Circio Holding ASA (CRNA) Earnings Call Transcript

Good morning, everyone, and welcome to Circio and our fourth quarter R&D and corporate update webcast. My name is Erik Digman Wiklund. I am the CEO of Circio. And with me today, I have Dr. Thomas Hansen, our Chief Technology Officer and responsible for building our unique and powerful circular RNA technology. And as a reminder, we have a long experience in this field and been involved with it since the very beginning, and we have published a number of scientific articles in this space. And this also positions us well to build and excel in the area of circular RNA. So let me start by explaining the fundamental aspect of what we do at Circio. We have built and developed a novel and powerful alternative to the main dogma of molecular biology. The main dogma of molecular biology is that DNA, the genetic code forms messenger RNA and that messenger RNA encodes a protein, DNA to RNA to protein. What we do at Circio is that we reengineered the DNA so that it makes a circular RNA on the way to make the protein. And this is a platform technology that you can use for any vector-based therapeutic. And importantly, it enhances and prolongs the expression of the protein. This is the key advantage of what we do, and we do not know of anyone else that can do this as well as we can do at Circio. So we have a unique expertise and IP and know-how in this particular area. So why does that matter? So that's shown on this slide here. We can get enhanced expression and improve durability. If you look at the top part of the slide, we're comparing to AAVs using circular RNA or mRNA benchmarks. And you can clearly see that the circVec, the circular RNA-based AAV dramatically improves on the protein expression up to 40-fold in AAVs that we target to the heart. What we believe we can achieve here is enhanced, safer and lower-cost AAV gene therapy. The improved durability, we are mainly harnessing in the format of DNA. So in this case, we're delivering synthetic DNA vectors that we normally formulate in a delivery system such as an LNP. And in the lower panel, we're comparing head-to-head mRNA versus circVec DNA and you can see here, we give a dramatically improved expression. We've shown more than 6 months durability now on the circRNA-based DNA vector in vivo versus less than 3 weeks per mRNA, a dramatically improved expression window. This we are now moving forward in vivo cell therapy, and we'll come back to that later. So to start on the AAV side. Our data suggests that we can fix maybe the biggest problems that are with AAV gene therapy today. I point here a particular case in a heart disease because heart is where we've shown the biggest advantage yet of circVec AAV. And there's a program in Danon disease that an early trials showed a clear clinical benefit. And this is a genetic disease which is devastating, and there is no therapeutic therapy available for these patients. So the therapy with AAV works. However, you need to give so high doses that it becomes too toxic. And unfortunately, there was a patient that died from treatment with this lead program, which is in the Phase II trial earlier this year. So the dose is what drives up the toxicity and the cost. What we can do with circVec is that we simply exchange the classic AAV to a circular RNA expressing AAV, and this will allow significant AAV reductions with the same clinical benefit. This should lead to better, safer and lower-cost AAV gene therapy. And this particular disease, Thomas will come back to, and this is what we have selected as our lead initial program. So here is our pipeline. We have our AAV platform. This is the most advanced technology, the most straightforward to bring towards the clinic. And then in parallel, we work on in vivo cell therapy with an emerging area of very substantial pharma interest. And here, we're targeting in vivo cell therapy for oncology. This is at a somewhat earlier stage, but we have important upcoming milestones. And then you can see on the right-hand side here that we have a clear program, clear plans and multiple important milestones that we will achieve in the next 6 to 9 months. So before I hand you over to Thomas for the technology updates, I'd like to quickly summarize what we're going to show you today. We have established a novel circVec generation 4. Generation 4 adds 50% boost versus generation 3. And I'll remind you all that generation 3 is already highly optimized. So the 50% booster is meaningful in absolute terms. We have also replicated, strengthened and broaden our data package for our circVec AAV platform, particularly in heart. That's where we have advanced the furthest. But we also have strong data in eye and we are pleased today to present your data showing that we also deliver an advantage in brain. For the LNP circVec for in vivo cell therapy, we now have established that we achieve expression for at least 6 months on one single dose. And I'm also very pleased to announce last week that we announced that we have entered our first fully funded R&D collaboration with a major pharma. And this is in the area of circVec AAV for a specific indication. So with that, I hand over to Thomas. Maybe before we move on to Thomas, I will add that we have received more questions than usual from our shareholders and investors. This is great. We've decided to try and cover this throughout the presentation rather than in a dedicated Q&A afterwards. So as much as we can, you will get responses throughout. It's still possible to submit questions in the Q&A if there are additional points you would like us to consider. So with that, over to you, Thomas.

Thank you, Erik. I hope you can all here me, and good morning. I'm pleased to see so many dialing in here at our R&D update with this webcast. So I will take you through our most recent platform development, what is sort of the critical features for the circVec 4.0 and how it performs. And then I'll spend most time on showcasing our most recent AAV gene therapy data in heart, eye and in the CNS. But first, just to make sure we're all on the same page. As Erik said, what we are doing at Circio, we are developing DNA cassettes that inside cells will express a circular RNA. The circular RNA is extremely stable. It's resistant to the normal decay pathways inside cells. So you can see this little Pac-Man schematics that are normally chewing up linear mRNA is unable to do anything with the circular RNA. So in turn, it stays in the cell for quite a long time. And as a consequence, you have a lot of protein being expressed. So how you can actually see this is like similar to the financial situation, if you earn a lot of money, but you don't spend any money, you'll increase your bank account, and this is the same what happens on the RNA. We're basically producing a lot of RNA. It's not being turned over, so it will accumulate inside cells. So you get a lot of RNA inside cells. And as a consequence, you get this higher protein expression. So that's the core aspect of the circVec technology, namely harnessing the longer half-life versus circular RNA, whereas the mRNA is intrinsically unstable. So it has to be replenished all the time. So that just means that you go shopping all the time, so you don't have that much money left sort of at the end of the month, basically. So that's the same for the mRNA. So we need to produce more mRNA to keep it running. So this can also be seen here actually based on some of our earlier in vivo data, where we actually try to measure the half-life of the RNA that's produced from our circVec vectors compared to linear normal mRNA. And what we observed here is that you go from 8-hour half-life to more than 600 hours, which is 75x improvement in half life. And that actually also translates to a significantly different profile of expression over time. So what you can see, and this is basically how much protein is being expressed from circVec compared to mVec. So we have circVec in purple and mVec in yellow. And what you see actually at the very early time point, if you measure immediately after you administer the DNA to cells, you may actually find higher expression from circVec. But that -- sorry, it's higher expression from mVec, but that changes quite rapidly because then the stability of the RNA makes it accumulate over time. So it takes a while for the equilibrium to be achieved. So you get that steady state where you have the same production and decay rate. So that actually this experiment took up to a month before we reached that steady-state level. And as you see, when that's being achieved, we have 15x higher expression in this experiment when we compare circVec to mVec. So this is an earlier circVec vector that was used. But of course, this is something that you can optimize, you can work on this. So we started a while back some years ago with sort of the circVec generation 1, which was more or less sequences that were copied from our human genome, you get circular RNAs are naturally being expressed inside our cells. So we can look at the cells. You can look at what circular RNAs are being expressed very efficiently, and you can copy that into the vector and see how it performs. So we did that at the first step, and that was basically the design of the circVec generation 1. So there's 2 different aspects we can basically optimize. One aspect is the production of the circular RNA from the DNA vector also, we refer to that as the biogenesis. So how much RNA, how much circular RNA do we get out per DNA copy. But then there's another aspect, and that's a protein production from the circular RNA. And there's 2 these different steps, and that's what you normally refer to as a translation of the translation efficiencies. And that's mostly covered by this IRES element. So as you can see here, as we go through the evolution, we've been optimizing this IRES element. So that's the first step that actually happened that we focused on, and we now have an IRES element that still is -- we have not been able to find anything that has outcompeted this IRES element. So this consistently shows extremely efficient transition from circular RNA. But that's sort of the translation aspect, but then we've been focusing a lot on optimizing the biogenesis. So just basically looking at these elements that are in the DNA, but outside of the sequence that will be circularized. So this is the stuff that governs and stimulate and basically determines the efficiency of biogenesis. So this is something we've been working on. We've been working on generation 3, where we added an additional element, further boosting the biogenesis. And then most recently in the generation 4 of circVec, we've been going back to putting things into the circRNA cells that may actually stimulate translation. So we've worked on relatively small elements that we can actually introduce that seems to give an additional boost on the translation. So here, you basically see how that translates into protein yield approaching quantities coming off the circVec platform. So if we started down here at 1, you can see we're up here at 37 now at generation 4. So that's quite a tremendous achievement and a very significant improvement over the initial design. And even though the last going from 3 to 4 is only around a 40% improvement, you can still see that it's quite substantial when you compare in absolute levels to the original design. So we were quite pleased with that. And of course, this is something we are consistently working on to drive up expression even further. Probably this is not the ceiling, but it's pretty high, if you ask me at least. So that's where we are with the sort of the core platform development at the generation 4. So how does this work in a more clinical relevant setting such as in an AAV gene therapy setting. So as you can imagine, the approach is quite simple. You have an AAV vector, so that's a DNA vector. So you put in -- normally, you will put in a vector that will express an mRNA and with those caveats that I've explained before, the RNA is unstable. It will be turned over quite rapidly. So there's a level of how much an upper limit or how much protein you would be able to achieve from that approach. But if we exchange the genome of the AAV vectors to express our circular RNA instead, we can -- we expect to drive up protein yield per vector copy quite significantly. So we had a little bit of an issue. I'm not going to go into that with the generation 3 that didn't perform as expected, but we were able to optimize that to actually see the same level of improvement in AAV as we saw in vitro with normal DNA expression. So one of the focus areas we've had at Circio is within the heart. So one approach you can have, you can have your AAV sort of being mostly expressed in the heart. So this is one of the experiments we've done. We've used an AAV that's called AAV9. The cassette, both in the mVec cassette and the circVec cassette is driven by a promoter that's mostly active in the half. So that is why you see sort of a mostly heart topic or heart centered expression profile across the vectors. But as I hope you can appreciate, there's a substantial difference between using the conventional benchmark AAVs and then using the Circio proprietary in this case, 3.2 AAVs in terms of the expression level. So you can compare the heart expression here. And if you look at the profile across the mouse, you can see you have a very specific expression predominantly or solely in the heart region of these mice. So that's extremely encouraging to see. So I would just also mention that this is done at a dose level that would probably be in a clinical setting be referred to as a low dose. So this is at the very low end of the dose spectrum that you will sort of consider in the clinic. And even here, we see a very, very substantial and a very effective expression in the heart. And if you just quantify this over time here, we see a 40x improvement over the mVec specifically in the heart. Now we have just embarked on testing the 4.0 in a similar setting. So this is an ongoing study. These mice are still running around, having a good time, and you can monitor the luminescence here in real time to follow protein expression as time progresses. So this is at a single injection, as always, at day 0 and then you follow at different time points here. So we are now 4 weeks into the experiment. And as you can see, even though the difference doesn't look that striking, it's a logarithmic scale, and this is as we saw in vitro is a roughly 50% increase over the 3.2. So we still think that is a meaningful improvement over earlier generations. So this is nice to see that what we've worked on in the lab, testing in cell cultures, this also seem to demonstrate the same full change in a mouse model. So that's where we are with the 4.0. It looks promising. But of course, this -- we will monitor this experiment, and we will also need to deploy this 4.0 in different areas to fully validate its performance. Now going back to the 3.2 experiments. We've done some additional work, so that experiment has come to an end. So we could sacrifice the mice. We can look at the expression specifically in heart after termination and an approach that's called ex vivo tissue analysis. So here, you just see the heart from the mice, and you can, of course, appreciate that there's a much higher expression in this case, when you quantify 35x more expression in the heart from circVec 3.2 compared to mVec. But another added feature with circVec is actually that we see more selective expression in the heart. If we compare what tissues are actually responsible for gene expression across the body of these mice. You can see almost 80% of all expression that we get from circVec emanates from the heart, whereas mVec seems to be more promiscuous, it's expressed more or less all over the mouse body. It's less specific to the heart, so only 40% in the heart, much more noticeable expression in the liver from circVec -- from mVec, sorry, very little liver expression from circVec. So that's sort of an added feature using the circVec technology, it seems to accumulate specifically in the heart, we get a very heart-centric expression profile, which is, of course, very desirable if you want to treat something specifically in the heart. On top of that, if we profile the transcriptome and look at all the different genes that are expressed within the heart of these mice and look at what is going on when a heart cell is encountering an AAV that is either expressing an mRNA or circRNA. And maybe a little bit to our surprise, considering the 40x improved gene expression in the heart, you actually see a low stress profile. So this plot basically shows you a little bit on how much stress, do you see in the mVec scenario compared to in the circVec scenario? So you compare these 2 different vectors to each other. And you can see genes that are involved in this cellular stress pathway called the UPR, the unfolded protein response pathway are significantly up in mVec compared to circVec. And this is actually also something we tend to see in cell lines in vitro that when you administer DNA into different cells, having these DNA expressed mRNA seem to be more stressful for the cell compared to the circRNA. And I think this boils down to the fact that we are working on the stability of the circRNA. We're not necessarily working on churning out more RNA, but it will accumulate slowly due to less decay. And that -- of course, at least in my mind, also suggest this must be less stressful because it doesn't require a lot of work for the cell to actually keep the expression high, but it just needs an RNA that's stable, so will not be degraded in the cell. And that's basically what you achieve using the circular RNA intermediate here. So I think there's a lot of benefit with our technology, specifically within the heart currently. So just to summarize what we observed in the heart, we see much higher expression, 40-fold high expression roughly in the heart. We tend to see higher specificities or lower expression in the liver, higher expression than the heart. And then addition, see, we -- in the cell, the stress response in the cell seems to be lower for circVec despite the higher expression. So I think all this combined, suggests that circVec could be a very attractive approach in heart-based rare diseases. Now we're also exploring circVec in other avenues. So one approach is in the eye, in opthalmology. So we've done so far, one experiment. We are setting up another one with more recent circVec designs. So this experiment was done with an earlier design. It takes a little while to engineer these AAV vectors and get them tested, but this was done with the circVec 2.0. So we believe we can actually do much better based on where we are today. But even in this scenario where we have the 2.0 compared to benchmark AAVs, we see a 7x improved expression in the eye of these mice. But as said, we are doing a follow-up experiment. We hope to initiate this year, but this is sort of set in motion, and we are looking very much forward to seeing the performance of these more advanced circVec vectors in eyes. So this is another avenue we believe is -- could be a very good fit for our technology. And most recently, we have just initiated an experiment in CNS using a local delivery approach called intra-cerebro-ventricular injections or ICV, as we normally call it, which is actually a clinically approved approach, but this is also something you can do in a mouse where you inject very generally a small volume directly inside the brain ventricles of mice. And then using the reported genes that we start off with always the [ iduronidase ] so we can monitor expression in real time. You can already see here at the first time point. So this is where we are. 10 days into the experiment, you have 4x more expression in the brain from circVec 2.1 in this case compared to mVec. So I think this is another avenue that we are quite excited about and we are eager to follow up with more circVec vectors and also just to see how this experiment develops over time, how much accumulation from the circVec design do we see in the CNS compared to the mVec scenario here. All right. So basically, our main focus in-house currently is to develop circVec within the heart. We have a focus on Danon disease, as Erik mentioned, this is a good opportunity. There's no approvals. It's a validated target. We know that AAV gene therapy works, but there are some issues with toxicity. So I think this is an excellent fit for circVec despite the relatively low population. So this could be a good clinical proof of concept for Circio because we clearly believe we can get better expression and likely at a lower dose with higher safety. So I think this basically addresses the issue currently observed with Danon disease and assuming this works, as expected, there's a whole range of rare diseases and the cardiovascular indications that you can consider with circVec basically using the same approach, better expression, higher specificity, lower toxicity. So overall, a better approach. In parallel, we have -- I could just mention, we are already now expressing in vitro, the therapeutic protein that's required for Danon disease. It's a protein called LAMP-2B. It already now works quite nicely in vitro. We will do a little bit of more optimization just to make sure we have the best possible design, but then we'll put that into an AAV and test and compare it to benchmark AAV. So this is work already put in motion and ongoing. Similarly, for eye, we see a 7x improvement for an outdated circVec generation. So we're working on testing the most recent designs in the eye, but at the same time, we are also developing now therapeutic proteins for wet AMD, which is basically an anti-VEGF which is a secreted protein. So that's a slightly different type of protein compared to LAMP-2B here. So also very-- we are very excited. So we are very excited to see how that performs in vitro and ultimately, how that performs in wet AMD mouse model that we also aim to set up next year, early next year. But finally, with the CNS data that we just recently obtained where we also see an improvement over existing AAV technology, I think that opens up a lot of opportunities as well. It's a huge unmet need. And I think this is currently where the highest pharma activity exists. So there's a lot of potential partnership opportunities and collaboration opportunities for us within the CNS. So specifically for Danon disease, this is the time line as we see it currently. We have developed and are working on the 3.2 and 4.0 already showed that those milestones and those data now we're working on the Danon disease circVec construct expressing the LAMP-2B and we hope to put that into an AAV and ultimately into a mouse early next year to test the expression and compare that to benchmark. And then by the end of next year, we hope to have efficacy data in a mouse model. Unfortunately, it takes a while to establish. It takes 6 months just for the model to demonstrate a phenotype. So this is something that we are actually initiating as we speak because the outlooks are quite long. It's a quite extensive mouse model that we need to establish here. But hopefully, next year, we aim to have some efficacy data in that model and then following the year after with some tox and NHP data. So that will be some main inflection point that we expect next year and the year after within the area of Danon disease. So I think that leads me to the take-home messages from the R&D update, specifically for the AAV work we've done. I hope you agree that AAV circVec outperforms conventional benchmark AAV approaches, both on expression level, specificity and toxicity. We've seen this advantage across different tissues, most notably in heart, but also in eye and CNS. And we are -- we foresee several commercial and partnering opportunities, both in the near-term future here. So just to remind you, in-house, we are establishing proof-of-concept Danon disease and wet AMD in heart and eye, respectively. And the next step for us is to test these specific circVec vectors that are expressing either the LAMP-2 or the anti-VEGF proteins responsible or required for the treatment of these 2 different indications here. And then as Erik said, we recently entered a partnership with a major global pharma company also within the context of AAV but we're also aiming on establishing and identifying collaboration with companies that are working on engineered AAV capsids. So that's basically the surface of the AAV that may enable more effective delivery to tissues of interest. One thing could be actually allowing CNS delivery by systemic delivery -- getting CNS targeting by systemic delivery. So this is something we are keen to establish. We believe that could be a key synergy between our technology and the engineered capsids. So this is something we're working intensively on establishing. And I think with that, that sort of concludes the AAV update. And I hope that covered it all. And I'll pass the word back to Erik.

Thank you very much, Thomas. Now I will provide you an update on our in vivo cell therapy program. In this area, we foresee quite a unique opportunity for our circVec in this case, with DNA, not AAV vectors. And there's been a lot of deal activity in the in vivo cell therapy space mainly centered around mRNA and circular RNA, synthetic RNA approaches. These are looking encouraging, but they only give you expression for a few days. With circVec, we can express for several months. We've actually shown now that we can express for 6 months in the spleen on a single dose. So we have a vastly extended expression window. The only other solution that could give you this type of expression window would be to use integrating DNA or viral vectors that permanently give you expression, but that carries a lot of safety and other issue aspects and other issues that you ideally want to avoid. So we can offer something that is safer, nongenome integrating, long duration of response and is redoseable. And on top of this, circVec naturally avoids the liver. So you don't have to deal with that issue as well. So here, we have a quite unique opportunity to move forward and we think this could have a particularly interesting use for cancer application in the in vivo CAR space, potentially also autoimmune disease. We are earlier in our development here, but we have some very interesting data in vivo showing that in this case here, where you compare on the left-hand side, mVec or mRNA vectors to our circVec vectors on the right-hand side, we get a completely different, both temporal and tissue-specific expression pattern when we use circVec. The mVec expresses mainly in the liver, lasts for about 2 weeks and then is gone. circVec, it takes a little bit longer to express like Thomas showed you before, it usually peaks after about a month, and that's the consistency we see here also. We reached a maximum expression after a bit over a month, and then it just keeps going. And now we've shown this in these animals in repeat experiments that it has a duration for up to 6 months, and we continue to track. We will see how long this window actually lasts. So you get a fundamentally improved expression pattern when you use circVec. The key next step here is to show that we can deliver this in a targeted manner to T cells. That's what is required to make this attractive for in vivo cell therapy in oncology. So we are now discussing with companies that can offer T cell-targeted LNP delivery solutions, and we expect to decide on that and initiate studies in Q1 next year to see whether and how well this performs when delivered in vivo directly to T cells. And shortly following that, we will start with the reporter. We then move to an approach where we deliver CAR -- actual CAR constructs. We already have this generated. So these are functional in vitro, they are available and that's the immediate next step. So these will be very important readouts for us in 2026. And this space has been an area of substantial deal activity. I will come back to that shortly. So key takeaways here, 6-month duration of expression on a single dose versus just a few days for mRNA-based in vivo CAR or a couple of weeks if you use mRNA DNA. circVec expression, we've shown our expression in the spleen emanates from both B cells and T cells. So we know we can express in these cells. We have technically validated a CAR expression construct that has application in vivo cell therapy for cancer. So as I said, next step, test active T cell delivery in vivo. And we're also actively looking for R&D collaborations to test this in the partners' validated targets and delivery systems in vivo initially. So with that, I move to the summary and outlook. And as I mentioned, there has been substantial deal activity in our areas. If we start from the left here, Lilly recently licensed technology from MeiraGTx which is an AAV company. So this is a deal for an AAV gene therapy targeted to the eye. It's an AAV engineering platform to increase the specificity for the eye. And this is a Phase I candidate but as you can see, very substantial economics. And here we see big pharma are actively investing into this space. There's actually been several also funding rounds for concepts now in recent months. The 2 other deals on the slide are for in vivo CAR approaches, BMS just now recently acquired Orbital Therapeutics. It's a synthetic circular RNA approach, so it's different to us. But again, it shows the potential there was a $1.5 billion cash buyout. And earlier this year, AbbVie acquired Capstan. Capstan was considered the most advanced in vivo CAR company, and they are now -- they have entered Phase I. And this deal was worth $2.1 billion. So you can see the big players are moving into both of these areas. We had important business development news. So recently, we entered a feasibility study with a major global pharmaceutical corporation. The identity of this company remains confidential, but I can say that it is one of the largest global pharmaceutical companies. We can't say much about the feasibility study, except that it's testing circVec AAV, novel designs of circVec AAVs, and it's in a specific disease area. And if this study is successful, we will be testing both new designs. We'll be testing them in vitro and then in vivo, if this works, as we expect, then the next step may be that we out-licensed the technology for these particular applications that we are testing them in. We also have a number of R&D collaborations ongoing, which we consider more 50:50 type collaborations. These mainly have 1 of 2 natures. One would be that we are looking for a vector type or a delivery technology that would be interesting for circVec to enable us to move into the direction we want to go or vice versa. It may be for the partner to utilize circVec for their particular applications of interest. And we have both of these ongoing. We have several that are active. When and how data is presented, depends on sort of the mutual interest to move forward. And I can say for Circio that we would anticipate to pick for the circVec DNA delivery, 1 or 2 of these to move forward with probably during the first half of next year. So that would be a time point where we expect to communicate results and partner selection for that area. And we're also now looking for new partnerships based on the data that we have seen. So we have now 3 areas that we've shown clear advantage of circVec, especially in heart and then eye and CNS were moving forward. And we remain very interested to form similar R&D collaborations like the one we now recently entered in areas that remain open. For in vivo cell therapy, we are actively looking for partnerships with companies that possess robust T cell targeted delivery system. So that's not going to be an important priority for the business development activities. And similarly, on the AAV space, we would really like to try to combine circVec in engineered AAV capsids that have specific targeting for the tissues we are interested in. For example, in the heart, you're usually relying on delivering your AAV systemically. So you can add a lot of advantage if you have a capsid that traffics to the heart more specifically than classic AAVs. And if we combine with the technology like this, we maybe can lift even tenfold further the advantage that we see today. So it can be a really powerful combination and engineered capsid together with circVec. So we are in dialogue with companies that have interesting technologies available in this particular space. So those are our BD activities. Of course, in general, now with the strong data we have and the broader data we have, we become interesting to a larger set of companies and also our own needs are much more concrete as we now know more clearly where it makes most sense for Circio to move forward. We remain very cost conscious. We have a cost base that is, I would say, very low considering substantial output we have. We're burning substantially less than we did before. It's well under NOK 4 million per month. We've streamlined our activities and we strictly prioritize R&D activities to make sure that the funds we use are value creating. We do have financing available from Atlas. So it was announced in Q3 that we extended on the previous financing commitment by 3 plus 3 tranches. So these cover NOK 4 million.They can be drawn once per month, and they're design so that one tranche is 1 month of operations. We've drawn 2 of these so far. That means 4 are available, and this would be available to be drawn at the company's discretion until the end of Q1 next year. So we have flexible access to capital on the need to basis, but we're very careful. We only use this as required when necessary, we want to minimize it as much as we can. And as probably many have seen Atlas has now converted all of their convertible bonds. So there is no outstanding debt in that regard. And naturally, we're exploring continuously other financing options to secure long-term funding through business development activities or through investments from specialist biotech investors or generalists with an interest in the area. And this is going to be required to ensure a long runway for the company and enable us to really accelerate our R&D. So coming to the end here, I summarize now the main milestones ahead. Thomas showed you, I think, very convincing data for circVec 3.2 especially in heart, we have also the emerging eye and CNS data and the 4.0 data is starting to emerge now towards the end of the year. In Q1, we expect to be able to test in vivo constructs, targeting the Danon disease and AMD, not in disease models, but testing them in healthy animals to see how the vectors perform. So this is the first step on the way to disease model data, but it's very important to understand how well the expression is. And we can, in this case, also test different dose levels and do head-to-head dose responses versus the conventional approach. And for example, for Danon disease in clinical trials, protein expression in the heart rather than actual efficacy data is what's being considered as endpoints for clinical trials. So I want to highlight that even though, as Thomas said, it will take a while to get the efficacy data because of the nature of the disease, already in Q1, we are expecting to see expression data of disease constructs that normally will be considered to have good predictability for therapeutic outcome. And the in vivo CAR delivery to T cells is also something that is starting. All of these data points, we channel into ongoing R&D discussions with partners, and we expect to enter more R&D collaborations like the one we announced with the big pharma in the coming months based on readouts now in novel tissues and in the disease context. Looking a bit further ahead, we think that the big milestones coming up are the actual efficacy data in disease models. The first one we anticipate to deliver would be in AMD. This is a quicker disease model set up. Then we will start exploring more disease context in CNS and finally, in heart, which takes a little bit longer due to the nature of the disease. And if you look back at the deals we have seen, I think this is data that when these emerge will put us in a position to be a contender in these larger transactions that we see in the space. So an exciting year ahead for Circio. With this, I wrap up. To finish the presentation, we have had a number of recent articles published in industry media about Circio, about our technology and how AAVs can be enhanced. So if you want to read further, I point you in this direction here, where you can learn more about Circio and what we do and how we're perceived in the industry more broadly. So with that, I thank you all for attending. We're very pleased we had a large audience today. As always, please send a follow-up questions. We're always here and answering every question that comes in as much as -- to the best of our ability. So thank you, Thomas. Thank you to our listeners, and we are signing off from Circio.