All right, we're ready to get started. Welcome everyone. I'm Josh Schimmer from the Cantor Biotech Equity Research Team and very pleased to introduce from four d Molecular Therapeutics. We have David Kern, Co Founder and Chief Executive Officer and Chris Sims, chief commercial officer, going to talk all about the gene therapy portfolio of four d molecular. David, start us off with a snapshot of where you are in advancing the programs.
Yeah. Thanks, Josh, for having us. It's great to be here. Great conference. So we are we have two lead programs in genetic medicines.
One is 4,150 for neovascular diseases of the retina, so wet AMD and diabetic eye disease. And then we have 4,710, which is our cystic fibrosis aerosol delivered vector for cystic fibrosis patients initially who are not amenable to modulators and then ultimately for a broad population. The four thousand one hundred fifty is in two Phase III trials, FORFORNTA I and FORFORNTA II. We're seeing excellent and very exciting enthusiasm and enrollment for those studies. We're thrilled with the recent PAT survey results from the ASRS showing that fifty percent of physicians queried, said gene therapy is their most exciting novel approach in wet AMD and that, again, durability of therapies remains the number
So we think we're well positioned. We're in a strong cash position. We have cash out into 2028, and we expect to complete enrollment of FARFront one in Q1 'twenty six, which means we'd hit the fifty two week endpoint in '7. That's a high level summary.
And then the CF program?
Yeah. So CF, very exciting program, 04/2010, aerosol delivered vector that we invented through directed evolution that expresses CFTR. There, we expect to we've been in a Phase I dose finding study where we've been looking for the optimal dose for that product. We actually found that we were overdosing at initial doses, that we were the product was overperforming in humans compared to nonhuman primates. So we've actually reduced the dose for now into dose level three, dose level four, much more physiologic expression and we expect to give readouts on lung function, lung architecture and quality of life in Q4 this year.
Great. Excellent. Well, why don't we dig in on forty one thousand five hundred first? Just kind of refresh us on the signals that you saw in Phase II that give you the confidence in the Phase III program.
Sure. So again, intravitreal administration, simple topical corticosteroid regimen for twenty weeks, so very established dose and regimen. We saw a strong dose response in the Phase I and Phase II and Phase IIa IIb as well as in DME, so we were very comfortable with the 3e10 dose level, showing a treatment burden reduction anywhere from eighty three percent to ninety four percent. We saw the highest, you know, reduction in anti VEGF utilization in the most recently treated patients, and that's the patient population we're going to focus on in Phase III. So safety is critical in this disease.
So we've had a very excellent safety profile, we believe, to date with less than three percent one plus transient inflammation seen and none of the bad things that we worry about in retina. Clean safety, injection burden reduction, strong expression of aflibercept that's been stable over time.
So in the PRIZM study, you did kind of identify the frequency of anti VEGF injections leading into the clinical trial. I don't think we've really seen a snapshot of their OCTs, their visual acuity, in those months leading up and what the triggers for the reinjections were, which may be informative for interpreting kind of the patient's before and after scenario. You have any of that data? Is it available to review? You know, it's not
collected in a way that we think we can make meaningful assessments. You know, on study, when they come on study, there's a third party independent Duke Reading Center, one of the world's top in neovascular retinal diseases, and they confirmed highly active disease. We confirmed response on bolus and aflibercept as the patients came on study, but we don't think we have useful data there. I think certainly, over time, though, we have very detailed OCT and BCVA measurements and with the Duke Readings Center.
Sorry, Josh, maybe one thing to add. Dave, I think and correct me if I'm wrong, I think those patients predicted a Phase I and IIa patients, so those that were getting, on average, 10 issued injections in the year prior, I think their baseline OCT coming into study was in the 400 range. $4.25, yes. Which would signal that they had fairly active Westerdahl: Westerdahl:
disease, so that despite having 10 plus injections in the preceding year, still had an OCT in the four twenty five range. Yeah, and confirmation of active disease from From
Yeah.
So, you've made a number of changes from phase two to phase three in terms of the design and reinjection criteria, etcetera, which you lay out nicely in the corporate deck as well. Maybe you can kind of review what the changes are, why you made the changes, and then number three, how do you think those changes may impact the results?
Yeah. So, this is a critical issue for any drug development program, right, is when you go from Phase II to Phase III, what, if anything, do you change and why, right, because you want to make sure that you, you know, set yourself up for success. So when we think about it, we start with, you know, what did we not change? So we're very excited to have the three ten dose levels locked down. It's shown great results and we've seen a clear dose response versus the 1e10 dose level in all wet AMD populations and in DME.
So the dose is not changing. The steroids, we've been very fortunate that we locked in on this topical Durezol twenty week taper that hasn't changed. So that's unchanged, excited about that. That derisks things. The endpoints in terms of following BCVA non inferiority and OCT are unchanged.
The patient population is refined. So we have, I believe, had a robust development program here where we started in the most high need patients, which, you know, may be the worst ten percent, which I think is good drug development first in human. But then we've also treated all the way to the other end of the spectrum patients treated in the last six months, and that's the population we're going to focus on in Phase III. So it's and then we've refined that population even more by saying we have to show that they respond to bolus anti VEGF on study to confirm that, so that will weed out maybe ten percent, fifteen percent of the patients at the most. So refining that population and then patients will come in treatment naive but get three bolus anti VEGFs on studies, so they'll ultimately get that standard of bolus.
So we've refined that population. That is a population where we've seen the best results. We think that could be because of, you know, they have the healthiest retina, the best transduction, certainly no fibrosis or atrophy in that population yet. So we've refined it there. And then we've made minor tweaks to the supplemental injection criteria using 10 letters or 100 CST or a combination of 55 for supplemental injections.
That's, again, a refinement that's very consistent with many other programs. So we think, you know, we try to change as little as possible. The reason we went into frontline patients is really FDA and EMA said, That's where we want you to start. The FORFront two does require some patients, forty percent, who've had prior therapy, but everyone is ultimately going to get three bolus therapies on study. And all of the statistics for non inferiority, all the regulatory precedents with Vabismo and high dose EYLEA and even EYLEA itself versus Lucentis all started in those naive patients.
So it's kind of counterintuitive because we know on the market, once it's commercialized, patients will probably start in the highest need patients, but for regulatory purposes, the naive population is definitely the strongest precedent and the most conservative approach and is what EMA and FDA asks us to do.
Can you explain the rationale for the change specifically in the reinjection criteria? And does that let the retinas run a little drier, wetter, or does it drive retinas to be a little drier relative
to Well, I think the baseline will be a little drier, particularly because this is a naive population, so we expect, you know, unlike that first recalcitrant population, these retinas should be pretty dry after three bolus injections, especially since we're enriching for patients who respond. But then we would expect that these criteria, the minor changes we made, would really protect BCVA as a primary endpoint, but still not materially reduce the injection burden reduction, which again, in this population was over ninety percent. So even if that were to drop to eighty percent or seventy five percent, that's still a target product profile. It's a huge commercial win, we believe.
So, in terms of the trial design, we've talked about this a lot, right? Your comparator arm isn't really matched to your treatment arm in terms of the EYLEA, nor is the comparator arm really reflective of what many consider standard of care today, we treat and extend, right? So, if you're comparing injection frequency versus a regimented Q2 month approach, that may or may not differentiate from the real world practice, which is a PRN. So what is the magnitude of reduction you think you need to show to not just differentiate in this trial but to make a compelling case relative to what many consider that standard of care to say, Hey, we're actually making that better. Because you're not studying that. You're going to have to infer it across trials.
Yeah, exactly. So, I mean, we're in the same position that Babaismo was and other products where you have to compare to the standard of care and currently FDA and EMA see that's Q8 Eylea. So that is what it is. We do allow actually rescues on the aflibercept arm as well in this study, which we think should be, you know Anyway, we felt that that was the best comparator to allow supplemental injections on either arm. But your point about real world versus a trial is absolutely true, and Chris can speak a great deal, I think, to the differences between what people see on trials and what happens in the real world.
For Bismol, I think, you know, in a trial supposedly was a Q16 drug, I believe, and in the real world, it's maybe two weeks benefit versus aflibercept. It's actually used more frequently. So there will be differences. But I think for ensuring regulatory approval, alignment with FDA and EMA, we really stuck to the, you know, comparing to the standard of care with the opportunity for supplemental injections. Chris, do want to speak to real world versus trials?
Yeah. I mean, it's not uncommon. Think, you know, in the trial setting, as David mentioned, you need to have some level of control and you've got to follow what's been precedent. And Josh, you know this space well. We've talked about it before.
Listen, post approval, I wouldn't be surprised if we, partner with the community to do a trial that's more reflective of what you just said. It's hard to do that for approval purposes, but the community often does investigator sponsored trials where you start to look at some of those questions. So it's something I think we can tackle in a post approval setting. And I think you're right, the real test at the end of the day is how these medicines perform in the real world. The first test, by the way, in the real world is on safety, as we know, right?
As someone who's launched medicines in this space before that has had safety issues out of the gate, the efficacy could look as good as you want it to. If you don't have a safety profile, then you're severely limited in the commercial potential. And I think, listen, we believe our data the breadth of our data from Phase II because keep in mind, that patient population we referred to earlier, that was a relatively real world type of history of getting 10 plus injections in the prior year. In fact, those patients are probably getting treated more heavily than the average patient in the real world. So we perform well against that patient population, which we think gives us confidence as to how we would perform in the real world.
But ultimately, you're right. Will come down to what does it do in a physician's hands and how does it compare to what the practice looks like.
What is the heterogeneity of practice within treat and extend, specifically for the criteria that they you've already found two, right? You had one in PRISM. You've got a different one in the phase three program. I think we've seen three, maybe, across all the different trials. And so that does make it a little challenging to benchmark, like, what is the natural treat and extend practice profile of, say, EYLEA or VIBISMO?
And again, because that's the bogey for you that reflects the real world to you. And it may even inform just sorry to ramble but you talk about those patients leading into PRISM getting frequent injection. Do you know what reinjection criteria they were using?
Yeah. Nothing standard, obviously, right? Because these are patients that came from different practices that wasn't a controlled setting. So you're absolutely right. I mean, there is no standardized protocol that drives treatment in this practice in the real world today.
Some physicians are, put patients on a Q8, twelve, sixteen regimen and treat them every Q8, twelve, sixteen regardless of what's going on anatomically. Some look for the presence of some level of fluid, be it intra or subretinal, and that guides your retreatment decision. Some have a higher tolerance for fluid, some don't, some just treat based upon vision. So, there is no standard. I think every physician tries to, you know, tailor a retreatment frequency that's best for that patient and what they can tolerate in terms of their ability to come back and trying to control the disease.
But where I think we have the advantage, though, is what no doctor wants to undertreat. Regardless of how they decide to treat, no one really wants to undertreat. So, the potential for a therapeutic like 4,150 to provide continuous expression, we think has significant appeal, particularly if you don't know for sure if you're hitting that retreatment kind of criteria precisely today with a bolus option.
So, presumably, the retreatment criteria will itself determine the reinjection cadence. Now, again, recognizing that there is a heterogeneity of everything here in kind of the real world, coming back to your target profile from phase three, just to illustrate a scenario, let's say you show a thirty three percent reduction in anti VEGF injections, but a doc says, well, wait a minute, I get that with my protocol as well. So I'm not convinced that there's anything. Where do you think you need to be hitting in terms of that decrease to make that statement that resonates with practitioners to say, Wow, I don't achieve this with my regimen?
Yeah, I'll start and then Chris can weigh in. We do know that Vovismo is about a 23% reduction or so versus afliberceptin. It's been a blockbuster. I think it's up to 4,000,000,000
annualized now. But that is more head to head. Do we have enough head to head on Vovismo relative to EYLEA to substantiate better drying? It's a little bit different context.
Sure. Sure. But I think just in terms of injection burden reduction, that's one metric. FDA signaled, look, we'd like you to be at or above 50% informal guidance, not required in the protocol, but certainly guidance. We've been in that eighty percent to 90%, over 90% reduction in the more recently diagnosed patients.
So I think we're set up to be well above that threshold that physicians would want to see. I don't know if you have any more granularity, Chris, on what they'd want to see generally. I
don't think there's a hard cutoff. The danger of putting a number there, it's like you're in a binary situation. But, I mean, internally, I think we would certainly hope to see a continuation of an 80 plus percent reduction in treatment burden. I think some physicians would tell you anything in that 50% range would be highly meaningful for them. If it's in the range of what you referenced, in the 30ish or lower, then that's noise.
And I think that will be a highly challenging way. We're not banking on that at all. But I think, certainly, if we continue to show overall treatment burden reduction in the 80 -plus range, then we're positioned very well.
Okay. Got it. Why don't we turn to the CF program, and there is an interim data update ahead, as you noted. Maybe kind of frame what we should be looking for in terms of number of patients, duration of follow-up, doses that you're exploring.
Yeah. So it's a Phase I dose exploration study, the ARROW study. This is, again, using 4,710 aerosol delivery, very standard routine aerosol delivery using a nebulizer device, and it utilizes a vector that we invented for improved delivery by aerosol to the lung airways throughout, from the largest airways to the smallest. That study, the readout in Q4 will focus will have long term data on the first seven patients at the higher dose levels, but we'll now have nine patients at the lower dose levels, the dose three and dose four, and we'll have anywhere from three to eighteen months of follow-up on those patients. I think what's critical here is, you know, when we first saw that data in terms of biopsy data at two months, we were blown away by the level of expression.
In fact, CF Foundation, our close partners, they kind of looked at that and said, That pattern is pretty unusual. You're seeing staining in the interstitium. You're seeing a higher level than is normally seen in a normal lung biopsy. We realized over time we were probably overdosing. And again, the vector just outperformed in humans.
We've seen that with other vectors as well. So we brought it down to now where we're seeing staining patterns and staining percentages much more in line with physiologic range, and so we do expect to see improvements in lung function based on FEV1 and lung clearance index, lung architecture based on high resolution CT, and then we hope to continue to see these quality of life benefits that we've seen anecdotally throughout.
So, one of the big questions around the early data set was the super high expression, sky high expression, but the FEV1 wasn't predictably moving in a favorable direction. I guess you suggest maybe you've overdosed those Is that would that explain why you didn't see a robust FEV1 benefit? And if so, how? Like, what would the explanation be?
Yeah, it's interesting. I think, look, it's CFTR biology is complex. It's, you know, the way it operates is it's not like a simple enzyme replacement, for example, where just more is better. So, you think we need to be in physiologic range. We don't want to be overdosing.
We don't want to be getting expression of cells in the interstitium, for example, where you can get fluid shifts that could lead to unwanted effects. And so the biology is such that we think you have to optimize dose and so we do think we're now in a therapeutic range. We also are going to be reporting out data with long term biopsies. So we've gotten great two month biopsies, all patients, very reproducible, but we also will now have anywhere from one to three year biopsies where patients volunteer to get a late biopsy to say, Okay, how is expression levels? How have they changed?
And so we think we're, you know, based on that data, you know, we'll share it. We'll be able to show durable expression and our plan would be to ultimately think about redosing every, eighteen months to three years, somewhere in that range based on the biopsy results.
So for those patients who receive the high dose who've got extended follow-up, presumably, as you've kind of alluded to, the expression starts to wane. Yes. And so in theory, you've gone from overdosing patients early on to actually normally dosing patients as the effect wears off. So you should be in that therapeutic window. Would you expect to then see the FEV1 benefit?
The only yeah, potentially, but the only caveat there is if you've overdosed and you've gotten vector into places where you don't necessarily want it, then that could be even counterproductive over the long run. So, again, I think it's going be critical to show safety over the long run, which we've shown to date, but I'd say for the optimal therapeutic benefit, it's really going to be about those lower doses. The other thing is, as you know, FEV1 is quite a variable endpoint, right? So Vertex has used hundreds of patients to show their benefit. The patients we're starting in, the majority of them are null mutations and non amenable, so it is a different patient population and we've had regulatory guidance and KOL guidance that three percent or more FEV1 benefit would be important in those patients.
We're also excited to look at other things like Lung Clearance Index, which looks at the smaller airways rather than just the large airways with FEV1. So it's really there's a lot of exciting emerging science and endpoints here that we can leverage to really understand the totality effect that we're having.
So, how many low dose patients do you expect So,
three at dose level three and six at dose level four, which is a quarter of where we started, 2.5.
Okay. And what kind of follow-up for the lower dose?
Three to eighteen months.
Okay, for those as well. Smaller, In early trial, a 3% FEV1 may simply be noise, right, averaged across hundreds of patients. That's a signal. But how do you think about an FEV1 threshold in a small trial that and it may just be hard to do in a small trial in CF, given the variability. But what in your mind would be a very compelling signal to say, Hey, we got the expression and we
got Yeah. The think it's going be the totality of the data, right? So if FEV1, you're seeing even anecdotal effects of, five plus percent or three on average. That's going to be meaningful, but small numbers and variable endpoint. We'd love to see the Lung Clearance Index also track in the same direction because that's much more reproducible and might ultimately it has been used for approval in pediatrics where they can't do the FEV1 and might ultimately become the de facto primary endpoint.
So I think that's important. We'd want to see some benefit in terms of high resolution CT scanning and the lung architecture. And then, of course, quality of life, uncontrolled, but I think it's going to be the totality of the data given it's only nine patients. And then, ultimately, together with CF Foundation, we'd have to see how that data matures and how Phase II expansion matures to then make a decision on Phase III. But you know, initial guidance on Phase III could be a relatively small, you know, 30 to 50 patient study.
Patients could act as their own control or we could have a small control, internal control arm, And then the CF Foundation also has something called the REACH study, which is like a third party independent control group for use in other studies, so we can look at all three of those as options.
Is of the metrics that we've talked about to capture clinical effect, is the Lung Clearance Index the least noisy? What do we know about that That's in terms of
the how our understanding disrupt
patient
is that's the least noisy and it's the least prone to effort because it's effort independent. It's just basically clearing an inert gas while you're breathing at room air. So it's felt to be the most reproducible and the least effort dependent. That's why it's particularly useful in kids. So we think that's probably going to be the tightest.
Do you have a sense well, actually, couple of questions here. Maybe you could elaborate a little bit to the extent that you have an understanding of it, how the Lung Clearance Index is actually measured and then what counts as a good result?
I'm going to ask for help from the crowd. It's one to 1.5, but what's the units, Julian, on that? Like number of breaths with a gas to be just threshold.
Okay. All right. Now, that presumably depends on FEC or effort, no?
No. Think they're breathing at room air or at normal kind of resting. They're not asked to blow it out at a certain amount of time, so it tends to be more reproduced. One, think it's a one to 1.5 units is felt, decrease is felt to be meaningful. Is that fair? Our threshold probably is minus eight. Minus zero eight. Okay. Minus 0.8.
Are you going to help us with that metric and kind of show us where other experiences
100%. Have And we'll try to do our best over the next several months to really educate folks and get that out in the public domain and share some of the publications from other studies, including approvals in pediatrics.
So, for the high dose cohort, one of the patients was kind of described as normal FEV1, but it was low normal, I think maybe like eighty percent. And I think the explanation then was that, oh, you know, like they may not have much room for improvement because they're already normal. But now I think you're maybe flagging the dose, maybe the overdose as a rationale for why we didn't see the FEV1 benefit. What kind of patients should we be looking for in the low dose cohort? Are they likely to be in that low normal range?
And do you still believe that there may not be an ability to show a meaningful increase when you're starting at, say, eighty percent?
Yeah, I think 50% to 90% is where we expect to be, which is similar to where we expect to be in the roughly in Phase III. There's certainly, with the modulators, you can have patients who are at ninety percent baseline who improved, you know, above that, but generally speaking, we want to be in that 50% to 90% range and we think that's a population where we could show a meaningful benefit. Ninety percent now kind
of does sound high, like if a patient doesn't improve from ninety percent there, I might say, Well, you're already pretty good and maybe there was no chance to get to one hundred percent. So, is that a little too high to show a benefit?
It could be. I mean, you know, there's different know, this will be something that we refine whether we go down ultimately to 80% instead of 90%. Right now, we're at 90% just based on, you know, physicians want to be able to offer it to their patients with no available therapy, so we aligned on that. But in Phase III, it's possible we could end up anywhere from 40 to 80 or 50 to 90.
Now, as you've focused your development efforts on ophthalmology platform and you've had some collaborations in the past. Would you are you considering additional collaborations? If so, what might they look like? And if you can talk to maybe some of the stumbling points that some of the other collaborations ran into.
Yeah, I think, you know, when we were very much a platform company looking for signals and validation, I think it made a ton of sense to do a lot of those, and in each case, we did come up with better vectors. I think what we've seen is a realization for much of the pharma world that a rare disease where you only get to treat a patient once isn't a great business model, so some of them have walked away from rare disease gene therapy. We've elected to focus on large market and our own products, so we, of course, would be open to platform deals if they weren't distracting us from our job one, two and three, which is developing our own high value products.
I think we're just about out of time. David and Chris, thanks so much for joining. Thanks everyone for, tuning in. We'll head over to our keynote session next.
Thanks, Jeff. Thanks, Josh.