Good morning, and welcome to the MeiraGTx program update. At this time, all attendees are in a listen-only mode, and a question -and- answer session will follow the formal presentations. As a reminder, this call is being recorded, and a replay will be made available on the MeiraGTx website following the conclusion of the event. I'd now like to turn the call over to Zandy Forbes, President and Chief Executive Officer of MeiraGTx. Please go ahead, Zandy.
Thank you everyone for joining us this morning. Today, we're presenting the full three-year data from our phase I study of AAV-AQP1 for the treatment of persistent moderate to severe radiation-induced xerostomia, or RIX, following radiation treatment for head and neck cancer. We will also provide an update of the commercial opportunity for this potential therapy. Be aware that we're making forward-looking statements, so please read this slide. On the agenda for today, I will first introduce the disease, go over the etiology of the condition, and the mechanism of action of AAV-AQP1 therapy. I will give a quick recap of the phase I 12-month data and then present the full three-year data, which shows robust durability and within-patient consistency over three years. We have two study investigators speaking today about the disease burden, the patient experience, and importantly, their hands-on experience of delivering this therapy in our clinical trials.
Following the physicians' talks, I will be back to provide an update overview of the commercial opportunity, and then we will open the call for questions. I thought we would start by considering how bad persistent radiation-induced xerostomia is. The answer is, it is very bad indeed. Persistent moderate to severe RIX isn't simply dry mouth. It is a severe lifelong condition with devastating consequences and no effective treatment. Neither does it just affect a few patients, as more than 30% of all patients who receive curative radiation treatment for head and neck cancer suffer for the rest of their life with this condition. What happens when you have persistent radiation-induced xerostomia? What are the symptoms? You have a very dry mouth. You can't swallow or chew, and you lose a sense of taste. There are major diet restrictions.
You may lose weight, and you may even need a feeding tube. There are oral health complications and frequent oral infections and sores in the mouth, constant pain, impaired speech, difficulty sleeping, and inability to exercise because faster breathing may lead to choking. Difficulty eating and speaking lead to reluctance to interact with others and social isolation. You can understand that poor nutrition, lack of sleep, inability to exercise, continual pain, and limited social interactions have significant life-changing impact and can lead to frailty and even premature death. I'm not sure that people who haven't witnessed or experienced persistent radiation-induced xerostomia can truly understand how horrible it is. We will hear later from two physicians who care for many of these patients about the lifelong disease burden, and you can see from the quotes on this slide some extremely unpleasant symptoms that can even lead to life-threatening consequences.
How does this condition arise in patients cured of head and neck cancer? All patients treated with radiation for head and neck cancer get acute xerostomia at the time of treatment. However, in 60%-70% of them, their xerostomia resolves or becomes manageable within about 12-18 months. However, in 30%-40% whose xerostomia persists after two years, xerostomia then persists for life, only getting worse over time. 80% of these patients do not respond to any available therapies. There are a lot of these patients with this severe, untreatable, lifelong condition whose trajectory is only further decline. Currently in the U.S., there are 165,000 of these patients with over 20,000 new patients each year. Globally, the number is large, 435,000 patients today and around 48,000 incident patients per year. This is a complete unmet need and a large commercial opportunity for an effective treatment.
Salivary glands are very vulnerable to radiation, leading to the acute loss of function and acute xerostomia in almost all patients at the time of radiation treatment. This disruption in salivary function can resolve over 12 or 18 months as the glands remodel and heal. However, often the damage is extensive, and the salivary glands don't recover. When salivary glands lose cells and structure, they lose the normal mechanism of saliva production. In normal glands, the polarity of the epithelial structure is critical for water flow as the water channel responsible for permeability, AQP5, requires polarity-dependent signals to function. Our approach is to deliver a one-time treatment with a viral vector that contains the human aquaporin -1 gene, AQP1.
The therapy is delivered by putting a little cannula through the opening of the parotid duct in the back of the mouth and instilling the appropriate volume directly into the lumen of the salivary duct. AQP1 encodes a non-polar water channel normally found in red blood cells and kidney expression of non-polar AQP1 in the remaining acinar cells and duct cells of the salivary gland results in these cells becoming permeable to water. This allows water to flow down the osmotic gradient across the cells from the interstitial fluid into the lumen of the salivary duct and into the mouth. Importantly, the delivery of AAV- AQP1 is quick and very easy. It involves virtually no pain, no anesthetic. It's a small dose locally delivered to the salivary gland, and it's a low cost of goods. David Owens will talk more about this later in the call.
Now onto the phase I study. First of all, let me remind you that xerostomia is entirely a patient-reported condition, and the severity of xerostomia is not correlated with the absolute amount of saliva. This is described in the ASCO guidelines on xerostomia. Every person experience a different degree of dry mouth depending on their own threshold for saliva. For example, one patient may have xerostomia with 100 mcL of saliva, and another will have xerostomia with a ml of saliva. The reason for this is that the shape and size of the mouth, the size of the teeth, the rate of breathing all impact the amount of saliva that is effective in stopping dryness in the mouth. However, xerostomia does result from too little saliva, which causes dryness and all the features I told you about earlier.
There's a relationship between xerostomia and saliva flow, but no correlation between the absolute values of either of these measures. In our clinical studies of AAV-AQP1, we look at measures of both xerostomia and saliva flow. First, the PROs, which is the patient-reported measure of the severity of xerostomia. In our pivotal study, the primary endpoint uses the xerostomia questionnaire PRO called the XQ, which is a standard tool to measure the severity of xerostomia. We also measure saliva flow, an objective endpoint that reflects the mechanism of action of the therapy. This measure is the unstimulated whole saliva flow rate, or UWSFR, which is the key secondary in our pivotal study. The phase I study design is as shown here. There are two dose-escalating groups. The first set of cohorts were treated in one gland unilaterally.
Having completed the unilateral dose escalation, we went on to complete a dose escalation treating both glands bilaterally. The primary in the phase I was safety, and the treatment in both unilateral and bilateral was remarkably safe, with only a few mild treatment-emergent treatment-related events, all of which resolved without sequelae. The phase I data at 12 months is very compelling, with both XQ and unstimulated whole saliva flow rate showing an unprecedented degree of improvement. Just as a reminder, here's the 12-month XQ data. In terms of XQ score, a decrease in XQ is a decrease in severity, which is good. On this graph to the right, the lower dotted line is the bilateral cohort, the same as in the primary in the pivotal study in which all patients are bilaterally treated.
The upper dotted line is the unilateral cohort, and the green line is the average of all treated patients bilaterally and unilaterally. The unilaterally treated patients do not have as strong a response as the bilateral. At 12 months, the bilateral patients had an average 21-point decline in the XQ. Consider this in the context of an 8-point decline being clinically meaningful and a 10-point decline transformative. In the bilateral cohort, an impressive 75% of the patients responded with a transformative 10-point or more change in the XQ. While xerostomia score and salivary flow rate are not correlated, we did see a strong improvement in saliva production at 12 months. This is an objective measure of the mechanism of action of the treatment and a key secondary in our pivotal study. On this slide is biopsy data one to three years after AAV-AQP1 treatment.
You can see that in six out of seven biopsies, DNA was detected with somewhat of a dose response, and in one patient there was sufficient material for protein staining, and AQP1 is clearly seen in the cells of the salivary gland. Now moving on to the three-year data. Here is the data for the XQ score over three years. When I saw this data, I must say I thought it looked remarkable. The 12-month time point, which we just looked at a little earlier, is shown with the arrow. The 21-point decline in the XQ at 12 months. Then if you go straight out to the far right to the 36-month time point, the XQ score is virtually the same as 12 months. Even the difference between the unilateral and bilateral are maintained at three years.
We think this demonstrates an extraordinary durability of benefit over three years, particularly as XQ is a PRO, which are sometimes considered potentially more variable than objective measures. Even XQ measures collected at 12 months apart, like two years and then three years, are very similar. From a population perspective, the effect of AAV-AQP1 treatment on the whole population is clearly maintained over three years. However, in order to understand that this average was not the result of some patients getting better and others getting worse, and this was not because of certain outliers or patients behaving strangely during those three years, we looked at the data for each individual patient at each time point. First of all, let's look at the waterfall plots at 12 months.
On the top left in dark red, you have the bilateral population, which remember, has a better XQ average score than the unilateral. Here you can see a really nice waterfall plot where everyone gets at least a small bit better and some people get extraordinarily better. If you look across at the level of the 10-point decrease on the Y-axis of the upper graph, you can see 75% of patients have a 10 or more XQ improvement. At the far right of both graphs are the patients improving by over 40 points on the XQ in both bilateral and unilateral cohorts. To the right is that waterfall plot for the XQ change from baseline in the unilateral patients who responded on average a bit less well than the bilateral.
While there are a couple of very high responders to the far right, there are a few patients with no responses as you get to the far left. The graphs shown below show the percentage change from baseline, and they are very similar to the absolute XQ scores from baseline above. When we looked over time and made little waterfall plots of the cohorts at each visit, the waterfall plots were very similar across time. There was some variability, and at the eight-month time point had more missed visits, but on the whole, the patients making up the population remained consistently distributed over time. This gives us comfort that the durability of the mean changes in XQ score from baseline reflects durability of responses in the individual patients in the population.
Finally, we looked at the durability and consistency of the XQ scores in each patient at every time point. For this analysis, we put the patients in the order of magnitude of the 12-month waterfall plot for XQ data. For each patient, the score at each visit is shown. From the left to right for each patient, the darkest red is XQ score at 12 months, the slightly lighter red is at 18 months, the next 24 months, and the bar on the right is at 36 months. What you can see at a glance is pretty good within-patient consistency. The patients with the biggest improvements at 12 months to the right of each group remained strong responders over time. Those with the worst responses to the far left hovered around zero over three years.
The good but not shockingly good responders in the middle stay around that level and may be a bit more variable than the non-responders to the left or the out of the ballpark responders at the far right of each group. One thing to note in the unilateral patients, who on average did not do as well at 12 months as the bilateral. While several of the non-responders at 12 months remained non-responders, two of them did respond to 18 months, patients 007 and 010, and then remain around that level of response out to three years. When you look at the graphs of all treated patients, both unilateral and bilateral together, as is shown here, it is really striking how remarkable the consistency of the level of XQ responses over time is. We think this is indicative of real responses to treatment that are very durable.
This increases our confidence in a positive outcome in our pivotal phase II study. In addition to the XQ, we also looked at water flow, unstimulated whole saliva flow rate. The arrow points to the absolute change in unstimulated whole saliva flow rate at 12 months. If you look to the far right, while there is variability and there are some missing data points, particularly at 18 months, we do see a very durable improvement in water flow in this entire population over a three-year period. When we look at the waterfall plot at 12 months and then the consistency of water flow rates within patient over time, we see a very similar pattern to the XQ. On the far right, patients that respond extremely well with increased water flow at 12 months also respond well at three years.
If you look to the far left, this is interesting because the patient that had the worst response, in fact a negative response at 12 months, reduced water flow. That patient continued to reduce water flow further over time. The patients who did not respond or had minimal responses at 12 months remained the same over three years with some variability around baseline. Those middle responders remain middle responders with a bit more variability than the best responder. Just one thing to note about the variability in water flow in these patients in the phase I is that they all had different hydration levels. If you look at patient S016, that patient has a pretty good UWSFR at 12 months but goes in completely the other direction, losing saliva production at 18 months.
This was surprising and in fact, the only time this had been seen, and it was determined that on the way to the 18-month visit, this particular patient had driven for many hours to get there, drinking only coffee, and when tested, they were found to be extremely dehydrated, explaining the inability to produce saliva. Because of this case, we included in the pivotal study protocol a requirement that before each saliva collection, the patient is tested for hydration. If they're dehydrated, they're required to sit and drink water until they're fully hydrated. We think that this might, to some extent, reduce the variability in water flow rates in the pivotal study.
Based on the durability and the intra-patient consistency of this data in both XQ, the primary endpoint in our pivotal, and in water flow, which is the objective measure of the drug function, we have increased confidence that the benefits we are seeing in response to AAV-AQP1 are real, and higher expectation of a positive outcome of the pivotal study. We include the three-year data in an application for breakthrough therapy, which we were awarded a couple of weeks ago. We also conducted a global physician and payer survey using the three-year data in the target product profile, which further improved the view of AAV-AQP1 for the treatment of persistent RIX in the eyes of both physicians and payers, and I'll talk about this later. Here is the design of the pivotal study. There are four escalating dose cohorts and two placebo cohorts. All patients are bilaterally treated.
It is double-blind and randomized and conducted at 30 sites in the U.S., U.K., and Canada. The study started as a phase II with three initial cohorts, 1:1:1 placebo to two different doses. In response to our IND amendment to open the phase II, the FDA wrote to us and said that as the manufacturer, if we manufactured the material used in this study to the specifications outlined in our filings, they would consider this a pivotal study. With this in hand, we filed for and were awarded RMAT status. In response to the briefing book for our RMAT meeting, the FDA provided positive answers to all of the clinical questions we asked regarding the current phase II being potentially the single pivotal study to support a BLA filing.
With these written responses, we were able to cancel the FDA meeting, and we have clear written alignment with the agency on this pivotal study. As we converted the phase II into a pivotal, in order to keep it double-blind and randomized and cover all the doses that we had seen responses in the phase II, we completed the initial 1:1:1 placebo versus two different low doses and then carried out a second set of cohorts, 1:1:1 placebo to two different higher doses. All patients are completely blinded until the last patient reaches the last 12-month visit. The study protocol has 276 patients randomized to one of four active doses or placebo. The primary endpoint is the change from baseline in the XQ score.
The study is 95% powered for success, with success being a difference in the change in XQ from baseline in any one of the treated arms versus the pooled placebo arms. The key secondary endpoint is change from baseline in unstimulated whole salivary flow rate. With that, I will turn the call over to two physicians, both of whom have experience in treating RIX with AAV-AQP1 in our clinical studies. The first is David Owens, a Consultant at the University Hospital of Wales in Cardiff. The second is Dr. Michael Brennan, Professor and Chair of the Department of Oral Medicine and Oral and Maxillofacial Surgery at Wake Forest University School of Medicine. Dr. Owens, over to you.
Thank you. Good morning. I'm David Owens, a Consultant Otolaryngologist at the University Hospital of Wales in Cardiff, and a former President of the International Head and Neck Cancer Conference. Along with my team, I'm one of a number of consultants in the U.K. participating in the double-blind, placebo-controlled AQUA x2 clinical trial of AAV- AQP1 in patients with persistent xerostomia following radiation treatment for head and neck cancer. Today, I'll be talking about my experience of delivering this treatment to patients in the ongoing AQUA x2 trial. As a Consultant Otolaryngologist, I look after cancer survivors with Grade 2 and Grade 3 radiation-induced xerostomia, which persists for many years after the cancer has been successfully treated.
It is a genuinely miserable condition, a lifelong one for which we have no good solutions, and in many patients, it is considered the worst consequences of curative radiation therapy for head and neck cancer. Anyone who has had treatment for head and neck cancer and suffered the consequences of dry mouth will know what an impact it has. Xerostomia is a social disability and can broadly affect the health and quality of the patient, as well as their caregiver. These patients live with constant dryness, pain, and sores in the mouth. Swallowing is hard. Taste is altered, making it difficult to eat, which can result in unwanted weight loss and frailty. Difficulty in eating and talking means that social interactions may be limited. In addition, the loss of saliva's antimicrobial functions lead to dental deterioration, which can be very hard to control.
The reality is we don't currently have treatment that works well for these patients. The drugs that can help in cases with less severe dry mouth are just not effective in this patient group. The only option may be gels and washes to reduce symptoms, which don't work well and provide minimal short-lived benefit at best. AAV2-hAQP1 is different because its actions aim to restore salivary gland function, addressing the cause of xerostomia, not just the symptoms. The phase I AQUAx data are really encouraging in terms of efficacy, safety, and tolerability, with transformative benefit observed in many of the treated patients. The recently released long-term data from AQUAx demonstrate that this level of benefit lasts for more than three years. The ongoing pivotal AQUAx2 trial, we mustn't forget, is completely blinded with one in three patients receiving placebo.
With that in mind, and not knowing who has received treatment and who has received placebo, some of our patients have reported very significant changes in symptom scores and saliva volumes. Importantly, the procedure of retrograde intraductal delivery of AAV2-AQP1 into the parotid gland is straightforward, minimally invasive, and quick. From the patient's perspective, tolerability has been excellent. The procedure involves minimal pain, with patients reporting that on a scale of one to 10, pain was somewhere between zero and one. For a group that has already been through significant treatment, that matters. It makes acceptance of the procedure much easier. From a clinician's point of view, the procedure fits easily into existing practice. The skill set is familiar to those of us working in this area, ENT surgeons, oral medicine specialists, maxillofacial surgeons, and interventional radiologists. With focused training, people can become competent quickly if they're not already so.
It doesn't require complex infrastructure or theater time, and I can do the procedure within my clinic in the clinic chair using only basic equipment I would expect to find in any standard outpatient clinic. It's also quick. Bilateral treatment can be performed in under an hour, which means it can realistically be incorporated into a normal working day without disrupting clinical flow. I could likely complete eight to 10 of these procedures in my clinic in a full day. Obviously, we need to wait for the unblinded results, but if AQUAx2 data is consistent with the phase I AQUAx data demonstrating AAV2-AQP1 is as effective and as tolerable as we believe it is, the practicality of a simple one-time treatment and ease of use in real clinical setting is very compelling.
When you put all this together, a simple one-time treatment, easily integrated into practice with high tolerability and durable effectiveness, it becomes something that is genuinely practical, not just in theory, but in everyday clinical work, and could become the standard of post-treatment care for survivors of head and neck cancer. Most importantly, this has the potential to make a real difference to patients who are currently living with this severe condition and have very limited options. For this group, as they will tell you themselves, that's long overdue.
Thank you, Dr. Owens. Good morning. I'm Dr. Michael Brennan, Chair of Oral Medicine and Oral Maxillofacial Surgery at Atrium Health Carolinas Medical Center in Charlotte, North Carolina. I'm an investigator in both the phase I AQUAx study as well as the pivotal AQUAx2 study. As a specialist in oral medicine, I care for individuals suffering from Grade 2 and Grade 3 radiation-induced xerostomia. During and shortly after radiation treatment for head and neck cancer, virtually all patients suffer from acute xerostomia, which is often severe. However, over the 12- to 18-month period following radiation, xerostomia resolves to a manageable level in 60%-70% of these patients. In the remaining 30%-40%, xerostomia persists as a severe lifelong condition for which there are no effective or tolerable treatments. This long-term persistent radiation-induced xerostomia is increasingly debilitating over time.
Patients may suffer from mouth sores and continual pain, oral infections, and uncontrolled dental caries. Difficulty swallowing and eating may cause weight loss and the potential need for a feeding tube. Dry mouth results in difficulty speaking, disrupted sleep, and even the inability to exercise, as faster breathing can lead to choking. Over time, these symptoms lead to nutritional challenges, frailty, social withdrawal, and a meaningful decline in overall health and well-being. For these 30% or more of head and neck cancer survivors, Grade 2 and Grade 3 radiation-induced persistent xerostomia is a lifelong, deeply burdensome consequence of curative radiation treatment. Despite its prevalence and its significant negative impact on so many areas of patients' overall health, there is no effective treatment, underscoring the severe unmet medical need.
In the open-label phase I study, I saw truly extraordinary life-changing improvements in many of my patients who were treated with the AAV2- AQP1. What makes this therapy particularly noteworthy is that it is designed to restore activity within the affected salivary gland epithelium to reestablish function at a biologic level. This represents a meaningful shift in therapeutic intent, one that aims to address the root cause of the condition rather than manage its downstream effects. The phase I study results I have observed in both the patient-reported outcome measures and the objective improvements of salivary flow are aligned with that ambition. The durability of these transformative changes in PROs and salivary flow over at least three years in the phase I patients is truly impressive. It reinforces my confidence and excitement in the pending data from the AQUAx2 pivotal study.
In many of my patients who have shown a positive response to treatment with AAV-AQP1, the level of improvement has been dramatic and frankly, unprecedented. They are regaining function that literally transforms their lives. They can eat comfortably, speak naturally, and participate more fully in professional and social settings. Many have shared that they feel a renewed sense of normalcy, something that had been lost following radiation treatment. From a clinical perspective, this kind of functional recovery is both uncommon and profound. It is also important to highlight the degree of these responses. The improvements I have observed are clear, consistent, and highly clinically meaningful. They are not subtle findings that require careful interpretation. The benefits are evident in both patient experience and objective measures. For example, in one of my patients who had been suffering with Grade 2/3 xerostomia for seven years, the response was incredible.
His overall xerostomia score showed a remarkable improvement in the first three months after treatment in many of the complications of xerostomia. Importantly, he reported to me that he was able to exercise for the first time since undergoing radiation therapy. Another of my severely impacted patients had difficulty speaking and eating normally. In addition to the overall health impact of the xerostomia, she had become severely socially isolated. Following treatment with AAV-AQP1 in the AQUAx study, the impressive improvements in her xerostomia allowed her to regain more normalcy in her life for the first time since radiation treatment for her cancer, meaningfully reducing her social isolation and improving her quality of life and overall health. I want to emphasize how important the three-year durability data from the phase I study are.
A key consideration is the persistence of the transformative benefits demonstrated at 12 months following the one-time treatment with AAV-AQP1. In a severe lifelong condition where permanent dysfunction is expected, more than three years of durability is particularly notable. From my perspective, the durable effects of treatment results from changes to the underlying biology of the condition itself, that is disease modification. The potential for an easy-to-administer one-time treatment to alter the course of radiation-induced xerostomia is especially compelling. Based on my firsthand clinical experience, treatment with AAV-AQP1 can have an unprecedented impact, radically altering the patient's life and well-being. A simple one-time treatment can have a dramatic disease-modifying effect on this otherwise untreatable, severe, lifelong condition. It is wonderful to be part of the development of such a potentially transformative treatment for this population of patients, and to witness its life-changing impact firsthand.
Thank you, Dr. Brennan. We conducted a global physician and payer survey using the three-year data in the target product profile of the therapy, which meaningfully improved the view of AAV-AQP1 for the treatment of persistent RIX in the eyes of both physicians and payers, compared to an earlier study using the 12-month phase I data. We looked at physician preference, potential commercial opportunity, pricing, and we also looked at the concentration of the market in the U.S. as we start thinking about launching this product. First of all, there was incredibly high physician preference. 70%-80%, wherever they were in the world, said that they would use this treatment. This is very high preference share globally. What were physicians most compelled by? The endpoints which we used are convincing and exactly what physicians want to see.
The effect sizes were unprecedented, very strong, and the durability to at least three years was very impressive, especially in the context of a one-time treatment with minimal safety concerns. It was important that the one-time treatment is quick and minimally invasive, and doctors can do it in their normal practice, as you heard from David Owens. You can see in some of the quotes from physicians here that long-lasting benefits of this single one-time treatment is very compelling. This therapy is now seen as a simple one-time treatment that results in a transformative disease-modifying effect on a severe, otherwise untreatable, lifelong condition. How does this convert into commercial opportunity? As we have seen, the physician preference is very high, around 78% globally based on the three-year durability data and one-time treatment.
However, we did not use that in our projections, rather the overstatement adjusted number of 52%, which has further decreased due to access and coverage and things like that. The number of prevalent patients today is large, and importantly for a one-time treatment, there is a significant annual incidence population. In the U.S., there are 165,000 patients existing today with around 20,000 new patients every year. Globally, it's a very big number, 435,000 with about 48,000 new patients a year. The three -year durability data also impacted payers and what they're willing to pay for therapy in the U.S. Previously, with only 12 months data, we had 90% coverage at $100,000 for a one-time treatment. Now we have a range of $100,000-$250,000. Here we're looking at $150,000, 15%-20% gross to net with 90% coverage, and it's a medical benefit.
With those numbers, you can get a global peak net revenue of $3.7 billion, reaching a steady state of about $3.2 billion because of that 480,000 new patients each year. In the U.S., the numbers I've gone through result in about $2 billion peak sales with a steady state of $1.8 billion, which is very impressive for a one-time treatment. These projections didn't come from treating every risk patient that exists. In these projections, around 32%-37% of available patients have been treated after 10 years. We think these are reasonable projections, and we're very excited to launch this product at the beginning of 2028, following, of course, positive pivotal data. We also learned something that was incredibly encouraging to us. We learned that over 60% of the U.S. population with this condition are within three hours' drive of only 15 metropolitan areas.
Within those areas, they're seen at 15 specific individual sites. These big centers are the usual suspects, and we had envisaged them being sites that we would go to anyway. We didn't quite understand the concentration of these patients to these large sites that treat over 60% of head and neck cancer. We feel confident that we can have a strong launch with little teams focused on each site that address every touch point for these patients. This may be the oncologist, the radiation oncologist, the ENT, the dental division, the nurse practitioner. It is different at every center. As there are only 15 of them, we feel that we can get to everyone who sees these patients. Finally, a summary of what I've shown today.
AAV-AQP1 is a one-time therapy delivered via a minimally invasive quick procedure which can be done as part of normal ENT clinical practice. Unprecedented improvements have been demonstrated in the XQ PRO and objective endpoint of unstimulated whole saliva flow in patients treated in the phase I study. Data from the long-term follow-up of all cohorts in the phase I shows durable and consistent responses in both PROs and saliva production out to three years. The consistency of the data over time gives us a lot of confidence in the outcome of our pivotal study, which uses as a primary XQ and as a secondary unstimulated whole salivary flow rate. Not only does it give us additional confidence, but the durability data had a meaningful impact on how physicians and payers see the product.
With the three-year data, the view of this potential therapy has changed from one that had a meaningful impact on xerostomia at 12 months to a product that is disease-modifying, life-changing, one-time simple treatment that transforms the life of these patients who've been cured of head and neck cancer, but without this therapy, are looking at the rest of their life suffering with persistent radiation-induced xerostomia and all of its debilitating consequences. This is a big unmet need and a large commercial opportunity with a very concentrated initial market in the U.S., with a significant incident population leading to substantial ongoing annual revenue of up to $2 billion in the U.S. alone. AAV-AQP1 has been granted orphan drug, RMAT, and breakthrough therapy designations by the FDA. We have written alignment with the FDA on clinical and CMC requirements for the BLA-supportive phase II study.
The data from the pivotal phase II study and the subsequent BLA filing are expected in Q2 2027, targeting an early 2028 launch in the U.S. With that, we will open the call up to questions.
Great. Thank you, Zandy. At this time, we'll be conducting a question and answer session with our speakers. Please hold for a brief moment while we poll for questions.
Our first question comes from Alec Stranahan at Bank of America. Please go ahead, Alec.
Hey, guys. Thanks for taking our questions, and congrats on all the updates today. Really fantastic to see the progress across the pipeline. I guess, two on xerostomia. You noted a consistent response with individual patients between the objective salivary flow and the subjective symptom scores. I guess, how do these findings strengthen your confidence in using XQ as the primary endpoint in the phase II? And then I've got a follow-up.
Thank you. Thank you for joining the call. I think one of the things that gave us great confidence is that if you look at individual patients, you see that whatever their response at 12 months, it was maintained largely all the way to three years. You can see that when you compare the 12-month water flow plot to the plot that has each time point for each individual patient. If you think about an XQ and you consider what may be a placebo response in a PRO, right, which is one of the concerns, if you have the same scores coming out of that PRO that is targeted specifically at xerostomia, and those scores are equivalent 12 months apart and over three years, it encourages you that those are real, actual responses, real changes in xerostomia that are not the result of a simple placebo response.
Okay. That makes sense. Zandy, I guess my follow-up is just around sort of the expectations heading into the phase II readout later this year, I guess. Based on the 12-month data you're seeing in the phase I, is sort of a 10-point or more mean change from baseline kind of the expectation in the phase II as well? I appreciate, I guess, you're using slightly different doses in the phase II. Thanks.
Okay. Thank you again. The method of dosing in the phase II is slightly different, you're right. However, we did four different dosing arms so that all of the actual doses that patients saw in the phase I are now covered in the phase III. We did that because we did not see a clear dose response to the way of dosing in the phase I. We wanted to cover everything that had given good responses in that phase I. Sorry, I've forgotten the beginning of the question. Can you say it again, Alec?
Yeah. Just around, I guess, the change in XQ you expect to see. I know it was kind of like.
Oh, the change. Yeah
10 points.
It was hard here because there's no evidence or understanding of what a placebo response would be in this treatment and in this sort of patient, right? There's very few studies and there's only one other drug approved. We actually based the phase III study on the data that we got from the phase I. We used the average of -17 as our expected change in the XQ. That was all patients treated, not just bilateral. Bilateral was -21, so that was high. We powered the study to be successful if we powered it for a 9-point change compared to placebo, and it would be successful with a 6.5 change. In that way, the placebo would have to be around or more than 10 points, and that would mean that the placebo alone would be transformative, right, in the XQ.
While we didn't know how to estimate the placebo, we used the data we had and our best understanding of the potential for the XQ to respond in powering the study.
Got it. Very clear. Thanks again for taking my questions.
Great. Thanks, Alec. Our next question comes from Yixin Zheng at Evercore. Please go ahead.
Yeah. Hi. Can you hear me?
Yes.
Yeah, great. Thanks, yeah, thanks for taking our questions. It's Yixin for Gavin. Yeah, really great update. I have a question about market opportunity. Out of those 20,000 U.S. newly diagnosed patients who are eligible to receiving the therapy, just wondering what percentage do you believe is immediately addressable based on your current payer and the physician checks? And secondly, I have a question about the saliva flow rate improvement. Just to help us contextualize this number, what magnitude of those improvements is considered as clinical meaningful? Thank you.
Okay. First of all, I'll do your second question, okay? Then we go back to the first question. Can you repeat the first question?
Yeah. First question about market opportunity, like what percentage are addressable?
Up to 20,000.
Yeah. Yeah, immediately addressable. Yeah.
Okay. To the first of your questions about the water flow and the good increase that we saw in water flow at both 12 months and 36 months. This is one of the reasons I emphasized that there is no direct correlation between xerostomia, which is the clinical measure, or sorry, the XQ, which is the patient-reported clinical measure of xerostomia. It is only a patient-reported condition. There is no direct correlation between the absolute amount of saliva and xerostomia. One person, as I said, with 100 mcL can have xerostomia, and another person with a mil can still have xerostomia. There is no clinically meaningful change in an absolute water flow or a percentage water flow. However, xerostomia is caused by too little saliva to make the mouth wet. It's a different amount for every patient, so there's no correlation.
However, it is related to water flow. The increase in water flow is the mechanism of action of our drug. The water flow data is not clinically meaningful. However, we have it as a key secondary because the FDA wanted us to show something that said that our drug was working via the mechanism that we suggested. Do you have anything more on that or is that okay? I'll then go on to the market question. You asked about the 20,000 new patients a year, and that 20,000 is the patients who are immediately available to the extent that they are all patients who have had radiation treatment in the previous or three or more years, before, and they all have moderate to severe Grade 2, 3 xerostomia.
They don't respond to pilocarpine or cevimeline or any other drug, and essentially those are the people who are immediately available for treatment. I would say probably more immediately available because they are more recently post-radiation treatment, and they are more frequently, arguably, seeing their physicians at the oncology center in order to be screened for recurrence of their cancer. that 20,000 is actually the number of patients for whom this drug is targeted.
Thank you.
Thank you for the questions. Our next question comes from Christopher Raymond at Raymond James. Please go ahead.
Hey, thanks. A couple questions, I guess. First, Zandy, I just wanted to zero in on those three patients that seemed to do worse at 36 months. Do you have any more color on, I guess, I know you mentioned there's no dose effect, but anything that happened between month 24 and 36 that you could sort of talk about or any conjecture on what happened there? Or did they get a lower dose? I have a follow-up.
Let me go back to the slide because I'm not sure that there were three patients that did worse. Sorry. Let me get the slides. Is this in the XQ?
Yes, I think it was the bilateral patients. Yeah, it was XQ.
Okay. Sorry. Can you
Zandy, what's the slide number? We can pull it up for you.
If you can-
I think it might be-
Chris?
31 maybe.
Okay. That would be great if you can go to the slide. Thank you.
Great. We'll pull it up.
Thank you. I'll start by saying, with respect to predicting who will respond and who won't respond or a dose-response, we looked in the phase I at many metrics, such as gland size, such as the actual titer that the patient saw, the total viral vector that the patient saw, and the time since radiation. We were unable to find anything that predicted the response.
Okay.
Okay.
It was slide 32, I think, actually. Yeah, I guess, well, really the more important question was maybe my second one is, you're going to have a variability, I guess, of sort of durability, and this might be maybe a cart before the horse question, but I know typically, AAV therapies are not redosable, but this is a low dose local delivery. Have you thought at all about the possibility for redosing?
Yes. Sorry, can we just go back to your other question?
Yeah, sure.
Okay. This is at 12 months. I don't have the slide.
Yeah. Sorry, that's not it. It's the waterfall. Yeah.
It's 12-month waterfall, and you're looking at the unilateral, the light pink population. Is that right? From my memory, the unilateral.
No, I'm talking about bilateral.
Okay. The 12-month waterfall plot, everyone gets a little bit better at 12 months.
Mm-hmm.
Later on, there's only one patient, which is the worst patient at 12 months, which is 009, who gets worse in the XQ.
Yeah. That's not the slide. It's the waterfall.
Okay. The waterfall, you have three patients that get worse.
Right. At month 36.
In the right-hand side, right?
Yes.
As I'm remembering, and they're unilateral. The one that is the worst remains a non-responder, but then the patients after that are interesting, because while they had a negative response, or they didn't get better, or a negative response at 12 months, what you can see in the waterfall plot of the unilaterals is those were the patients that didn't respond at 12 months, but then responded at 18 months. After responding at 18 months, they kept that response until the end of treatment, right? Those ones that went up in the light pink waterfall plot.
Yeah. I'm actually asking. Sorry, Zandy. I'm talking about the bilateral cohort-
Yeah
XQ.
Yeah.
There were three patients. One was marked, the other two were only minimally worse between month 24 and month 36. My original question was there anything that seems remarkable, right? If you see these patients doing better for two years, and my question is what might have happened at three years? That's it. It's the bilateral.
Yeah. Okay. We're looking now at the entire population, and we're looking at the bilateral, right?
Yep.
The worst patient in the bilateral, right, did get worse.
Mm-hmm
At the light pink line, which is at 36. Then you've got another patient who was the second-worst responder, and they are basically flat or a tiny bit worse.
Right.
That is just within variability.
Got it.
Now, I will say that the patients who didn't respond hovered up and down around baseline and tended to have the biggest variability, right?
Mm-hmm.
That was also true of middle responders. You can see in patient 024, for example, they went in the other direction. They actually improved more at 12 months. There's definitely some variability, and we don't know why certain patients respond at certain times. There is variability in this score. Overall, for an open label study and a three-year PRO, I would say this is generally very consistent.
Yep
Particularly at both ends of response, right?
Got it.
The really good and the really bad.
Great. Okay. Good. Maybe just the other question that I thought was.
Yeah
This is kind of out of nowhere, but have you thought about possibly redosing, just given the low dose local delivery, et cetera? I know that this is AAV gene therapy. There's the common wisdom that due to immune tolerance against the capsid, you can't redose, but this is kind of a different thing. Any thoughts there or is that just not possible?
No. It's a really good question. Yes, we think you can redose. We obviously haven't done it.
Right.
However, this is AAV2, and it is very local. It transduces just the cells that it touches. This doesn't really get very systemic, right? In some cases, you don't even get antibodies to AAV2. Even where there are preexisting antibodies, it doesn't change the response. This was also shown, by the way, for adeno in the first study ever of aquaporin-1. Patients who had already antibodies to the adenovirus, how they responded was not dependent on preexisting antibodies in the serum. It seems that the salivary gland is somewhat immune protected.
Just, I suppose anecdotally, we developed this program with the NIH, and at the NIH Dental Division, this was the work of John Chiorini, who is one of the original studiers of AAV, the discoverer of AAV5, et cetera, who has spent his career looking at AAV and response, immunological response, as well as clearing of the virus by the immune system. He mentioned that when I asked him this question, that all of the neutralizing antibodies he'd ever seen are IgG. The antibodies that are secreted into the duct of the salivary gland are IgA. He doesn't know if that's one of the reasons. There doesn't seem to be an effect of antibodies on transduction or function in the salivary gland, but it was his response to me when I asked that question.
Given that there's no difference if a patient already has preexisting antibodies, we think we could redose.
No, perfect. Thank you so much.
Great. Thank you for the questions. Our next question comes from Ashleigh Acker at Piper Sandler. Please go ahead, Ashleigh.
Morning. Can you guys hear me?
Yes.
Great. Well, I'm asking a question on behalf of Allison Bratzel. First, congratulations on the data today. Really impressive. From us, just two questions. We know that FDA has emphasized wanting to see a clinically meaningful PRO and also wanting to see an objective measure, which would be the saliva flow. We're just wondering if the phase II trial hits statistical significance on the primary XQ but misses on the secondary saliva flow rate, is a BLA path still viable, or would this change any regulatory interactions? Just the impact there. Second from us, you've also highlighted potentially entering into Sjögren's and radioligand-induced xerostomia. Are there any gating steps and timelines to getting these expansion indications into the clinic, or is the focus really on getting this into RIX first? Thank you.
Okay. Thank you for those questions. First of all, on the phase II pivotal study, right? There was a discussion within the company of the potential need to do a combined primary, and that would mean that we would have to hit both the XQ and water flow. I think that pilocarpine, they were supposed to hit a combined PRO. It wasn't XQ and water flow. I'm not sure if it could be either/or, but they only hit the PRO and not water flow, and they got approval. Now, in our case, we very clearly have advice that the primary is purely the XQ and the secondary, a secondary, the key secondary we've made it is water flow. Yes, if we hit on the XQ and we have a change or something non-statistical in water flow, this is still an approval-supporting study.
Is that okay for your question?
Yes.
Okay.
That's great. Thank you.
Okay. With regards to Sjögren's and PSMA radioligands, so yeah, you're right. Both of these forms of xerostomia, you can potentially treat with exactly the same product, AAV-AQP1, which obviously, when this is approved, if this is approved, will be manufactured as a commercial product. It would be a label expansion. Yes, the preclinical data for Sjögren's is very clear, and it seems to have a really good response in mouse models of Sjögren's in restoring salivary function and also actually having an impact on the systemic immune system. It is a really good second indication, and it is something we will continue to look at, and as we get further on in the RIX study and our efforts for approval there, we will probably look next at Sjögren's.
With respect to PSMA, this is very interesting because we did ask doctors about this in one of our surveys, and yes, xerostomia is one of the dose-limiting side effects in the use of PSMA radioligands. There was an interest from those physicians who obviously don't have the same degree of understanding of radiation-induced xerostomia as the traditional head and neck treaters who've used radiation for a long time, and there was a big interest in anything that could prevent the severity of xerostomia during treatment. We do have some data that is preclinical that suggests that pretreatment with AAV-AQP1 could prevent or reduce the later appearance of radiation-induced xerostomia and potentially reduce that dose-limiting side effect, which would be really helpful.
That's, again, preclinical data, and it's a very interesting opportunity to think about, and we do have interest from that group of physicians in exactly that.
That's great. Yeah. Thank you.
Thanks for the questions, Ashleigh. Our next question comes from Elemer Piros at Lucid Capital Markets. Please go ahead.
Yes, good morning. Zandy, how do we know that these patients who enter this study are at their steady state, meaning they are not prone to spontaneous improvement?
Okay. Thank you for that, Elemer. We are treating only persistent Grade 2/3 moderate to severe xerostomia, and something is defined as persistent if it's in our hands, if it's two to three years after radiation treatment. In our phase II study, the pivotal study, we actually had a three-year limit before patients could come into the study. Now, when you have radiation, everyone who gets radiation therapy for head and neck gets xerostomia acutely, and it can be very severe. Of those patients, about 60%-70% resolve their xerostomia over 12-18 months.
Wow.
Right?
Yep.
Once you get 18 months to two years, and even to three years, xerostomia just becomes persistent. We're treating those people with persistent xerostomia, which only goes down from, let's say, two years after radiation treatment. It is persistent Grade 2/3 because they can't respond to any of the available therapies. We only treat patients at a time when resolution just isn't part of the natural history. It just isn't seen. That's why we don't think we'll have spontaneous resolution in these patients.
Yeah. Thank you for that. I think previously, you also looked at another instrument, the Global Rating of Change questionnaire.
Yeah.
Have you looked at that, and have you seen any corroboration with the XQ? Will you also use this as a secondary endpoint in the pivotal study?
That and a single question are in the pivotal study. The McMaster is not xerostomia specific.
Okay.
It is, do you have a disorder? Has your disorder changed? It could be Parkinson's, it could be whatever it is. Has your disorder changed? If it has changed, is it better or worse, by how many points, right?
Mm-hmm.
Yes, we did that. Actually, I'm not sure about the three-year data, but it's a PRO that doesn't, as reliably as XQ, demonstrate real actual xerostomia. It's general
Mm
feelings of your symptoms of the disease, whatever that may be. Yes, it is a secondary. Yes, it did correlate in the phase I with the XQ, as did the single question, I think the dry mouth question. Yes, I seem to remember that dry mouth was very closely correlated to xerostomia score, which makes sense.
Yeah. Thank you so much.
Great. Thanks for the questions, Elemer. Our next question comes from Daniil Gataullin at Chardan. Please go ahead.
Yeah. Hey, good morning, guys. Thanks for the questions and congrats on all the progress announced this morning. I have a question. With respect to durability for patients who underwent several biopsies, how did the vector copy number change over time?
Okay. The biopsies were in an NIH study that is still going on of the AAV AQP1 vector. It's actually the same batch. A few patients agreed to one biopsy, right? There are no serial biopsies over time. It's actually quite difficult to get xerostomia patients to have biopsies of their minimally existing salivary glands. The data on biopsies is a one-time biopsy, and most of that is just looking for the DNA of the vector, and one of them, there was enough tissue to see protein, and that was between one and three years post AAV- AQP1 treatment.
With respect to durability of the vector, these cells that are not dividing, the durability that we've shown as well as durability that has been shown for the original adeno study, which was out to five years functionally in very few patients, but the study was supposed to be 45 days, right? Because it was so immunogenic. The functional durability seems to be a consequence of the DNA and the protein still being present in the gland. It's very hard to get these patients to do more biopsies than they've done already.
Okay. Got it. That makes sense. Another question with respect to variability. You mentioned hydration levels could play a role, and you're accounting for it now. What are some of the other factors that could result in variability in response, and how are you managing those?
There are many factors that lead to people's variable responses to PROs and many factors that lead to variation in water flow over time, right? Emotion, for example. If we knew precisely what those were, we would try and control for them. However, other than everyone being trained in how to deliver a score, everyone being trained in how to collect saliva and hydration, from a practical point of view, it's very difficult to specifically address the variability in PRO or in water flow. What I will say, because we haven't been able to identify, in general, a particular feature that can predict a good response or a bad response, is a lot of the variability may be due to the structure of the gland, and this is something that we don't necessarily know.
Remember, these patients' glands have all received side effect radiation, so they're each going to get a different amount of radiation, and in some cases, different parts of the glands will be destroyed, and they can be destroyed to different extents. I'm guessing, and we think that the variability in the actual structure of the glands between different patients is just something that leads to the variability in response. The long-term data, we think, showed us that the XQ, if it does respond well at a particular time point, it will continue to respond well. The inter-patient variability is something that we have seen is good in very good responders and also is probably true in those who don't respond.
We've conducted the study powered in a way to take into account variability and conducted it with scores and training that we hope in any pivotal study will keep that variability in the range.
Got it. That makes sense. Thanks again for taking my questions.
Thank you for joining.
Yes, thank you for the questions, Daniil. Our final question comes from Lisa Walter at RBC Capital Markets. Please go ahead, Lisa.
Oh, good morning. Thanks for taking our question, and congrats on the progress. Zandy, maybe just wondering if you can comment on patient discontinuations. It looks like patient 18 and patient 13 do not have data past 12 months. Just wondering if these patients discontinued due to an adverse event or another personal reason. Any color here would be helpful.
Okay. Thank you, Lisa. Well, number one, we really didn't see any adverse events that were of particular clinical importance. As I mentioned in this study, there were six treatment emergent, treatment-related, mild AEs, and they all resolved without sequelae. From the perspective of safety, this was very safe. Of course, it's a one-time treatment, so you don't have ongoing toxicity that in some drugs can make the patient withdraw after a certain amount of time when they just don't want that anymore. They have a one-time treatment. They now have that for the rest of their life in this case. We think probably people just didn't turn up for visits after the 12-month. It was in a long-term follow-up, and you can see that at 18 months, for some reason, there weren't as many people that came in.
Many of them came back, right, at 24 months. There can be personal reasons. There can just be motivation reasons for why people don't come to particular visits. I don't think it's anything to do with a negative feeling about the treatment.
Got it. Thanks, Zandy. Maybe just one last one on your regulatory plans here. I know it's still early, and of course, we have to wait for the pivotal results to come out. I'm just curious if your plan is to file a BLA for both the high and the low doses, or if you're thinking of just moving forward with the dose level that has the best benefit risk profile.
With respect to dosing, as I've mentioned, we haven't seen any dose responses when we look at vector genome per patient, titer that the patient sees, gland size, anything like that. That's the reason we did four doses in our pivotal study. We're powered such that the study is successful if any one of those doses improves in the XQ compared to the placebo, right? It doesn't matter which dose that is. Now, I personally think that they're most likely to all respond in a similar way based on previous data. If that happens in the pivotal study, what normally occurs is that you file the data, and you get approval for the lowest dose that works. In that case, that's what we would do. I think it's normal practice when you have a multiple dose phase III or pivotal.
Got it. Thanks so much, Zandy, and congrats again on the progress.
Thank you, Lisa.
Yes. Thank you for the questions, Lisa. This concludes our Q&A session for today and also our event. We thank everyone for joining. You may now disconnect.