All right. Good afternoon, everyone, and thanks for joining us at the Morgan Stanley Global Healthcare Conference. I'm Mike Ulz, one of the biotech analysts here, and it's my pleasure to introduce Geoff MacKay, CEO of AVROBIO. Just a reminder, format for today is a fireside chat, but before we get started, I just need to read a quick disclaimer. For important disclosures, please see the Morgan Stanley research disclosure website at www.morganstanley.com/researchdisclosures. If you have any questions, please reach out to your Morgan Stanley sales representative. With that, Jeff, thanks for joining us today. Maybe just to kick off the Q&A, if you can just talk about your gene therapy platform for people that might not be familiar with it. Yeah, sure. We call it hematopoietic stem cell gene therapy, also known, aka lentiviral gene therapy. The approach is it's an autologous procedure, so we take the patient's blood, select out the CD34 hematopoietic stem cell population, and then outside the body, transduce these stem cells, the patient's own stem cells, with our proprietary four-plasmid lentiviral vector to deliver the healthy gene. The cells are then cryopreserved, which allows for a battery of safety tests and also convenience, because then we can ship the frozen drug product to the center when it's thawed and reinfused into the patient. That's really the procedure. We can talk about how we've automated it and really turned an academic process into a pharmaceutical-grade medicine. The mechanism, the reason that we do this is because if we target diseases where what you need is a lifelong distribution of an active protein, we think that this is really a good fit. Once these cells are infused into the body, they migrate to the bone marrow, where they engraft. If they engraft properly, there's no reason to think they wouldn't be there for the life of the patient. The effect comes not from them, but from their progeny, the blood lineage. In an adult, for example, billions of blood cells would be manufacturing and distributing active protein, in most cases, in our diseases, enzyme, through the body, head to toe, both above and below the neck. Got it. You touched on sort of the manufacturing process, but I know you've, you guys have spent quite some time there. Maybe you can highlight some of the initial challenges you faced in manufacturing and ways you were able to overcome those. Yeah. You can imagine that it could be very complex, very laborious, and therefore very expensive just navigating each of those steps, having a one drug product produced for a single patient. We, when we founded the company in 2015, 2016, really just put an enormous amount of energy to try to get CMC analytics done right to begin with. There's a few things that really drive your ability to scale and drive the COGS, the cost structure. If I sort of break it into two, there's one critical component, the vector itself, and we now scale and produce that in large-scale bioreactors. In our field, that means a 200-liter serum-free suspension bioreactor. With 10 to the 9th titer, that is what we produce regularly. One batch of vector produces enough to treat 40-50 patients. That is now what we've always aspired it to be, is a long lead time critical raw material, but nothing more than that, something that is not too expensive. Because the vector's frozen, it's just part. It's one raw material for the manufacturing process. Getting there is critical. Getting there early is critical because, one of the themes that everybody's aware of is, there have been so many CMC-related delays in the Yeah Gene therapy field. One of the key triggers is trying to scale up into larger bioreactors late, either mid or after a phase three trial. One issue that we've done well is getting into large bioreactors for the vector. Remember, this is a cell therapy. This is a gene-augmented cell therapy. We're now the only company in the gene therapy space that manufactures, meaning our cell bioprocessing is in a fully closed, automated system. Little robots about the size of a dishwasher that have three-day cycle time to produce a drug product. That not only allows us to scale, because you can have rows of these robots, but it also drives down cost. Maybe one simple way of saying it is that if you manufacture drug product without it being in a closed system, you can produce one patient per big clean room, big expensive clean room. It's rate limiting for scale, but it also costs us a fortune. You can increase scale tenfold and drive down cost tenfold by this automated step. There's a lot more to scale and COGS, but I think the two big-ticket items are large-scale vector manufacturing and automation. Makes sense. Maybe you can just talk about, how if you've got engraftment and the importance of vector copy number and maybe, how that's changed over time. Yeah. We in the field use vector copy number almost as a surrogate for engraftment. People aren't born knowing what vector copy number is. But for us, even a very low vector copy number in vivo, vector copy number of 0.1, for example, which is much lower than what we typically see. But even a low number like that means 5% or 10% of all nucleated cells are carrying 1-2 copies of the transgene. Again, if you turn that into numbers that we can understand, what that means in an adult patient, for example, an adult Gaucher patient or an adult cystinosis patient. 2-3 billion blood cells producing protein, manufacturing and distributing protein. Our vector copy number, when we look at it, what we really want to see is stability. You want because once you can confidently predict that the VCN is stable, plateaued, then that long-term durability is what you would expect. In our field, it does look choppy in the first several months, and that's because what you're seeing. To go back to what we try to do, we collect as many CD34 cells as we can get our hands on, transduce all of them, but we know that many of the majority of them are progenitor cells. They will go away in the first 6, 9, 12 months. The long-term engraftment comes from a small subpopulation of true HSCs, what we call long-term engrafters. As long as they engraft, then that noisiness in year one abates and you see a plateau of vector copy number. Makes sense. You've mentioned durability. Maybe you can just talk about, across your programs, like currently, what's the longest patient you've had on that sort of had a stable vector copy number and, how long could that continue? Yeah. We gave a cystinosis update in May at ASGCT, so we of course have data subsequent to that. But the most recent update, the longest patient out was 31 months, with exactly what we would hope to see, a very stable plateaued vector copy number. All the biomarkers are down and stay down. We're seeing control, and the patients remain off their standard of care, off cysteamine, all five patients measured. Gaucher, we haven't given an update in over a year. Mm-hmm. The last update was just Patient 1 out 6 months. I'd call your attention. We're pretty excited to finally give a pretty significant Gaucher update in Q4 of this year. We're calling it a Gaucher R&D Day, focused exclusively on our Gaucher program with a lot more clinical data. Patient 1 will share data out 2+ years and of course, multiple patients following. The clinical update will be quite material. We've also timed this R&D event to be after our meetings with FDA and MHRA, where we're proposing a registration trial, a Phase 2/3 trial in Gaucher Type 3. The R&D Day is really focused on clearing the cupboards of all the clinical data that we've accumulated and sharing it. Perhaps even more important, getting into the trial design after achieving alignment with regulators. Maybe I could just ask one more question, just broadly in the gene therapy field. There's been some concerns related to safety, so maybe you can just share your thoughts specific to your program and maybe what you've learned from some of the recent FDA advisory committees, which I think were somewhat positive. Yeah, so it's a big topic. Yeah. I'll make some big comments first and then get right into- Yep Our issue. First of all, hematopoietic stem cell gene therapy has a unique profile different than AAV gene therapy. Just to begin with, because there's a lot of confusion and mixing up of the two, the known safety issues related to hepatotoxicity, immunogenicity on the AAV side of the ledger and dose-related challenges, we don't suffer from any of that. That's sort of a separate unique issue. In the hematopoietic stem cell gene therapy field, really there's two questions. One, vector safety, and two, the safety around the concomitant conditioning regimen. The vector safety is the one where there's probably the most confusion. The reason for that primarily is because the bluebird bio cerebral adrenoleukodystrophy data came out last August, showing a few patients with insertional oncogenesis, which raised, of course, the legitimate question, is this a class effect? Is this something unique to ALD? I think that we believe strongly, and we are very, very satisfied with FDA adcom supporting this view, that all of these vectors have different profiles, and that it's not reasonable to extrapolate from the ALD to others. FDA didn't to beta thalassemia, nor would we expect them to other programs. Not just because. I think because the ALD program has a unique vector approach and a more aggressive promoter, the MND promoter. Even in genotoxicity tests, the MND promoter comes out as more genotoxic. I think there's a lot of preclinical assays that support that it's sort of a unique standalone. But the much more compelling data is that if you look at the 55 clinical trials using lentiviral gene therapy, and which comprises anywhere from 300-500 patients, we can document, publish 314 patients. There's zero cases of insertional oncogenesis apart from that ALD program, which uses the different promoter. The modern application of lentiviral gene therapy in, quite a formidable dataset, a dozen-plus indications, 55 trials, zero insertional oncogenesis, reported. I think that the data speaks for itself. The second topic, conditioning. Keep in mind that the conditioning is related to. Typically, in our field, monotherapy busulfan is used, a mid-range dose of monotherapy busulfan. Less toxic, less harsh than the oncology. Mm-hmm Triple drug cocktails that tend to be used. What we see consistently across all of our studies is that the patients feel completely fine for a week following the conditioning, and then by design, the neutrophils and platelets drop to a low nadir. For 2, 3, 4, 5 days and then rebound. During that week 2, there's a predictable and a transient set of adverse events, nausea, vomiting, temporary hair loss, as the neutrophils and platelets abate, the patients recover. You never wanna trivialize any adverse events, but I think what's key here is clinicians like to know when they're going to arise and when they're going to abate, and this is very, very predictable. Because it's predictable, you can proactively address with hydration, with antiemetics, nausea, magic mouthwash for mucositis, and, turn what may have been a grade three into a grade one or two. Yeah. So. Okay, great. That was very helpful. May we shift gears to cystinosis? You touched on it a little bit earlier, but maybe you can talk more broadly about the disease and what causes it and the current unmet need there. Cystinosis patients have a genetic defect related to the cystinosin protein, which is a transporter protein. All of our diseases are lysosomal storage disorders. In this case, with a defective gene, either a protein or an enzyme is missing and substrate starts to accumulate. In the case of cystinosis, it tends to be crystals that aggregate in cells throughout the body leading to devastating results for the patient. Just to throw a few statistics out, even on today's standard of care, which is cysteamine, 90% of the kids will have a kidney transplant by their teens. They suffer from severe visual impairment, so not only do they have to take 30 pills a day of cysteamine, but eye drops of cysteamine every few hours just to try to control or abate the crystals that form in the cornea. Life is cut short. Roughly, lifespan is cut in half. The bigger challenge to the community is just the quality of life. It is impacted, it's really quite devastating. Cognitive challenges, endocrinology challenges, and so forth. We're very hopeful that a single infusion of the gene therapy with the corrected gene should have a very transformative effect. What we have demonstrated to date is a proof of concept phase 1/2 trial in adult cystinosis patients. We believe already we can sort of check the box that we've demonstrated proof of concept in adults. Because this is a full systemic disease, we measure many measures, but we've seen the gene therapy work. To go back to vector copy number, very strong, stable vector copy numbers suggest the gene is present in the CD34 hematopoietic stem cells having grafted and are doing their job. We've taken biopsies, rectal biopsies, skin biopsies, we've looked at the cornea, we've taken central nervous system measures, grip strength measures, and across the board, I think what we see is control. In this study, the patients stopped their standard of care a few weeks prior to gene therapy. We're measuring the pure effect of gene therapy. In every one of those measures that I referenced, we're either seeing stability, but in many cases, improvement versus what today's standard of care could produce. I'd also highlight that today's standard of care PROCYSBI, Horizon product costs $700,000 a year every year for life. It really not only doesn't control the disease, but the patients and the parents talk about a sulfur smell. They characterize it as rotten eggs in body and breath. So much so that if a young child enters a classroom, the classroom fills up. We think that there's certainly a big opportunity to do better. It's very encouraging that we've demonstrated a pronounced effect in adults, and that's why AVROBIO is now very busy working on our own next study that we hope to be in front of regulators, literally in the very short term to initiate our trial next year. Yep. Maybe you can talk a little bit about the design of that study and, is there a specific primary endpoint you have in mind or is it maybe a few key endpoints? Well, you can imagine that we do, but we're being a bit cagey. Yep until we align with regulators. First of all, 'cause we don't want our guidance to bounce around, but also we don't wanna antagonize the regulators. Yep As soon as we have our meetings with them, we will give an update. What I can say is that we have targeted the first phase of the study to be pre-kidney transplant, so pediatric cystinosis patients. Of course, it's such a devastating progressive disease that so many functional measures move early, that we believe that we can design a trial finite in size, looking at multiple functional measures and hopefully show improvement in the pediatric setting. We feel we've already done our job in adults, so if we can demonstrate in the pre-kidney population, we do believe that that would be a registration enabling study. Is the goal to stabilize some of those parameters or to show improvement? Secondly, is this going to be a controlled study or can you use a natural history control? Or how do you think about that? Yeah. We haven't disclosed that much, but I would say that, sometimes you're forced to look at biomarkers. Yep. This is not one of those cases. This is a disease that real functional measures related to central nervous system, to the kidney, to the eye, and other measures, grip strength, involuntary muscle weakness, move in these kids so dramatically at an early stage that we do believe we can look at functional measures. We'll show more when we have the regulators agree with us. Yeah. Maybe just back to the reduction in crystals in the proof of concept. Do you have a sense of, what level of reduction you need to maybe translate into sort of benefit on these other parameters? We do. keep in mind, we looked everywhere that we could. Yep Just because we wanted to not only answer are we making these kids better or adults in this case better, but also we wanted to look at the biodistribution of the gene therapy. A skin biopsy and looking at intestinal mucosa via rectal biopsy is important, but that's not really a transformative functional measure. What we showed in both of those, by the way, is a significant improvement in reduction in crystals versus standard of care using within-patient control. Something that is much more relevant looking at crystal reduction is the cornea. We looked at the cornea baseline and then throughout the trial. Importantly, that alone has served as a measure for approval for other medicines. What it relates to directly is photophobia. What we have shown so far in our study is by controlling and reducing the number of crystals that form in the cornea, we can reduce the photophobia scale for the kids. We reduce it 1 point, a 4-point scale. What that means for the kids is that, for example, some of these patients can now leave the house without dark thick glasses, so it really does transform their life. Okay, great. Maybe we can switch to Gaucher. You talked a little bit about it earlier, and you'll have a sort of very comprehensive update in the fourth quarter. Maybe just to start, if you can talk about the disease, what causes it, some of the symptoms, and sort of what are the key endpoints and maybe the difference between Type 2 and Type 3. Sorry, a lot of questions there. Yeah. No, but it is. I think that we ourselves at AVROBIO, because we have delineated Type 1 and 3 what gets lost is that it's absolutely one disease. Yep. These patients have a genetic defect of failure to encode for glucocerebrosidase. What that again, a lysosomal disorder. In the absence of this enzyme, substrate accumulates in the cells, and in the case of Gaucher patients, it results in organomegaly, so enlarged liver and spleen, bone pain, bone crises, just a number of challenges. The patients themselves complain the most about fatigue. The difference between Gaucher Type 1 and Gaucher Type 3 is Gaucher Type 3 has all of the above, but with a more devastating cognitive deficit, so CNS manifestation. It's not always even related to the particular mutation. That's why when we look at our clinical development program for Gaucher, we are developing one comprehensive program where we have our Guard1 trial, which is looking at Gaucher Type 1, and we are in front of FDA and MHRA this season. That's what we will update, where we're proposing a Phase 2/3 trial targeting Gaucher Type 3 patients. Our goal overall is a broad Gaucher clinical development program, including Type 1 and Type 3. We often get the question, what about Type 2? Really Type 2 is incredibly rare and incredibly devastating. Almost by the time these kids are born, it's too late for them. We don't know that gene therapy couldn't play a role, but it's not until we understand the profile more, the focus is Type 1 and 3. I would highlight, I know because of the old Shire and Genzyme, a lot of people know Gaucher disease quite well. It is the most common lysosomal disorder. There are about 30,000 patients worldwide. In our addressable priority markets of North America, Europe, Japan, that number is about 14,000-15,000 patients. For Gaucher Type 3, it is important to understand that although it is not always indicated, ERT is what is typically used. These patients have to go to a hospital and get infusions every 2 weeks to receive a medicine that does not cross the blood-brain barrier. A medicine that has no hope. Mm. ... of addressing the key manifestation. We certainly hope and expect that gene therapy could outperform today's standard of care. Gotcha. Maybe just going back to Type 1, you're in the Phase 1/2 study. You're going to give update sort of 4Q, but you've also shared a single patient's worth of data that, had some promising effects there. Maybe you can walk us through that patient and then how should we- Yeah. think about the update and what are the key areas to focus on? Yeah. Just to give a little bit of color, I'll talk about that first patient. This was a 32-year-old female Gaucher Type 1 patient who was on enzyme replacement therapy since the age of 3. She's on the VPRIV, the Shire now Takeda enzyme replacement therapy. Biweekly infusions from the age of 3 until 32. Importantly, because she's human, she went through a period of noncompliance, which so many of these patients do. It was very informative because one of the measures that the clinicians look at, a biomarker, is chitotriosidase, activated macrophage. What we see is VPRIV, the Takeda product, was controlling her levels of chitotriosidase at about 150. As soon as she started skipping infusions, it spiked up to 3,000. That's the background. Now, if we take that same patient and a second time stop ERT, but in this time give her a single infusion of our gene therapy. Looking just at that same biomarker, not only did chitotriosidase not spike up to 3,000, but it was cut in half versus what the VPRIV, the Takeda enzyme replacement therapy was able to deliver. Very similar with the other biomarker, lyso-Gb1, which is the toxic metabolite itself, essentially cut in half versus the within-patient control VPRIV levels of control. VPRIV did do a good job controlling hemoglobin and platelets. It was normal off the ERT. It remained normal on gene therapy, and no adverse events related to the gene therapy and the kind that I've already described related to the conditioning regimen. It's an N of one only out six months. We'll have a lot more to say in our next R&D update, going out 2+ years, more patients and so forth. We're encouraged with this early data, and we just hope to be able to replicate it. Yep. Assuming you replicate the data, what would be the next steps for Type 1? Would it be, advancing into a phase 3 study, or how should we think about that? We again will cover it with our R&D day, but I think I would reiterate that Gaucher is one disease, and we're not convinced and we don't believe that you need to run more than one randomized control trial. Sure. It's the same genetic defect. It's the same drug product, Gaucher Type 1 and 3. We're looking at one comprehensive clinical development program for Type 1 and 3. I gotcha. I don't know if you can share your current thinking on that phase three program. I know you're waiting to get some feedback from the FDA. Are there any particular areas where you need feedback? Seems like there's approved drugs, so the clinical path should be pretty defined here. Yeah, I think unlike, we at one time had a Fabry program, as you're aware of, Mike, where one of the big challenges was can you suss out a primary endpoint that can be measured in a reasonable period of time. Yeah ... without the statistics exploding into hundreds of patients. With Gaucher, we think that is absolutely the case, that the endpoints there are multiple functional measures that you can look at in a short period of follow-up. We expect and hope a 12-month follow-up. We hope the regulators agree with us, but what we're proposing, we think, is a very rigorous, well-designed trial looking at primary endpoints that are already well established in the field. Makes sense. Maybe in the last few minutes here, you also have a number of second wave programs. Maybe just talk about some of those programs, current status and next steps from here. Yeah. We would love, to be able to advance everything concurrently. Of course, with the cost of capital right now, with the financial environment, what we're trying to really do is to place the majority of our chips on entering two late-stage clinical trials in 2023, which is of course, Gaucher and cystinosis. We have Hunter syndrome and Pompe disease as what we now call our second wave of programs. Hunter, we expect to enter the phase 1/2 trial in an investigator-sponsored trial imminently. We would expect a CTA from our Manchester partners in the UK, in the very near future in order to pivot to a phase 1/2 trial. That also really, although it's a smaller disease, it really hits the sweet spot of hematopoietic stem cell gene therapy because it needs that, repopulation of the microglia compartment, so a CNS effect as well as the full systemic effect. We do expect to be dosing patients in short order, certainly in 2023, but more than likely in the front end of 2023. Pompe disease, we've shared some very compelling preclinical data already. We had guided towards initiating a trial in 2023, but, as I said, with the current environment, we've had to stagger our programs a little bit. Yep. We're still actively advancing Pompe disease, but we're no longer guiding that it will be in the clinic in 2023. Okay, great. It looks like we're just about out of time, so why don't we end it there? Thanks so much, Geoff, for your time. Really appreciate it. Yeah, thank you, Mike. Appreciate it. Thank you.