Welcome to the UBS virtual event. I would now like to pass the call over to Eliana Merle.
Hey, guys. Thanks for joining us. Happy to have Opthea out here with us today for a fireside chat. Joining us from Opthea is Chief Executive Officer Fréd Guerard. Thanks so much for making the time. Maybe just to start off, could you talk a bit high level about your mechanism and approach in wet AMD?
Yes, certainly. And thank you very much, Eli, for having us today present the story here. So, we're working on a molecule called sozinibercept, and sozinibercept basically work by providing a broader inhibition of the VEGF pathway for patients suffering from wet age-related macular degeneration or wet AMD. A number of drugs already have been, you know, launched twenty years ago are currently on the market. But all these drugs currently available to patients to treat wet AMD are all blocking VEGF-A, VEGF-B and PlGF, but they leave wide open the receptors for activation by VEGF-C and VEGF-D.
And what we have achieved with sozinibercept is to design a trap, which is very similar to the structure of aflibercept, which is designed to block VEGF-A, except sozinibercept targets VEGF-C and VEGF-D. And basically by combining these two mechanism of action, so baseline standard of care, anti-VEGF-A, and sozinibercept, we have demonstrated in a well powered large phase II-B trial that this combination of sozinibercept with standard of care provides a better visual outcome for patients suffering from wet AMD. So that's the mechanism of action. It's not a new concept.
It's been well described over the last twenty years in numerous animal models, and you can see the effect of VEGF-C and VEGF-D in the retina is very similar to what VEGF-A does in wet AMD, which is promote this abnormal growth of blood vessels in the back of the eye and promote vascular permeability, resulting in fluid accumulation in the retina, which over time, if not treated properly, can result in blindness and death of the photoreceptors. We now know as well that patients suffering from the disease have this upregulation of all forms of VEGF, but especially VEGF-C and VEGF-D. Then when the patients get treated with an anti-VEGF-A, which is the treatment they receive today in most countries around the world, you have a further upregulation of that VEGF-C expression.
So, blocking, basically, this whole pathway makes a lot of sense, and that's what we are doing with sozinibercept.
Great. Makes sense. And, you know, as we head towards phase III readouts next year, particularly as you're dosing in combination with standard of care, maybe just high level, where do you see this fitting in terms of the treatment landscape, and how could this, you know, change some of the, you know, treatment paradigms that we're seeing today commercially?
Yeah, it's a very interesting time and for patients, because for a very long time, patients only had biologics, anti-VEGF-A at their disposals. And when I say patients, it's obviously retina specialists and patients. And it was basically one class of drug, and none of them have demonstrated superior efficacy to the original benchmark, which was Lucentis, launched almost 20 years ago. So these drugs, some of them last a bit longer in clinical practice, and we see now data coming from real-world evidence showing that Vabysmo, for instance, lasts around 10 days longer than Eylea, which and Vabysmo in the real practice in the U.S. lasts around 54 days in. That's well published now, a couple of years after it was made available to retina specialists.
So this whole class of current treatment is basically one class of drug, all doing the same thing, all having the same efficacy on the vision for the patients, which is what patients really care about. Now, there is upcoming new technologies coming. We have small molecule approaches on, different delivery platforms with tyrosine kinase inhibitors, and these are quite interesting because they could potentially block more than VEGF-A. But of course, they are small molecules, and they have a different potency and bioavailability compared to, very powerful biologics like, you know, Eylea or Lucentis or Vabysmo. And finally, gene therapies are emerging, and some of them express, well-known anti-VEGF-A. Some of them, actually have, RNAi VEGF-C, expressed as well in their, in their cassettes, but they are more early stage.
Obviously, the risk-benefit for a gene therapy is very different from a biological small molecule that gets eliminated over time. Gene therapies cannot be switched off, and therefore, if there is a safety signal in the eye, it's actually much harder to control than other molecules. As you can see, the landscape may be evolving if some of these new emerging technologies are successful in showing non-inferiority to standard of care. All of them are developed as non-inferior to standard of care. We are the only program going for additional efficacy on vision outcomes for the patients. I think that is something that needs to be well understood. All the others are trying to be non-inferior to existing treatments, which are broadly now becoming generic, and biosimilars are already available in the U.S.
... Mm-hmm. Understood. That's helpful context, and maybe if you could tell us a bit about some of the data that you've reported to date, and the designs of your phase II program.
Yeah, no, certainly so the phase II-B trial was very well done. It was a large trial of over three hundred and sixty patients randomized into three arms. All the patients were getting the standard of care as per label, so Lucentis every four weeks for six months, which is how Lucentis is indicated to be used, and then patients were either receiving a sham injection or sozinibercept low dose or sozinibercept high dose, 2 mg. So it was around a hundred and twenty patients per arm. The inclusion/exclusion criteria were very standard for this type of trial, with one exception, which is the baseline, the best-corrected visual acuity.
In this trial, in order to be able to show an additional effect and the superiority of sozinibercept added to standard of care, patients were selected with a lower vision to start from, and this is simply to be able to show a separation of the curve. If you take patients that have a perfect vision to start from, obviously, you cannot show superior vision in this trial. That's why the inclusion criteria were designed this way, and they are the same in our phase III, with one exception, which I will show later. The primary endpoint, it was a 24-week trial, six-month.
The primary endpoint was the same endpoint as we have in phase III, which is the mean change from baseline in best-corrected visual acuity, which is a validated endpoint for pivotal programs, except the pivotal programs require that endpoint to be later. In our case, it will be at one year. The phase II-B had also secondary endpoints, very traditional functional and anatomical endpoints that we are very common in these wet AMD trials. And finally, it was. The thinking about the lesion type has been around for quite some time, and from the Lucentis days, it was very well known that some patients with some type of lesions are harder to treat than others.
So these subgroups of different lesions in the back of the eye were actually pre-specified in the phase II-B to allow to see where is the combination benefiting the patients the most. And we have the answer, and that's why that has been reflected in the design of our phase III program. Maybe baseline characteristics, so patients were very standard. They were late in their seventies. It was mainly carried out in Caucasian population, so we have a bit more diversity in our pivotal program. But you could see here the mean baseline BCVA was around 50 letters. So that was as per design and inclusion criteria. These patients had active disease. You can see that they had elevated central subfield thickness, which is a marker, a biomarker of disease activity.
Normally, patients should be around 300 micrometers or a little bit less. You can see here, they were over 400 micrometers, and they had presence of fluid in different compartments of the retina, which is also a biomarker of the disease. The primary endpoint was on all comers. So you could see here that there was a clear superiority of the high-dose sozinibercept in combination to standard of care compared to Lucentis alone, and the low-dose sozinibercept did not demonstrate additional efficacy on vision. It did demonstrate additional efficacy on anatomy, but not on visual acuity.
More interestingly, in the patients we knew were harder to treat with standard of care, anti-VEGF-A, we had actually a bigger improvement of the number of letter gain from baseline, in the combination arm with a high dose sozinibercept compared to a standard of care alone. As you can see here, we had 5.7-letter increase versus standard of care. It's an incremental 5.7 letters on top of 10 letters of increase of vision in this wet AMD patient with a very good p-value. This is the on the graph here, you have the patient population we're going to study first in our phase III trial.
We are enrolling all comers except RAP patients, and simply because these RAP patients did not benefit a lot from the additional injection of sozinibercept, and therefore, they're excluded from the phase III trial. But otherwise, we enroll all comers in the phase III trial, but we focus the primary statistical analysis on that exact patient population you have on the slide here. That patient population is around 75% of wet AMD patients in clinical practice around the world, so that's a vast majority of wet AMD patients. Moving to, you know, a very visual representation of what the combination with sozinibercept 2 mg did to the patient population. You can see here we color-coded.
Every bar is a patient, and we color-coded their baseline best-corrected visual acuity improvement over baseline on different colors. So every yellow was a loss of vision of less than a line, less than five letters. Orange was five to nine letters, and then you can see red was ten to fifteen or more than fifteen letter loss. And the green, green was the vision gain. So you could see clearly the whole patient population in the combination arm with a high-dose sozinibercept, basically moving to the right, moving to vision gain, with a lot less patient having vision loss and a lot more patients gaining extreme level of visions, six lines or more of vision on the reading chart.
So you can see here very visually the benefit of adding our drug to a standard of care. It's interesting to see so many patients on the control arm losing vision despite getting Lucentis every four weeks, which is, as of today, the unbeaten benchmark in terms of efficacy in wet AMD. And yet you can see here, the addition of sozinibercept 2 mg delivered a lot better in terms of reduction of vision loss and extreme vision gain. Another interesting outcome is how many patients were getting to 20/40 vision or better. 20/40 is the minimum legal threshold you need to have in at least one eye to be able to drive in the United States and many other countries.
You can see here, basically, adding sozinibercept increased the proportion of patients being able to drive again with that eye that was treated, from a third of patients with standard of care alone to half of patients with a combination with sozinibercept 2 mg. That is a massive impact on quality of life of these patients. It means they can not only drive, but to perform all sort of other tasks that are absolutely required to be able to live independently and not depend on other people. This is clearly a very interesting patient benefit that was observed in the trial.
Moving to anatomy, there is no very perfect biomarker that allows to extrapolate anatomy into vision of the patients, but what we have here is a combination treatment arm, which showed consistent improvement of the combination over a standard of care alone in all anatomical parameters, whether it was the reduction in central thickness, so basically the drying of the retina at this particular point where it is measured, the presence of subretinal fluid, the presence of intraretinal cyst, the lesion area reduction, as well as the choroidal neovascularization area reduction. All biomarkers of better disease control with the combination as opposed to standard of care alone, resulting in that superior visual outcomes demonstrated with the combination of sozinibercept.
Maybe one patient population interesting is this very hard to treat PCV patients, so these polypoidal choroidal vasculopathy patients are very common in Asia, but also not rare in Caucasians. We had around 20% of our patients in our clinical program in phase II-B had these lesion types. You can see here, standard of care, instead of delivering the 10-letter vision gain, in these harder-to-treat patients, there were around 20 of them per arm. You can see that they only got a 7-letter improvement with Lucentis alone, but they got almost 14 letters improvement with the combination of sozinibercept 2 milligram and ranibizumab. This is really interesting because these patients normally still get laser in some instances because they are very hard to treat, and sometimes anti-VEGF-A are not enough to control them.
So we have these patients enrolled in the phase III, and we will be looking at that subgroup as well in our phase III trials. And then finally, on safety, nothing remarkable on safety. Very similar safety profile of sozinibercept and anti-VEGF-A injections, even on inflammation, which in the eye is always the thing we look at. You can see here the rates of inflammation were very similar across the three-arm, and no remarkable difference between sozinibercept and standard of care.
Great. That's a, that's a helpful overview of the data. So with the phase III ongoing, you have the COAST and the ShORe pivotal phase III trials. Can you walk us through the designs of these?
Yes. So we embark into this very large pivotal program, which is made of two different trials with similar design, but using two different drugs as a comparator. The first one is COAST. Both are around a thousand patients each. They are fully enrolled, and we expect the data next year. I'll cover the details a bit later. Basically, patients are randomized one to one to one into three arms. They get, in COAST, Eylea 2 mg as a standard of care injected as per label in all arms. The patients get every four-week injections for the first three injections, and then every eight-week injections of Eylea, which is the label.
And they either get a sham injection, or they get a sozinibercept injection every four weeks, or they get a sozinibercept injection every eight weeks after the loading phase. And the idea is to show two things. One is to confirm the superiority on efficacy, which is visual outcomes, BCVA improvement versus baseline, for patients benefiting from sozinibercept 2 mg, either every four or every eight weeks, and then have an idea of duration of treatment by testing two different injection frequencies for sozinibercept, either every four or eight weeks. The primary endpoint is for both trials at week 52, and we expect the top-line data of COAST in the second quarter of calendar year 2025.
ShORe is the same concept, except the comparator is Lucentis, injected as per label every four weeks for two years. We also have the primary endpoint at week 52. It is expected a bit later for ShORe. We expect it in the middle of next year, and I forgot to mention these trials are two years, so the second year of data is mainly captured for safety purposes, but we can file to the FDA with the first year of data and then have a supplementary filing with the second year of safety data during the review process.
Great. I'm sorry, did you have more?
Maybe the primary endpoint, so the idea was to change as little as possible from phase II to phase III, so the patients' inclusion and exclusion criteria will be very similar. The exception is we excluded, as I said before, the RAP patients from phase III to maximize the probability of success of the clinical trial. The primary endpoint is the same as in phase II-B, mean change from baseline in BCVA, except it's at week 52, and then the secondary endpoints are a mix of functional and anatomical endpoints, which are very standard in this disease.
And I guess from a statistical perspective, what's the threshold that you need for statistical success? And just if you can walk us through the design there in terms of the stats testing. And then from a commercial perspective, what do you view as sort of the bar in terms of what's clinically meaningful as an improvement?
Yeah. There is no superiority margin defined by the FDA because nobody has ever shown superiority compared to standard of care in that disease. As much as all the other programs need to show non-inferiority to standard of care, and that non-inferiority margin is very well defined by the FDA, we need to show a p-value here and show superiority, statistical superiority. Because of the power of the trials, which are very large, as you can see, over 300 patients per arm, we can detect less than three letters of difference between the arms, between the treatment arms and the combo arms and the standard of care.
So, the idea, obviously, we look at the totality of the evidence, we look at the p-values, we look at the number of letters we demonstrate over standard of care. We also look at safety, obviously, and make a judgment on the totality of this evidence for us to be approved.
Now, on the clinical meaningfulness and how retina specialists think about what we need to demonstrate to be used basically in clinical practice, we ran that survey in the summer of this year, and we asked 125 retina specialists in the U.S., what would be their uptake of a second injection of an anti-VEGF -C and -D, which is obviously sozinibercept, used on the same day as an anti-VEGF-A standard of care, based on how many letters of incremental vision we deliver versus the standard of care. You can see we tested less than three letters. We tested three to four, four to five, and over five letters of improvement over standard of care. You can see the answer. It was a mix of retina specialists.
Half of them were considered to be high prescribers, the other half were lower prescribers, so it's kind of a blend of different clinical practices. But you can see it ranges from around 24% of patients that would be offered that second injection if the improvement would be under 3 letters. But it's over 40% if the improvement is over 5 letters. So it gives some sort of a range of how physicians look at the clinical meaningfulness of these different increments of efficacy over standard of care. What's very interesting, and that this is quite remarkable, this market is around $10 billion a year. The wet AMD market alone is around $10 billion a year of potential sales around the world.
A point of market share is around $100 million a year. So it's a very significant opportunity for every point of share we can gain here.
Mm-hmm. Great, makes sense. And just in terms of the label, since you're studying versus Lucentis and versus Eylea, I guess, do you expect to have sort of a VEGF-A agnostic label in terms of combination? And I guess just your any color from the regulatory interactions around that.
Yes. So all drugs on the market today being all anti-VEGF-A, that's the discussion that took place before initiating the pivotal program with the FDA, is running these trials with using two different comparators would allow us to have a class effect. Basically, the label could be something along the lines of, you know, sozinibercept in combination with any anti-VEGF-A demonstrated superior efficacy in wet AMD or something along these lines. So the idea would be not to be niched with one particular molecule, but basically using this pivotal program design to be offered the opportunity to be combined with all drugs on the market today. And the market is becoming a lot busier in this baseline standard of care treatment family, simply because of biosimilars coming as well on the market.
The idea here is to have the broadest label possible.
... Makes sense. And I guess just in practice, how do you think this will be used in terms of the dosing frequency, particularly as, you know, there's a lot of treat and extend currently being used? Do you expect that this is what would be done, as well with your drug, and do you have any studies that have looked at that?
Our trials are done on a fixed regimen. We do not have a treat and extend arm. We are keeping these dosage frequencies you have here on this slide for the two years. We're not gonna have that data at pivotal program readout. But we know from experience that retina specialists do indeed treat and extend. They individualize the treatment based on the patient needs. The way we've been told retina specialists would use our drug is basically first on naive patients, because we're gonna get the data. All this data is generated on naive patients, so we would certainly become somewhat the standard of care in combination from day one for these patients that are newly diagnosed.
But there is a large number of patients currently under treatment who are not completely dry, and actually so dry or even have not recovered a significant vision. There is still a large unmet medical need, which is very, very well known. You know, half of patients or most do not achieve significant vision gains. They still have fluid in the back of the eye. The majority of patients do not go back to driving level of vision, as you've seen before, even in the phase II trial, we've seen that. So, we believe our drug will be used on these existing patients as well, added to the anti-VEGF-A on the day of injection of the anti-VEGF-A.
Patients, instead of coming for one injection, they would be coming for two injections, but not necessarily asking the patients to come back another time. It should not be a patient burden. Patients are already in the chair. They're already... Their eyes are already prepared for the injection. That just takes another thirty seconds to inject a second drug. The volume of injection we will do between the anti-VEGF-A and sozinibercept is similar to the approved volume of injection of drugs in geographic atrophy. It is now well known that this volume is actually safe, and because it's been approved by the FDA. We should not have volume concerns during the injection for these patients.
Makes sense. And what was your rationale for studying in naive patients versus in, say, experienced patients who have an inadequate response, given your ability to improve efficacy?
It's a very good question, so you remove a lot of variability going into naive patients, and this is why most, if not all, clinical trials in wet AMD are done on naive patients. Because patients that have experience already treatment will come with very big differences. You know, they could have been diagnosed five years ago or two years ago. They could have received fifty injections of anti-VEGF-A or only five, so there is a huge variability in these patients that are pre-treated, and we know some of them, because of the suboptimal efficacy of anti-VEGF-A, may already have developed some sort of fibrosis or some sort of resistance to anti-VEGF-A, so you introduce, by using non-naive patients, you introduce a lot of variability, which is not, for us, suitable for getting the drug approved.
But indeed, those are questions we would want to answer in subsequent clinical trials when the drug is approved. For naive patients, we would absolutely run studies to see what benefits patients that are resisting to anti-VEGF-A or have been receiving anti-VEGF-A, but have not benefited from these injections very much. How do these patients do when they are added sozinibercept on top of their standard of care?
Mm-hmm. And so your expectation for the initial label would be indicated for previously untreated patients?
It would be probably for naive.
Right.
But, as insurance companies have never limited the use of these drugs in the past, despite the trials being done all on naive patients. So all drugs on the market today have been trialed in naive patients for their pivotal profile. So we expect the exact same thing for sozinibercept.
Mm-hmm. And last question, for you, since we're heading into phase III data and potential, you know, regulatory filings and then, you know, commercial launch, I guess, where do you stand in, you know, your commercial preparations, and also, where are you from a cash runway perspective?
Yes. So we are starting to build a commercial team. We had one person for a very long time. We are adding market access and marketing experts to the team. So now we have what we need to be prepared before top-line data, and of course, we will need to expand that team after top-line data. So we're in a good place. We are doing the normal prep work in terms of talking to insurance companies and payers around the world to make sure they're all aware the data is coming, and that they should expect us to come back to them asking for them to cover this drug.
On the cash runway, so we have a very good cash balance at the end of June, which is the end of the financial year in Australia. We had $207 million US dollars pro forma, so that includes a second tranche of a financing we did over the summer. And that is enough cash for us to sustain our operations into the third quarter of next year, so beyond the two top-line data readouts.
Great. Well, Fred, thank you so much for joining us today, and we look forward to seeing the phase III readouts.
Thank you very much, Eli.
Yeah. Thank you.
Have a great night.
Bye.
Take care.