Good morning, good afternoon, and welcome. Thank you for being here, those of you in the room with us and for those of you joining online. The one reason we brought you together today is to talk about SENS-601, our gene therapy program for GJB2-related hearing loss, which is entering the clinic. This is an important milestone for us. It is also the first time we are disclosing the details of the program in one place, the disease, the biology, the preclinical program, the regulatory progress, and the trial we are about to run. Very quickly, the logistics. We will run about 90 minutes in total, including roughly 30 minutes of questions at the end. Everyone is in listen-only mode during the presentations. When we reach the Q&A, please dial in on the teleconference number. The line will then open, and we will take the questions.
The slides and the replay will be available on our website, and if you would like to follow afterwards, our investor relations contact is on the panel slide at the end. You have seen this slide already. You know the drill. I am not going to read it, but I will point out four things on it. First, some of what we say today is forward-looking, our development plan, our regulatory submission, the timing of data. Those are intention, not assurances. The risk factors are set out actually in our 2025 Full Year Report published on the 18th of March. Second, SENS-601 is an investigational product. It has not been approved by any regulatory authorities, and its safety and efficacy have not been yet established. Third, everything you will see today is preclinical and natural history data. Fourth, nothing in this presentation is an offer for securities.
Here is how we will spend the time. Our clinical application has been approved in France by ANSM, and we will share more details about that. Canada is under review, and filing in the U.S. and Australia are expected to follow, with the first patient dosing by early 2027. We will take it in order the science took, the patient first, then we will talk about the biology and the evidences, and then the trial. Finally, we will close with a panel and your questions with all of the speakers. Today's speakers, two from outside Sensorion, three from the company. First, Dr. Sharon Cushing joins us from SickKids in Toronto, where she is a Pediatric ENT Surgeon and directs the Cochlear Implant Program. She knows the current standard of care as well as anyone because she delivers it on a daily basis.
She will open with the patient journey, and then, as she is also the Coordinating Investigator for the HearConnex study, she will come back later on the clinical trial. Professor Christine Petit is a Professor Emeritus at Collège de France, Professor at Institut Pasteur, and the Founder of the Hearing Institute, Institut de l'Audition in Paris. She is also the Kavli Prize laureate. Christine Petit is one of the pioneers of the genetics of hearing. She has been instrumental in the exclusive research collaboration Sensorion has with Institut Pasteur, and she will tell us more about the science behind everything we will present today. Laurent Désiré heads our preclinical work. He took SENS-601 from target identification to the clinical candidate. There is no one better placed to show you the construct, what it restores, and what we have seen in animals to date.
Then Valérie Salentey, Head of Regulatory Affairs and Quality. She built all our regulatory strategy across four jurisdictions, and with her team, run effectively all the interaction with the agency behind it. Myself, I joined as a CEO in June after 30 years across biotech and pharma. I spent the first month watching what the team has been building over the years, and this is the kind of day you take the job for, I would say. I will be your host for today, and I will moderate the questions at the end. One slide to talk about where the program comes from, and because years of work has been accumulated behind it. Sensorion started to develop inner ear expertise from its inception in 2009, working on biology, cochlear delivery, translational models, and audiology endpoint, built across small molecule and gene therapy.
Obviously, the collaboration we have with Institut Pasteur has been one of the building blocks of our gene therapy programs. Our technology, the AAV vector design to target the cochlear cells, the microRNA-based mechanisms to keep the expression outside of the hair cells, the injection system built specifically for intracochlear administration, and we have also an in-house process development and analytics. The clinical education is not a new thing for Sensorion. We have already performed this surgical procedure in children in the Audiogene trial, and our OTOCONEX natural history study is feeding patient identification and recruitment today. Which brings us to where it leads. HearConnex study is approved in France, as I said before. Hopefully, we will make quick progress in other countries very soon, and we will deliver data throughout 2027. The world of and scope for today.
Just to make sure we are very precise about it, today it's about congenital GJB2-related hearing loss, also called DFNB1A, autosomal recessive non-syndromic hearing loss caused by biallelic variant in GJB2. It affects around 2,000 newborns a year across the U.S., the United States, EU4, and the U.K. It's roughly 32 patients out of 100,000 births. Our ambition might not stop there. Pediatric and adult [progressive] forms are of interest. These programs are at preclinical stage today, and we are working on patient identification. Step by step, what we are aiming to do is to build an hearing loss franchise expanding across indications. Now I'm going to hand it over to Dr. Cushing.
Hi, everyone. It's a pleasure to be here today. I'm really excited to be part of this team. What you're going to hear as the slides come up is that this story is really a foundational one. What you're going to see are different building blocks, and my goal is to present the first part of the story upon which this is built. That's really the patient journey. It is at the foundation of everything that we do, both in the lab and in the clinics. I'm so pleased to start there. Next slide. As we heard, GJB2 -related hearing loss is the most common recessive genetic cause of hearing loss, and we've got the estimated incidence up here. I'd like to take it one step back.
When we think about congenital deficits that babies can be born with, if we take all congenital deafness, what we hear is that it is the most common deficit that a child can be born with. Of that, GJB2 is indeed the most common. If we estimate that in the U.S., the EU and the U.K., that is about 2,000 babies every year that are going to be affected by GJB2- related hearing loss, with an annual incidence of about 32 per 100,000. This is a large component of our hearing loss population. Next slide, please. When we think about the subcategories of children who are affected by this, severe to profound deafness at birth is really the current target. That is children who have hearing levels that are worse than 71 dB.
These children are really separated from their world by this deficit. Let me walk you through this slide. In terms of the diagnosis, often the initial physiologic impairment is diagnosed on newborn hearing screen, and we're going to talk a little bit more about that. Then we are able to underline the etiology, typically through GJB2 phenotyping. That's done in different parts of the world in different ways. I'm going to shortly present to you how we do that in Ontario, Canada. Let's skip to the consequences, because that is really what we're trying to stave off with the therapies that we are creating for this deafness. There is a huge impact on development, particularly as we consider language acquisition for these children.
When we think about our interventions, we really need to think that as clinicians, as scientists, as investors, we are helping these children build their brain. It's a really huge responsibility. What we do and what we don't do in these roles can have a huge impact on how these children are building their brain. What we're doing now is creating additional treatments in the care pathway, which is already complex. It involves screening, it involves diagnostic testing, it involves genotyping and phenotyping, and ultimately involves therapy. Our toolbox is expanding, as you're going to hear today, which is a very exciting component of it. If we think about that concept of building a brain, we know that children build their brains on a day-to-day basis.
This is not a linear. It is a linear component, and while we know that ideally we want to intervene by three years of age, even by the time we get to three, there are going to be delays. We're really going to place emphasis on keeping that duration of hearing deprivation as small as possible, and I think we're really well-poised to do that with this study. Next slide. How do we identify children today? We alluded to this, and this is going to be done differently in different jurisdictions. Often, typically, these children are identified early through newborn screening, and that is done differently in different parts of the world. Then layered on top of that, we then try and identify the etiology of the hearing loss, the most common genetic etiology of which is going to be GJB2.
Now, if we go to the next slide, I am going to share with you a little bit of what we have been doing in Ontario. Essentially since 2019, we have co-coupled our physiologic hearing screening. That is the newborn hearing screening test, which in Ontario is distortion product otoacoustic emissions. We have added onto that a second track looking for etiology. This includes CMV, congenital CMV, but it also includes a genetic track. I will pause here just to make sure that I underline that it is every baby born in Ontario who is getting both the physiologic screen and the risk factor screening. Even if they pass their newborn hearing screening, they are still getting tested for congenital CMV and genetics in our province. Let me share some of the details of that program with you on the next slide.
We will focus on the genetics, and a lot of that emphasis is actually on GJB2. This is the program that we started in 2019, and 10 of the variants that we screen for are GJB2-related. It is not all-encompassing. This began as a pilot program and has been wildly successful, and it is the backbone upon which we are going to continue to expand universal genetic screening. Let us look at some of the results on the next slide. Here, let me walk you through this. This is a period of three years. Since the program began in July of 2019 to March of 2022. What you can see, I have put on the scale just so that you can compare it to the geographic area that you are coming from.
The population of Ontario is about 15 million, and there are just under 450,000 babies born every year. The first number I want to point out here is the 98.8%, and this was a consented process. I think sometimes as clinicians, we worry about the acceptability of universal genetic screening to our family. In this context today, we might even worry about how acceptable is a genetic therapy to a family. What I can tell you is our consent rate was nearly 99%. Families had no trouble with this. They wanted their baby to be screened universally for genetics. What I can tell you is the flip side is that families themselves are very anxious to have access to genetic therapies for their child. I can say that many of them will be much braver than we are as clinicians and researchers. Absolutely.
Now, if we break down those numbers a little further, what we can see is that just over 93 patients, or so 93 patients were positive for the genetic screen. Okay? Now we are going to flip to that bottom line, because when we look at those 93 children, let us see how they did on their newborn hearing screen. Many of them referred, almost 80, but many of them missed their newborn screening because no screening program is perfect, and several of them passed. In that 93, ultimately 72 of them were identified with permanent sensorineural hearing loss. It was very high yield, and we have a cohort of children that we are also following to see if they go on to develop hearing loss.
And so this is one of the best ways that we can identify children with hearing loss related to GJB2, because we do so within the first few weeks of life, which gives us a long runway in terms of finding the best treatment for them. Next slide. Here are the results of this study, and what I'll point you to is the very top bar graph. This is 92 patients that were identified here. What you can see is that only three-quarters of them actually fall outside the profound range, which is that darkest of blue. Okay? Those are children who traditionally would have benefited from cochlear implant. In the lighter shades of blue, we have moderate and moderately severe, which is a relevant population for who we're going to discuss, and we'll get to that in the clinical component.
But what you can see is that there is a large variability in the phenotype, and that varies depending on the genetic mutations themselves. When we have highly penetrant biallelic mutations, we see the worst outcomes, and then we have much milder hearing losses in different variables. How we are going to look at this variability when it comes to the study and reduce it is by offering this therapy for a very specific component of those bar graphs, which is the child with severe hearing loss. Next slide. What do you need to do to get a hearing intervention? From this, I'm going to borrow from my clinical practice in cochlear implantation, and this is some data out of our program at SickKids.
What I'm showing you here on the x-axis is all of the things a family needs to do in order to get to an intervention, in this case, a cochlear implant. In this instance, these are a group of children where we were able to implant them quite early, under one year of age. But what you can see is there's lots of stepping stones in order to get to where we're going. The first of which is confirming the hearing loss diagnosis, and we know that that actually happens quicker based on our data when it's coupled with a genetic etiology. What you can see is as we follow kids through this pathway, the duration of deafness increases, and it's very important for us to get to this final component at the bottom, which is either implantation and hopefully in the future, gene therapy. Next slide.
Cochlear implants are an incredible discovery. They're incredible technology, and they have changed the lives of so many of our patients, but they're not perfect. There are limitations in terms of what cochlear implants can offer, and I'm going to just spend the next part of this talk discussing what some of those limitations are and potentially how we may be able to overcome some of those with gene therapy. Next slide. Again, this is a graphic, a little bit of what I showed you based on our data in terms of what happens when a child is born. They're identified on newborn screening. They get put into the right program. The diagnosis is confirmed. The etiology ideally is determined. We fit them with hearing aids. We evaluate their speech, and ultimately they get on to cochlear implant surgery.
And you can imagine that there's lots of things that can get in the way of this pathway. The obvious limitation of cochlear implantation, and it's one of the first things that families bring up, is the dependency on a device. Essentially, these children are going to be dependent on technology. They're going to need their external processor always. Things happen to that processor, especially in children. Many of our kids often have one that's broken. Every summer, I get a call about processors that are at the bottom of the lake. When their processors are not on and not working, they no longer have access to sound. It really is a very vulnerable dependency, and it's the first thing that families consider when they're considering cochlear implantation. Many will accept it, most will accept it, because it's our only solution.
I think that is one of the components that I think really has where gene therapy may be able to give us or will be able to give us a significant advantage. Next slide. The other component is that hearing with an implant is not normal hearing. One of the ways that we can think about this in the visual system is to think about low-resolution images versus a well-focused, clear image. Most of us can make out what this image is. We use brain pathways in order to do it, and that is the same way in which children who have a degraded signal from a cochlear implant use it to hear what they need to do to get through their days. But there's a cost to that. The brain does need to create additional pathways.
The brain does need to create additional time and effort in order to get the job done, and that is really well reflected on the next slide. This is some data from our lab where we actually look at objective measures of listening and the effort that individuals put into play. Here we have some examples of reaction time in emotional-related tasks, where we look at the difference in terms of capacity to identify level differences between the ears with children in the cochlear implant configuration in green and normal hearing children in black. In addition, when we look at additional reaction times when it comes to spatial release of masking. Listening with an implant is effortful. The data is there, and it is highly reflective of the effort people put in. But I think it's very important to also listen to our patients.
If we go to the next slide, I'll share with you what some of our patients have said. We have a podcast through our lab at SickKids, and a lot of it is about science. That being said, Season 3 and Season 4 are our patient stories. They very clearly will tell you where their cochlear implants fall short. If we go to the next slide, I have just a couple of excerpts to give you a flavor of what these now- adult individuals have. These are children who received the standard of care. They got bilateral cochlear implants at under one year of age. They got excellent rehabilitation therapy. They are living their best lives. But let's see what Sophia says.
She says, "I consider myself to be a very extroverted person, but if I'm at a restaurant with my family or my friends, after the first hour, I'm absolutely exhausted." She's really speaking to that communication fatigue. Then we have another patient, this is a patient of mine, May, and she says, "The hearing fatigue is always in the background. It's hard to identify. Is it fatigue related to hearing or is it fatigue really related to other activities?" These are the stories you hear again and again in the clinic, and it really tells us that implants are amazing, but we need to push farther for the sake of our patients. Next slide. If we contrast what the journey towards an implant looks like relative to the journey towards genetic therapy, much of the initial part is the same.
I think, again, we talked about that this is a story of building blocks. This is something we know how to do and how to do well as clinicians is how to get a child who was born with deafness to the therapy that they need. What's changing is the therapy, and that's really the exciting component of it. Certainly, in an ideal world, we're getting children to gene therapy even as early as six months of age. As I said, the pathways are in place. This is something as clinicians and as programs that we really know how to do. The benefits are in counter distinction to the challenges of implantation. You don't have that ongoing need for devices. It is always- on hearing, which has an incredible impact on their day-to-day lives.
There's no need for ongoing technological upkeep, and so has the capacity to really allow them to live a life where hearing is not as effortful. Next slide. Now, one of the final concepts I want to leave you with, and I honestly think this is the most important one, and it's worth understanding the difference between bilateral and binaural hearing. I think oftentimes these terms get used interchangeably, and they are at polar opposites. When we think about bilateral hearing, that is two ears. What we do with cochlear implants or hearing aids is we provide devices to both ears. But this is something that happens independently. We know that there are brain stem pathways that connect both of our ears that allow us to really realize the full spectrum of sound in the world that we live in.
This is not something that cochlear implant candidates can do, nor will they ever be able to do with cochlear implants. They are two separate devices, two separate ears. When we think of what we want to achieve and what allows us to listen in a difficult listening environment, to know where sound is coming from, to know the where of sound, that is binaural hearing. We have no hope of achieving that with technology such as cochlear implants or hearing aids. What we can do with gene therapy, however, is restore that physiologic hearing that allows us to use those binaural pathways and achieve what we're not achieving with cochlear implants. To me, this is really the greatest benefit to providing physiologic hearing without any of the downsides of listening effort for our children who are born with hearing loss. Next slide.
So let's pause here with this first building block of who we're trying to help and summarize. We know that GJB2-related hearing loss is common. It's identified at birth, and if we're going to be successful, the window for intervention is likely going to be measured in months. Cochlear implantation is amazing. It restores access to sound, which is no small feat, but it is not physiologic hearing, and gene therapy will hopefully pick up where cochlear implants leave off in specific patient populations. I think the key to the benefits of gene therapy are going to be in that capacity to provide binaural integration, listening in noise, reducing listening effort, so that really there's less unmet needs for these children who receive gene therapy. That is a synopsis of the patient population that we're going to target.
It is my great pleasure to now pass it over to Professor Christine Petit. Her work is foundational at the Institut Pasteur, and you're going to now get to hear where all of this comes from. Professor Petit.
Many thanks. I am delighted to share with you our recent advances in the development of gene therapy for the early onset form of GJB2 gene-related deafness. This program is taking place in our institute, the Paris Hearing Institute, an Institut Pasteur research center associated with a dedicated medical center for hearing located in the Institut Pasteur campus. Like vision, hearing-
The other way. Oh. The other way.
Like vision, hearing alerts us to dangers, but not only during daylight, but also during the night. Moreover, hearing is a cornerstone of a feature unique to our species, language, which may have led Immanuel Kant to say that "Not seeing separates us from seeing, not hearing from our fellow man." In 1990s, much was already known about the sense of vision, smell, and taste, but there was no molecular data for hearing.
The molecular mechanism underlying audition remained a mystery despite solid understanding of the physics and to some extent, physiology of the auditory system. The primary obstacle lie within the cochlea. This auditory sense organ contains too few cells of each cochlear cell type to yield sufficient material for analysis by biochemical methods available at the time. A retina contains more than 100 million of photoreceptor cells, whereas a cochlea contains only about 3,000 inner hair cells, the genuine sensory cell of hearing. This extreme scarcity made it impossible to identify proteins essential for audition through the conventional biochemical and molecular biology approaches. The identification of the causal gene for deafness could be an alternative strategy because genes carrying defects that result in deafness must encode protein essential for hearing. However, genes causing isolated, that is non-syndromic form of deafness, deafness non-associated with other symptoms, were entirely unknown.
Crucially, not even one locus for a deafness gene had ever been mapped onto human chromosome, a prerequisite for the identification of the gene at the time. To cut a long story short, taking into account the expectation that different deafness gene will result in different forms of deafness that no clinical examination allowed to distinguish from one another, we searched for condition in which a single gene causing recessive deafness is probably present in the studied deaf families. Large consanguineous family with profound deaf members living in geographic isolate could satisfy this requirement. Such geographic isolates are generally be founded by a small number of individual, thus maximizing the probability of a single mutation being inherited and then easily traced. Consanguinity increases the likelihood of homozygosity of a causal mutation present in deaf children, making its finding easier.
To this end, we develop close collaboration with ENT specialists, audiologists, geneticists from several countries around the Mediterranean Sea. At the top, we can see parents and children from deaf family living on edge of a desert who generously agreed to participate in this research. Below the first cohort we study, a large kindred of over 200 individuals from North Tunisia. Close square and circle represent affected, that is deaf individual, whereas open symbols indicate unaffected individual that is normal hearing. This study led us to identify the location of the first causal deafness gene on human chromosome. The corresponding deafness form was named DFNB1. DFNB indicates that it display an autosomal recessive inheritance, and one identifies it as the first deafness form to be mapped on human chromosomes.
Today, more than 150 genes are known to be responsible for monogenic form of deafness, as shown by this curve with green circle representing a mean discovery rate of about seven new gene per year. The green arrow indicates the discovery in 1990 of the first human deafness gene by our [lab] in collaboration with Steve Brown, a mouse geneticist. This finding provides the first evidence that defect of the same gene could cause isolated deafness, both in humans and mice. The subsequent discovery of the other causal deafness gene confirmed this conclusion, and this finding has had an incredible, unprecedented impact in the field of audition and audiology. Clinical examination of patient can provide only rudimentary information.
Furthermore, the impossibility of observing the human cochlea directly is a huge disadvantage, particularly with respect to retina, for which noninvasive examination, such as optical coherence tomography, can be used to observe individual photoreceptor cell in the retina. In this context, studies of mouse model through multidisciplinary approaches have provided valuable insight into the pathogenic process of each hereditary and genetic form of deafness. The genetic architecture of monogenic form of deafness in human can be summarized as follows. Together, 157 genes have been identified as underlying isolated non-syndromic deafness form in humans and about 300 genes underlying syndromic deafness. Recessive deafness form of isolated monogenic form of deafness are mostly congenital, and they have an early onset. They are generally the most severe form. Dominant form of isolated monogenic deafness have a later onset, are progressive, and are generally less severe.
Several of the genes can cause either dominant or recessive form. Similarly, some of the gene responsible for syndromic form of deafness can also cause isolated deafness. Over the last 10 years, the predisposing genes to deafness have also been discovered. For example, 100 or so genes predisposing to noise-induced hearing loss and 150 gene predisposing to age-related hearing loss have been identified. Now I will focus on DFNB1A. This slide illustrates the contribution of the various deafness genes to the DFNB form in Algeria. Although we detected mutation in 24 deafness gene, a single gene, GJB2 in green, accounts for half the cases of congenital deafness. It encodes a connexin 26. Connexin constitute a multi-gene transmembrane protein family with a common topology as shown on top left. This family has 21 members differing principally by their intracellular domain.
Six connexin, as shown bottom left, combine to form a connexon, which may consist of several different connexin. Two connexins from neighboring cell can assemble to form intercellular channel between cell, allowing the bidirectional transit of cytoplasmic ions such as calcium or potassium, or let's say inositol trisphosphate, microRNA, glucose, and other molecules. Finally, several intercellular channel can group together, as you can see at the bottom right, to form a gap junction plaque. The cochlea contains two main type of connexin, connexin 26 and connexin 30, that are similar in sequence, but differ in permeability to ion and other molecules. At the top right in green, we can see cochlear gap junction in black detected by antibody against connexin 26, which connect adjacent cells.
Most of the cell in the cochlea express connexin 26, as shown on the top left in green, and in a greater detail at the top right. In the top center, a cochlear section, the sensory hair cell are shown in red. The surrounding cells, the supporting cell in blue, are connected to each other via gap junction. In addition, in the stria vascularis, two of the three layers are connected by gap junction. The cochlea thus contain two large gap junction network. The epithelial gap junction network, shown bottom left, connect all the cells surrounding the sensory hair cell and their adjacent epithelial cell.
The other gap junction network is a connective tissue gap junction network, shown bottom left, which consists of two types of cells, the fibrocyte of the spiral limbus and spiral ligament in green in the middle, and the basal and intermediate cell of the stria vascularis, just in orange. These two networks are separated from each other, but all of the cell in each individual network communicate. During development, the epithelial gap junction network form first on embryonic day 16 in mice. The fibrocyte network develop next, followed by the stria vascularis network, and the fibrocyte and the stria vascularis network then fuse to form the connective tissue gap junction network. As shown on top right, the hair cell, in red, are not connected to the surrounding epithelial gap junction network. This slide illustrate the hypothetical potassium recycling pathway within the gap junction network in the cochlea.
Several roles have been proposed for cochlear gap junction network, including potassium buffering and recycling. Indeed, mechanoelectrical transduction leads to the extrusion of a potassium ion from the hair cell. These ions, which are toxic for the hair cell, enter the supporting cell and move from cell to cell via channel composed of connexin 26 and connexin 30. Potassium ion also enter connective gap junction network, generating the endocochlear potential and resulting in potassium secretion in the endolymph. Connexin 26, connexin 30 channel are also essential during development, notably for the formation of the tunnel of Corti. There is also some evidence suggesting that this channel are involved in the protection of cochlear hair cell in adulthood. Studies in mouse model have shown that the lack of connexin 26 prevent the opening of the tunnel of Corti, which play a crucial role in hearing physiology.
Lack of connexin 26 abolish the endocochlear potential, preventing mechanotransduction in the hair cell and resulting in hearing loss. Hair cell may acquire functional defect and directly resulting from mild to profound hearing impairment. The key missing information for the development and the testing of gene therapy product concern when and where connexin is essential for cochlear cell survival and auditory function. This lack of information is largely due to the embryonic lethality of GJB2-deficient mice, resulting from the critical role of connexin 26 in the mouse placenta. To bypass this obstacle, conditional or inducible knockout mouse model are required. To this end, we engineered conditional mouse model by crossing Cre driver mice with a mouse line carrying two loxP site recognized by the Cre recombinase, such that the Cre recombinase will delete the GJB2 gene.
In order to generate this Cre driver mice, we develop a single-cell transcriptomic atlas of the mouse cochlea at postnatal stage P8, P12, and P20, identified the time point and cell in which GJB2 is expressed. This allowed, as I said, to design specific cochlear cell promoter for our Cre driver mice. We engineered three mouse models for investigating the role of connexin 23 in the two gap junction network. The first two model focused on connective tissue gap junction network, initially in the fibrocyte and then in the intermediate cell of the stria vascularis. The third model was designed to evaluate the role of connexin in the various cell type of the epithelial gap junction network. In the fibrocyte, connexin 26 is expressed in fibrocyte I, type I, II, and V.
As shown top center, in mutants, connexin 26 is in white, is no longer expressed in the spiral ligament and the spiral limbus, but persists in the neuroepithelium. No morphological defect were observed in these mice. Below, on the left, in the control, connexin 26 is shown is yellow and is present in the spiral limbus and the spiral ligament. Let's move on to the basal and intermediate cell of the stria vascularis. Connexin 26, in yellow, is present in the intermediate cell of in the control animals, but has disappeared from those cell in the mutant top left. Below, at higher magnification, connexin 26 can be seen in white in the intermediate cell of the stria vascularis in the control. However, it is no longer present in the mutants.
Next to the control, the expression in the control, the expression persists in the fibrocyte of the spiral ligament and the spiral limbus, showing the specific deletion of the gene in the intermediate cell of the stria vascularis. In the last model, addressing the role of connexin 26 in the epithelial gap junction network, we observed three different phenotype based on ABR profile. Top left, we can see the controls with connexin 26 detected in the epithelial gap junction network. In the mutant, on the right, connexin 26 persists in the stria vascularis, but is not detected in the epithelial gap junction network. We classified the mutant mice according to the ABR threshold at postnatal day 15 and at one month. Phenotype 1, the curve in purple, correspond to profound deafness phenotype.
Phenotype 2, in red, correspond to severe deafness, and phenotype 3 on average in orange, correspond to normal hearing at one month, followed by the progressive development of deafness between months two and six. As shown here, the three phenotype were correlated with the labeling for connexin 26. In phenotype 1, connexin 26 was undetectable in the epithelial gap junction network, and the tunnel of Corti collapses. In phenotype 2, connexin 26 expression was weaker, but the protein was still detected to some extent in the epithelial gap junction network, and the tunnel of Corti is open. In phenotype 3, connexin 26 labeling is weaker, and the tunnel of Corti is open. No inner hair cell loss was detected in any of these three phenotype for up to six months.
At this time point, the loss of outer hair cell was limited to the base and mid base of the cochlea, and only in phenotype 1. Distortion product otoacoustic emission, which evaluate outer hair cell functions, have shown that the outer hair cell of model 1 and 2 are dysfunctional, as shown here at the top. Whereas in model 3, the outer hair cell dysfunction emerged progressively and was almost complete at six months, the curve in orange. This GJB2 model with three phenotypes provide a unique experimental platform for assessing the efficacy of gene therapy in different cochlear cell type within the epithelial gap junction network. To summarize, we have initiated a new area of innovative therapeutic solution for patient by discovering the gene responsible for hearing loss.
We have identified cochlear cell in which connexin 26 expression is essential for normal hearing, and have generated mouse model for testing the requirement for connexin 26 expression in the other cochlear cell type. Our biallelic GJB2 deletion model, in which mice have congenital severe to profound deafness, mimic the most common form of DFNB1A. By decoding DFNB1A, we are driving therapeutic innovation and paving the way for the development of gene therapy for the benefit of patient. Here are the member. Sorry, here are the member of our team working on this project. Many thanks to all of them, and many thanks to all of you for your attention.
Thank you.
Thank you. Now we are moving to the third session of this webcast, and with my colleague, Valérie Salentey, on the regulatory department at Sensorion, we would like to present the preclinical package that we have generated so far. This package, again, comes from the longstanding collaboration with Institut Pasteur, for which Sensorion plans to have an exclusive collaboration and access to an exclusive license option for the program. This is built on the co-development model on the GJB2 gene therapy program that was developed in the laboratory of Christine Petit. The two teams have worked together to advance the program from target identification and then tailoring a clinical candidate. Our collaboration provided two main pillars.
First, as said before, a very deep mechanistic understanding on the biology of connexin 26, which are the cells that are critical to target, and that allowed us to build a very good vector to address these most relevant cell types. The second pillar is, as said also, very relevant disease mouse models, highly relevant for the disease with the biallelic deletion of connexin 26. This mattered all for the design of the program because really, target cell addressing, silencing the expression where it's not supposed to be, and the right models, and the right efficacy results are critical for success. Let's start first with the construct and the design rationale from SENS-601. What we want to achieve is restore connexin 26 in cells that critically express the protein and not in cells where it's not supposed to be.
For that, we selected a capsid called AAV-DJ, which provides at the basis a broad tropism for this supporting cell that we are targeting. To drive the expression of the human connexin, we selected a constitutive promoter because it ensures a strong and durable expression. It is well- characterized in the cochlea. The transgene is a human connexin. To drive the selectivity of expression and in particular silence the transgene in cells that don't express connexin, we are using a biological mechanism called microRNA. We have added regulatory sequences against the microRNA-183 in our gene construct. Along the way, we patented both the vector and the microRNA target sequence for the program. How this silencing work? We took advantage of the fact that in the cochlea, hair cells selectively express some microRNA, including microRNA-183.
This is showing up in red in the right bottom section of a mouse cochlea, age one year old. When the construct is transducing the cells, two options. This is hair cells. There is a microRNA inside, and this microRNA will bind to the microRNA target sequence, which is in the three prime of the transcript. That will lead to prevention of the production of the protein in cells that express this microRNA, hence the hair cells. In cells that do not express hair cells, the supporting cell here, there is no microRNA, so there is no silencing, and there is steady and stable expression. Our expression system is selective off in sensory hair cells, on in target supporting cells. It is highly efficient.
We have demonstrated the transgene silencing efficacy goes from 95% up to almost 100% from in vitro and in vivo, so it is very conserved in term of efficacy. It is sustained because the sequence remains expressed in the aged cochlea. It was demonstrated by RT-qPCR in mice and human, and we demonstrated locally by in situ hybridization, as shown here below. It is also durable because we have gathered data showing that the silencing is durable for at least one year in mice. Not to mitigate, it is also important to note that this type of gene control is well-tolerated in the cochlea. Let us see how well the vector transduces the cochlea. For that, we used mice and non-human primates, so mice on the left, non-human primates on the right.
We use a surrogate vector of SENS-601 that express GFP here to easily detect the cells that are transduced and express the transgene. What we did is compare a construct that bear or not a microRNA target sequence in the construct. When you compare these two, it is apparent in mice that without the target sequence, the vector has a good tropism for the supporting cells. It is shown in green. But also a very good transduction efficiency for hair cells at the layer in yellow, highlighted by the three white arrows. When we add the target sequence, you can see on the right part that this is not changing the tropism toward the supporting cells in green, but there is almost no detection at the level of hair cells.
What we see on the image, it is just the end feet of the supporting cells surrounding hair cells, but not the hair cells themselves. That was quantified with a histogram, and you can see virtually no transgene expression in hair cells. This was also recapitulated on the right part in non-human primate cochlea. Without the target sequence, we have good tropism for supporting cells and expression in hair cells, and fully de-targeting in the right part when the regulatory sequence is added. So we have a very good rate of the target cell population and a very good de-targeting of the hair cells. To push this further, we examine cochlea of monkeys injected with SENS-601, in which the connexin bears a FLAG-tag on the 3′ , which allow us to discriminate between the endogenous connexin of the monkeys and the transgenic one, which has a tag.
So we stain in red and green these two proteins, and the resulting signal shows up in yellow here, and it is really apparent that at the level of supporting cells on the top and below of the image, we have a localization of the protein at the membrane with typical staining relevant for gap junction detection. And at the level of the hair cell layer, the outer hair cell or OHC, inner hair cell or IHC, they appear completely black in term of staining because they have no endogenous connexin and no transgene protein is expressed. So this is a very nice confirmation of the correct addressing of the protein at the cell membrane and de-targeting of the hair cells. Then we looked at the durability of the transgene expression in the cochlear context. So we did experiments in mice and non-human primates.
The picture shown on the top of this slide are mice that are stained six months after injection of GJB2 bearing connexin 26 and the FLAG-tag. And you can see the same pattern is expressed, a nice expression in the supporting cells in green and no expression in hair cells, but there is a maintenance of the expression of the protein in the long term. This was recapitulated in non-human primates. And on the right, these are RT-qPCR quantification, demonstrating that the mRNA also remains expressed with no loss of expression in the cochlear context. Now moving to efficacy part of the package. SENS-601 has been tested both in vitro and in vivo for efficacy studies. And here we are showing data on cells that have been cultured in vitro, in which we monitor the reconstitution of functional hemichannels that will assemble into gap junctions.
So we are using HeLa cells that are deficient in connexin, and inherently, these cells are not able to internalize a dye such as a propidium iodide, which is a fluorescent dye. When we provide back some connexin, these cells will be able to generate functional hemichannels, and they will internalize the dye. If we focus on the histograms in the middle of this slide and on the gray bars, you see that in the control non-transduced cells, there is no dye internalization, so none of the cells are positive for the fluorescent dye. When these cells are transduced or when these cells are made of making a stable clone, making connexin 26, in both situation, we have dye internalization and a strong percentage of cells positive for the dye. And this can be blocked by two types of inhibitors of gap junctions, such as high calcium or carbenoxolone.
And on the right part, interestingly, what we did is generate cell lines that bear a well-known pathological variance of GJB2 gene found in human, either truncating mutation or missense variants responsible for profound or severe hearing loss or progressive hearing loss. And upon the transduction of the vector, we show here that these cells are able again to internalize the dye, hence they have a functional hemichannel due to the expression of the transgenic connexin 26 protein. And below, what we wanted to confirm is the fact that separate hemichannels are able to assemble into gap junctions. So what we did is use another dye, microinject it into a cell, and then follow the dye transfer from one cell to the receiver cells, and that was done upon transduction of SENS-601.
And with that type of experiment, we were able to demonstrate that not only we can reconstitute hemichannels, but the hemichannels can make gap junction that connects to adjacent cells. Now moving to in vivo efficacy studies. For that, we used the knockout model developed by the laboratory of Christine Petit. This model was internalized at Sensorion, and we recapitulated the initial proof of concept that were made by the Christine Petit teams, and that was done in-house. Two types of experiment were performed. On the left, we injected a number of cohort with different dose of SENS-601, and we measured the hearing restoration using ABR. The lower the ABR threshold is, the better the hearing is.
When we compare the orange mice, which are the severe to profound congenitally impaired mice, which have ABR threshold at 80 dB, we saw that most of our mice cohort develop a statistically significant amelioration of the ABR threshold. From that, we were able to derive a responder criteria for these mice, which is when the mice have 20 dB improvement on at least one sequence, we call it responder, and that's a pink curve. You see that the responder mice, so 60% of the mice have a very good hearing that almost overlaps with the normal hearing mice control, which are shown in blue. On the right part, we engaged to several dose- ranging studies to determine the efficacious dose, and this dose was determined as, and we call it the 1x dose.
We observe overall efficacy within three weeks of injection of the mice at P0 or P2. As you can see, this is occurring across all tested frequencies. This was reproduced across multiple batches, ranging from R&D through the one made with the final manufacturing process. That's ABR. But fully in line with the restoration of connexin 26 biology, we're able to show that there is also an improvement of otoacoustic emissions and endocochlear potential in these mice, for which Professor Petit has shown that they are deficient in this model. We were able to confirm that the effective dose is really reproducible in the mice. Now, what we wanted to achieve is demonstration that there is an improvement in the connexin 26 levels and its network. On this slide, what I am showing here is a connexin 26 staining in the mutant. That's the middle panel.
This is fully depleted, for example, in the outer sulcus, as we show here, and the lower image show that upon SENS-601 treatment, we have a comparable staining as the control mice, which is on the top of the slide. On the right part, we quantified through image-based analysis using artificial intelligence and algorithm that the level of connexin 26 using our efficacy dose is almost back to normal when we compare the green histogram and the red histogram, the red histogram being achieved with the 1x dose, which is efficacious dose. That was the level of the connexin protein, and then what we wanted to show is whether we can restore the connexin network in the cochlea because it is complex. For that, we reduce the complexity of the network into various network metrics, which are shown here.
So that is the density of the network loop, the length of the loop, the length of the network, all of these parameters were analyzed on over 600 images of the different cohorts. The outcome of these studies is that we can achieve network improvement across these multiple metrics, which are characteristic of a good connexin network. Importantly, this was achieved not only in the outer sulcus, but in other key elements of the cochlea that critically express connexin 26. Below, this is to show that these metrics are associated with the ABR outcome. So really the network improvement tracks with the auditory recovery in these mice. Now I would like to focus on specific topics that relate to the delivery approach. Over the years, we have developed an approach, which is made of a catheter and a pump to inject the product.
This was developed in non-human primates and then used in the clinic for previous gene therapy programs. This is designed to allow a good dispersion of the vector into the inner ear fluid to minimize overpressure and no backflow is seen with this device. This has been shown to be well-tolerated in the clinic with no serious adverse event reported. It was reproducible across multiple administrations and across clinical sites. When you use this technique in the non-human primates with SENS-601 surrogates, for example, on the right, this is a SENS-601 vector that drives GFP expression. We can observe that at the dose, which is equivalent in the NHP as effective dose determined in mice, we have a strong transgene detection in the cochlea from the base where we inject up to the apex.
This is also confirmed using a FLAG-tag surrogate, where we detect the FLAG separately of the endogenous connexin in monkeys. Using the same approach that we used in the mice, through image analysis, we were able to demonstrate that the network is also put back to normal situation with a different compartment of interest and along the tonotopic axis in the monkey. Here, again, using this technology, we are able to demonstrate that the transgene expression pattern in NHP is fully consistent to the one we observe in mice, which is our disease model. So this gives me the opportunity to speak about an important point at this stage. We have been determining in mice an effective dose, and this dose has been translated into a human equivalent dose as a starting point of our clinical trial.
This is really important to say that in between mice and human, with this NHP equivalent dose, we are able to demonstrate very good tropism for the supporting cells across the tonotopic axis. So we are confident that the dose is supported by mice efficacy studies and very good exposure in the non-human primates. Now in terms of safety package, for this program, we have been performing two types of studies in mice and non-human primates. Rapidly in mice, this was IV injection to mimic unwanted systemic exposure due to, for example, failed surgery. In non-human primates, that was based on the device that we plan to use in the clinic. In two studies, it was two combined three and six months GLP toxicity studies and biodistribution studies. We used two cohorts, a low dose and high dose in monkeys.
The dose were designed to provide a suitable safety margin for the clinic. The message is that the tolerability in both species is very good. No mortality. In monkeys, we found a small and transient response against the transgene, sorry, against the capsid, but not against the transgene. Very minor observations were made on these species. We found dose-related biodistribution, which was very limited outside the target organ. Outside this target organ, the exposure rapidly decreased with time. The key message for this part is that we've worked on the cellular targets. We've worked on the expression of the transgene in the right cells, and this was successful. Based on that, in the right models, we were able to demonstrate the compelling efficacy, normalization of the connexin 26 network, and various sign of functional restoration in mice models.
No signals of safety have been identified in the toxicity studies. The surgery is known to be well-tolerated in the clinic. We are very hopeful that this program will be successful in the clinic, and we are very excited to bring it there. With that, I will stop and hand it over to Valérie for the regulatory part. Thank you.
Thank you, Laurent. Before turning to clinical development, let me briefly cover the regulatory framework that was behind our clinical trial preparation. Alongside the preclinical data package that we've just seen, Sensorion has developed a robust regulatory strategy built on proactive engagements with health authorities, and this is driven by two strategic consideration. First, for a gene therapy medicinal product, the preclinical data package is typically mature at the time of a first in-human clinical trial planning. This makes early alignment on marketing authorization requirement both feasible and very valuable. Second, early discussion with health authorities offers a clear and efficient path to clinical trial authorization. In this context, regulatory interaction were initiated early in 2025 with the U.S. FDA and the European Medicines Agency. This initial feedback was then followed by additional consultation closer to the clinical trial application.
In Europe, pre-submission meeting occurred with the French health authority, ANSM. This exchange resulted in the resolution of outstanding comments, and more importantly, it also confirmed the eligibility of the clinical trial application to the Fast Track procedure in France. With our global package ready, what was just mentioned before is that our first in-human clinical study with SENS-601 was submitted in Canada and France just before the summer. We received the authorization in France. Thanks to the Fast Track procedure, the review timeline were shortened significantly. The application remained under review in Canada, and Sensorion plans the next wave of submission in the U.S. and Australia by year-end. This regulatory foundation supports the clinical trial that is underway, and I will now hand it over to Sharon again to present to you the clinical development and the study design. Thank you.
There we go. Thank you very much. We've started in the clinic, we've gone to the bench, and now we're going to go back to the bedside with this final short component. As I mentioned, the Sensorion program is really a series of well-planned building blocks, and if we go to the first slide, what I first want to talk to you about is this backbone, which is a natural history study called OTOCONEX, upon which the genetic therapy, HearConnex, is built. Next slide. What this is it's a longitudinal clinical, audiologic, and genetic data support study, which really is meant to feed into and feed from the HearConnex study. The data that's collected, again, is within all of these domains, the genetic phenotype, the audiologic phenotype, as well as functional hearing assessments.
There are several populations that this encompasses in children age 16 and under, as well as children 10 years and under who have an implant. It's really a reflection of our current data set of individuals, along with children who have a variety of different phenotypes of hearing loss who are pre-lingual and post-lingual. It really covers the entire variability of the population. Next slide. These are the assets that are provided in OTOCONEX, and this really fed into the design of the clinical trial, which is why this is so foundational, is that we use this data in order to design the clinical trial as opposed to basing it on assumption in this patient population. If we think about the patients, first of all, what it allows us to have is a large group of genotyped patients.
Some of these will be candidates, some of these will not be candidates, but both are valuable to our understanding of what we need to do with HearConnex. It's provided clinical characterization of the phenotypes involved, just because we know that there is variability in the phenotype, and we want to use that narrow band of individuals that we're targeting. It also provides longitudinal follow-up in terms of what is the natural history, so that when we add therapies overlying this, we have a sense of what is consistent with natural history of disease versus what is not consistent, what is therapeutic effect. The second part is the sites, and so it's allowed the development of relationships with experienced investigators, as well as referral pathways and a readiness in the clinic to have early phenotypic and genotypic data available very early on in the pathway.
That is really the foundation for patients that are going to then proceed to an intervention like HearConnex. It's also allowed for a standardization of the assessments, and this is really important when we're thinking about outcomes, especially in children, where we can't always test everything that we want. It's provided a good sense of what we should be testing to determine the clinical efficacy. The final component is the data asset. It's really taught us a lot about the natural history of the data. Much of what we knew in the setting of connexin-related hearing loss was biased by our children with severe to profound hearing loss, and we're getting a much more fulsome picture as a result of this natural history.
It also shows us the power of the outcome measures that were chosen and how this disease progresses and the impact that will potentially have based on the therapeutic. Let's move forward to the next slide. HearConnex. I think a lot of people are anxious to hear this part in terms of the clinical study design. Let's get right on to the next slide. As we talked about, GJB2 is very common and it's frequent. The standard of care for the patients with severe to profound hearing loss due to connexin currently is cochlear implantation, and we've gone over those limitations. What we're hoping to do with SENS-601 is really to facilitate physiologic hearing without any of the drawbacks and really reduce that duration of deafness in children born with connexin-related hearing loss. Next slide.
This is a multicenter phase I/II trial that is focused on safety, tolerability, and efficacy of SENS-601. The trial objectives, the primary objective is dose escalation. This will be a two-pronged escalation that will really have as a fundamental outcome safety and tolerability of both the drug and the delivery system. It will be very closely followed by dose expansion trial, where the outcomes will be the efficacy component on the ABR and pure tone audiometry response with secondary view on efficacy and safety of both the drug SENS-601, as well as the injection system, and how usable it is along with the clinical benefit.
As Valérie did a beautiful job, regulatory pathways are well, well underway, and the hope is that the first patients will be enrolled in early 2027 and that throughout the course of the year, data will be generated and released with each of the cohorts. Let's take a little closer look at the details. This table really tells you everything that you want to know. Part one is dose escalation, and these are again children with non-syndromic bilateral severe sensorineural hearing loss, who would otherwise be eligible for cochlear implantation. They're in that severe range as we want to have enough neural substrate to demonstrate efficacy. The first part one, will comprise of eight children. We have a beautiful age range.
If we go back to what we talked about in terms of limiting that duration of deafness, the youngest patients will be six months all the way up to 31 months if they have not previously received a cochlear implant. That age broadens up to six years if children have been unilaterally implanted within 18 months of the trial. These are going to be unilateral intracochlear injections, and we talked a lot about the importance of bilateral, and that is why it's important that this component be efficient so that we can get on to bilateral injections safely so that we can achieve those benefits that we discussed in terms of binaural hearing.
There's going to be two dose levels. Laurent did a beautiful job explaining the rationale for those dose choices and the confidence with which we are going to be in the correct part of the dose -response curve. The primary objective of part one is safety and tolerability. Very quickly we're then going to go into dose expansion with additional 10 children. Here, these will all be cochlear implant naive children. Again, very young age, six months to 31 months, aimed at reducing that duration of deafness and maximizing the potential outcome of the therapeutic. These will be bilateral injections. They will be done at the optimal dose as based on part one. This is really where we're going to be assessing efficacy of the primary endpoints in terms of ABR and PTA. Next slide. Here's a graphic of how that's going to be done.
Really, this was designed to allow very rapid transition to the pivotal trial. That section of cohort three, where we know we want to eke out bilateral benefit, which is the true power of genetic therapies. The expectation is that data about each cohort will be communicated in turn. Next slide. In terms of the centers that are already involved, this is an international study. I'll point out Necker in Paris as well as The Children's Hospital at Westmead, who were involved in SENS-501. This is by design. These are two experienced centers that have experience with this specific procedure. We're adding our site in Canada at SickKids, as well as a site in the U.S. Next slide. If we go through the positive benefit and the risk assessment, this is that critical partnership between safety and efficacy.
If we look at efficacy first, because that is our ultimate goal, it really is grounded in the mechanism. Professor Petit and Laurent did such a beautiful job of explaining all of the work that's gone in to suggest that we're at the right stage to go into the clinic with respect to efficacy. If we go to the bottom slide, this is a surgical procedure that is already performed in this population. Again, we're using the building block of what was learned with SENS-501. In addition, this is something that is absolutely in the toolbox of surgeons who use cochlear implants on a regular basis. In terms of toxicity and safety, that is the fundamental component, particularly when we talk to parents. The first question they're going to ask me is: Is this safe?
Certainly, both the mice and non-human primate data, has really reported an absence of any concerning safety signal. The final block is that our options remain preserved. What we've learnt, based on previous trials, is that our opportunity to go to cochlear implant remains when we don't achieve a therapeutic result in a given child. This is very reassuring for myself as a surgeon, it's reassuring for parents. What we need to make sure that we do is we keep that duration short. The earlier we can go with the therapeutic, the earlier we can pivot in those cases where it's required. Next slide.
I am going to sum up. The key message here is that really, OTOCONEX is really the backbone upon which HearConnex is built and has helped define the endpoints, the disease trajectory, as well as the patient survey before the trial was even designed. This data fed into the trial design, which was not based on assumptions. HearConnex really leverages the experience. This is, as I said, it is a building block, and many things that have come before have really laid a very solid foundation to create success in this domain. The final take-home message is that this is really a phase I/II trial that allows flexibility to yield early data regarding dose selection as well as confirmatory data, that will get us to where we want to go with respect to bilateral administration.
This is really an exciting time, and I think 2027 is going to yield lots of answers, certainly more questions. I will now hand it back to Fred to start the discussion. Thank you very much.
Well, thank you all. That is the end of the presentation part. We are running a little bit behind, but I think there was a lot to cover. The team will now take your questions. As I said, to ask one, please dial in on the conference number, and I think the lines are now open.
If you wish to ask a question, please dial pound key five on your telephone keypad to enter the queue. If you wish to withdraw your question, please dial pound key six. The next question comes from Yasmeen Rahimi from Piper Sandler. Please go ahead.
Good morning, team. Congrats on the really excellent presentation and very comprehensive review for us. Thanks also to walking us through the HearConnex detailed trial design. The first question for you is, as we think about cohort dose level 1 and dose level 2, if you could maybe comment on sort of the separation, what magnitude difference it is between dose one and dose two. Then for our very thoughtful KOL, per dose expansion is going to be really helpful to detect some key efficacy markers. It would be great to hear your thoughts at what could we expect to detect changes in ABR. If you could maybe also comment on PTA and maybe also comment on what is considered a clinically meaningful differences in those assessments that we will be looking for to part two of the study.
Thank you, Yasmeen, for your multiple questions. I think maybe the first question is on the rationale between dose one and dose two. Laurent, you want to take this one?
There will be a threefold difference between the two dose.
Does that answer your question, Yasmeen? I mean the first part.
Yeah. Thank you.
Okay. Then you had a question about the dose extension and what we can expect on a clinical standpoint, regarding from this data. So maybe, Sharon, maybe you can comment on that?
Yeah. I think certainly the primary thing we're going to be looking from part one is a clean safety signal. Okay? And so that's always going to be the fundamental baseline. When we think of clinically meaningful differences in terms of outcome, I think Laurent did a really good job of demonstrating that in the mouse model. Certainly when we look at improvements in hearing, we recognize through prior genetic therapies that we're not going to get every child to normal hearing, but getting them to the place where they may be able to wear a hearing aid is also of benefit. I think that there is certainly changes in PTA and ABR that get us better than a 70 dB hearing loss, all have different elements of success.
You also asked a question about what's going to be the timeline to see those changes, and I think that it's something every clinician and every parent wants to know. I'm an eternal optimist, and I hope that what we've learned with otoferlin stays true for GJB2 in that oftentimes there were very early responses detected. They often were not formally measured until three months post-dosing, but the anecdotal evidence suggests that it was much before. It's difficult to say whether that will be the same with connexin. We think about ABR responses, but they're not the be- all and end- all either in terms of behavioral responses, in my mind, are the most important outcome measure for these children.
Thank you, Sharon. Yasmeen, I hope we answered your question, and maybe we can—
Yeah, that was very, very helpful. Thank you. I'll jump back in the queue.
Thank you very much.
The next question comes from Cecilia Hernandez from Van Lanschot Kempen. Please go ahead.
Hi, this is Zoe on for Cecilia from Van Lanschot Kempen. Thank you so much for a great presentation and for answering all our questions. My first question would be on GJB2. What is the ideal treatment window to treat these patients, and what is the overarching goal, and what can patients realistically achieve with this gene therapy? My second question would be on the patient population. Why did you specifically decide to go for the six-month to 18-month range, if I remember correctly? Thank you so much.
I think what we can expect clinically, I think Sharon already answered, but maybe, Sharon, you want to talk about the ideal timing for the treatment, and maybe what was the second part of the question? I forgot, sorry.
The 18 months' difference—
Oh, yeah.
—apart from the first cochlear implant.
Right. The way I think about. Oh, go ahead.
Cecilia?
No, sorry. The second question would be on the patient population and the age that you chose for this group.
Age. Okay. Sorry.
Okay.
You broke up a little.
Sorry.
Yeah. I think this is such an important question, and it is fundamental to success because when we think about duration of deafness, every day with deafness is not a great day for the brain in terms of development. While there may be these critical windows that as humans, scientists, surgeons, parents, we define as as long as we do this before two or 31 months, which is the endpoint, the truth is that any additional day is an additional day of delay. I think that is where it is so critical. We know that we can get cochlear implants in as young as six months of age, and so it is important in terms of the overall outcome. I think it is also important in terms of the plan B.
In the child where it does not work, and we need to pivot to cochlear implant, we do not want to lose time, and we want to leave ourselves enough time to measure efficacy. To me, that is where that 6- 31 months comes into play. As a clinician and as a mom, I want to be as close to six months as I can possibly be.
Thank you, Sharon.
Thank you very much.
Thank you for your question, Cecilia. Sorry. Maybe we can take the next question.
The next question comes from Joe Pantginis from H.C. Wainwright. Please go ahead.
Hey, everybody. Thanks for all the. So two questions, please. First, I wanted to get some additional color and/or confirmation about the clinical program and the delivery method. Is this the same surgical technique and delivery method as the OTOF program, and will you be looking at any additional means of surgical implantation? Number one. Then number two, what are your near or intermediate terms with regards to manufacturing capacity needs? Thanks a lot.
Okay, so I think I saw the first question about the. Yes. That's the same delivery as the OTOF program. Are we looking at others? Not for the time being. I'm sure you would understand that bringing a different delivery system in such a small trial and first in mind would not be the right thing to do, but that's potentially something we can consider for the future if we believe there are some benefits. I think your second question on manufacturing I would say, we have manufactured all the product we need for the part of the trial you have seen on the slides, but we have already started to build the right capabilities and capacities for the next steps already.
Thank you.
My pleasure. Maybe we can take a few more minutes for a few other questions.
The next question comes from Clémence Thiers from Stifel. Please go ahead.
Hi. Thanks very much for the presentation today. It's been very helpful. Just a couple of questions on my side. First, on the preclinical data. You showed 21 responders among 35 treated animals. Do you understand what differentiates responders from non-responders, and how could you translate it in the clinical trial? Second, you show a 6 - 12 months follow-up before dose selection. Is that follow-up primarily driven by safety, or would you expect some efficacy signal to evolve or to just show up? Just the third, you highlighted that the objective goes beyond simply restoring access to sound and is really about binaural hearing. Is that something that you would be able to demonstrate in the study, or is that benefit something you just would expect from gene therapy? That's all. Thank you.
Okay. Thank you for your questions. The first question, on the preclinical data and the responder rate.
Yeah.
Yeah.
Yeah, happy to answer that. We have this criteria, which is stringent, 20 dB at one frequency. It means that if mice demonstrated 19 dB, it was not selected as responder. That's one thing. We have a 10 dB variability in the ABR method measurement, so that can explain. But also remember, the mice has a very small inner ear. It's 2.5 µL in term of volumes, and we inject 1 µL. Any variability into the surgery or administration in pups under the microscope can explain some inefficient transduction and hence the responder status.
Then you had a question on the 6 - 12 month follow-up, and maybe Valérie.
Yes, I can take this one. After the second dose, indeed in the study design, there is 6 - 12 months follow-up. This is the result of our discussion with the authorities that they want to see a longer follow-up. Of course, during this follow-up, we will have the data, the efficacy data coming in, and it's open label. We'll be able to appreciate what is the efficacy day by day.
If I understood correctly, the third question, because unfortunately you were breaking up a little bit, but I think it's a question about neuronal connection and maybe, Sharon, you can take it.
Yeah. I love your question. The binaural benefit—
Binaural, yes.
—and how do we measure that?
Yes
There are lots of different ways to measure binaural benefit. Certainly, some of those will be standard measures in the protocol. Things like binaural summation, head shadow effect, signal and noise, all of those. There is also some fancier methods of doing it, looking at EEG. That is something that we do in the lab at SickKids. There is a whole spectrum, even including questionnaires that get at that binaural benefit. If you think about the data that I demonstrated in our kids with cochlear implant, with all of the green slides, what we would expect is when we repeat those measures in children who have been dosed bilaterally, that they would look much more like typically hearing children. So lots of different ways, and I cannot wait to measure it.
Good. Thank you. I hope we answered your questions. Maybe we can take a last one.
Thank you very much.
Yeah, my pleasure. Do we have more questions to come?
The next question comes from Jack Allen from Baird. Please go ahead.
Great. Thanks for taking the questions, and I really appreciate all the presentations today. Two quick topics from our end. I know one of my colleagues on the line asked about the delta between the two doses that you are going to test in the ongoing study. I just wanted to clarify what the low dose is or any initial thoughts you have around what that low dose is. I know for SENS-501, it was [1.5E 11 vg per ear], but that is with a dual vector approach. I am curious if there is any ability to reduce that dose level or if you are staying in a similar kind of dose range. That is my first question, and then my second question was more kind of on the commercial side of things.
I would love to hear the physicians' thoughts around the GJB2 prevalence and the use of cochlear implant in this indication and how it compares maybe to otoferlin. I know a lot of investors are familiar with the otoferlin product just being approved, the gene therapy from Regeneron. I would love to hear any thoughts around the use of cochlear implant in the GJB2 space.
Thank you, Jack, for your questions. Laurent, do you want to answer the first question?
Yeah, for the first question, at this stage, we do not disclose the dose in the clinic. I explained how it was rationalized. But what I can say is when you look at the spread of dose tested in preclinical models in the literature, there are maybe 10 papers demonstrating some attempts of hearing restoration over the years, and we are in the range of the dose, so it is not significantly higher and lower. Also, you are right, there is no need to compare to the otoferlin because it is a fully different context and setting and vector design.
But we have done a lot of work on the dose and how to calculate the dose based on the data we got on mice and monkeys, and we feel very comfortable about the dose we are going to take to the clinic in dose one and dose two. The second question was a kind of a commercial question from physician point of view, so maybe Sharon, if you want to take this one again.
With genetic trials waiting in the wings, we have patients who come into my clinic every day with GJB2-related hearing loss, and I think the biggest comparison to otoferlin is just the higher prevalence of this type of genetic hearing loss. I think the approach is very sound in that we are looking currently to start, and I will emphasize start, because this is truly just the beginning of exploring what gene therapy could do in this disease entity with children with severe hearing loss, where that risk-benefit ratio is optimized. These children would otherwise get a cochlear implant or would be leaning towards a cochlear implant, but they have enough neural substrate.
Children with profound hearing loss, I think if they came to see me today, are still going to get a cochlear implant, but there's also lots of other phenotypes of GJB2 that are milder, some of which are progressive. I really think that when we look at the entire demographic of children presenting with GJB2-related hearing loss, this initial population is really a starting point. It's very accessible. It will be very easy to recruit for part one and part two across the centers. That to me is a real advantage as a clinician in terms of knowing what's going to be our next step beyond this first one.
Thank you.
Great. Thanks. All the color.
Good. Thank you all for your questions. I'm sure there are many more questions, and there will be many more questions to answer. I'd like to move to the conclusion of our presentation first. Before we close, I would like to thank the speakers, Dr. Cushing, for the patient and the clinical view. I think it was very good to remind us what it looks like from a family standpoint, and from the disease when you are a patient in consultation. Thanks very much for Professor Petit for all this explanation on biology, the development of these quite unique models on which our programs are resting. Thank you, Laurent, for the preclinical package, and Valérie for the fantastic regulatory path we have done so far. Before we finish, I would like to leave all of you with some closing remarks.
The first one is, you heard the biology has been mapped, is very well mapped. Thanks to the experts we have been able to work with. The cell types that are critical for hearing in DFNB1A have been very well characterized in disease-relevant models, and we've developed a SENS-601, or we designed a SENS-601 around that. We believe it's a very good step to the clinic. Secondly, on the preclinical package, Laurent showed a lot of functional restoration. Connexin 26 expression and network has been restored. ABR threshold recovery across frequencies in the severe to profound model. Third, the trial is designed based on our own experience. Obviously, OTOCONEX gives us a lot of endpoints, disease trajectory, and helping a lot for disease identification, so we feel we are very well-equipped here.
The surgical procedure, it has been said, has been performed in children, and two of the participating centers are already trained. When you put all these things together, plus the learning from SENS-501, I think what comes next is quite straightforward. As we disclosed publicly, we are planning to dose the first patients by early 2027 and to disclose data throughout 2027. Thank you very much for your time this morning or this afternoon, and we are looking forward to updating you in the future. Bye.