Tenaya Therapeutics, Inc. (TNYA) Earnings Call Transcript & Summary

Hello, and thank you for standing by. My name is Lacey, and I will be your conference operator today. At this time, I would like to welcome everyone to the Measuring Protein Expression in Cardiac Gene Therapy, KOL webcast event. [Operator Instructions]. I would now like to turn the conference over to Michelle Corral, Vice President, Investor Relations and Corporate Communications. You may begin.

Thank you so much, Lacey, and good morning to everyone. I appreciate you joining us once again for our KOL webinar and hope that everybody can see and hear our presentation this time. Welcome to our KOL webinar event to discuss the measurement of protein expression in cardiac gene therapy. Joining me today on today's call from Team Tenaya are Faraz Ali, Tenaya's Chief Executive Officer; Dr. Whit Tingley, our Chief Medical Officer; and Kathy Ivey, our Senior Vice President of Research. As well as our special guest. We have with us Dr. Mike Previs of the University of Vermont. As a reminder, the information discussed during this call will include forward-looking statements, which represent the company's view as of today, August 26, 2025. These statements involve certain assumptions, and we caution investors not to place undue reliance on this information. Please refer to our filings with the SEC for information concerning risk factors that could cause actual results to differ materially from those expressed or implied by these statements. I would also like to note that any opinions expressed by our guest speaker, Dr. Michael Previs are his own and do not necessarily reflect the opinions of the company. And with that out of the way, I will pass the mic to Faraz Ali for some introductory remarks. Faraz?

Thank you, Michelle, and good morning, everyone. I'm Faraz Ali, Chief Executive Officer of Tenaya Therapeutics. At Tenaya, we're driven by bold and urgent mission to discover, develop and deliver curative therapies that address the underlying drivers of heart disease. This mission is not just aspirational, but it's foundational to everything we do. So on behalf of the Tenaya team, and thank you for joining us today for our session focused on measuring protein expression in cardiac gene therapy, during which we will discuss the ground breaking work being done at Tenaya to transform the standard of care for patients living with genetic cardiomyopathies. Today, I'm proud to share how we are translating that mission into meaningful progress for patients and families affected by devastating genetic cardiomyopathies. We were founded in 2016 with a singular focus, and the company was filled very intentionally with deep expertise and integrated capabilities. In addition to our core value of patients first, Tenaya's culture is rooted in scientific excellence and tenacity, which together have resulted in a pipeline of 3 clinical stage programs, including 2 novel gene therapies that each have near-term data readouts. We're in a pivotal moment in our journey as an emerging leader in gene therapy for inherited heart conditions, we're advancing a pipeline that targets the root cause of disease, not just the symptoms. These are not incremental steps. They represent a paradigm shift in how we think about treating and ultimately, hopefully, curing cardiomyopathies. And they are made possible by the deep scientific expertise, relentless innovation and unwavering commitment of our team and collaborators. Here's our pipeline. Our lead program is focused on hypertrophic cardiomyopathy, or HCM, where we're developing TN-201 and adeno-associated virus or AAV based gene therapy designed to treat adults and children with HCM due to MYBPC3 gene mutations, the most prevalent form of genetic HCM. This program is advancing through the clinic and progressing steadily towards pivotal studies. A data readout encompassing the first 6 patients to receive TN-201 is planned for the fourth quarter of this year. Our second gene therapy program is TN-401, another AAV9-based gene therapy being developed for the treatment of arrhythmogenic right ventricular cardiomyopathy or ARVC caused by mutations of the Plakophilin-2 or PKP2 gene. Our RIDGE-1 clinical study for TN-401 is ongoing with an initial data readout for the first 3 patients coming in the fourth quarter of this year. Both MYBPC3 -associated HCM and PKP2 associated ARVC are the result of protein haploinsufficiency, which is the lack of certain proteins that are crucial for the heart to be properly is the cause of the condition. Unlike other gene therapies, we may be more familiar with in MYBPC3 associated HCM and PKP2 associated ARVC, the vast majority of patients are heterozygous and producing some protein with their 1 working gene. That also means that the protein products produced by TN-201 and TN-401 are indistinguishable from the background protein. As a result, the task of understanding how much background protein already exists and discerning whether gene therapy is increasing the patient's protein levels and ameliorating the cause of disease, is more complicated versus other gene therapies. This is also quite important in the context of seeking potential accelerated approvals based using protein as at least 1 surrogate marker for efficacy, for which there is recent and relevant FDA precedents. We have learned a lot from TN-201 program that we're now applying to our second TN-401 program. Today, we hope to illuminate our learnings as we've set out to conquer the challenge of protein measurement in cardiac gene therapy. In today's session, Dr. Whit Tingley, Tenaya's Chief Medical Officer, will review the clinical status of TN-201 and TN-401 and outline expectations for the near-term data readouts. In order to offer additional context with those readouts, Dr. Kathy Ivey, Tenaya's Senior Vice President of Research, will be joined by Dr. Michael Previs to discuss the rigorous methodology Tenaya has undertaken to measure proteins. Evidence of protein expression is a critical early sign of our gene therapy success in addressing the underlying cause of each of these conditions as well as providing a valuable clue to the therapy's durability over time. Dr. Michael Previs is a world-leading expert in heart muscle disease and specifically, the characterizations of key proteins critical to healthy heart function. Professor of Molecular Physiology and Biophysics at the University of Vermont's, Larner College of Medicine. Dr. Previs lab uses a combination of mass spectrometry-based proteomic strategy and state-of-the-art single-molecule imaging techniques to characterize the structure and function of muscle protein complexes in health and disease. Specifically, he is focused on understanding the molecular mechanisms by which myosin binding protein C regulates the heart's ability to contract. Finally, we will open up the call to questions from our covering analysts for Dr. Tingley, Ivey and Previs. We also invite questions from investors, which may be submitted into the chat box on your screen. With that brief overview, let me turn our event over to Dr. Tingley. Whit?

Thanks Faraz. As Faraz mentioned, this is an exciting time for Tenaya and for our lead gene therapy programs, both TN-201 for MYBPC3-associated hypertrophic cardiomyopathy and TN-401 for PKP2 associated ARVC. As of now midway through the year, we have achieved significant progress against our goals for both programs. For TN-201, we presented initial data at the American College of Cardiology in March, covering the first dose cohort, Cohort 1. And we've announced we have fully enrolled the second dose cohort and that the DSMB has reviewed all available data from these cohorts and recommended expanding the study per protocol. For TN-401, we presented initial data on the natural history at the Heart Rhythm Society this year. Cohort 1 is fully enrolled, the DSMB reviewed the Cohort 1 data and recommended we proceed per protocol to expand that dose level and escalate to the next dose level at which we have initiated. We will have data readouts from each program in the fourth quarter of this year. But before providing details on what to expect, let's take a moment to cover important characteristics of each condition and the aims of the current studies. Let's start with MYBPC3-associated hypertrophic cardiomyopathy. MYBPC3 is the most common genetic cause of HCM, estimated to affect 120,000 patients in the United States and many more around the world. As far as mentioned, the vast majority of patients have heterozygous mutations, though on rare occasions, infants are born with 2 MYBPC3 mutations. Their disease is so severe, they require heart transplantation to survive even the first year of life. Heterozygous mutations can affect people at any age, affected children such as Gabe pictured here typically of earlier progression and a higher overall burden of disease than adults. We look forward to sharing a natural history data from our MyClimb study of pediatric MYBPC3+HCM patients this weekend at the European Society of Cardiology Congress. In this disease, the muscle of the left ventricle thickens and is unable to relax and fill properly limiting cardiac output and the ability to meet the body's demand for blood supply. Initial symptoms are shortness of breath, chest pain, fatigue, syncope and palpitations from arrhythmias. Patients are at high risk of sudden cardiac arrest and death. Fortunately, for patients, the treatment of HCM has gotten more attention in recent years. However, there's still no treatment that addresses the underlying genetic cause of disease. First-line therapy involves generic heart failure medications. Recently, cardiac myosin inhibitors have emerged, but these are only available for the subpopulation of patients with the obstructive subtype of HCM. Only about 30% of patients with MYBPC3 associated HCM have obstructive HCM. So the majority are not eligible for current cardiac myosin inhibitors. MYBPC3 HCM is caused by protein insufficiency, mutations to the MYBPC3 gene result in a lack of myosin binding protein C, which is critical for regulating the contraction and relaxation of the heart. Protein C deficiency results in hypercontractility, stiffness, thickening of the ventricle and disorganization of the muscle cells themselves, which places patients at greater risk for lethal arrhythmias and progressive heart failure. TN-201 gene therapy is designed to address this protein deficit by inserting a full-length working MYBPC3 gene into the heart where it can produce more C protein and restore contraction and relaxation. This is expected to halt disease progression and potentially reverse disease, ultimately improving patient outcomes, symptoms and quality of life. In November 2023, we dosed the first patient with TN-201 gene therapy as part of our MyPEAK-1 Phase Ib/II clinical trial. The primary objective of the study is to establish the safety profile of TN-201 and evaluate tolerability and performance at 2 different dose levels. In addition to safety, we are taking heart biopsies to assess the transduction and expression of TN-201 using cardiac imaging to examine changes in heart structure and function. Following levels of plasma biomarkers associated with disease and with heart strain, monitoring heart failure functional class and exercise capacity and measuring patient-reported outcomes for symptom improvement and quality of life. The ultimate goal is to see many of these parameters of disease improving together consistently. We are often asked what will be included in our planned Q4 data readout and what success would look like. For Cohort 1, we plan to share longer-term follow-up data building on our prior presentation at ACC in March. The 3 patients who received TN-201 at the 3E13 dose should have assessment out to 1 year and more, including heart biopsies. Success here is continuation of the positive observations we have shared thus far. We shared at ACC that TN-201 has been well tolerated with robust evidence of DNA transduction and increasing RNA expression as well as improvements in protein levels for the first 2 patients. 2 of the Cohort 1 patients saw improvements in one or more measure of cardiac hypertrophy and all 3 achieved New York Heart Association Class 1 status, meaning they were no longer expressing -- experiencing heart failure symptoms, that interfered in daily activities. For Cohort 2, our focus is going to be on safety and biopsy results. This will be the first data for the 6E13 vg per kilogram dose. We look forward to reporting baseline and post-dose biopsy data at this dose. Success here is continued good tolerability, and we'd hope to see dose-dependent increases in DNA, RNA and protein levels. We'll be looking at clinical endpoints, but for these patients, it may be too soon to see post dose changes. The goals of TN-401 treatment in our RIDGE-1 clinical trial are similar to what we've been sharing with TN-201. PKP2-associated ARVC is a devastating disease characterized by life-threatening arrhythmias in adolescents and young adults. PKP2 mutations are responsible for approximately 40% of ARVC cases. Current estimates are that the disease affects about 70,000 patients in the U.S. and many more globally. We believe this disease is underdiagnosed. These numbers may be higher. Nearly 1/4 of those affected present with their first a symptom being sudden cardiac death. Other early symptoms of disease include palpitations, lightheadedness and fainting. Patients often experience constant premature ventricular beats and other ventricular arrhythmias physical exertion and exercise such as sports aggravate arrhythmias and accelerate disease progression, so patients are placed on tight exercise restrictions. Due to the risk of sudden cardiac death, most patients have implanted cardioverter-defibrillator devices, these devices restore normal rhythms during life-threatening events but leave significant emotional burden for patients sometimes described as posttraumatic stress disorder. PKP2 encodes the plakophilin-2 protein. This is an essential protein within the desmosomes complex. The desmosomes is like a fastener between heart cells, helping to hold heart muscle cells together. They also support proteins responsible for the electrical signaling in the heart coordinating each heartbeat. A deficiency in the PKP2 protein leads to the collapse of the desmosome cell adhesion structure. Heart muscle cells are damaged, electrical signaling is disturbed and heart muscle is gradually replaced by nonfunctional fibrofatty tissue. Over time, the ventricle can distend and weaken leading to heart failure symptoms and then even greater risk of life-threatening arrhythmias. Much like we discussed for TN-201, TN-401 delivers a full-length copy of the PKP2 gene to heart muscle cells where it produces the Plakophilin-2 protein, with the goal of restoring the structural integrity of the desmosomes and improving the heart function. Improving electrical stability, reducing arrhythmias and slowing or even halting the progression of the fibro-fatty replacement that leads to heart failure. In November 2024, we dosed the first patient in our RIDGE-1 Phase Ib clinical trial of TN-401 for the treatment of PKP2 associated ARVC. The primary objective of the study is to establish the safety profile and evaluate the tolerability and performance of 2 different doses. We've completed enrollment of the first cohort at the 3E13 vector genome per kilogram dose and dosing is underway in Cohort 2 at 6E13. In addition to safety, we are taking biopsies at baseline 8 weeks and 1 year post dose to assess transduction and expression of TN-401, we're monitoring ICD activity and gathering arrhythmia data. We are imaging the heart for structural and functional changes. We are assessing changes in plasma biomarkers, and we're surveying for patient-reported outcomes and quality of life measures. The treatment goal for TN-401 gene therapy is to reduce arrhythmic events and halt progression of heart failure by restoring the structural integrity of the desmosome. Our initial data from the first cohort of 3 patients dosed at 3E13 will focus on safety and tolerability and whether our immunological regimen is working appropriately. We're also going to look closely at heart biopsy data at this stage from which we'll be able to see the levels of TN-401 DNA transduction, mRNA transcription and whether these increase Plakophilin-2 protein expression, success at this stage is seeing all 3 of these measures increasing from baseline. We are also monitoring for any early changes in arrhythmic activity. And with that, I'd like to hand the call over to Dr. Kathy Ivey, who leads our research group, including a translational medicine team responsible for overseeing the analysis of our biopsy data. Kathy?

Thank you, Whit, and good morning to everyone joining us. As we've introduced, cardiac biopsies are collected as part of Tenaya's gene therapy clinical trials. And next, I'll describe what those biopsies are being used for. In the TN-201 and TN-401 clinical trials, biopsy sample analysis and the resulting measurements are our first opportunity to affirm that our gene therapy is reaching the heart that it is first entering the heart cell through the process of transduction that the healthy working gene being delivered is transcribed by the cell's machinery to produce messenger RNA. This mRNA provides the instructions needed by the cells to produce protein. Biopsies are taken at various time points in each of our clinical trials. Using a catheter, a small net of tissue, just 1 to 2 millimeters in diameter is collected from the septum. A number of these tissue samples are collected in the cardiac cath lab, and that's where they first received an initial visual inspection for quality. And then each of these precious tissue samples is preserved and earmarked for specific quantitative analysis of either DNA, RNA or protein. For each of the components of the biopsy that we are trying to measure specific assays have been qualified. Our DNA assay utilizes digital droplet PCR or ddPCR, and is able to distinguish TN-201 or TN-401 DNA from endogenous DNA encoding MYBPC3 or PKP2, and quantify the number of vector copies per cell. This measure provides the first evidence that our gene therapy is reaching the heart and is reported as the VCN or vector copy number. Shortly after infusion, DNA levels in the heart should be at their highest. Over time, a decrease is anticipated as DNA is cleared from non-cardiomyocytes such as fibroblast and a durable steady state is reached that represents long-term retention of the TN-201 DNA inside cardiac myocyte. Using reverse transcript based quantitative PCR or RT-qPCR, we can quantify the expression of TN-201 or TN-401 messenger RNA. The gene therapy-derived mRNA can also be distinguished from the patient's endogenous mRNA, encoding MYBPC3 or PKP2, RNA expression provides an important proxy for potential protein expression via gene therapy since without one, we certainly won't see the other. Over time, we expect to see the levels of gene therapy mRNA increase and then stabilize, which brings us to the focus of today's discussion, protein measurement and cardiomyocytes, for this, we utilize liquid chromatography - mass spectrometry or LCMS, which can quantify the amount of particular proteins in a sample. Similar to RNA, we expect to see an initial rise in protein levels, but anticipate that over time, the increase will level off and endure. In many cases, the aim of gene therapy is to introduce a protein that is entirely absent with the most established example of this being the approved gene therapy Zolgensma for treatment of spinal muscular atrophy or SMA, which occurs only when both copies of the SMN1 gene are dysfunctional. In homozygous conditions like SMA, there's no protein being produced, so the measurement is from 0 to something, in diseases of haploinsufficiency such as MYBPC3-HCM and PKP2-ARVC, patients are producing some of the needed protein from their one healthy copy of the gene is not enough of the protein. So if we're successful, the protein produced from the gene therapy will be identical to the protein produced from the patient's own single healthy copy of the gene. And while the gene therapy RNA can be distinguished from the endogenous RNA, the resulting gene therapy proteins are indistinguishable from their corresponding endogenous protein. With this in mind, we knew at the outset of these programs that we needed a highly sensitive tool for measuring these protein levels from cardiac biopsies, and that's where the work of Dr. Previs and his lab comes in. Dr. Previs. Thank you so much for joining us today. Before we dive into the methodology, can you tell us a little bit about how you came to be an expert in this field?

Sure. Thanks for having me, Kathy. It's great to be here with you today. So I did my PhD in cellular, molecular, biology at the University of Vermont, but I subspecialize in analytical chemistry. During grad school, I developed a mass spectrometry based platform to measure levels of protein phosphorylation, which really sparked my interest in muscle protein structure and function. To expand this expertise, I did postdoc in biophysics where I developed single molecule techniques to study your favorite protein, myosin binding protein C. At the time, much less was known about myosin binding protein C and my work contributed to a fundamental understanding of its function and healthy hearts. When I started my own lab, I wanted to move things in a more translational direction. The big question at the time was whether the most common variant in the MYBPC3 gene resulted in hypertrophic cardiomyopathy by creating toxic protein [ fragments ] called poison peptides. Or is the variance impact in myosin binding protein C translation, resulting in a lack of functional protein or haploinsufficiency. To address this question, I teamed up with Dr. Sharlene Day at the University of Michigan and examined the expression of myosin binding protein C and what we consider it to be a large cohort of human samples that she collected in her clinic. The results from that study were Faraz and Whit referenced earlier and brought us here today.

Thanks, Mike. Can you talk us through the role and function of myosin binding protein C?

Yes, absolutely. So the cardiac version of myosin binding protein C that you're interested in is solely found in specialized heart muscle cells called cardiomyocytes. These cells are what contract for the heart to pump blood. Each heartbeat is produced by sarcomeres within these cells, which are tiny contractual units. You can think about these as microscopic machines capable of performing mechanical work. So we returned here to that slide that -- which showed earlier, we can see that myosin and actin form filaments and these filaments slide past one another to shorten the length of sarcomere to produce each contraction. So the myosin head shown in green here, extend away from the filament backbone kind of like hands, and these are clearly the business end of the molecule. Actin is more like a rope that is pulled by these hands and the game a tug-of-war to contract the sarcomere. So myosin binding protein C though is a long tether like protein that sits in specific part of myosin filament called the C-zone. Its positioning and interactions allow myosin binding protein C to modulating both force and speed of contraction. So while myosin binding protein C doesn't generate force by itself. It's critically involved in modulating how this sarcomere behaves and therefore, how the heart beats.

How does MYBPC interact with other components of the sarcomere to regulate contraction of the heart.

Yes. So that's a great question, and I'd love to give you a straightforward answer. But the truth is, I'm 1 of the world's experts in this area, and they're still very active way with the lots and lots of multiple questions. What we do know is that one end of myosin binding protein C is anchored to the myosin filament as shown in this cartoon. While the other end could actually be untethered to anyone it can interact with myosin head as shown here or it can interact with the actin filament. So there's the 3 options for interaction. The balance between these interactions influences how quickly myosin heads bind to the actin filament. So we can think about this as how quickly those hands can grab the rope and start pulling. And they also modulate how long the hand stay attached to the rope and continue to pull. So myosin binding protein C essentially serves as a master regulator of both the timing and magnitude of each contraction. That being said, there are many models in the current literature that try to explain exactly how this regulation works. But as an expert, I really want to use this platform to say, don't believe in all the hype of these models. These models often ignore data that don't fit and oversimplify things that are clearly very dynamic, and there's a complex set of interactions. The reality is, my field is still working towards consensus, if there's really a need for more detailed studies to tease these mechanisms apart.

And what's happening in the heart with MYBPC3 mutation? And how does that insufficiency of the protein result in HCM?

Yes. That's a really important question that I think everybody wants to answer, and it's one that really continues to drive my lab's research in new directions. What we do know from our previous work is that most variants in MYBPC3 gene lead to these premature stop codons in the mRNA as Faraz and Whit said, which is responsible for making proteins. So I think if we go to the next slide. Here. Yes, that one. Okay. So if we go to this slide, our work from tissue -- this is from our work from tissue with patients with obstructive hypertrophic cardiomyopathy. And this work shows that there's a 40% decrease in the levels of myosin binding protein C compared to donor control. This reduction in our hands appears to be independent of the location of where that premature stop codon happened in the mRNA. And we don't see any evidence of a poison peptide resulting from the translation of the variant mRNA that's associated with the disease. We know for many animal models, human iPSCs and human tissue, that a lack of myosin binding protein C disrupt the function of sarcomere. This contractile defect then at the level of the sarcomere appears to set off this cascade of changes in the heart over time and the heart begins to remodel itself. Unfortunately, this remodeling is what is impacting people's lives and driving them into the clinic. So I think the short simple answer here is insufficient levels of myosin binding protein C clearly leads to poorly regulated contractions, which the heart actually may be trying to fix by remodeling but the remodeling is what causes the symptoms and complications of an HCM.

And what are some approaches available for measuring protein levels in cardiomyocytes. Can you talk about that and then there benefits and limitations?

Yes, sure. I can talk about this for a day, if you want. But so -- so right now, the main approach is for measuring protein levels in any system are generally antibody-based or mass spectrometry based. Each of these methods has its own nuances and pros and cons. Antibody-based methods rely on the generation of an antibody that combines specifically to a protein of interest. You then detect that binding as an indirect readout of how much protein is present. These methods, they can be pretty straightforward and accessible at very low cost, but they depend heavily on the quality and specificity of the antibody that you're producing. If the antibody isn't great or the binding affinity in samples, differs from what you tested this in from something like post-translational modification of your protein of interest, you could run into issues with accuracy. In contrast, mass spectrometry is much, much, much more direct, robust and unbiased approach, but it's really expensive and require specialized expertise. In most cases, proteins for mass spectrometry assays are going to be digested into peptides and those peptides will then go on to be identified and measured by the mass spectrometer as proxies for protein abundance. The technique is really powerful, especially for the use in complex tissues like the heart, where you have many different proteins in the same tissue. With both antibody and mass spectrometry-based approaches, the readouts need to be normalized, which is just the reality of both of these assays. And if I think we take a look at the next slide, right? Oh, yes, there we go. We can see the biggest challenge with normalization. So when working with small samples from patients with heart disease, such as a sample from your trial, the sample of the size of a tiny crystal and sea salts. So these are tiny, tiny biopsies. The size is actually indicated by the circles in this image. So you have a yellow, blue and red circle. As you can see, the number of cardiomyocytes within these circles being those cells that are in pink, can really vary in the biopsies relative to the other material in that image. Therefore, if we use the weight of the biopsy or the total protein content of the biopsy for normalization, this isn't very desirable. So with Western blotting what you typically do is you're going to normalize your readout of your protein of interest against the readout of a secondary antibody, it's a housekeeping protein. So here again, you're at the mercy of your primary antibody to your protein of interest and to the secondary antibody to your housekeeping protein. And there's also going to be the potential that the expression of the housekeeping protein varies in your disease say, which will further skew your data. One big advantage to the MS-based approaches is that in a single run, you can get quantitative information on hundreds of proteins giving you a wide yet detailed view of the entire sample. This allow unbias look at the overall composition of the sample and provides many, many opportunities for normalization. If the assay is designed properly, it only increases the confidence in the quality of the results .

Okay, Mike. So based on your background, you're clearly a fan of mass spec-based approaches. Can you walk us through your approach using data from human heart samples as a sort of case study?

Yes, yes, absolutely. This has really become our approach now because we've collectively -- we collectively worked to refine techniques. So mass spectrometry, many forms of mass spectrometry or the cornerstone of my labs work for many different reasons, just aside from what we're talking about here. And we really have refined the techniques specific to this application, working in collaboration with your team over the past few years. So with your samples, what we do is we started the process by inspecting each biopsy under dissecting microscope to get a sense of its quality and then we loosen the tissue with forceps in the presence of surfactants, so we kind of loosen up those samples. We then relax the proteins chemically and then digest them into peptides using enzyme. Again, we do this because peptide abundances, are much more manageable measure with mass spectrometry. Next, we inject those peptides onto a chromatography column, which separates them based on their chemical properties over a 2-hour period. During this entire period, those peptides are fed directly into the high resolution mass spectrometer, which allows us to identify them and record their abundance with high accuracy and precision. One of the unique aspects of our approach is really how we normalize the data. We normalize the abundance of myosin binding protein C to peptides shared between cardiac myosin isoforms because cardiac myosin is found exclusively in cardiomyocytes. While we validated this approach over the years in our own lab, when we started working with your team, we're asked to systematically reevaluate our methods and fully understand our reproducibility. And I think it would go to the next slide, we have some data. Yes, there we go. So if we take a look at the slide, we see an example of where we performed the assay on several biological replicates. So these are individual samples from both donor control hearts where we had bigger pieces tissue and for the septal myectomy biopsies, again, where we have bigger pieces of tissue that are unlike what we get from your clinical trial. Of course, if we look at these individual data points, there's always going to be some variability in the number -- the numbers. But the measurements for multiple pieces of the same heart are very consistent. So across the x-axis here is individual samples, and we see this consistency in these measurements. If we compare these data from the donor controls, which are shown in green, to that from 3 separate septal myectomy samples, which are shown in red, we can see that the variability between repeat measurements from the same heart is similar. So again, if we make repeat measurements from the same heart, we have very, very little variability. But there is some variability between septal myectomy patients, those data points shown in red. This is similar to -- this variability is really similar to what we originally reported in 2019 in our initial study. Internal studies like these with your team have really increased my confidence in the robustness of our methods and that we are detecting real biological differences rather than just measuring technical noise. So I think if we go to the next slide, yes, here. Here, we have an example of what happens if we try to normalize the data differently. So these are examples of similar data set that I just showed you. But the readout from myosin binding protein C is normalized either myosin -- peptide for myosin or GAPDH independently. So the data from myosin are on the left in this graph and the data for GAPDH normalization are on the right of this graph. So -- and we selected GAPDH here because this is a common house keeping protein that's used for normalization of Western blotting data and other biological systems. And what we see is that when the readout is normalized to GAPDH, there's high variability in measurements of myosin binding protein C within pieces of the same heart and this variability between hearts is even higher. There's more disparity in these data point. Based on this data set in red, we would no longer be able to detect a difference myosin binding protein C levels between the HCM samples and donor controls using the GAPDH normalization strategy. So to me, these data really highlights the importance of rigorously testing these methods like your team has forced me to do, to make sure that these preparations are really consistent and reproducible. And they also really introduced the concept that you need to get the normalization strategy right because even when you're working with a piece of a very expensive piece of equipment like mass spectrometer, that's capable of making these high accuracy and precision measurements. If you're normalizing the data incorrectly, you're still going to get the wrong answers.

So then, Mike, what gives you the confidence or reassurance that when there are seemingly modest increases in protein levels that those changes that we observe are not just noise?

Yes. So it's important here to say. So first and foremost, I'm a scientist. So I'm trained in skepticism. It's run deep inside of me. So I question our data constantly, which my trainees absolutely love and appreciate, they're excited to show me data and then I just start picking it apart, right? But the unfortunate reality with the collaboration with you is that when we're working with human samples, there's no real gold standard, and we've really learned this through this internal testing. It's not like we can just take a gold clock and place it on a scale and get repeat measurements and really, really tell you what our precision and accuracy are because each piece of tissue is unique. And so there's no perfect benchmark for these measurements, right? That being said, the 1 thing that I've really appreciated working with your team is that you guys bring the same level of skepticism to these projects, which has pushed me to be even more rigorous and really created learning opportunities for my own lab. The in-house testing has really allowed us to understand the limits of our detection methods which provides me comfort and reassurance that we're providing top quality data here. This gives me confidence and pride that every measurement I make in Vermont is good as you'll get anywhere in the world. Other big advantages in our interactions really do come by chance and by both chance and design. In some cases, surgeon gets a large enough biopsy that we can analyze biological replicates from the same time point, right? Only a few of your biopsies are coming to me, you're using them for other things. But I really like when we have these analytical replicates from that same time point. But even in the absence of these replicates, we always get the longitudinal biopsy is from those same patients. So when I see consistent numbers from replicate measurements and directional increases and protein levels changing over time, I believe these measurements are reflective of the underlying biology. And well, for me, as a basic scientist, the early data from Cohort 1 have been really great and super exciting. And they make me believe that maybe -- is really -- we are getting protein expression from that AAV -- but really, they've just increased my excitement for Cohort 2 and the higher doses of the AAV.

Tell us then how imaging or other measures can serve to supplement or complement the mass spec analysis?

Yes. I mean so that's another really great question. And one of my early mentors during my PhD is cardiologist, and he said to me, if you need statistics, it's not real. And I know that people don't want to hear that, but the statement is really stuck with me through the years, and I think it's real. And it's something that really we carry into all of our homework in our labs to kind of build confidence right? So when we're working on projects, we really aim for conclusions that aren't just supported by statistical measurements, I lab and the statistics are the lasting that we usually do, but our data are usually supported by complementary assays that are going to provide multiple new points or multiple angles to address that same question and that's where I get compliments. So for instant in my lab, we often pair our mass spectrometry data with fluorescence imaging or functional assays to validate the findings from mass spec. So the really exciting thing for me about our collaboration is that our data that we're collecting proteomic data are continually being integrated with other key measurements like RNA expression levels and functional outcomes. Unfortunately here for me, unbinding this data, and I learned about them in things like this presentation. So that's the limitation I have with our relationship. From what I've seen so far from the change of that were reported that even though that the myosin binding protein C levels have been modest, we've seen a modest increase in this levels they're reproducible and they seem to align really well with the rest of what you're been seeing at the RNA level and at the functional level. So to me, it's super exciting. And I think that this kind of orthogonal validation that's -- it's really this convergence of data when multiple methods were telling the same story that always gives me the most confidence that we really understand the underlying biology because biology is really complicated.

So given what you know about myosin's function in the sarcomere, what do you make up the range of protein levels in the healthy samples? And how does that impact your thinking about myosin binding protein C levels in HCM patients?

Yes, this question keeps coming up. It's come offline a bunch of times, and it's a really complex question, I think, and it's much harder to talk about in the details of this protein assay, but I'll take a shot. So if I think back to my days as postdoc, one of our key findings was that even a modest change in the phosphorylation level, the phosphorylation status of myosin binding protein C could have a significant effect on actin and myosin interaction. So how that sarcomere shortens. So to me, it's not just about how much myosin binding protein C is present. It's about how many of those molecules are functionally active and interacting with those binding partners that I talked about at any given time. So this leads me to think that normal protein levels might not be telling the whole story. While the full replacement of myosin binding protein C volume protein C in your clinical trial would obviously be a home run. I'm not entirely sure that it's not possible that we don't even need these full levels of myosin binding protein C to restore it's function sarcomere, to get back to what appears to be a normal heart. But I think for me and the reality here is that your team, not my work, it's our team that's doing the incredibly important work of doing a study of myosin binding protein C replacement in these humans that have disease that I could have really only dreamed that as postdocs, right? So I really, really appreciate the collaboration here. So -- in the end, I'm very confident that my data, our mass spectrometry data will provide a very, very accurate picture of myosin binding protein C levels during all stages of your trial, but these data really are going to be transformed when they're paired with their clinical outcomes. I think it's this critical combination of data that will ultimately help us to find that question of how much myosin binding protein C is not to restore function.

All right. Mike, last question here. How does the methodology apply to measuring other cardiomyocyte proteins such as PKP2? Can you talk about what's normal for PKP2? And is there also a range?

Yes, yes, yes, absolutely. So just as mentioned multiple times earlier, mass spectrometry is an incredibly powerful tool, right? And if it's really used properly, it can quantify the abundance of hundreds of proteins across the wide dynamic range with high accuracy and precision, super powerful. So while our interactions or my interactions with their team, initially were centered around myosin binding protein C to my combined expertise in this protein and mass spectrometry it was super easy for me to extract PKP2 data from the controlled data sets on that and expert in cardiac muscle protein expression. So this is easy for me to do. However, in true form, your team wanted more validation on my end because I'm not an expert in PKP2. So I think if we go to the next slide, we could see why your team is correct here. So what we did is we ran many additional samples. We look back at our C protein samples, and then we ran many additional samples and tested various normalization strategies that we talked about in many meetings. And although PKP2 is localized to the interpolated disc, as what said rather than sarcomere, we still settled in on normalizing its readout peptide shared between cardiac myosin isoforms. Because, again, myosin is most reliable marker for these cardiomyocytes in the samples, right? And seeing that you're trying to express this protein in cardiomyocytes, this normalization, again, gives us confidence that we're comparing apples-to-apples across samples. So we're normalizing things correctly. So in our new internal studies like the ones I'm showing on the slide, a few really interesting things popped out. First, the variability in PKP2 levels when measured normalizing these things these 2 -- those peptide share myosin isoforms, was similar when measured in multiple pieces for -- from the same samples. So if we look, I believe we have 4 donors here, and you see there's 4 -- multiple measurements from those 4 donors and the variability is very tight. So this was good and again, giving me confidence in the reproducibility of the assay. However, the variability between donor control patients was surprisingly large. You can see donor one levels were much, much lower. And again, to my colleagues dismay, this was statistically significant, which did give me pause here. So we don't really fully understand what's driving this variability, but all signs point to this variability being biological. So another challenge for measuring PKP2 is that we don't have a large cohort of examples from patients with the disease. So we can't get say anything about the variability in the disease cohort. So while I'm fully confident in our ability to measure PKP2 accurately, we're still really learning what normal looks like and how normal is going to differ with disease. And I think by the end of the TN-401 study, we're really going to be able to answer these questions. However, in the short term, just like in the TN-201 trial, we're analyzing longitudinal samples. So each patient essentially acts as their own internal control. So we really don't need these information to determine if things are working. So in summary, I'm again confident that we'll be able to tell if the AAV is increasing the PKP2 levels in this clinical trial. So -- but again, here, your team will also benefit from pairing our proteomic data. So the protein levels might not be telling those story. So our proteomic data with the same orthogonal approaches that you're using in the TN-201 trial. And it's again, as integrated view that will really help us move from just measuring a change in protein expression to really understanding the therapeutic response. And that's really what I'm excited about.

Okay. Mike, I'd like to thank you again for walking us through this work. It's been a truly productive collaboration with you and your lab and we've learned a lot along the way about the strength of mass spec to discern potentially really subtle changes in protein levels, which is critical evidence of gene therapy's potential for success. We're convinced that mass spectrometry is the best method to obtain direct measurements of the proteins of interest and also to quantify their levels in the sample. And it's apparent from today's explanation, the steps that have been taken to navigate the challenges of tissue quality and sample variability, we know as you pointed out that literature has examples of different conclusions being reached based on the selection of proteins for normalization and by selecting myosin heavy chain, which is expressed only in cardiomyocytes but not in other cells of the heart, we can be confident that the measurements are directly applicable to the goals of treatment. The work in normal control heart shared here also bring home the fact that there isn't just one threshold that we need to get to, the biology of the individual matters and it will be the changes in each patient over time, that will be the most important way of looking at and interpreting biopsy data. Thanks so much again for this discussion, Dr. Previs. Operator, I think we're ready for Q&A.

[Operator Instructions] Your first question comes from the line of Cory Jubinville with LifeSci Capital.

This one for Dr. Previs. So you emphasize normally MYBPC3 levels to myosin heavy chain 6 and 7 to help kind of mitigate that biopsy, heterogeneity. Does HCM status impact the stoichiometry of the sarcomere or in the case of PKP2, the desmosome? And if so, how do you validate that myosin heavy chain 6 and 7, content is stable or comparable, whether it be across time points or disease states or sampling sites?

Thanks for the question. As always, we know that you're going to go deep, and I'm so glad that we have Dr. Previs here to talk about the methods that we use for normalization. So Dr. Previs why didn't you get started? And then after that, I'll invite either Kathy or Whit to add any more on top of your comments.

Yes. So I mean that's a great question, right? So for myosin binding protein C, the answer is very simple that myosin is in complex with myosin binding protein C, we know under normal conditions, exactly how many molecules of myosin are in thick filament. We know how many molecules myosin binding protein C are thick filament, this is rock-solid from everything from zebrafish to mice to humans, and we've measured this in all of these things. So yes, in HCM in that case is clearly a -- due to a lack of stoichiometry of myosin binding protein C to myosin. We also get quantitative data on actin, the regulatory proteins on thin filament and all of those things are highly regulated and precise and we're only seeing loss of myosin binding protein C in the case of disease triggered by that, by mutations and MYBPC3 gene. So PKP2 is a little bit more complicated, right, because it's in the interpolated disc. And as we showed from those donor controls, there is definitely differences from one control to the other, and we were very surprised. But if we try to say normalized to some other protein interpolated disc, which is what we've also tried to do, we're concerned that we don't know enough about biology, the normal biology of PKP2. So we -- our most consistent data are when we normalize to myosin heavy chain because it seems like the myosin heavy chain is the most representative of what is in the cardiomyocyte. And again, from my lab, if we image, we see very -- the fluorescence imaging, we see very consistent reproducibility and the expression of myosin within the cardiomyocyte, no, there's not big gaps in expression. So I hope that answers your question, but it's a great question.

A great question. Yes, so far in HCM due to loss of MYBPC3 the goal is to restore the stoichiometry. And so the appropriate number of hands are physiologically regulated. And of course, as Kathy mentioned, the protein we express is indistinguishable from wild-type protein, so we should have all the regulatory capabilities. So that Stoichiometry that we're measuring, the ratio we're measuring is what we're trying to improve. Now yes, for ARVC and PKP2, it's a little more complicated. You could think, well, we want to improve the stoichiometry of the desmosome. But in that case, PKP2 is a critical anchoring protein for the structure. So the whole structure falls apart. So we hope and expect that all the desmosomal proteins will increase an expression, and there'll be more on the cell surface, as we express the PKP2 protein so not appropriate to normalize to those. And I'll say the same is actually true of the gap junctions, the critical [ tingling ] molecule in the heart, those also drop off without enough PKP2 to anchor the desmosome. So we do actually -- we wouldn't want to normalize to any of those, and that's why normalizing the myosin heavy chain makes the most sense..

Got it. That's helpful. And as a follow-up on PKP2 expression. So on the final slide, it notes the variability between -- of PKP2 expression within a patient is low, but across patients is high, but when you outlined some of that historical data across donors on MYBPC3 protein expression, it seemed pretty remarkably consistent at about 60% of normal. I guess what's the biological rationale behind this discrepancy between MYBPC3 and PKP2? Assuming these are relatively healthy donors on PKP2 expression and how should we make sense of this as it relates to evaluating the efficacy of a PKP2 gene therapy? Is it kind of less about hitting the specific threshold relative to normal? Or is it more about an individual specific improvement from baseline?

Thanks again for -- Cory for the question. I think for this one, maybe Whit, I'm going to ask you to go first to just talk about the different sort of thresholds and how we think about thresholds versus absolute protein and how we think about sort of expression relative to the normal range?

Yes. And both of our studies will really help inform how much protein makes a difference in the phenotype, but going in we don't think that there is a threshold effect. We do think that the variability across both healthy people and in these patient populations suggest that everybody has their own sensitivity to protein loss and protein reduction. So in another way of saying the same amount of protein loss of reduction could cause disease in 1 person and not in the next person. So our goal is to increase the protein level in patients towards normal, but not achieve any specific threshold.

Maybe just to add 1 more comment to that. If it's possible to go back to that slide, the last slide. Cory, it's an important question, right? This is the first time we're showing this data. I believe the first time data like this has been presented, showing sort of a range of normal using this very, very precise method. Enjoying PKP2 protein in this way. And it is interesting to see this wide range of variability. And I think a few things are true here, right? One is that we don't think that we need to get to the top end of this range, probably don't even need to get to the middle of this range if there was like a median or an average that right? People at the very low end of this range have perfectly normal hearts. They have no disease, right? And this is from what end of '12, right? If we were to run this end of '24, we'd probably see some clustering towards a mean, and we might also pick up even more variability at the top and bottom end. For the purposes, we hope to get patients closer to the lower end of that range because we know that at that lower end of that range, patients have completely normal heart, but we don't think it's a magical threshold effect even at that level every little bit that we give to these patients can go a very long way and we've seen this from other genetic cardiomyopathies, including from some of our peers, but a little bit can go a long way. So the general goal is to get in the direction of the lower end of this range. But the absolute goal is to give each patient more than what they have because it appears that the heart is very sensitive to the loss of this protein for each patient in their own way. So if we give them a little bit more back, that should go a long way. So would you agree with that statement?

Yes.

Your next question comes from the line of Ritu Baral with TD Cowen.

This is Joshua Fleishman on the line for Ritu. Curious, from prior data, are there any preliminary covariates that you guys have identified, which can help predict the quantity of transgene expression needed to provide a meaningful functional benefit?

Joshua, it's a good question. And are you referring to the TN-201 MYBPC3 program?

Yes, sorry.

Okay. So are there any covariants, other proteins that are changing that might be predictive? You're the experts, Dr. Previs. I'm going to give you a first shot at answering that question, whether we see other proteins changing at the same time with MYBPC3. And then Whitt, I'll kind of ask you to follow up with that about how we think about predictors of success.

Yes. So that's a complicated question, again. And we published papers on this. The -- at the protein level, there are many proteins that are also changing. Myosin-binding protein C is lost, but there's metabolic proteins that are changing because there's complex remodeling that is happening, right? In previous studies, we have tried to team up with geneticists to take our protein data and correlate it with their genetic data to answer that exact question, right? So is there -- is any protein that is decreased, also in combination with myosin-binding protein C? Does that have another mutation in that other protein? And my understanding is that there is no other protein that has been shown to be -- or other genes to be co-affected to trigger the disease. But definitely 5 years ago, I went down the same path, and maybe we just don't have the technology yet for the informatic ability to kind of go through the genetic data. Whitt might be much better answering this than myself.

Yes, Just additional thoughts. So we don't have good clinical predictors of the protein level. We have looked at the heterozygous patients and compared the protein level to the age of onset and other things, didn't find a clear pattern, but we are very much underpowered for that analysis. So more to be done as we get more data. As Mike was saying, the best predictor is the genetics and particularly the type of mutation, number of mutations occurring in the MYBPC3 gene. So if you have two truncating mutations, you probably have zero protein. We look forward to more data coming out about that, but that's the presumption of the field. There are leaky truncation mutation. So if you have a little bit of leak, you may have a little bit of protein. And -- but those with zero protein have very severe disease. The infants I talked about need heart transplant in the first year of life. If you have one truncation mutation and other is normal, that's the patients we've been talking about here with roughly 60%, but some variability there. And then, of course, if you have 2 wild-type copies, you averaged at 100%, but there is a range there as well. So that is the biggest predictor of protein level as the number of truncating or loss of function, mutations and their leakiness. My understanding is that we have not found polymorphisms that regulate the level of C protein, so quantitative trait low side like that. So we can't build a polygenic model of C protein expression at this point.

Okay. And then I just have one follow-up question, please. Just to confirm, using the LCMS approach over other traditional approaches, it appears that the major benefit is just the higher sensitivity and higher specificity of the assay. But it looks like both approaches are still limited by only measuring total protein, not protein expressed solely from 201 or 401, correct.

Yes, correct. So...

Go ahead, Whitt.

So because the protein products of our gene therapies are full-length, wild-type sequences, they are identical to the endogenous. And so these techniques cannot distinguish where that protein came from, transgene or the endogenous gene. But there are many more advantages to mass spec that Michael listed. We're measuring many peptides across the length of the protein, and we can much better control than when we're measuring a single epitope with a single antibody.

Yes, the only I'd add to that, Joshua, is on Slide 27 that was presented. It really is so telling that the -- how methods matter. We really, really have swept the small stuff here with Dr. Previs over now many years. And really, this is why we decided to select mass spec. And even within mass specs, selecting what is the -- what are we going to do myosin -- normalized to myosin, not only myosin, but specific peptides. We actually had a whole family of peptides to select from within myosin and looking for stability, looking for things that are not going to vary as much to make sure that we're picking the right anchor against which we were measuring the changes in the Myosin-III protein. And it was a similar method that went into PKP2. If you compare the methods, you're normalizing the GAPDH versus myosin, we just wouldn't even come to the right conclusions about who is normal and who has disease. That is the central insight for us here and why we decided to go the direction of mass spec and within that, be nearly obsessive and beyond the scope of today's presentation about which specific peptides we're measuring for each biopsy samples within the myosin. So a lot of detail here beyond the scope of today's presentation, but gives us a lot of confidence about the mass spec method and the normalization to myosin from both programs. Operator, let's take the next question.

Your next question comes from the line of Yasmeen Rahimi with Piper Sandler.

This is Dominic on for Yasmeen Rahimi. Thank you for the great presentation and the helpful insights. So we just had a quick question. Considering the MyPEAK data will be helpful to inform dosing selection, how informative do you expect the additional Cohort 1 and new Cohort 2 data to be? Specifically, how do you plan to use the totality of data across the protein improvements and the safety measures to inform continued development?

Dominic, thanks for the question. Whitt, I'll turn it over to you, how do we think about the incremental data that we'll be getting and presenting in Q4 of this year relative to what we presented to date and how that informs our future direction?

Yes. The best way to choose a dose is to compare doses, and we really look forward to sharing the first data from the higher dose 6E13 vector genomes per kilogram. And of course, that comparison will be the starting work to help select the dose for future pivotal study. Safety, of course, is paramount. We want to choose the dose that is definitely safe and very well tolerated. We are encouraged that our DSMB has reviewed the data and endorsed proceeding, indicating that both doses are well tolerated. The -- we will be sharing biopsy results from the high-dose cohort. And the biopsies for all the reasons we've said today provide very quantitative and early in the post-treatment time course data on the expression of protein, which we predict will lead to efficacy. So the dose protein expression relationship will be very informative for dose selection. But we also want to confirm that with clinical endpoints, which take longer to mature, we are certainly following those the end of the year readout might be early to really compare the clinical efficacy across the dose cohorts. So later next year might be the time for that of that analysis. But all these factors will go into the ultimate dose selection to take forward to subsequent trials.

Yes, Dominic, the only thing I'll add to that is, it's a good question, right? We're at an exciting moment in the program. We'll be presenting early additional data from the dose Cohort 1. We're already pleased with what we're seeing from dose Cohort 1, right? We were quite pleased with what we're able to share at the ACC earlier this year, and we summarized that at the top of this call. So we're looking forward to sharing even more from longer follow-up from these patients. What we've seen in some of our peers and what early evidence seems to bear out here is that the longer we wait, the more we see both the durability of the effect and even some additional effects over time. So we're excited with Cohort 1, we're excited for dose Cohort 2 with the biopsy. From here, there are many directions we can go, right, from a program perspective. And we've said this for a while, we can continue exploring adults. Right now, the patients dosed to date have been mostly nonobstructive patients. There's also obstructive adults, but then there's also the children. And we're quite excited to be presenting data, the first ever data presentation from our MyClimb natural history study of more than 200 children, both retrospective and prospective data that will be presented at the upcoming European Society of Cardiology, what will be there with the other members of the team to share that data and to really bring a spotlight on the very severe children, both the homozygous infants as well as compound heterozygotes and other severe heterozygotes. So that's another exciting direction that this program can go. The data we're generating now helps us decide what's the right dose, irrespective of the population we're going after, whether nonobstructive or obstructive adults or children, what's the right dose that strikes the right balance between safety, protein expression and signals of efficacy. So in Q4, we look forward to presenting the data, but we won't be sharing the direction that we're going with the program at that time, that would be more of a 2026 discussion. Hopefully, that answers your question, Dominic. And operator, ready for the next question.

Your next question comes from the line of Mike Ulz with Morgan Stanley.

Thanks for all the details on your methods here. So I guess, maybe just with your more precise methods in terms of protein expression, I guess what's the lowest level threshold over time in the same patient where you're confident you're seeing a real change in the expression versus just noise? It sounds like maybe very low percentage difference would be meaningful with these methods.

Yes. Great question, Mike. Thanks for asking it. And for here, I'll ask Michelle to wind back one slide. And Dr. Previs, if you don't mind speaking about how do you think about the noise, right? There is a certain level of difference in the samples on Slide 29. So Dr. Previs, maybe you can speak about that for each protein in each program.

Yes, I can definitely do that. So what slide are we going to go back to?

This one. Just the variability.

Yes. Yes. Yes, yes. So I don't want to give you an actual number for what the level is because I don't want to mislead you. I've been shocked, we've done many mixtures assays where we're mixing known levels of proteins together, and our methods are even more robust than I would have ever thought or I thought at the beginning of the process here. But you could see for donor 1 for PKP2, which, again, PKP2, I want just to keep in context here, is expressed at a much lower level in the cardiomyocyte than something like Myosin-binding Protein C or myosin because it's only in the interpolated disk rather than distributed everywhere. So our absolute signal is always a little bit lower for PKP2. So there's probably a little bit more variability in that measurement. But the variability between repeat measurements, as shown here on the left, so these are individual pieces of heart from the same patient; is quite low. It seems like a shock on blast that we could see. And we can tell the difference between donor 1, donor 2, donor 3 and donor 4 really when the data are lined up like this. So for me to really address your question, I think we showed for TN-201 maybe a 3% to 5% change for multiple samples. To me, that seemed real because those data came from repeat measurements of biopsies, as shown here, when they were available, and we saw that increase over time. I think that in Cohort 2, I think we have both the access to the tissue is a little bit better than when we started, and we're repurposing them for mass spec a little bit better. And I do think that the higher dose is going to be telling. So I don't know if that quite addresses your question.

Yes. No, that was very helpful.

Your next question comes from the line of Sami Corwin with William Blair.

I have one for Dr. Previs and a couple for the company. Dr. Previs, just curious, how low cardiomyocyte levels in a biopsy could affect the mass spectroscopy sensitivity and accuracy? And then for the company, can you validate or clarify that mass spec was used for protein quantification in the initial TN-201 data? And then obviously, this presentation highlighted how powerful of a tool mass spec is. But given some of your competitors who are also developing PKP2-based gene therapies, have presented protein data using other methodologies, do you plan on supplementing the use of mass spec with other methods for measuring protein expression?

So those are three questions there. And we'll try to do them justice one by one. I will -- maybe Dr. Previs, you can confirm the answer to one of the questions, which is the methods. Have the methods changed at all between 201 and 401? Or have we generally used the same methods?

Yes. So the method is the sample prep, the LCMS, how the sample goes through the LCMS is all identical. Obviously, we're selecting different peptides for PKP2 versus myosin. In the background, we're also looking at peptides from collagen, from albumin to really assess the quality of that biopsy and get a sense of how many party of cardiomyocytes are there. Faraz, would you like to answer the question about how low we can go?

Yes. Well, so the other one is -- the question was now we have seen that other companies -- this is not in the MYBPC3 space, Dr. Previs, but this has been in the PKP2 side, there are two other companies, and they have different methods. And we're not commenting on the quality of their data or anything like that. But methodologically, there are some differences using Western blot versus mass spec, normalizing to GAPDH versus normalizing to myosin, two very obvious differences between our methods were there. So the question was, would we be changing our methods, now that we know that others have presented this in a different way? And I think the answer is no on our side. No, we're not changing our methods. But do you want to add to that in any way about reaffirming the metrics that we've used compared to others.

Yes. I mean without -- as an expert here, without seeing the actual -- how the sample was handled to the output that came about, it's really hard for me to complement or comment on any other method. Kind of I make this look easy often, even with your team, right? But it's from 20-plus years of experience of handling cardiac muscle biopsy, right? And there's a lot that goes into the front end of the sample prep to make sure that these things are right. And with your team, we've looked at -- we've taken data that was run in 2017 that was published in 2019, that I didn't even touch as a technician in my lab, and we've compared those data to the sample sitting in the freezer for God knows how many years and me doing the preparation, me running the sample and the data are virtually identical. We can't tell the difference between biological variance versus the technical variance on the data, which is fantastic. So I think it would be premature to comment on anybody other -- anybody else's math because Western blotting is not Western blotting, the same thing. I try to make a point that mass spectrometry is not mass spectrometry. You could buy $700,000 piece of equipment, you could try to do this yourself, but you need the expertise to do that in that area. And it's -- it kills me when faculty members tell the student, "Oh, just do mass spec on this." It's its own specialty in that.

So the last part of the question -- go ahead.

As a grad student, I did a lot of Western blotting back in the day, and I'm really impressed with mass spec and its level of -- not the sensitivity, but accuracy overall for so many different reasons. But they're are apples and orange, very difficult to compare and different labs, different samples, that would be a very hard crosswalk to do. But please do, Mike, go ahead, and I'll answer the other question about how we...

Well, and also, I think that -- if you don't mind going to Slide 24 because there was also this question about the sample. And if depending on the content of that sample, how does the method handle that? And so I think this was sort of glance at the very beginning 3 hypothetical pinches with varying levels of cardiomyocytes in there. And so Dr. Previs, do you mind just sharing how the method adjusts to this in mass spec better than other methods at quantifying the protein in the backdrop of fibrofatty replacement or fibrosis?

Yes. I mean so you -- these are very critical. They're very special samples. There's a lot of anxiety whenever one of these samples arrive, right? And I think from your end, we had a sample that produced no data, right? And when we look at that sample, there is no -- there's very little myosin heavy chain in that sample, which is the most abundant protein, right? And that raised an alarm in my mind that the quality of the sample is really critical. The surgeon is blind when they go in and take these biopsies, right? They're sampling wall, they can get some fatty material, they can get fibrotic material, right, and they're tiny. So that's exactly why we use myosin because, again, it's a marker of how many cardiomyocytes in there. And we are now in the process of just revising exactly, so we know exactly what level of myosin we need to be able to see on the mass spectrometer to be able to know that we're detect -- we're able to detect PKP2 or Myosin-binding Protein C. So to your question, it's very low. Like it's very sensitive, but we do need cardiomyocytes. If the biopsy is a vessel or a collagen or fatty material, we are not going to get data from that biopsy. That's the reality.

And I believe what you're referring to is in our TN-201 program, the third patient [ C1 P3 ] had a baseline biopsy, the initial post-dose biopsy, we weren't able to really quantify protein from that because there were simply just not enough cardiomyocytes in there. But of course, we have a second shot at a biopsy of that patient, which is the second biopsy, which will be part of the data readout in Q4 of this year, where we do have meaningful data for that patient as well as subsequent patients in the high-dose cohort. So Whitt, was there something else that you were trying to add before I jumped to the slide?

No, this was exactly it.

I think we may have time for one last question before we get to the end of the allotted time. Operator?

Your final question comes from the line of Joe Pantginis with H.C. Wainwright.

So my first question out of two is a bit layered, and I'll qualify it by saying your views as of today because it could certainly change tomorrow. As the program moves forward, I wanted to get your sense of the role of biopsy, say, the level and frequency in later-stage studies versus just assessing clinical parameters. And what your views might be regarding the commercial and regulatory scales for mass spec?

That's an interesting question, Joe. Thanks for asking that. Let me first turn the second part of your question. I'll address the first part of your question with Whitt's help. But for the second part, the scalability of mass spec, imagine a future world, Dr. Previs, where we're analyzing a lot more samples in parallel. Is this -- are these methods scalable?

Yes. The short answer is, absolutely. The method is scalable, it's transferable to different mass spectrometer, different [ tolerogenic ] setup where the data could be produced even faster, right? It's very scalable.

Go ahead, Whitt.

Yes. The vision is not to need a mass spec or tissue samples for commercial use. These protein levels are very, very valuable for showing the efficacy of the gene therapies, and we're very encouraged by increasing enthusiasm at the FDA for mechanism-based approvals for these types of therapies. So replacing protein in disease is caused by protein, so very valuable for dose selection, for demonstrating activity and for approvals of gene therapies at this point. But at commercial launch, we'll use the genetics. As I said that the genotype really predicts whether there is insufficient protein causing disease. And so it will be the genotype pathogenic, likely pathogenic mutations in MYBPC3 or PKP2 that would determine eligibility for clinical use without requiring heart biopsy.

I appreciate that. And my second question is...

Joe, I do want to be fair. There's I think somebody else is waiting in queue. And the other thing I just want to be clear, in a commercial setting, we think that protein biopsies are going to be really important for the potential for accelerated approval based on protein as a surrogate marker that's been done with many gene therapy programs, including some of our peers, as well as, obviously, there have been some high-profile cases in the Duchenne muscular dystrophy space. So protein is really important when you're seeking that initial regulatory approval based on a surrogate marker. However, in a commercial setting, right, we don't see that biopsies continue to be necessary. So it's really important at this stage when we're trying to understand the relationship between protein and benefit, but not something that we would be doing at commercial scale later. So I just wanted to sort of put that to rest because that was -- I thought that might be linked to your question about scalability of the method. We would not be suggesting to do baseline biopsies and post-dose biopsies on thousands and thousands of patients in a commercial setting. We do -- we are over, but there was one more question from one of our analysts. And operator, if you don't mind opening it one last time.

We have a question from the line of Geulah Livshits with Chardan.

Maybe just another question on normalization. Again, you showed the nice intrapatient and a little consistency across the different samples. So when you think about the trial interpretation, I'm wondering if the longitudinal data showing consistency over time for MYBPC3 and perhaps less so than for PKP2? And also for the trial, could there be an impact of the immune regimen on the protein expression dynamics, for example, myosin heavy chain or the MYBPC3 and the PKP2 that could affect the interpretation of the data?

Good question. Maybe, Whitt, can I first ask you just immune regimen and whether you had any reason to believe from our work or the work of others that the immune regimen might in any way complicate these measurements? And I'll ask you, Dr. Previs, to also add to that.

Yes. So great question. Thank you. Obviously, in the gene therapy field, there has been a question about the immune system removing cells that are expressing the therapeutic protein. We don't believe that will be a concern. We have not seen any evidence of reaction against the protein probably because these are heterozygous patients that are [indiscernible] positive, meaning they're already tolerant to all the peptides in the MYBPC3 and PKP2 proteins. Also these are intracellular proteins. So that makes it that issue less likely. Now to your question of inflammatory cells potentially being there, of course, we do heart biopsies, we do not really see that. So that's not going to be a problem. Theoretically, that would be compensated by the methodology that Mike has developed to normalize to that myosin heavy chain. So I guess no across the board.

Dr. Previs?

Yes, I don't think I can add anything to that. But again, that's the reason why we're normalizing to myosin because it is in the cardiomyocyte. And I think it does take care of those questions.

To address the other part of your question, Geulah, it's a good one, and it's come up before, so it may be worth putting to rest. Michelle, do you mind going to Slide 22? Because this is your original work, Dr. Previs. And one thing that struck us when we first saw this, and while we approached you even about using this method and our -- prospectively in our studies is how remarkably consistent it is, but isn't one insight from here. These are different patients, all had MYBPC3 mutations, all had myectomies. They were at different stages of their disease, right? And so some might have been 30 years old, somebody might be 40 years old, somebody might have been 50 years old. So this idea that these data may fluctuate a bit over time longitudinally, it feels like the tightness of this measurement suggests that if there is any such thing, it would be within a very tight range. Is that a fair assessment of this data set?

Yes. I mean absolutely. I am always shocked in healthy animals, healthy individuals. I am shocked by how consistent that number is between mice, again, rats, humans. And the one thing that did come out of my work post publishing this paper with you is I do think that the data points that are high here above this blue line were consistently high from biopsies. I think that's just what's in the person, what's in that person. Again, we don't have multiple biopsies from that person over time. So we don't know how much it will fluctuate. I think the coolest thing about the collaboration with you is we get those multiple labs to describe, so we'll see if the level increases, if it continues to be increased, how it varies between biopsies and that sort of thing. So although you're doing clinical trial, me as a basic scientist, I am super excited about basic science, the basic science aspect of this also, which is kind of what some of these questions are touched.

Yes. And the only last thing I'll add to that, it is important, it's a relevant question. I would say that Obviously, we'll have, Geulah, two shots on goal here for each patient for which we have the baseline biopsy and one post-dose biopsy as well as sort of more immediate, for example, in the PKP2 program, 8-week biopsy. And then we'll have another d at the 52-week. So we'll be able to see if there are any changes with time, but there's always going to be a certain amount of pinch-to-pinch variability. That has nothing to do with the time course of the disease, the changes in the disease. So we just have to have the humility there that there will be changes from pinch to pinch over time. But we'll have a chance to see that. I'll also remind you that it's not just a protein and this normalization of myosin, but there's other things like RNA that we're also looking with the totality of the data that we're looking at. And so there may be small differences in RNA from time to time or protein from time to time. We'll be able to see some of that. But overall, we don't think that there is meaningful changes over time, and that's what this data set seems to tell us. Final thing I would say is in our peers who published in the New England Journal last year, had some excellent work done in Danon disease; they were able to show longitudinally over many years, they showed RNA and protein, and there were changes over time. Some of that may have been biological variability, some of that may have been methods. And so the importance of getting your methods right upfront cannot be overstated, and that has been the whole point of today's presentation. So I think we're over time, we're so glad that there was some good Q&A here, an opportunity for back and forth with our speakers. I want to thank you, Dr. Previs, for lending your time and your expertise to us today and being such a wonderful -- for committing your career to these methods. And and that enables us and our mission to move forward towards patients and really enable us to make sense of the data that we're getting from these studies. So thank you for your work. Thank you for your partnership over many years. And thanks to Whitt and Kathy for adding their voice to this, and thanks to all of you for attending today. Our analysts had a chance to ask questions, and we are able to respond not everybody on the webcast. If you submitted a question, had a chance to do so, but you have an opportunity to do so later. You know how to reach us, Michelle Corral, our VP of Investor Relations, or e-mail is in the public domain. Please, if you have burning questions that came from today's presentation, please don't hesitate to reach out to her, and then we'll find a way to answer that. This is just setting some stage for data releases in Q4, more to come. It's an exciting time for Tenaya. We thank you for your attention, and looking forward to the second half of the year and these data releases and the discussion that will follow. With that, operator, I think we can close.

Thank you. You may disconnect.