long as 5 years, we've seen no known long-term safety concerns with deramiocel, which I think is really important to point out. Next slide. So with that, I'd like to transition to Dr. Nathaniel Hogan in Biostatistics from Capricor Therapeutics.

Thank you, Dr. McDonald, and good morning or afternoon, wherever you may be. Over the next few slides, I'll walk you through 2 pieces of work that we've delivered to the FDA after the advisory committee. The question behind both is pretty simple. Is the reported benefit of deramiocel real? Or could it depend on how the analysis was planned and conducted? Next slide, please. We know the inventory Committee left many of you with questions about our statistical choices and we've heard that, that concern from families and clinicians. We take this very seriously in our aim today is to show you exactly what we did and what we found. The word that came up most at the advisory committee was fragility which is the worry that a result might rest on a few patients or 1 analysis decision. We responded with 2 deliverables. The first is a robustness report on HOPE-3 alone which whether the primary result holds when the assumptions and choices behind any change. The second is an integrated summary of efficacy, which pools hope to and hope so we can look at the effect with more total patients in a single analysis when we take those 2 in order. Dr. McDonald discussed earlier, the primary endpoint in HOPE-3 is the PUL 2.0. It's a validated sort of upper limb function, can [indiscernible] lifted arms, bring it up to its mouth, use his hands, et cetera. Our pretested analysis showed deramiocel slowed the loss of that function, as Dr. McDonald mentioned by about 1 point over a year. And 1 point, as we've mentioned on the scale is clinically meaningful and relevant. Next slide, please. Now here's the key idea. The hallmark of a fragile result is a treatment effect that moves drastically when you change analysis decisions, whether you swap the model, handle missing data, maybe the benefit shrinks materially or even disappears. So we tested the many choices and options in the HOPE-3 analysis to see if the treatment benefit disappears or even favors placebo. These analyses are agnostic to any statistical analysis plan Instead, they explore all the different choices made when modeling the data to see if any of those choices display for Jit. We ran more than 600 alternative analyses. We are the model, the missing data approach, the patients included and how progression is defined. Every thought on this chart on this slide represents one of those analysis results. the position left to right marks the magnitude of the treatment effect that's the resulting of the analysis results in. The size of the dot represents a number of analyses that produced the estimated treatment effect. You can see that the vast majority of analysis show a treatment benefit at or above 1 point with none even approaching the 0 line, which represents no benefit. In other words, HOPE-3 simply does not show the signature or hallmark of fragility. What that tells us is that the differences in P values presented by us and by the FDA at the advisory committee are more a result of disease heterogeneity and the sample size are not the result of an unstable or serious treatment effect. The size of the benefit is stable and a larger study would make these analysis choices less impactful on the PVIs. This is exactly what the innovative analysis that are shown next lets us test by combining data from both HOPE-2 and HOPE-3. Next slide, please. HOPE-2 and HOPE-3 were both placebo-controlled trials of deramiocel and DMD with the same dosing administration and assessment so they can be analyzed together. The table shows who contributes to each 62 boys on deramiocel and 64 on placebo had PUL assessments, 45 versus 50 with the valuable cardiac MRIs and 35 in both groups who already had cardiomyopathy at baseline. For a rare disease, this is a very meaningful sample. Next slide, please. This is the main PUL result of the PUL 2.0 total score. Blue as deramiocel and red as placebo and the vertical axis is estimated change from baseline to down net losing function or progression. Both groups declined but at 12 months, patients on placebo have lost about 2 points and those on deramiocel less than 1. The difference is 1.29 points with a p-value of 0.0087. Another way of saying that is if the drug did nothing, a gap is large would appear by chance less than once in 100 times. As a percentage of the score deramiocel preserved about 5 percentage points more function with a p-value near 0.005. This is the same effect size that we saw in HOPE-3, but now just with more patients and a clearer p-value. That's precisely in line with the picture we just presented by the robustness analysis on HOPE-3. It's important to note we see similarly strong results in the middle PUL when [indiscernible] studies, which is also consistent with the HOPE-3 results. Next slide, please. Now the same stress tests on the pool data, and this is where we address the advisory committee concerns 1 by one. Each row here is a prespecified alternative to a question raised in the committee, how to treat missing composite values, recomputing the flow progressor flag, a different correlation structure keeping visits after primitive therapy and 2 conservative missing data assumptions that assume dropouts behave more like placebo patients. All these are modeling or analysis decisions. In this force plot, the dot and cap lines are the estimated benefit and accompanying 95% confidence intervals. 0 means no effect, and the dash line at 1 signifies the clinically meaningful threshold. Every estimate, as you can see, of the treatment benefit lands between 1.1 and 1.3 points above the clinically meaningful threshold and as important, every p-value is below 0.2 all analysis on the PUL data demonstrate an even more consistent treatment effect, supporting the benefit of deramiocel for treating skeletal impairment in DMD. Next slide, please. The last piece is cardiac function, which for most families, we know, matters a great deal. Left ventricular ejection fraction or LVEF is the share of blood the heart pumps with each beat and in DMD, it falls steadily, as shown by natural history, generally by Dr. McDonald mentioned about 2 to 3 points per year. The last plot shows the full population. In this population, terminal is essentially flat at 12 months, while placebo declines about 2 points. This results in an estimated difference of 2.2 percentage points with a p-value of 0.05. Importantly, the sickest 5 patients, those with the lowest LVEF were all randomized to deramiocel in HOPE-3, which makes this result even more striking. On the right, the trajectory of patients who already had cardiomyopathy at baseline is also displayed. The difference at month 12 is 3.1 points favoring deramiocel with a p-value of 0.04. Again, macro history and nonambulant DMD consistently shows a decline of a couple of points per year, so holding out the ES steady as a real change in their trajectory. Next slide, please. summer, we tested the HOPE-3 result more than 600 ways and the effect did not materially move, which is the opposite behavior of a fragile result. Furthermore, we increased the sample by pooling with HOPE-2 and the same treatment effect appeared with an even stronger p-value confirming the original uncertainty was not about the efficacy of the drug, but rather about sample size. This is a common concern in rare and heterogenous diseases. We ran every since [indiscernible] analysis that was a topic in the committee, and all 8 agreed confirming the benefit of deramiocel. Cardiac data points the same way. We wanted to put all that in front of you transparently and -- and now I'll turn the time back over to Linda.

Thank you, Craig, Dr. McDonald. Thank you, Nathan. These are important data that we've made available, I know there's a lot of it. So happy to take some questions. And also, please feel free to reach out individually and independently I guess what I'd like to close is to extend our gratitude to the Duchenne community for signing up for the clinical trial for showing up at meetings for reaching out to the FDA for continuing to be a voice of advocacy for yourselves, for the unmet medical need. For those of you that are in this later-stage Duchenne population, I looked out at your faces when I presented at MDA. I looked at your families and your faces at the open public hearing. And I can tell you that -- we will continue to work as hard as we can to make termite available to you and your physicians decide that would be part of your care paradigm. We also want to thank PPMD and all of the other organizations that have worked with us already many years now. as we continue to develop deramiocel. So with that, I'll stop, and I will ask for questions which we can hopefully answer and clarify for you.

[Operator Instructions] And there's a lot of data. I think walk back through some of it just to make sure to understand, but that was pretty robust. So if we could go back to maybe the baseline characteristics slide, and there's a few I kind of want to walk through quickly. While we're doing that, though, Linda on the webinar a few times as mentioned, PUL versus the mid-level PUL. And just for families who maybe don't know kind of what that is and the difference there. Could you just talk about that and the significance of PUL?

Yes. So I think for those of you that aren't aware, the performance of the upper limb and metrics that was developed based on activities of daily living to assess upper limb function when angulation or ambulatory cars cannot be performed adequately. It's both in his redemand soldier, which is obviously your shoulder and sort of start with the ability to raise your arms above your head to do just something to do on camera. Mid, which is your arm, which would be representative of being able to move essentially your elbow function, and that would be like eat 10 bites or being able to drink or do other tasks that involve your arm. And then wrist in hand, which are the fine motor movements that are controlled by wrist and hand. Mid-level PUL is very important because it's considered to be sort of a Goldilock zone, shoulder declined very rapidly and sometimes it's hard to catch changes in risk to hand muscle function declines very slowly. So you need longer time to study it. Mid-level PUL with the arm changes are typically able to be cost at least in some version within a year and are also really relevant to quality of life for reasons that Dr. McDonald mentioned. So I hope that clarifies, Eric.

Yes. No, that's very helpful. And then I wanted to start here, there was a question in the chat around kind of number of patients that were not ambulatory. And so I just want to make sure that they saw that it was around 84%. There's also a question around the age range of what was kind of the less patients in the study for the understand list.

The youngest patient in the HOPE-3 study with 10 years of age, the youngest patients in the HOPE-2 study was 9 years of age.

And then just to clarify here, so I know Craig had mentioned it. But when we're talking about slow progressor as one of the characteristics, can we just define that really quickly for everyone?

Craig or Nathan, I don't know which one of you would be better able to take on that slow progressive term definition. Maybe, Craig, from a clinical perspective.

Yes. From a clinical perspective, that would essentially include genetic mutations that were predicted to be more associated with a slowly progressive disease course. So for instance, there's a literature on exon 44 skip amenable mutations. Some of those exon 44 skip [indiscernible] patients have very characteristic Duchenne like progression. Some of them actually have a bit more slow progression. We didn't want to exclude particular genetic mutations in the trial, but we did actually define some of those mutations as associated with sole progressor. So that defines this sort of slow progressor subgroup. We didn't necessarily originally stratify based on that. We stratified based on PUL entry scores. So I think that's another really important point to make with regard to the importance of looking at the primary endpoint, which is upper limb function the whole study was powered and stratified really to show an effect on upper limb function, not necessarily cardiac MRI function. But essentially, if you look at the placebo group, in the original trial. There were actually 5 patients in placebo that actually had mutations that were prespecified to be associated with more slow disease progression. There was only 1 patient in the deramiocel treatment group. That actually would typically what -- this would favor actually the placebo group over deramiocel. But even with that imbalance 5 patients versus 1 patient we still were able to show a statistically significant treatment effect in the randomized placebo-controlled trial with deramiocel treatment.

Thanks, Craig.

Yes. That's really helpful. On the -- it might be the next slide is basically the first -- the PUL data. Okay, maybe 1 more. Yes. So here, I just wanted to make sure that families understood because I think it's a different way sometimes of some of the data being presented with kind of the overlap here. So on the left-hand side, where you presented this data, just so they're kind of aware that, that first dotted line on the delayed start, those are the patients who were in the placebo group to not receiving anything and we see that there's no -- roughly, we'll see kind of that 2-point decline. And then the red line is at that point after the 12 months started on treatment. And then that -- what was the more sharp decline slows to kind of this, again, roughly 0.5 maybe. So we're seeing a change in that the rates of decline essentially between those the year 1 treated and then when they are put on treatment.

That's right. And that's what we've seen consistently. And I think my last slide, which don't -- do I think this is an important line to say shows that we consistently see slow the rate of decline year over year over year and some of the OLE patients on HOT, we've seen that for 5 consistent years. to look at this graph, and I know it's complicated. It's even hard for us sometimes is to look at the left graph of the red line is the same patients, red line is year 1 on placebo. Dotted line is year 2 on deramiocel -- and yes. Sorry, the opposite. Dotted line is year 1 on placebo, red line is the placebo patients that have shifted over to treatment with deramiocel. And what we can show using 0 is the baseline is a delay in the rate of decline. On the right side, again, you're seeing the exact same paradigm, but these are patients that have had 2 years of deramiocel treatment. And the most important thing is that the dotted line is their first year on deramiocel during the randomized portion of the trial and the blue line is during the opening label extension on the second year of deramiocel. So those guys have 2 years of treatment. And what you can see is there's no difference in their response between year 1 and year 2. And so that's really important to acknowledge here that there's really no difference in the treatment between year 1 and year 2 with a significant delay in disease progression over both years.

Yes. So the rate of decline is staying consistent win on therapy, which, as Craig mentioned, the hope of something like this is -- and we talked about this, we try to convey that to the regulators a lot from the community perspective, and I think, Craig, you did that nicely in terms of the cumulative effect that if we can slow the disease, these things might drastically push out to these milestones, even though it's kind of still a decline, but the rate of decline is changing, which is important. And then if we go forward, maybe 1 more on modeling. I just wanted to make sure families understood. So the black dotted line here is actual patient data from a natural history versus the dashed lines, which are modeled data. So what was projected to be similar to the patients in the study?

Yes.

I could take this one. So again, I think the black dotted line is the published natural history data, which I think shows a linear decline occurring continually over over 3 years. When you take natural history data like this and what was done here is data from Eric Nick, who has a really robust data set actually took that data. And then essentially, what they did is they developed a prognostic model for what an individual patient would likely have what their trajectory or their progression would likely be over the course of 24 months. Based on their baseline characteristics, how strong they were at baseline, what their initial PUL score was, their PUL entry score, the -- and other age-related factors and so forth. And so for each individual patient, you can actually get a projected disease decline or trajectory over the next 24 months, kind of what each patient would be predicted to do over the next 24 months. And so the data -- the solid line data here is the whole 3 OLE data, the blue line is the patients on dermisel for all 2 years of the follow-up. The red line, which is you see that hockey stick phenomenon, with patients slow their progression or the placebo patients that had delayed start in their treatment. So rather than taking an external control population and try to cherry pick out a group that would be so-called propensity matched to that group. A very non-biased way of looking at a treatment effect over a longer-term open-label extension time is to apply a prognostic model to those patients that you follow for a long period of time. And in this case, you can see pretty clearly that both the deramiocel 2-year treated group and the delayed start group both have a much slower rate of decline than the prognostic model shown in the dashed lines would predict them to experience.