Good afternoon, everyone. Thanks for joining us. I'm Salveen Richter, biotechnology analyst at Goldman Sachs, and really pleased to have the 4D Molecular Therapeutics team with us. Maybe to start here, David, I'd have you just introduce the company, kind of the situation that's played out through, you know, IPO till now with regard to the directed evolution platform, the disease verticals, and where we've seen the progress play out across the programs to date.
Perfect. Well, thanks for having us here today. We're pleased to be here. I'm David Kirn, Co-founder and CEO. This is Julian Pei , our Head of Investor Relations. We're thrilled to be here. 4D Molecular Therapeutics is the next generation gene therapy company that's built on the power of the directed evolution platform to invent customized AAV delivery vectors for any tissue in the body, essentially. What this allows us to do is have best-in-class products, best-in-class vectors, which translates into best-in-class products. It's an extremely versatile platform, we can use directed evolution to invent customized vectors for essentially any tissue in the body by any route of administration.
What we do for every therapeutic area is we start with what's called the Target Vector Profile, which is basically, you know, what features do we want this vector to have to be best-in-class and serve the unmet needs with the existing vectors. That's everything from route of administration, the cells we wanna transduce, the dose levels we wanna be in, immunogenicity, and ramping that down. Where we are today is we have three vectors that we've invented that have now entered the clinic across five different products. We're in ophthalmology, lung, and cardiology.
We're really excited to say that, you know, over the last nine months, we started releasing clinical data that shows that we're validating these vectors, validating these products in the clinic, showing not only the promise of each of these products, but also of the platform. We've released data on our product 4D-150 for wet AMD. This is a unique and we believe, best-in-class gene therapy in the sense of being intravitreal, an outstanding safety profile with no evidence of inflammation across three different dose levels. The transgenes that we're expressing are secreted of aflibercept, so vectorized aflibercept, which hits three different molecular targets. We've also inserted into the same transgene payload, a second transgene that knocks out VEGF-C, and that's in a large randomized study at Opthea.
They've shown that that's an attractive target to knock out as well. We, we believe we're the only product out there that has, you know, inhibitors of four different VEGF family members. It's expressed directly right in the macula, right where it's needed, right on top of the choroid, and it's obviously expressed 24 hours a day, seven days a week. We think we really have an opportunity not just to replace aflibercept, but to actually be superior there. We've also recently announced data on the first 3 patients with cystic fibrosis treated by our aerosol-delivered 4D-710 product. There, we're excited to say we had an outstanding safety profile, no related AEs after completion of the routine aerosol delivery using a commercially approved aerosol device.
We also showed on the order of 92%-99% of the cells in the lung were expressing CFTR, which is just a remarkable finding, and the intensity of staining and the percent of cell staining were significantly above levels in normal lung controls. I think this is probably the first time this has ever happened in gene therapy. We actually got super physiologic delivery and expression, and that translated into really nice improvements in quality of life with a well-validated instrument called the CFQ-R instrument , showing improvements of six, 11, and 22 points. The MCID on that is about 4 points, so well above that. We also saw in the one patient who started with a low FEV1, they showed a significant improvement of 7 points.
Small data set, we're now enrolling in the second dose level, but certainly very compelling expression levels and efficacy. The third vector that we've taken to the clinic is an IV vector for the heart, and there we showed in the first three patients with 4D-310 for Fabry disease, significant benefit in terms of everything from echo contractility, cardiopulmonary exercise testing, and quality of life as well. You know, approximately 50% of cardiomyocytes staining positively for transduction at a very low dose of 1e13 . We feel like across these three therapeutic areas, we validated the vector, which we think has significant read-through to other products in those areas, as well as to the platform.
Maybe jumping into the portfolio at this point, starting with the ophthalmology vertical, what do you think might be overlooked by the investors with regard to your recently provided data in wet age-related macular degeneration? Perhaps you could also just talk about the physician feedback that you received post-presentation?
At the ARVO meeting in April, we released the data with six to nine months follow-up on all three dose cohorts in the phase I. I think what's really remarkable is the four patients at the high dose level who would be eligible essentially for phase II or phase III, were all injection-free at 9 months, so a really stellar effect there. No evidence of inflammation and, you know, very robust gene expression as determined by aqueous humor taps. I think what may have been underappreciated with that data set is the fact that these lower doses are also very active. You know, we have a dose response, which we were trying to show, and FDA wants us to show.
Those lower doses maintain the outstanding safety, and they're still highly active, with about a 75% reduction in need for supplemental injections. About 70% of patients either had zero or only one injection over the 6 months timeframe of follow-up. I think that's really exciting data. I think that was maybe underappreciated, is that activity at lower dose levels.
Can you speak to the cataract that was seen in patient one, and what may have caused it, and how you're thinking about the risk of cataracts due to steroid use, which is generally required here for.
Yeah.
Gene therapy?
Exactly. Just to take a step back and ask a question about steroids. We felt that when we were developing this product, we thought, look, because of this efficiency of this vector and the intravitreal nature, we should be able to get by with very low doses. In contrast to other programs of gene therapy in the eye, which have had doses anywhere from, you know, 10 to 100-fold higher and have had significant inflammation. We thought we would have very little, if any, inflammation, but we thought, to be safe, let's start with a steroid eye drop regimen over about 20 weeks of tapering down. We saw outstanding safety there. One patient, first patient ever treated with the agent, the physician elected to keep...
continue those steroid drops and did probably induce a cataract in that patient, which can happen when you're on steroids, you know, for 40 weeks+. You know, our standard is 20 weeks. We think the risk of cataract formation with the 20 week regimen is probably pretty minor, not zero, but it's minor. In that patient, you know, once the cataract developed, the cataract was removed, and the patient's visual acuity went right back to baseline, which was great to see, which is a huge win out that far after with no supplemental injections. Going forward, we need to decide, do we, you know, are we comfortable with the risk of cataract, or do we want to shorten that steroid regimen since we've had no inflammation? Do we want to, you know, potentially.
Some groups have actually enrolled only patients who've already had their cataract and their lens removed into phase III. That's another option. I think for us, probably the most reasonable thing is just to shorten that steroid course before phase III.
There was a patient who had a low BCVA and lost about 18 letters. How variable, I guess, is BCVA in severe patients? How, you know, how can you be sure that it wasn't due to drug?
It definitely was not due to the drug, and it was definitely due to a cataract which formed right in the center of the visual field. That patient already had a pre-existing cataract, that progressed very quickly, we think it was an extension of that original cataract, which basically was right in the middle of the visual field, which pushed the BCVA down. To your point, BCVA is a variable endpoint, especially in these patients with very low starting BCVA. This patient had essentially been legally blind in the preceding year, certainly, we would all agree, I think that patient should not have been enrolled. This does happen in phase I.
I mean, it's part of phase I. We say, "Okay, that patient would never enroll into a phase II or phase III." They're certainly valuable for safety. There was zero evidence of toxicity to the retina, no evidence of inflammation. It's definitely not a toxic event. It's explainable by this central cataract. Nevertheless, those kind of patients would be excluded in phase II and phase III.
When you then look at the remaining three patients, you know, what gives you confidence for the dose expansion portion, and I guess how would that differ from what you did in the phase I in terms of inclusion?
Yeah. With the phase II, we brought up the baseline BCVA required to get on the study. The first patient would have been excluded, if patients stick to the if physicians stick to the 20-week regimen, which I think they will, given the lack of inflammation, then cataract development really shouldn't be a significant risk there. We're confident we're gonna be able to replicate the efficacy and safety we saw in the phase I.
You mentioned that you could have an in-line profile to Eylea or potentially a superior profile to Eylea. What is it about this data set that you saw that leads you to believe that could play out?
That's a great question. First of all, you know, these patients were heavily pretreated. These patients had all been diagnosed on average about three years earlier, and you know, had been receiving on average 10-11 injections in the preceding 12 months. Some of them had had 13 injections, so more than one a month, with drugs, approved agents like aflibercept and Lucentis. The fact that we could treat them once and see that they were injection-free for a total of nine months, I think is very strong evidence that, okay, this could bear out. I think there are certainly patients out there, not the ones on this study, but there are patients out there who do respond quite well to Eylea, and might be fine with every eight weeks, or might be maybe every 16 weeks with high-dose Eylea.
There will definitely be patients as well you know, just are relatively resistant to that and need frequent dosing, where we, I think, could be superior in that population.
I think that the question that comes up a lot, just given the history of this field, is the safety profile and how to get comfort with it and when you would get comfort with it. When you look at this study, how long does it typically take for inflammation and other adverse events or serious events, such as blindness, to occur? And then have you passed this threshold, and then with how many patients? And when would there be enough to say, "Okay, we get it. We understand this profile. We've got a relatively safe asset.
You know, I think, with gene therapy in the eye, a number of groups have run into trouble with inflammation, and again, they're starting at much higher doses, typically, whether it's subretinal or intravitreal with seven and eight, with Adverum or suprachoroidal with the AbbVie REGENXBIO Program. We have not seen that inflammation. Now, when they see it, they typically see it pretty early, as early as two months, and then that can last, in some cases, a few months, or in the case of Adverum, out beyond one year. We believe with our data, if we get to 20 weeks and patients are, you know, four or five months out and they're off of steroids, the likelihood that inflammation pops up later is exceedingly unlikely and has never been reported.
The late hypotony that the Adverum program had in terms of low pressures in the eye, leading to serious complications and blindness, that didn't occur till nine to 10 months. It was preceded by a tremendous amount of inflammation, especially in front of the eye inflammation with pigment release and pigmentary changes in the front of the eye, which eventually the theory is that that knocked out the ciliary body and then dropped production of the fluid, and that's what led to it. While that eventual hypotony took 9 months to be diagnosed, the precursors to that were very evident very early on, and we're just not seeing anything like that. Again, the likelihood that we would suddenly get a significant adverse event out beyond six months is very, very low.
How well understood is the chance for an idiosyncratic event to play out in this population?
You know, that's just unknown until we get more data. There's nothing we can do about that. You know, we just have to treat more patients and get more follow-up. Yeah.
You had a program in choroideremia and XLRP. How translatable is the profile that played out there to wet AMD with regard to pigment dispersion and uveitis?
Yeah. You know, those two rare inherited retinal disease programs, we're excited about those programs. Each of those enrolled on the order of 14-16 patients, we've reported on that, showing early evidence of activity in the subset of patients who are valuable. The XLRP program of 4D-125 patients received up to 1e12 VG per eye, anywhere from 30-100-fold higher than what we use in wet AMD. In wet AMD, we're much lower dose, that's again because the aflibercept is secreted, therefore, we don't need to get in every cell the way we might want to in XLRP. We know that the capsid itself is very well tolerated up to 1e12 .
I think that's a direct read through to say, wet AMD, that capsid itself is safe and well tolerated at doses 100-fold higher, we're in great shape. The choroideremia patients, that's a very unique patient population with a unique immune system because of the mutation. It's a very pleiotropic transgene that turns on all sorts of different pathways. With that product, we did see a REP1 transgene-associated pigment release syndrome from the iris. We have not seen that in the XLRP program, we're confident we won't see that at these doses and with the transgenes in wet AMD.
Can you speak to when we might see the first data in diabetic macular edema and the translatability of the data that you've seen here in wet AMD over to that population?
Yeah, we're really excited about that program as well. You know, there is some competition with gene therapy in wet AMD, but as far as we know, we believe we're the only company in DME now with AAV gene therapy. It's a very large market, and it's growing rapidly with the increase in diabetes in this country. Historically, all the agents that have worked in wet AMD have translated and worked just as well or even better in DME, and frequently at even lower doses. I think the odds that we see a translation of the efficacy over to DME is very high, historically, based on historical data. We've given guidance that the IND is open there. We expect to treat our first patient with DME in Q3.
We would, you know, by early next year, we'd have initial data sets available to discuss safety and efficacy there.
You're also going to move forward in geographic atrophy. What gives you confidence on the translation, I guess, over to complement factor H and the ability to have read through in that population?
You know, I'll have Julian speak to the complement factor H itself. I would say we're starting from a position where we have a validated vector. What we've found is that the modularity of these vectors makes the drug development and building of a portfolio in a therapeutic area very efficient, right? When we go in with that GA asset, we use the same exact manufacturing process as 4D-150 for wet AMD. We'll use 90%-95% of the release tests will be the same. The dose ranging in non-human primates will be the same. It's a very efficient way to develop products, and then we can leverage all the experience we have with safety and efficacy of this and wet AMD clinically.
Then we do like to go after either genetically or therapeutically validated targets to decrease the increase the probability of technical success with our products. Maybe Julian can speak to the advantages of the Complement factor H that we acquired.
Yeah, absolutely. Regarding the complement factor H, I think, you know, the genetic validation, as David mentioned, is really important. We see that in 40% of cases of genetic geographic atrophy, they actually have mutations that reduce the function of that naturally human occurring factor that regulates the complement pathway. I think that's a very clear sign that there's a mechanistic validation there. We think when we go into clinical development, we would likely enrich for that population, which would also increase the chance of success. Specifically on this construct, it was invented by the University of Pennsylvania, a professor there, then it was licensed to a company called Aevitas Therapeutics. We mentioned, kind of DME will be dosed, that's going to be the first half next year.
Regarding wet AMD, we will have completion of enrollment of the phase II, 50 patients on the randomized phase II expansion. That data will read out initially, interim data in first half of next year. You can expect anywhere between six to 12 months of follow-up.
Great. Remind us on the partnership aspect, how you're thinking about this vertical, 'cause you did have a partner with the rare diseases with Roche, but now you have an unpartnered portfolio.
Yeah, we're really excited about the potential to build very significant value in larger market ophthalmology. You know, now that we have a validated vector that we can just swap in and out different transient payloads, we can build significant value, and we think that we could build a U.S. sales force and commercialize ourselves in the U.S. I think the base case for 4D-150 for wet AMD and DME would be for us to, you know, complete the phase II, build more value in the asset, get initial DME data, and then look at partnership opportunities in 2024. Our goal would be to retain the U.S. market and partner ex-U.S.
In parallel with that, we've found that once we validate these vectors in humans, the value that others perceive in that vector goes up significantly, as you might imagine. We are open to partnering in ophthalmology outside of large market. You know, if other companies want to do rare disease targets and leverage the R100 vector, we're open to those kind of partnerships.
Moving to pulmonology. You recently presented new data at on cystic fibrosis with your aerosol gene therapy, 4D-710. Can you just outline the key takeaways here and what next steps are with this program?
Sure, I can kick it off, and you can add some color. 4D-710 is our cystic fibrosis product candidate. It's based on a vector called the A101 that we invented at 4D, using the directed evolution technology. The profile there is we wanted something that we could deliver via a routine, commercially approved device, nebulization device for aerosol delivery. We wanted something that could penetrate through the thickened mucus in cystic fibrosis, and we wanted the vector to be highly resistant to antibodies in the human population because those antibodies can be present in the mucus barrier, which could lead to inhibition.
Finally, we wanted to make sure we transduce all cell types, because, again, it's believed for CF and many other diseases, hitting as many different cell types in the airway that you can, the better. That's what we had going in, and the CFTR transgene that we deliver is an optimized version of that fits in AAV. We reported out the first three patients with CF will have a mutation that's not amenable to a modulator, usually because it's a null mutation, there is no protein. These patients have very aggressive disease, very rapidly proliferating progressive disease. We also can enroll patients who are have severe toxicities to the modulators, so say hypersens. After completion of the administration, we had no related adverse events.
Really remarkable safety profile, especially given the fact that these patients have sick lungs and we're putting all this vector in there. Again, validation of the platform that our vectors are generally very well tolerated. Safety was excellent. We then looked at gene expression on biopsies throughout the lung and brushing throughout the lung at week four to eight, in that dose, in that time frame. We saw widespread gene expression in all samples. When we looked by the protein immunohistochemistry, anywhere from 92% to 99% of the cells were positive for CFTR. Really remarkable. I don't think that's ever been shown in gene therapy. We actually had expression levels in terms of the intensity that was significantly higher than in the normal control samples.
All this data is analyzed by something called the Visiopharm machine learning assisted software. This is not, you know, a human doing this. It's confirmed by humans, but it's actually derived from this device and application. Remarkable levels of gene expression throughout the lung, and we saw expression in all three major cell types, including the basal cells, which are very long-lived cells in the lung, which you really want to target. That was all very exciting. You know, the two clinical activity endpoints that most people look at are either FEV1, which is how fast someone can blow out over a course of a 1 second, so it's kind of a airway resistance measure. The other is the CFQR quality of life instrument.
All three patients had very significant improvements in quality of life. Again, anywhere from 6 to 11 to 22, versus four being the threshold for activity. Then FEV1, we had one patient who started at a significantly lower level, where we could show that improvement of 7 points. It was durable out through nine months. Early data, small data set, but each of these different endpoints is telling us the same thing, which is exciting.
If you look, I mean, it was a small N. If you look, there was variability with those two measures. There was one patient who kind of bounced around on an FEV1, and it's an open label design, right? I guess when you look at the effect that you saw, how comfortable can you be on that, you know, on the quality of life or FEV1? And how does that kind of correlate then with the gene expression aspect?
Yeah, well, I can kinda kick this off and then turn over to Julian for additional color. I think, first of all, when I look at small data sets, what's important to look at is different orthogonal endpoints. You know, are they telling you the same thing or not? Here we're looking at very, very different things, a biopsy looking at protein, a biopsy at looking at RNA, safety, a quality of life instrument, and then FEV1, which is a function of how quickly you can blow out on spirometry. They're all telling you there's clearly clinical activity here, which I think is really important with small data sets. Julian can maybe speak to the, you know, CFQR and our interpretation of single-arm data on that.
Yeah, no, I think, really importantly, this measure, you know, it's used in all Vertex modulator studies. You know, control arms do not seem to have a placebo effect. Generally speaking, it decreases anywhere from 2 to 4 points on placebo, even in Trikafta studies. You know, Trikafta showed a 20-point delta on that measure in their approval study. When we show a 22-point improvement in the 1 patient that most resembles the average patient on the Trikafta trial, we think that's a, you know, almost ambiguous way to see that there is an effect, at least in the type of patients that we may see in the phase III.
Great. With regard to the higher dose, do you think there's an ability to add more benefit, given you were almost at 100% expression?
It's. We'll see, but it's really hard to imagine that you could. You know, it looks like we've already saturated expression levels. We'll see. You know, there's gonna be variability based on different mutation status and the severity of the patient's disease when they come in. I think when you're talking about 99% of cells expressing and the intensity being above normal, I think you've probably maxed it out, you know. I think if nothing else, that higher dose can give us greater confidence in the safety window, and then we'll just have to see if it translates into higher clinical benefit than we've already seen.
Remind us, as you think about going into the other 85% of patients, are you looking at a combination with Trikafta approach? Do you have a sense of just the regulatory path for both of these subsets?
Yeah. Starting with the population that we're in today, who have no modulator therapy, there, we think there's a real opportunity, working closely with the Cystic Fibrosis Foundation and the FDA, to think about accelerated approval or fast-to-market options. There, we could imagine some sort of blended endpoint looking at, you know, quality of life, stabilization or improvement in FEV1, and, you know, supportive data from biomarkers. You know, there is precedent for that in these smaller, rare disease populations. Here we're talking about 10%-15% of CF. CF is pretty big, but you know, we're talking about 10%-15%. We're looking forward to having those conversations with FDA, and we think there's a fast-to-market opportunity there for that population, which would be just amazing for our company.
Because, again, you know, that's something with all the support and relationship we have with all these sites and the Therapeutics Development Network of the Cystic Fibrosis Foundation, we think we could commercialize that ourself. The combination with modulators is very exciting. Not only is it a bigger population of patients, but we really have an opportunity to see synergy there. The reason we believe that could be the case is the modulators should really thin out the mucus. They should improve aeration of all parts of the lung. The modulators can bind and positively increase the function of wild-type CFTR. We think there's a real opportunity for potential synergy there.
That's a route that would definitely require a randomized trial, and we could envision someone being on, you know, stable doses of a modulator for three to six months, stable FEV1, stable, quality of life, and then you just randomize, and half the patients will get 4D-710 on top of that. We could look at changes in all those endpoints over the following six to 12 months.
Maybe jumping to the cardiology vertical. You've actually had some pretty good data in Fabry disease, but it does seem like with vectors, particularly AAV9-like vectors, you know, there is an inflammatory response that needs to be dealt with, and you're working with the FDA on a steroid regimen. Maybe help us understand how that's progressing, how the FDA is thinking about this, and what that could mean for the future of this program.
Yeah. We're excited about developing a cardiology vector. We think there's a lot of interest in the field for that, and there are a number of great indications. We selected Fabry disease as our first indication. It's a lysosomal storage disease where there's a monogenic recessive mutation in an enzyme that breaks down sphingolipids in the cells in the body. We thought that was great. First target, it's a secreted product, it's highly active. We know that high levels of that are safe as an enzyme replacement therapy. We also know that the enzyme replacement therapy, or ERT, does not affect the heart. We thought this here's leading cause of death in these patients. There's nothing for it. It's a great target.
We evolved the vector for low-dose IV delivery to the cardiomyocytes, and in fact, in the clinical trial, we've shown the first three patients significant benefit on multiple endpoints. In the one cardiac biopsy that we have to date, shows on the order of 15% of heart cells transduced. Excuse me. The vector overall has had a very favorable safety profile and no evidence of generic systemic inflammation. What we have seen is what's been seen with other AAV vectors, is you can have a rapid increase in the IgM antibody, and that binds the capsid, and that can activate complement and cause what's called aHUS. It's a very specific. It's not really like an inflammation, the way we think about it. It's really just antibody and complement.
Some really clever investigators at University of Florida came up with a regimen that uses rituximab, which is great at suppressing B cell activation and antibody generation, and then sirolimus. That combination has been outstanding at blocking that aHUS effect. We were already moving in the direction of moving to that new immune regimen, which I think is gonna be standard for IV AAV vectors. You know, FDA supports that strongly, those are the discussions we're having now, is moving in that direction. You know, in the next quarter or so, we'll have a one year follow-up on all 6 patients, and we'll be able to look at that program and make some decisions.
Great. We have time for maybe one question from the audience.
As you think about patient enrollment in the wet AMD and DME trials, how do you think about balancing the exclusion of the severe patients in order to achieve higher probability of success versus showing a benefit in the unmet need population?
In the current study, we started with the most difficult to treat. We think these patients are maybe, like, 15% highest users of anti-VEGF. You know, as we wanna make sure when we get to the market and on phase III, that we have a broad range of patients. We wanna start moving into that earlier stage population and get safety and activity there. We will exclude patients who are so far advanced that, you know, BCVA really can't be rescued, and I think that will give us a stronger signal in phase III. We'll still be treating, you know, 95% of the population with those criteria.
Great. Well, with that, David and Julian, thank you so much. We appreciate it.
Thanks for having us.