Taysha Gene Therapies, Inc. (TSHA) Earnings Call Transcript & Summary

Hello, and welcome to Taysha Gene Therapies GAN Program Update Webcast. [Operator Instructions] As a reminder, this webcast is being recorded today, January 31, 2022. I will now turn the call over to Dr. Kimberly Lee, Senior Vice President of Corporate Communications and Investor Relations. Dr. Lee, Please go ahead.

Thank you. Good morning, and welcome to Taysha's GAN program update conference call and webcast. Earlier this morning, Taysha issued a press release reporting clinical efficacy and safety data for the high-dose cohort and long-term durability data for TSHA-120 and giant axonal neuropathy or GAN. A copy of this press release is available on the company's website and through our SEC filings. On this call, we will review the reported data and discuss next steps for our clinical development program. Joining me on today's call are RA Session, II, Taysha's President, CEO and Founder; Dr. Suyash Prasad, Chief Medical Officer and Head of Research and Development; Sean McAuliffe, Chief Commercial Officer; and Fred Porter, our Chief Technical Officer. We will hold a question-and-answer session following our prepared remarks, where Kamran Alam, our Chief Financial Officer, will join as well. Please note that on today's call, we will be making forward-looking statements, including statements relating to the existing clinical data and therapeutic and commercial potential of GAN and our investigational product candidates. These statements may include the expected timing and results of clinical trials for our product candidates and the regulatory status and market opportunity for those programs. This call may also contain forward-looking statements relating to Taysha's growth and future operating results, discovery and development of product candidates, strategic alliances and intellectual property as well as matters that are not of historical facts or information. Various risks may cause Taysha's actual results to differ materially from those stated or implied in such forward-looking statements. These risks include uncertainties related to the timing and results of clinical trials and preclinical studies of our product candidates, our dependence upon strategic alliances and other third-party relationships; our ability to obtain patent protection for our discoveries; limitations imposed by patents owned or controlled by third parties and the requirements of substantial funding to conduct our research and development activities. For a list and description of the risks and uncertainties that we face, please see the reports we file with the Securities and Exchange Commission. This conference call contains time-sensitive information that is accurate only as of the date of this live broadcast, January 31, 2022. Taysha undertakes no obligation to revise or update any forward-looking statements to reflect events or circumstances after the date of this conference call, except as may be required by applicable securities laws. With that, I would now like to turn the call over to our President, CEO and Founder, RA Session, II.

Thank you, Kim, and thank you, everyone, for joining us on the call this morning. We are very excited to share positive clinical efficacy and safety data for the high-dose cohort and long-term durability data for TSHA-120 and giant axonal neuropathy, or GAN. I will start off by providing a brief overview. And then Suyash Prasad, our Chief Medical Officer and Head of R&D, will review the biology of disease and natural history data before discussing the clinical data. Sean McAuliffe, our Chief Commercial Officer, will then discuss the commercial initiatives for TSHA-120; followed by Fred Porter, our Chief Technical Officer, who will share our manufacturing strategy. Afterwards, I will review upcoming milestones before opening the call for questions. We are excited to share the data from the high-dose cohort of 3.5x10^14 total vg, along with long-term data from all therapeutic dose cohorts, including 1.2x10^14, 1.8x10^14 and 3.5x10^14 total vg. The data we will present today demonstrates a favorable safety and tolerability profile for TSHA-120 that is supported by 53 patient years of clinical data. Treatment with TSHA-120 achieved a clinically meaningful and statistically significant slowing or halting of disease progression seen in the highest dose cohort of 3.5x10^14 total vg and across all therapeutic dose cohorts. At the highest dose, TSHA-120 demonstrated clinically meaningful and statistically significant improvement in the MFM32 score by year 1 compared to the natural history. Additionally, long-term durability data across all therapeutic dose cohorts demonstrated a 10-point improvement and the mean change from baseline in MFM32 score by Year 3 compared to the estimated natural history decline of 24 points. These long-term data confirmed the disease-modifying effects and sustained durability of TSHA-120. Notably, nerve biopsy data provided new evidence for active regeneration of nerve fibers following treatment with TSHA-120. In addition, we observed preservation of visual acuity as measured by the LogMAR scale and optical coherence tomography. The totality of these data support our plans to engage with regulatory agencies to discuss possible pathways for registration. We believe that TSHA-120 has the potential to be a transformative therapy for an estimated 5,000 addressable patients, which represent a significant commercial opportunity. With that, I will now turn the call over to Suyash to provide an overview of our GAN program. Suyash?

Thank you, RA, and Good morning, everyone. First, let's start with a brief overview of GAN and the rationale for why we think this is good gene therapy approach. Next slide. GAN results from mutations that affect production of the protein gigaxonin, which leads to abnormal accumulation of neurofilaments in axons, increasing their size and causing interruption of the electrochemical signal along the nerve fiber, which then results in the clinical phenotype of GAN. Gene therapy is an attractive approach for the treatment of GAN since the gene is relatively small and fits well into an AAV9 capsid that does high target organ transduction. Moreover, the low level of expression is expected to be sufficient to restore function. Next slide. There are 2 phenotypes for GAN: early and late onset. Early onset GAN is the classic phenotype, which is characterized by rapid progression. Signs and symptoms usually begin before the age of 2 and can include delayed early motor development and an unsteady gait due to the sensory ataxia resulting from the longer nerves of the body being affected early in the disease course. Between the ages of 3 to 8 years, patients demonstrate progressive motor and sensory weakness that results in ataxia, dysarthria, nystagmus, optic neuropathy and decreased visual acuity. By age 8 to 11, progressively worsening scoliosis and contractures are common features, and most children lose the ability to independently ambulate and become wheelchair dependent in the second decade of life. As time progresses, serious CNS dysfunction and respiratory failure requiring tracheostomy and ventilator support develop, and death occurs in the late-teens or early 20s. Currently, there are no approved therapies for GAN. Late onset GAN is detected in the teenage years. Individuals with a late onset form have a phenotype, which is slower in progression with respect to motor and sensory dysfunction. And although there is a significant disease burden, may not be life limiting. Due to nonspecific features, these individuals are often categorized as having Charcot-Marie-Tooth Type 2 or CMT2 disease. Next slide, please. Disease progression in patients with GAN has been characterized in a natural history study by a number of clinical assessments, most notably by the Motor Function Measure 32 or MFM32. The MFM32 is a well-known validated scale for motor function measurement that has been used in multiple regulatory approvals. It uses a 32-item scale to evaluate motor function measurement developed for neuromuscular diseases and assesses severity and progression of motor function across a broad spectrum and in 3 functional domains, including standing, transfers and ambulation, proximal and axial function and distal function. A 4-point change is considered clinically meaningful in several indications, including DMD, SMA, LAMA2-related muscular dystrophy and cerebral palsy. Next slide. There is robust natural history data for GAN that includes an ongoing study that was initiated in 2013 following 45 patients with GAN age 3 to 21 years. These data serve as a dependable comparator for treatment studies. The MFM32 was selected as the primary endpoint in the interventional trial because it reflected the endpoints collected in the natural history study reliably and is often used in confirmatory trials. Per natural history and as you can see on the graph on the right, the MFM32 total score declined 8 points per year across all patients with GAN regardless of age. As a reminder, a 4-point change in the MFM32 is considered clinically meaningful. Next slide. Now let's discuss the design of the TSHA-120 construct and our clinical approach. The construct was originally invented in the lab of Dr. Steven Gray, our Chief Scientific Advisor. It is on the AAV9 viral vector engineered to deliver a transgene encoding the human gigaxonin protein. The self-complementary AAV capsid enables rapid activation, stable expression and optimal tropism, while the JeT promoter drives ubiquitous expression at a low to moderate level. TSHA-120 is designed to deliver a functional copy of the GAN gene to the CNS and PNS. The drug has already received orphan drug and rare pediatric disease designations from the FDA and is being evaluated in the ongoing Phase I/II clinical trial at the NIH, led by Dr. Carsten Bönnemann. Next slide. TSHA-120 is the first gene therapy to be intrathecally dosed and is currently being evaluated as part of a groundbreaking historic dose escalation clinical trial. The primary endpoint for the trial is safety. The secondary endpoints evaluate efficacy, including pathological, physiological, functional and clinical markers. 14 subjects have been dosed, all of whom are older than the 5 years of age. TSHA-120 was administered intrathecally at 4 different dose levels that include 2 patients at 3.5x10^13 total vg, 4 patients at 1.2x10^14 total vg, 5 patients at 1.8x10^14 total vg and 3 patients at 3.5x10^14 total vg. Patients were placed on an immunosuppressive regimen, which comprised prednisolone and sirolimus, and oversight was provided by a Data Safety Monitoring Board. Next slide. We now have 53 patient years of clinical data that support TSHA-120's favorable safety and tolerability profile. There were no significant safety issues and no increase in the incidence of adverse events at higher doses. All adverse events related to immunosuppression or study procedures were comparable to other gene therapies and transient in nature. There were no dose limiting toxicities reported following treatment with TSHA-120 and no signs of acute or subacute inflammation, including encephalopathy, persistent headaches, seizures or vision changes. There's no evidence of nerve root or dorsal root ganglion inflammation and no evidence of thrombocytopenia or persisting transaminitis. Next slide. Importantly, we were able to safely dose in the presence of neutralizing antibodies. This pie chart depicts the 22% of patients who were seropositive at baseline and the 21% who were considered borderline. That's 44% of patients who ordinarily would not have been able to be dosed. The route of administration, combined with the dosing and immunosuppression regime, allow for effective dosing in patients with neutralizing antibodies. Next slide. Now let's delve into the efficacy data. At 1 year post gene transfer are clinically meaningful and statistically significant slowing or halting of disease progression was seen with TSHA-120 at the highest dose of 3.5x10^14 total vg, n of 3. As a reminder, the prespecified efficacy endpoint is the change in rate of decline in MFM32 score by year 1 compared to natural history. In the graph on the right, natural history data is represented as the gray bar and the high dose cohort is the blue bar. The Y axis of the graph represents the decline in slope of the MFM32 score within 1 year. The change in the rate of decline in the MFM32 score improved by 5 points in the 3.5x10^14 total vg cohort compared to an 8-point decline in natural history. Again, a 4-point change in MFM32 is clinically meaningful. Next slide. Now although the change in MFM32 score was clinically meaningful, we might have expected a greater change in the MFM32 score compared to natural history in the first year, but 1 patient in the high dose cohort was a delayed responder. At the 12-month follow-up visit, the patient had a 7-point decline in the MFM32 total score that was similar to the slope of a natural history curve as you can see in the image in the top right-hand corner. Notably, from year 1 post gene transfer to year 2, this patient's change in the MFM32 score remained unchanged, suggesting a stabilization of disease at 2 years post treatment. About 2-year post treatment time point, there was a 9-point improvement in the patient's MFM32 score compared to the estimated natural history decline of 16 points. Note that the annualized estimate of natural history over time assumes the same rate of decline as in year 1. Next slide. An additional analysis was performed to examine the change in the rate of decline in the MFM32 score of all therapeutic doses combined, including the 1.2x10^14 total vg, the 1.8x10^14 total vg and the 3.5x10^14 total vg dose cohorts with a total n of 12. The change in the rate of decline in the MFM32 score improved by 7 points by year 1 compared to the natural history decline in the MFM32 score of 8 points. This result was clinically meaningful and statistically significant. Next slide. A Bayesian analysis was conducted on the 1.2x10^14 total vg, 1.8x10^14 total vg and 3.5x10^14 total vg cohorts at year 1 to assess the probability of clinically meaningful slowing of disease progression as compared to natural history. This type of statistical analysis enables direct probability statements to be made and is both useful and accepted by regulatory agencies in interventional studies. We saw a 10-point improvement in the mean change from baseline in MFM32 score for all patients in the therapeutic dose cohorts. Next slide. In addition to the compelling 3-year data, this slide highlights 1 patient at year 5 whose MFM32 changed from baseline, improved by nearly 26 points in the 1.2x10^14 total vg dose cohort compared to the estimated natural history decline of 40 points by this time point. As a reminder, 4-point change is considered clinically meaningful. Next slide. Here we provide an additional analysis of the mean change from baseline in MFM32 score at the last visit for the mean of all patients at the last visit in the therapeutic dose cohorts compared to natural history. As shown, TSHA-120 demonstrated increasing improvement in the mean change in MFM32 score from baseline over time. Next slide. Moving on to secondary endpoints. We are happy to share pathology data from biopsies of the superficial radial sensory nerve in 5 out of 6 patient samples analyzed. The remaining patient samples are currently being analyzed. These biopsies confirm that treatment with TSHA-120 can stimulate active regeneration of axons. There's an increase in the number of regenerative clusters observed at Year 1 compared to baseline, indicating active regeneration of nerve fibers following TSHA-120 treatment. On the right side of this slide, we have included a representative patient case study showing the pathology of the superficial radial sensory nerve at baseline and 1 year post the treatment with TSHA-120. The photo on the left denotes baseline with the black arrow identifying a giant degenerating axon and a star identifying a regenerating cluster. On the right is what the nerve looks like 1 year after treatment. The yellow arrows indicate regenerative clusters, which as you can see, are notable in number. Next slide. Loss of vision has been frequently cited by patients and caregivers as a symptom they find particularly debilitating. I would like to see improvement in following treatment. Patients were analyzed for visual acuity using a standard logarithm of the minimum angle of resolution, or LogMAR, with an increase in LogMAR score representing a decrease in visual acuity. And to remind you, a LogMAR score of 0 means normal vision, approximately 0.3 and you'll need eyeglasses and a score value of 1.0 is blind. Based on natural history, individuals with GAN experienced a progressive loss in visual function reflected by an increase in the LogMAR score. Ophthalmologic assessments following treatment with TSHA-120 demonstrated preservation of visual acuity over time compared to the loss of visual acuity observed in natural history. In fact, stabilization of visual acuity was observed following treatment with TSHA-120. The graph on the slide depicts the mean LogMAR by therapeutic dose cohort with LogMAR score on the Y axis and time frame dosing on the X axis. And for completeness, we looked at each eye separately. Baseline or BL on the graph represents the initial treatment time point and values for the left eye increasing in time are on the left side of the graph and values for the right side -- the right eye increasing from time on the right side of the graph. Collectively, what these data demonstrate is that treatment with TSHA-120 resulted in stabilization of visual acuity. For the patient and family, preservation of visual function and ability to see has a significant impact on the child's ability to communicate and on their quality of life. Next slide. Similarly, we examined the thickness of the retinal nerve fiber layer, or RNFL, which is an objective biomarker of visual system involvement and overall nervous system degeneration in GAN. Treatment with TSHA-120 resulted in stabilization of RNFL thickness and prevention of axonal nerve degeneration compared to diffuse thinning of RNFL observed in natural history as measured by optical coherence tomography or OCT. Analysis by individual dose groups as seen on the graph demonstrated relatively stable RNFL thickness, which is in contrast to the natural history of GAN, where RNFL decreases. Overall, these data provide new evidence of TSHA-120's ability to generate nerve fibers and preserve visual acuity. Notably, we reviewed these data with a group of ophthalmology key opinion leaders who felt that visual acuity and the RNFL data were very compelling. Next slide. This data set is the most comprehensive gene therapy data set in GAN, offering TSHA-120 a potentially derisked regulatory path. This program currently meets the majority of registration requirements based on FDA and EMA's guidance for gene therapy for neurodegenerative diseases. This table summarizes key registrational requirements from regulatory agencies, including the FDA and EMA. And as you can see, the GAN program checks nearly all the boxes. We look forward to our continued discussions with major regulatory agencies on potential registration pathways for TSHA-120 and anticipate having a regulatory update by the end of the second quarter. Next slide. As previously indicated, we believe the current data set supports a pathway to registration. We envision a different global regulatory scenarios for this derisked path to BLA and MAA submissions. In Europe, there is potential to file for conditional approval with the current data set based on EMA guidance documents. In the U.S., there are 3 possible scenarios: the first is an immediate filing for approval based on the current data set and the comparability of commercial grade material. What we believe is a reasonable unlikely scenario is that we may need to dose a few more patients to demonstrate comparability of clinical effects between clinical and commercial grade material, which was a similar approval pathway for Zolgensma in spinal muscular atrophy. The final scenario is the need to initiate a new pivotal trial, which we believe is unlikely given the recently published guidance document on gene therapies for neurodegenerative diseases and the extensive long-term safety and efficacy data set available for TSHA-120. I will now turn this over to my colleague, Sean McAuliffe, our Chief Commercial Officer, to discuss the commercial opportunity. Sean?

Thank you, Suyash, and Good morning, everyone. Next slide. We are very excited about the potential of TSHA-120 to be a transformative therapy for patients suffering from GAN and believe it represents a significant commercial opportunity. Based on epidemiology and work Taysha has done to understand where patients are currently being treated, TSHA-120 has the potential to address an estimated 5,000 patients with a GAN mutation globally. While the U.S. is a major market, Europe, other parts of North America and Latin America have a significant number of patients that may benefit from TSHA-120. Next slide. Preparing for a launch is no easy task, but I believe Taysha has the right teams in place. This includes a seasoned commercial team with extensive global gene therapy product launch, preparation and execution experience. As you can see, the team brings significant experience from AveXis, where we all play key leadership roles in preparing for and executing the launch of Zolgensma. We are very proud of the number of patients we were able to help gain access to the first onetime dose gene replacement therapy for SMA. As a result of obtaining broad access quickly, that gene replacement therapy exceeded $1 billion in sales within 2 years of the first regulatory approval. In addition, the team also has considerable rare disease experience from Novartis, Baxalta, Sarepta, Lundbeck and Shire. Next slide. The commercial team, working in close collaboration with our cross-functional counterparts who also all have extensive gene therapy and rare disease experience, have developed and begun implementing the launch readiness plan to maximize patient identification, site readiness and market access by focusing on initiatives across 3 key imperatives: the first is education with the overarching goal of shortening the diagnostic odyssey and help patients gain a diagnosis. As Taysha currently has no approved products, our patient awareness initiatives are done with the goal of helping health care professionals, caregivers and broader patient communities understand GAN, the burden of the disease, clinical development and availability of clinical trials. Our educational initiatives led by Taysha's medical affairs team are raising awareness of GAN and the precision medicine tools second aid in diagnosis. These early efforts also afford Taysha the opportunity to explore all potential access pathways in ex U.S. markets. Our second key initiative is genetic testing and diagnosis. Currently, there are several genetic testing laboratories that allow industry sponsorship to enable ordering of certain genetic tests by a treating physician based on their independent medical judgment at no charge to patients. Taysha has ongoing partnerships in place with Invitae and GeneDx to sponsor 4 panels that screen for over 480 genes. We also have ongoing collaborations with the Hereditary Neuropathy Foundation and Charcot-Marie-Tooth Association Centers of Excellence, health care professionals and patient advocacy groups to increase access to genetic testing and help identify patients with GAN mutations worldwide. Ideally, every rare disease would be included in newborn screening. While patient medical and patient affairs teams are continuing to work towards that goal, our medical affairs group is working with key opinion leaders and our patient affairs team is working with advocacy groups to increase awareness of GAN, genetic testing and the ongoing clinical trials. Once TSHA-120 is approved, these precision diagnostic tools will continue to further develop patient awareness opportunities and help health care providers diagnose patients. Finally, we believe in early engagement with payers. It's critical to ensure a timely and appropriate access across the various reimbursement approaches found around the world. We intend to implement a comprehensive value-based strategy to ensure timely and sustainable access for TSHA-120. Our access strategy will be supported by a global health economics plan with emphasis on real-world evidence and burden of illness to support the value proposition. I will now turn it over to Fred Porter, our Chief Technical Officer, to discuss manufacturing. Fred?

Thank you, Sean, and Good morning, everyone. Next slide, please. At Taysha, we have established a platform for the development and commercialization of AAV therapies that we leverage across our pipeline of CNS-targeted programs. All programs utilize a proven AAV9 capsid and intrathecal route of administration to effectively deliver our therapies to the target organ, the CNS, and minimize off-target effects of systemic delivery. To support this approach, we've also developed a highly scalable, commercial-ready manufacturing process for our lead clinical candidates, including TSHA-120, that utilize suspension HEK293 production process and comprehensive analytical testing panel that allows us to rapidly advance programs from clinical to commercial production and produce our drugs in a cost-effective manner. Next slide. To support our commercial launch for TSHA-120, Taysha has moved rapidly to establish commercial manufacturing for the program. We've recently initiated a collaboration with the leading gene therapy CDMO Viralgen, the contract manufacturing arm of AskBio, who produce the initial clinical material. Viralgen is currently testing clinical lots to ensure consistency, efficacy and safety of the commercial material. Their platform process is suited to produce a comparable product of the previous vector lots that will meet regulatory requirements. The recently announced commercial manufacturing facility in San Sebastian, Spain is positioned to support supply of commercial material for GAN. Tech transfer and manufacturing activities are already underway to complete process and analytical validation work for TSHA-120 to support future clinical needs and BLA and MAA submissions. I'll now turn it over to RA to discuss next steps. RA?

Thank you, Fred. Next slide. Collectively, these results demonstrate clinically meaningful and statistically significant slowing of disease progression in patients with GAN following treatment with TSHA-120 across all therapeutic dose cohorts. Importantly, long-term analysis confirms the sustained durability of TSHA-120 treatment. These data also provide new evidence of TSHA-120's ability to generate nerve fibers and preserve visual acuity, which are significant findings for patients affected by neurodegeneration. 53 patient years of clinical safety data support the safety and tolerability profile of TSHA-120. The totality of data generated by TSHA-120 to date support our plans to engage with regulatory agencies to discuss pathways for registration. Next slide. As for next steps, our final engineering run is currently underway to support the commercial validation run, which will be released in the third quarter. We expect to complete our comparability assessment in the near term to support potential regulatory submission. The company plans to continue discussion with major regulatory agencies regarding possible pathways for registration. And on the commercial front, we will continue our institutional readiness, patient identification and payer engagement activities. Lastly, we expect several publications from the GAN data set by the NIH in collaboration with Taysha. Next slide. We would like to thank the patients, families and caregivers for their participation in our clinical study. We'd also like to thank our partners at the NIH and UT Southwestern, notably, Dr. Carsten Bönnemann and Dr. Steve Gray for their collaboration and dedication. Finally, we'd like to thank our advocacy partners for their continued support. With that, I will now ask the operator to begin our Q&A session. Operator?

[Operator Instructions] Our first question today is coming from Debjit Chattopadhyay from Guggenheim Securities.

This is Robert on for Debjit. Congrats on the good data for TSHA-120. Two questions from me at this point...

Thanks, Robert.

Of course. What is the risk the FDA could ask for randomized data considering the pace of disease progression in GAN? And then secondly, can you detail the importance of visual acuity and nerve regeneration as endpoints? And how the secondary endpoints may play when approaching the FDA?

Thanks, Robert, for the question. Maybe I'll tackle the first one and Suyash can discuss the secondary endpoints and the potential risk of an additional study. So for -- in my opinion, to be quite honest, I think being asked to do an additional study with a data set like this, where this is the first intrathecally dosed gene therapy trial in history that initiated in 2015 that is continuing to be assessed. We now have 53 years of patients' data with consistent data across the years, either with halting or clinically meaningful slowing of disease with a very favorable safety and tolerability profile. I think this would essentially be going against their own guidance document, which they issued around this time last year. They're really focused on a couple of things that, really, the GAN program hits on. The selected endpoints, validated endpoints, meaningful endpoints of selection of how to manufacture the material, when it's appropriate to use natural history, which essentially this program checks all the boxes. I'll stop there and turn it over to Suyash to provide his comments. Suyash?

Yes. Thanks, RA, and thanks, Robert, for the question. With regard to the risks of being asked about doing a randomized trial, it's always there. But as RA said, there was this very nice guidance that was drafted about a year ago. And they really -- they tackled a specific point about the need for a control and what kind of control would be appropriate for a rare disease for a gene therapy trial. And they talk about natural history studies, but they also talk in more detail about how the natural history should be run. And this study that's been run by the NIH started in 2013. So we have 7 or 8 years' worth of comparative data. We have 45 patients in the baseline data set that were published in -- it was about July last year in the [indiscernible]. And we have 50 patients in the study currently. So it was a really extensive data set for a disease that's so rare. We collect all the endpoints in the natural history data set that have been used in interventional studies. We have a very specific mashed control process. So for all of those reasons, it would really be going against their guidance that they asked us to do, a new study that brought in randomization and blinding. So we're looking forward to having these discussions with the FDA and seeing where they end up. But in line with their guidance of those things that would be really something they would really ask for -- shouldn't ask for based on the guidance. Having said all that, it is the FDA and they can ask for what they would like. With regard to the secondary end points, I think visual acuity and the pathology are both really powerful endpoints. And regulators view data, especially rare diseases, not in terms of a single registration enabling end point but in terms of the totality of data, especially with these multifunctional diseases, where they like to look at the breadth of assessment. So we have -- they have a very, very broad breadth of assessments in the interventional trial, but several which we haven't even spoken about, there'll be more end point data coming as the year emerges -- as the year progresses. On the visual acuity side, we spent some time speaking with patients and found this through our formalized workshop process. And they've told us repeatedly that the visual acuity -- the loss of visual acuity is incredibly upsetting and causes huge burden on patients and families, specifically in the realm of communication because as these children are losing their capabilities in terms of speech, language, movement, the last way that parent can communicate with the child is nonverbally. And when the child loses their vision, that's devastate the patient and family. So from a humanistic burden perspective, visual acuity is incredibly important, preserving as incredibly important, and the regulator will definitely appreciate that. As I mentioned on the call, we shared the data both on LogMAR, visual acuity preservation and RNFL thickness with a group of experts -- ophthalmologists. And they were very compelled by this data set. With regard to the biopsies of the nerve, also very powerful data. In fact, it's probably the most powerful data you can have for a gene therapy study. And we're very fortunate in that for most CNS type diseases, you can't actually get the biopsies performed because you need to biopsy the brain or a nerve. But it's the NIH and they're able to get some of these processes and procedures through IRBs. And we're very fortunate that every patient had a pre-gene therapy and post gene therapy nerve biopsy. And as I mentioned on the call, we can see the emergence of these regenerative nerve clusters, demonstrating that the nerve is actually starting to regenerate, which is in line with the efficacy data we're seeing both on visual acuity and on -- movement on the MFM32. So I think those 2 endpoints will be really critical. They look at the disease from a different perspective than the MFM32. So they complement the data, the primary endpoint very nicely. And I think that will be appreciated by all regulators. I'll stop there, Robert.

The only thing that I'll add to Suyash's point is every patient that enrolled in the interventional trial was also enrolled in the natural history study. So we're fortunate in that case that each one of those patients serve as their own historical control arm, which is quite nice, too. So I think with all that being said, including what Suyash mentioned, I think we're in a pretty good spot.

Our next question is coming from Gil Blum from Needham and Company.

Congrats on the evolving results. Just one from us. So maybe to orient us against natural history and based on the cumulative data that you have so far, what would've the hypothetical TSHA-120 treated patient disease progress look like? So would this be like patients who have late onset disease? Or would this even translate into a lower disease burden than what's seen in late onset disease?

Thanks, Gil. It's a really good question. Maybe I'll start with a couple of comments and turn it over to Suyash. I think just as in our experience with Zolgensma, earlier treatment is going to always be better. And the hope is that if and when this therapy is approved, that we would be able to identify patients either in utero or shortly after birth. And the goal is to actually stop degeneration before it happens. And that's really the goal, I think, of all CNS gene therapy. With that being said, I think the level of appropriateness for patients to be treated would be both in the early onset and the late onset disease states. And the reason why I say that is because at birth, you can't tell which phenotype that a patient is going to get. Essentially, they suffer from a GAN mutation. And so again, I think the level of appropriateness for both phenotypes is going to be high when using this therapy. I'll stop. Suyash, maybe you want to give your comments.

Sure. Thank you. Thanks, Gil. So in terms of -- your question is really about the hypothetical disease progress of these patients. And I guess it's not so much hypothetical. It's more based on the data we're seeing, in fact, in the natural history study, the 50 patients that have been characterizing the natural history study. And we know by looking at these data sets that these children progress. We know that when you look at the primary endpoint, the MFM32, these patients drop their MFM32 scores by about 8 points per year. So the MFM32 is a 32-question scale that's well used by neuromuscular physicians. It has been used in a number of different neuromuscular diseases and is validated and well understood and has been used as an end point in other regulatory submissions. What we know is that a 4-point change in this scale is clinically meaningful. So these children -- healthy child will start at 100 and they'll drop roughly from about 2 or 3 years onwards by about 8 points per year, so a clinically meaningful decline year-on-year. By the time these children hit about 70, they're needing a walking aid. And by the time they hit about 50 or 55, they need a wheelchair. And then they decline towards -- sadly towards death. So these children have been treated from a mutational perspective or have the classic form of GAN or the early onset form of GAN. So their progress would have been not dissimilar from what I'm outlining here, i.e. [indiscernible] around 2 or 3 with unsteadiness of gait, stumbling, starting to lose motor function, sensory function in a wheelchair at the age of 10, on a ventilator by 15 and usually death by the late-teens or early 20s. There's nothing to suggest these children wouldn't have followed that progress based on the well understood natural history now. And I think what we're seeing with the gene therapy is that we're halting that progress. And as RA mentioned, if we're going to halt that progress, we'll significantly slow that progress down at an earlier time point when the child is younger and they have retained greater function, that will result in a better outcome for the child and family. I'll stop there, Gil.

Our next question is coming from Salveen Richter from Goldman Sachs.

This is Elizabeth on for Salveen. Just wondering if you could just remind us of the immunosuppression regimen for TSHA-120? And then is there a potential? Or are you thinking of maybe tweaking the protocol at all just in light of what was seen in the GMT study data?

Elizabeth, thanks for the question. I think we're very fortunate in this case that this study initiated almost 7 years ago and used as an almost identical immune suppression regimen as we use across multiple of our studies. So we actually have really good patient experience with this immune suppression regimen. Again, just to remind you, we have 53 years of patient safety and tolerability data just from this data set alone, not to mention the other programs that we currently have in the clinic, including the CLN7 and GM2. And so for us, we feel really fortunate that we are able to take from this data set and then to be able to apply it across our portfolio and across our platform. I'll stop there and let Suyash go into some of the specifics around immunosuppression regimen.

Yes. Thanks, RA, and thanks for the question, Elizabeth. Yes, it's an important question. The thing -- a high-level comment about immunosuppression in gene therapy studies, Elizabeth, we all know that we need to have patients be immunosuppressed. But the degree, the duration, the types of medications that are to be used, people still discuss and there's no clear common approach. And likely, it does vary from disease-to-disease and patient-to-patient to some extent. But RA is quite correct. We've been using this regime of prednisolone and sirolimus also known as rapamycin, in this study for -- really since the interception, the first few patients were just dosed on prednisolone. And what the principal investigator found was that there was a degree of what has been termed CSF pleocytosis, i.e., an elevation in white cell count, which was fully asymptomatic. But in view of the fact that this may be some kind of a T-cell mediated reaction, the PI cost embodiment are very well known, very well respected pediatric neuromuscular and neurological expert of the MAAs decided in conjunction with the DSMB years ago to add in some rapamycin. So the way the study is currently run is that we generally give this combination of prednisolone for a shorter period of time, perhaps 4 to 6 months and rapamycin for up to a year. And we dose prednisolone in a specific way for a period of time then we tail the prednisolone dose off. And we give rapamycin at a particular dose. And we tailor the rapamycin to a specific serum level. And we're very comfortable with that regime. When you look at all the different regimes that are being used in gene therapy, this is a more moderate type of regime. It's not -- it's minimally immunosuppressed. It's not as maximum as some companies use, but it's kind of in the middle. We're very comfortable with that. In terms of tweaks, given the sad situation of the GM2 patient passing away that we talked about last week due to a concomitant infection, I think that child passed away -- I think the immunosuppression regime may have contributed to the susceptibility to the infection and the child's ability to fight off the infection. But don't forget, he suffered from a very severe degenerative disease, and he had a little reserve capacity. So we are constantly talking about with our immunology advisers and of DSMBs across our programs what is the right in the suppression regime. In general, we're going to stick to prednisolone and sirolimus. What we might do is we might modify the doses slightly and we might modify the duration slightly. And likely, we'll bring in a little more PI discretion and flexibility in treating each patient. But it's a complicated -- it's a complex question. We have lots of thoughts about it. But hopefully, I've answered your question, Elizabeth.

Our next question is coming from Laura Chico from Wedbush.

I wanted to switch directions a little bit, maybe a minute on patient identification efforts. I thought Sean's geographic breakdown was helpful. I'm wondering if you could just remind us, A, how you're getting to that 1,500 patient number on an addressable market number? And then it sounded like there could be about 1,200 in the U.S. Are you willing to share any identified patient numbers today?

Laura, thanks for the question. Essentially, we're not willing to share identified patient numbers today. But what we would say is -- and Suyash has mentioned this before. There's about 50 patients currently in the natural history data set. So that just gives you a sense of where the patient numbers are coming from if we were able to enroll and the NIH was able to enroll 50 patients or so in that natural history. So we're pretty confident. This is more of a prevalent disease than an incidence disease. And so what I mean by that is these patients in the early onset form of the disease tend to succumb around the late-teens, early 20s. Patients with the Charcot-Marie Type 2 later onset version of the disease tend to live within their -- the fifth decade of life and may not necessarily be life limiting. So when you go back and actually do the calculations, the combined incidence is somewhere between around 1 in 200,000 to around 1 in 215,000 or so from an incidence perspective. And then obviously, you just multiply that times the survival rates of both phenotypes. And that's how we get to the number that we've gotten to. So we're quite confident in that particular calculation. Maybe, Sean, you just want to briefly cover our patient identification activities. I believe we have about 20 more minutes and a number of questions to get to. So maybe we should just keep it brief, Sean.

Yes. Thanks, RA. Laura, I think it's a great question and RA summed it up pretty well. The only thing I would add, in working with the advocacy groups and the centers of excellence I talked about, if you look at the literature, 65% of that CMT2 patients population has not been genetically tested. But I think those efforts have been very fruitful in helping us understand more about that patient population and give us additional confidence that there are patients out there that don't have the specific genetic diagnosis that we're going to be looking for now and in the future. So thanks for your question.

Thanks, Sean. The only thing that I would add, Laura, is there are a number of hotspots from -- or founder mutations that we've actually located, one of which is actually in Turkey, which is, from a commercialization perspective and reimbursement perspective, certainly an addressable market for ultra-orphan diseases and gene therapy. So with that, we're quite confident in these worldwide numbers. And as you recall, I actually had a number of -- a great deal of experience of actually commercializing rare disease products ex U.S. as I was actually doing that in a previous life. And so a number of these initiatives where you would reference a major market approval in order to gain reimburse early access is quite tried and true, and I have personal experience being able to execute on that.

Our next question is coming from Joon Lee from Truist Securities.

Congrats on the data. Regarding the natural history study, which data that you're using as a proxy for placebo, is the rate of decline the same across different age groups? And if you were to match the age of the patients in your trial to that of the same age group in the natural history data, would you get the same outcome? I'm asking because your including criteria had a wide range of 3 to 43 age. And I'm just curious what the age has -- may have impact -- the ages impact on [indiscernible] decline -- rate of decline?

Thanks, Joon, for the question. So it's a really insightful question. Keep in mind, this study was started in 2015. So this is kind of the infancy of clinical trials from a gene therapy perspective. And also keep in mind, this was the first intrathecally dosed gene therapy trial in history, which initiated in 2015. So I think the sector's thinking has evolved on how you would normally do a clinical trial. I think back then, if you would look, it was around starting a little bit later in disease and making sure that the product was safe. And then once you were able to tackle that bar, then you would be able to kind of move down. We recently actually amended the protocol to actually go down to age 3. And so for that, obviously, the earlier treatment you have, the more effective the therapy is going to be. This is particularly important in monogenic CNS disease. I'll stop there and, Suyash, maybe you want to give your comments on kind of how the correlation of age and disease progression from the natural history perspective.

Sure. Joon, thanks for the question. I think in general, yes, what we've noticed -- and we have a nice diagram in one of our slides. I think it was Slide 7 earlier today that showed individual plots of the MFM32 for patients. And what we've seen is that in general terms, for the ages at which the MFM scale is validated, that's from 2 years and up, so I'm excluding the 0 to 2 year olds from this. In general, it looks as though this 8-point decline is relatively consistent across the years. So a 4-year-old who starts with the baseline score of 90 may drop -- likely drop approximately 8 points per year. A 5-year-old who is at a score of 70 or 75 may drop points per year. So in general, that seems to be absolutely true. The group of patients who is not true for are the later onset form. And we talked a little bit about how many of the late onset form -- many of the late onset GAN patients who are misdiagnosed or mischaracterized with Charcot-Marie-Tooth type disease. The vast majority of patients in natural history study are actually the early onset form that drop out by about 8 points a year. Looking at the small number of patients from the later onset form, they drop by about 4 points a year on the MFM32. So they do have a slower rate of decline. But I would say that across the age range, that 8 points per year decline looks relatively stable. And RA is quite right. When the study started -- when the interventional study started, the older patients were being dosed first. So the study started in patients, the 8-, 9-, 10-year olds were being dosed, that's dropped now to 5-year-olds. So the youngest patient that's been dosed thus far is a 5-year-old. And that's just been dropped now to 3. So we now dose children at the age of 3. So I think if the study had been designed to set up now would have probably dosed younger earlier. But that's the way the processes have been going. And I think the key thing here is let's identify these patients early. Sean has talked about some of the approaches to doing that. And let's treat these patients as early as possible when they have a high MFM32 score, they're doing well on other endpoints, and we can stabilize them earlier and maintain function for long term.

Our next question today is coming from Mike Ulz from Morgan Stanley.

Congratulations on the data as well. Just one quick one for me on the biopsy data. I noticed you had 5 of 6 patients that experienced nerve regeneration. Just curious on the 1 patient that didn't, was there anything unique there? Like for example, they were on the lower dose or maybe they were the late responder?

Thanks, Mike, for the question. I'll turn it over to Suyash to address. Suyash?

Yes. I think one of the challenges -- I think with that particular patient, I think it was less about what was happening histopathologically and more about the actual quality of the sample. It is very tricky getting a high quality sample of nerve tissue. It's a very -- these are very technically challenging. It's a superficial nerve. It's very thin. You've got to process it in a particular way. So I think it wasn't that the last -- the sixth out of the 5 -- the sixth sample didn't show any regeneration because it was more of an uninterpretable biopsy. But I think the 5 biopsies we talked about earlier clearly showed the emerging presence of a large number of these regeneration clusters. And as I mentioned on the call earlier, we'll be looking at -- we'll currently analyze with additional patients. And my expectation is you will see even better data because the later patients have been on the higher dose. So I'll stop there, Mike.

Our next question is coming from Eun Yang from Jefferies.

So I have a question on natural history study. So can you talk about how long the study has been followed when it started? And is that only in the U.S or is that international? If so, can you kind of break down how many patients by geographic area?

Thanks Eun, and thanks for the question. Suyash, do you want to comment?

Sure. Yes. So the study started in 2013. So it's been going for 9 years actually. And it started 2 years before the first patient in the interventional trial was dosed. 50 patients have been enrolled. There's a really nice in-depth paper that was published in 2017 in Brain that talked about the baseline data set in the first 45 patients. And in that paper, it actually outlines a lot of detailed demographic characteristics of that -- of the population in the study. Now the study is a single-center study. It's only being run at the MAA, but patients are flying from all over the world to this particular center. So I forget the number of countries, but we have U.S. patients, we have Canadian patients, we have patients from Mexico, patients from Turkey, patients from Europe. So I think a large proportion of the patients are from the U.S. We have representation from all over the world. So I hope that answers your question, Eun.

Our next question is coming from Yun Zhong from BTIG.

I wanted to confirm if you have seen a dose response between the highest dose cohort versus the previous dose cohorts? And is that -- because when you combine all doses, the MFM32 data seems to be better than the highest dose cohort alone. Is that mainly due to the late responders?

It's a really good question. I would -- most likely, the answer is absolutely. It's due to the late responder. If you can see -- and these are pretty small ends. And I'll let Suyash tackle this after I finish. But if you look at the late responder patient, that patient essentially performed almost identical to the natural history decline. I believe that patient had a 7-point decline in year 1. Natural history declines by 8 points from year 1 to year 2. That patient was essentially flat, which is -- which for us is demonstration of disease stabilization. And the hope is that, that patient continues on that pathway. And so if you would think about this -- and Suyash mentioned this before. If that patient would actually have had responded immediately, I think you would see -- you would be able to see a dose response -- a more significant dose response emerging. Keep in mind, this was only 3 patients that were dosed at the high dose. So 1 patient obviously is going to skew the data. Most importantly for me is that the mean data was statistically significant and clinically meaningful even with that late responder patient included. And so for us, that just gives us a lot of confidence that the therapy is doing exactly what we hoped it would do. Suyash, maybe you want to comment?

Sure. Yes, I'll make a couple of comments to build on RA's thoughts. So when you look at the study as a whole, it doesn't show dose response unless absolutely. So we look at a very low dose of 3.5x10^13, these patients didn't really improve on their MFM32 score, although parents and families said that there were some suggestion of physical improvement and greater energy. And then when you jump from that 3.5x10^13 to the higher doses, you could see clearly -- there seems to be some kind of threshold difference. But once you hit the medium-low dose, you're seeing quite significant clinical effect. What I will say is that the medium-low, medium-high and the high doses are all relatively close to each other. And once again, if we design the study now, I think we would have designed it in a way where you had greater jumps in dose because 1.2x10^14 to 1.8x10^14 to 3.5x10^14, there's not that much difference in the amount of drug you're giving for a biological agent. So you're only going up threefold from 1.2x10^14 to 3.5x10^14. So in some respects, you're not really increasing the dose that much. So we -- which is one of the reasons we decided to look at the 3 dose cohorts together because there's not that much difference in the dose that's been given. With that 3.5x10^14 dose, there was the late responder and it was quite interesting. I had a long discussion with the PI about this particular patient. And he wasn't entirely sure why that patient didn't show much improvement to start off with but then did show significant improvement from year 1 to year 2. He did say it was in the middle of COVID and he wasn't seeing the patient as much and they weren't accessing medical services as much. So if that maybe they had some impacts, maybe not, we don't know. It could be just one of those things. But I definitely think that late-responding patients probably drove the score in the high dose cohort down a little bit, which brought down the mean. But once again, you can see that even the high dose cohort is statistically significant, the difference is 5 points -- 5.3 points in the MFM32. We know a 4-point change is clinically meaningful. And then when you sum the 3 therapeutic dose cohorts together, you see a 7-point change over the course of the year in the MFM32. So once again, a very highly clinically meaningful and also statistical significance.

Our next question today is coming from Kristen Kluska from Cantor Fitzgerald.

Congratulations on these data. Just a question, as this data set reinforces your plans to engage with regulatory agencies, is your understanding still that the priority review about your program has been extended until September 2024? And if approved by this time, how would you consider valuing a voucher given the remaining of your pipeline is also targeting rare diseases where you have relevant designations in place?

Thanks, Kristen, for the question, and maybe I'll address it. It is our understanding that the pediatric review voucher is extended until that time period into 2024. And this program actually has a rare pediatric disease designation that would enable us to qualify for that review voucher. So obviously, that's of extreme importance to our program itself and one that makes us even more excited about the potential commercial opportunity. Obviously, the value of those vouchers have stayed somewhat stable over the last, I would probably say, 3 years or so. Most recently, I believe the last voucher sold for a little bit over $100 million. And so obviously, that would be our expectation if we were able or ever decided to monetize that upon potential approval of this program. But currently, that's our thinking once this program does have rare disease pediatric designation. Also, the extension of the pediatric review voucher would extend into 2024. And kind of the current data set supports that the value of those vouchers are around $100 million. That's the way we've looked at it. Hopefully, that addresses your question. Thanks, Kristen.

Our next question today is coming from Raju Prasad from William Blair.

This is Sami on for Raj. Congrats on data. You previously discussed that you might be able to readminister TSHA-120 through the vagus nerve. Are you guys planning on doing any additional Phase II examined redosing? And based on today's data, do you think redosing will be necessary in certain patients?

Thanks, Sami. It's a great question. Maybe I'll start and then I'll turn it over to Suyash. We've done some really interesting work. And when I say we, it's the collective we, but it's really our partners at UT Southwestern, which is led Dr. Steven Gray and Dr. Rachel Bailey. But they've done some extensive work around the redosing of gene therapy using the route of administration direct injection to the vagus nerve. And recently, we've actually achieved animal proof of concept in that -- in a large animal model, in a dog model, where we were able to successfully achieve target transduction efficiency in the presence of a previously dosed animal, which we were quite excited on, which provides proof of concept for being able to dose directly to the vagus nerve, which would obviously have an effect on autonomic nervous production. And so for us, we're quite excited, not only for this program but also has the potential to do dual routes of administration, but particularly across our platform because the model actually used AAV9 dosed animals. So we're quite excited about that data. And more of that data will be coming out later this year. But I'll stop there. And Suyash, maybe you have some thoughts on that?

Sure. Thank you, and thanks for the question, Sami. Yes. We have -- redosing is clearly one of those hot topics in the field at the moment. And it's interesting when you think about the CNS approach because in reality, theoretically, I don't think we're going to need to redose in general gene therapies that target the CNS and the PNS. And the reason is that you don't get a turnover of neuronal cells or brain cells in the way that the liver turns over or the muscle turns over. Once a brain cell neuron is transduced, it should stay transduced, producing the protein in perpetuity. So theoretically, you wouldn't necessarily need redosing. Having said all that, we do have, as RA has mentioned, a very, very robust effort to think about redosing and also think about different routes of administration. And RA is correct, we've done proof of concept in the animals, both in NHPs and in dogs. And we've actually used the GAN construct for some of those studies. We've been able to dose successfully in combination with -- and this relates to one of the questions earlier, in combination with the same immunosuppression regime that we've been using clinically i.e. this combination of prednisolone and rapamycin. So yes, that's moving forward nicely. Our program is moving forward. Whether we're going to need to redose patients either in the GAN study or any of our other programs remains to be seen. You've seen in our longest-serving patient here, out 5 years and the difference in their clinical situation in comparison to their natural history is significant. It's 25 or 26 points on the MFM32 score, and that's 5 years out. So far, it looks as though the gene therapy in GAN is persisting. But we've got a great robust approach to looking at redosing and that's moving forward nicely. And hopefully, we'll be able to share more information on that as time progresses.

Our next question is coming from Silvan Tuerkcan from JMP Securities.

Congrats on the data. I just want to ask about what is the potential path into earlier patients. So right now, you said the trial is enrolling patients down to age 3. How do you get into younger patients? Would you think that, that would just be extended? Or is there another trial needed for that? And then just in terms of the market opportunity, I suspect that since you call this more of a prevalent disease, that most of the patient populations will be older than 3 years?

Thanks, Silvan, for the question. Suyash, do you want to tackle it?

Sure, I can do that. Thanks, Silvan. Yes. It's a really good question, Silvan. And in line with our general approach, we want to get to as young patients as possible. So yes, the study -- the youngest patient is 5. We've just had the protocol amendment approved down to 3. Actually, the protocol approval was down to 2, but there's a logistical challenge at the NIH in that we actually need to open up a partner site at Children's National to dose younger than 3. So actually, that's in the process of happening. So very soon after experts will be going down to 2 years of age. The challenge with children younger than 2 is that they generally don't come to our attention. They usually present around the age of 2. So I think once we've dosed 3-year-olds and 2-year-olds, we'll absolutely be able to open younger than that. But the issue becomes how do we diagnose them. Now if we start newborn screening or if there's a pilot somewhere, we would be able to identify patients early or if it's a family member, so they had a child with GAN and had another baby and that baby turns out to have GAN, I think they'll then be eligible for the study. So my expectation is we can go -- we will be able to go down as young as possible relatively quickly. The challenge becomes those 0, 1 and 2 or just identifying them in the first place. Hopefully, that answers your question, Silvan.

And just to add to what Suyash mentioned, Silvan, you mentioned that the majority of the patients that would be aggressively commercially are going to be older patients or at the very least older than 4. And this is certainly where we have experience in the clinical trial. So I think that experience correlates nicely to the addressable commercial patients that would be out there. So I think you're absolutely right. The goal is always going to be to treat earlier. Newborn screening efforts would be underway shortly before approval. But certainly, the goal is to always be able to treat sooner.

We reached the end of our question-and-answer session. I'd like to turn the floor back over to Mr. Session for any further or closing comments.

Thank you, operator. And today, we just are really excited to share with you this groundbreaking clinical trial, this data set which we're quite excited about and the potential commercial opportunity for GAN patients and the ability to be able to help GAN patients around the world. Again, we'd like to thank our partners at the NIH, which is under the leadership of Dr. Carsten Bönnemann, our partners at UT Southwestern as well as the patients in the clinical trial. So for that, we have a large dose of gratitude. We really appreciate your interest in joining us today, and we hope you guys have a great rest of the week.

Thank you. That does conclude today's teleconference and webcast. You may disconnect your lines at this time, and have a wonderful day. We thank you for your participation today.