Good morning, and welcome to the ProQR Therapeutics Analyst and Investor R&D event. At this time, all attendees are in a listen-only mode. A question and answer session will follow the formal presentations. If you would like to submit a question, you may do so anytime throughout the webinar by using the Q&A text box at the bottom of your webcast player. To our analysts, please raise your hand to indicate you would like to join the queue. At this time, I would like to turn the call over to your host, Sarah Kiely , Vice President of Investor Relations and Corporate Affairs at ProQR Therapeutics. Please go ahead, Sarah.
Thank you. Good day, everyone. We appreciate you joining our event today. I'm Sarah Kiely , Vice President of Investor Relations and Corporate Affairs at ProQR. Today's event will take an in-depth look at our Axiomer RNA editing platform technology and our plans to advance it. On slide two, you'll find the agenda for the event and our speakers. From the management team are Daniel de Boer, our founder and CEO, Gerard Platenburg , our Chief Scientific Officer, and René Beukema, our Chief Corporate Development Officer and General Counsel. We're also very pleased to have Dr. Peter Beal of UC Davis present on ADAR. Following the presentations, we will have a management team Q&A session with covering analysts before we conclude the event. Today's event is being recorded. We will have a replay available on our website following the event. During the presentations today, we will make forward-looking statements.
There are risks and uncertainties associated with an investment in ProQR, which are described in detail in our SEC filings. I'll now turn the presentation over to Daniel. Daniel?
Thank you, Sarah, and good morning and good afternoon, everyone, and thank you for joining us today. We're excited to share with you today a comprehensive update on the company following the decision taken last summer to fully focus our strategy on advancing our proprietary Axiomer RNA editing platform technology. We believe this platform holds great promise to treat diseases that are otherwise untreatable. ProQR is exclusively focused on the development of our proprietary Axiomer RNA editing platform, which has broad applicability across multiple therapeutic areas. Our initial focus has been on CNS and liver, and we will announce today our initial pipeline targets during the event. These targets will focus on diseases that originate in the liver. The Axiomer technology was invented at the ProQR labs back in 2014 and uses the well-proven modality of oligonucleotides to recruit a novel mechanism of action.
Axiomer uses an editing oligonucleotide or EON to recruit endogenous ADAR, which can microsurgically edit the RNA, changing an A or an adenosine into an I or an inosine, which is subsequently read as if it was a guanosine or a G. ADAR is present in all human cells, and RNA editing is a naturally occurring process, and in fact, it is happening in all of us right now as we sit here. Our Axiomer platform makes use of the ADAR machinery that nature has developed and recruits it to edit specific adenosines in a very targeted way. Our strategy includes both in-house development of pipeline programs using this technology, as well as selectively partnering on non-core targets, allowing us to capture the full value of this platform technology.
As we will also delve into today, preclinical platform data demonstrated that ProQR's Axiomer RNA editing technology is broadly validated across multiple genes. At ProQR, we believe that RNA editing will become an important pillar within the field of genetic medicines. Over the last years, we have seen the field of genetic medicines maturing and leading to more and more important medicines for patients. RNA plays an important role in that. We've seen multiple approvals over the years of RNA medicines, and of course, most recently, the mRNA COVID vaccines. Many large companies are making significant investments in genetic medicine, placing an important bet for the future on this class of medicines. Within that field of genetic medicine, we are convinced that RNA editing will become as impactful as, for example, siRNA, by adding a novel endogenous tool to the arsenal.
As the leader of the RNA editing field, we have spent the last 8+ years since the discovery of the technology optimizing the Axiomer platform in our ProQR labs and in partnership with academic partners to drive a robust understanding of ADAR biology and the design rules. We see RNA editing as the next evolution in the RNA medicine field. Similar to RNAi, we are recruiting an endogenous mechanism that's already present in human cells. Instead of knocking down an entire gene, we can now edit individual bases in the RNA. This opens the door to develop treatments for diseases that are otherwise not treatable. In these instances, we have unique advantages over other technologies.
Axiomer is a very versatile platform, allowing us to target a wide range of disease and can potentially lead to hundreds of medicines in very different disease types with a variety of molecular mechanisms. As we can make edits to the RNA, we don't have to touch the DNA or make any permanent changes, which comes with significant safety advantages. With transient editing, we can expect to dose quite infrequently, maybe two to four times a year. As our editing is transient, it can also be applied in situations where permanent changes would be deleterious. Axiomer editing is done with high specificity, minimizing off-target effects, and we don't have to use viral vectors.
We can lean on well-established delivery routes and moieties that have been developed and proven over the last decades. We see Axiomer as an elegant platform using endogenous ADAR machinery that essentially leverages the body's own potential to treat disease. ProQR has quite literally led the field of RNA editing since 2014, when ProQR scientists invented the RNA editing using endogenous ADAR and performed the first experiments using our editing oligonucleotides to recruit natural and endogenously expressed ADARs. These experiments also led to our first IP filings for this technology back in 2014, which laid the foundation for our leading IP estate today. Over the subsequent years, we have done considerable work to optimize and enable the development of highly specific, safe, and efficacious platform. Today, we will share extensive data demonstrating it's capable of efficient in vivo editing for therapeutic uses.
We plan to capture the value of our platform across two key strategies. First, through the development of an internal pipeline of high-impact medicines, and second, through selective partnering. As we will announce today, ProQR will initially focus on translating the Axiomer technology into the clinic in two separate development programs that can yield potentially very important therapies. The initial programs also bear the burden of validating the broader Axiomer platform technology in humans. Our pipeline will focus initially on the delivery of our editing oligonucleotides to liver, as delivery to that organ is largely de-risked, and we can build on proven delivery technology to deliver. For the avoidance of doubt, this does not mean that we will focus on liver disease only.
Many diseases are caused by proteins that are expressed in the liver, by targeting the liver, we can treat diseases throughout the entire human body, as we will present to you today. As we will advance these first two and subsequently more programs into the clinic, we plan to select some of these programs for development to markets by ourselves. For large indications, we plan to advance these through clinical proof of concept and then seek to partner this out for continued development. This speaks to the second pillar of our strategy, partnering. Partnerships are a great way to advance applications of the platform beyond the targets that ProQR intends to develop independently. In addition, these partnerships bring important validation, external eyes on the science, and non-dilutive funding to help us fund our operations.
We will selectively enter into partnerships that help us advance the platform and the business. For example, we have partnered with Eli Lilly on concurrently 10 targets on the Axiomer platform technology, where ProQR leads the discovery phase and Lilly leads all phases beyond that. This partnership so far brought in $125 million in upfront payments, and we expect to receive another $50 million from Lilly, likely this calendar year, for an option that they hold to expand the partnership to 15 targets in total. In addition to this, ProQR is eligible to receive $3.75 billion in milestone payments, including discovery, non-clinical, early clinical milestones, plus royalties. We're very pleased with this partnership and with Eli Lilly as a partner, and we continue to execute on this partnership with high priority.
Given the vast opportunity with the platform, we have appetite and capacities to selectively form additional multi-target discovery partnerships. As we advance our pipeline, our R&D strategy takes a risk mitigated approach in the translation into the clinic to maximize the probability of success in this phase. It's important to acknowledge the learnings we have had in the previous 10 years in translating multiple therapeutics from the lab into the clinic, as those equip us for success in the future. We have selected targets that have deep rooting in human genetics, where we know that the post-editing RNA product is associated with beneficial impact on human health. We will prioritize targets that are expressed in liver, as liver delivery of oligonucleotides through hepatocytes has been de-risked, allowing us to study this new mechanism of action in isolation of other variables.
For the initial targets in our in-house pipeline, for programs that are based on Axiomer, we have selected targets that have strong translational tools available, including animal models that are representative of the clinical situation and clinical biomarkers that allow to get objective confirmation of target engagement via a blood draw to gain solid validating platform data early in the platform development. Incorporating all of our translational and drug development experience, we are confident that this risk-mitigating translational strategy, we can study the new mechanism of action of RNA editing with high probability of success into the clinic. Today, we are announcing our internal initial pipeline programs, Gerard will go through these in quite some detail later in today's program. On this slide, we are displaying our pipeline.
As you will see, our pipeline contains the mix of targets for rare and prevalent diseases, as well as wholly owned and partnered programs. We will initially prioritize the following two programs to move forward into the clinic. Programs that we believe carry all the characteristics to translate the platform into the clinic with high probability of success, including a solid rooting in human genetics. Firstly, we will develop AX-0 810 for cholestatic diseases targeting the NTCP channel. Secondly, we will advance AX- 1412 into development, which will be developed for cardiovascular disease, targeting the B4GALT1 gene. With AX-1412, our intention is to develop this to an early clinical proof of concept and then partner for further development, given this is a large population indication.
Gerard will detail these programs later on into the event, and in addition to these programs, you will find a range of programs for other indications, including NASH and other cardiovascular targets, metabolic disease targets, neuro disease targets, and also other rare diseases. All programs that at some point in the future we will provide further guidance on. For today's session, we will zoom in on the two programs, AX-0810 and AX-1412 . Although the RNA editing field is early, ProQR is well-positioned to execute on the Axiomer business plan. Back in 2014, we invented this technology, and over the years, we've been able to really focus on the basic science underlying the ADAR biology and optimizing the design rules for the Axiomer platform. As we'll describe today, there are a large number of potential therapeutic applications in both prevalent and rare diseases.
We will also present a range of pre-clinical proof of concept data, including in vitro and in vivo studies involving organoids, mice, non-human primate experiments in liver and CNS. For better or worse, ProQR has gained quite a bit of translational and development experience in RNA oligonucleotides over the last decade. We can integrate all of our lessons learned into translating the Axiomer platform into medicines for patients. That same vein, we have a lot of experience with target hunting and triaging. We have therefore selected targets for our pipeline that both have high probability of technical success and could lead to important products with viable business cases. Axiomer is protected by more than 10 patent families that have multiple granted claims, protecting the recruitment of endogenous ADAR with editing oligonucleotides in a very broad way, dating back to the invention in 2014.
These patents have been opposed and survived that opposition. Hence, we feel very confident about the strength of our leading IP estate in the field of endogenous ADAR editing with editing oligonucleotides. Our partnership with Lilly provides us with strong validation, a great partner to advance potentially important medicines, and is providing a source of non-dilutive funding to help us fund our operations. Beyond our existing partnership with Eli Lilly, we plan to selectively enter into additional multi-target discovery partnerships. We have adapted our team, management board, and advisors for the Axiomer business plan. This includes John Maraganore, joining our board as a strategic advisor back in 2022. John built Alnylam into a successful company over the last two decades, and we can leverage his experience to apply Alnylam's playbook to build out the Axiomer platform.
Our scientific advisory board has great expertise in RNA, including Phil Zamore, an inventor of the Nobel Prize-winning RNAi technology, Martin Maier from Alnylam, Art Levin from Avidity, and Peter Beal, one of the leading ADAR experts in the field, who will present today as well. James Shannon, the former Chief Medical Officer at Novartis, and Yi-Tao Yu, professor at Rochester and expert in the field of RNA editing, complete our scientific advisory board. Last but not least, ProQR is well-funded with a runway into mid-2026, in which timeframe we anticipate having clinical data from multiple development programs. Against this backdrop, I'm very pleased to now hand the call over to Gerard Platenburg, who will guide us on a deep dive through ADAR science, highlight our platform and a wide range of pre-clinical proof of concept data, and present our pipeline. Gerard, over to you.
Thank you, Daniel. It's my pleasure to introduce our RNA-based editing technology, Axiomer. Axiomer, as you can see on the left of the slide, uses endogenous ADAR, or adenosine deaminase acting on RNA, to convert adenosine into inosine, a process which takes place in every cell in our body with different functions, as you will learn. We will start a short video on ADAR, followed by a presentation by Dr. Beal, who is a world-known expert on ADAR. Now we will start the video. Enjoy.
How does ADAR work? ADAR enzymes, which are naturally present in the cells, can be used as powerful tools in treating diseases. To understand how, we first need to explore the functions of ADARs in the human body. RNAs are produced in a process called transcription, where the genetic information in DNA is copied into RNA. The information in RNA serves as a blueprint to produce protein via translation. Before translation occurs, the RNA can be processed in several ways. One way involves changing specific nucleotides or letters in the RNA code. This is called RNA editing. RNA editing helps ensure that the produced protein functions normally. It can also create slightly differently functioning proteins. One common type of editing is A-to-I editing, where A nucleotides, also known as adenosines, are changed into I nucleotides or inosines. How exactly does this work?
RNA strands are never completely linear. The nucleotides can pair together to create double-stranded structures within the RNA. This is when a specific cell machinery enters the story, the ADAR enzymes. ADARs find the double-stranded structures and bind to them. They then edit As, adenosines, into Is, inosines. Later in the translation, when the cell machinery called ribosome reads the RNA blueprint, it sees the Is, inosines, as if they were Gs, guanosines. After additional post-translational modifications, the functional protein is produced. What does this A-to-I editing performed by ADARs accomplish? One of the natural roles of ADARs is to help the immune system protect the body against external aggression, such as viruses and bacteria. To be specific, ADARs allow the immune system to differentiate between the body's own RNAs and the RNAs of intruders. How does this differentiation happen?
To fight against viruses, certain immune proteins attach to the double-stranded structure of the virus RNA. This sends an alert signal to the immune system, which then cleans the infected cell from the body. However, in the body's own RNA, ADARs perform A-to-I editing. This breaks the double-stranded structure, which prevents the immune proteins from attaching to it. As a result, the immune system is not activated. Another example of ADARs biological function is their vital role in the maturation of brain cells, the neurons. Neurons share essential information with each other via channels. During the development of the brain, some channels need to be replaced by new ones. At the premature stage, the RNAs for new channels do not contain inosines. This allows for a certain activity in the channels at this stage.
Once a channel needs to be changed into the mature stage, ADARs perform a very specific A-to-I editing in the RNA. This changes the specificity and the function of the newly created channel. Of course, the immune system control and maturation of neurons are just two examples of the many essential activities performed by ADARs. ADAR enzymes are present in all human cells, and it is a broadly active mechanism. ADARs have the ability to target adenosines with high specificity to turn them into inosines. There are over 16 million known locations in the RNA where ADARs perform A-to-I editing throughout the human body. This makes A-to-I editing a potentially powerful therapeutic mechanism for multiple disease areas. ProQR scientists have been at the forefront of ADAR research for years and will continue to push the boundaries of this promising area.
Now that you know more about the general information about ADAR, we are very pleased to have Dr. Beal today, who will give us much more details about ADAR science and learnings in the field. Dr. Beal is a professor in Department of Chemistry at the University of California, Davis and a world-known expert in ADAR science.
Thanks, Gerard, for the introduction, and I want to thank the organizers for the opportunity to present our work on RNA editing by the ADAR enzymes and discuss our collaborative efforts with ProQR. What I want to do today is give you some background information about RNA editing, A-to-I RNA editing, and the ADAR enzymes, and then talk about how we are using structural studies coupled with biochemical analysis to optimize the guiding strands for the ADAR enzymes for therapeutic editing work that we're doing in collaboration with ProQR. Before I start, though, on that, I want to talk a little bit about A-to-I RNA editing. The reaction we're talking about is this reaction shown in this slide.
This is the deamination of adenosine in an RNA molecule, and it can be in an mRNA, and the conversion of adenosine via deamination reaction in mRNA generates inosine, a minor RNA base. Because of its structural similarity to guanosine, inosine is actually translated as if it were guanosine. If this A is in the proper location in a codon, it can change the meaning of the codon, and in this way, it edits the information in the RNA. ADARs are human proteins. These enzymes that do this reaction are human proteins, so there are many known human substrates for the ADAR enzymes. What I'm showing you now on the bottom of this slide is one of my favorite human substrates for the ADAR enzymes. This is the pre-mRNA for encoding the protein NEIL1 , okay?
At the junction between intron 5 and exon 6, the RNA folds into a structure that is extensively duplex, and it has an A at the specific location indicated by the red letter. This A is in an AAA codon for lysine, and when the A, the red A is deaminated by the editing enzyme, it generates a inosine A, which is a codon for arginine. The NEIL1 protein actually comes in two forms. It comes in the genomically encoded form that has lysine, but it also comes in the edited form that has arginine. We published a paper, you know, quite a while ago now showing this is in collaboration with my colleague, Sheila David, who is an expert on this NEIL1 protein. It's a DNA repair enzyme.
We showed that the two forms of the protein are biochemically distinct. This is an example of how this ADAR enzyme, the enzyme that does this reaction, is able to tune the properties of a protein by changing the amino acid in the protein. The side chain of the amino acid at position 242 is in direct contact with the substrate for this enzyme. In that way, it's tuning the properties of that protein. The enzymes that do this reaction, I've already introduced them as ADARs for adenosine deaminases that act on RNA. There are three human genes that encode ADARs, the ADAR1, ADAR2, and ADAR3 genes. There are two forms of ADAR1, the long form p150 and the short form p110.
ADAR3 is not a functional deaminase. It actually has a defective deaminase domain and is not capable of carrying out the reaction, but is capable of binding RNA substrates. ADAR1 and 2 are functional deaminases, right? These are the proteins that we've been focused on in our lab. These enzymes can bind duplex RNA, and that's at least partially explained by the presence of what are referred to as dsRBDS or double-stranded RNA binding domains that are part of the protein structure. ADAR1 has three dsRBDS. ADAR2 has two dsRBDS. Each of the ADARs also have a C-terminal deaminase domain, where the reaction takes place. Our lab has actually shown that the deaminase domain is also a duplex RNA-dependent component.
It's important to think about in the context of directed RNA editing, why we're here today, right, that the ADARs require duplex structure. Okay? A duplex is essential for the substrate. The reaction of adenosine to inosine conversion in duplex RNA has multiple consequences. I've already talked about the recoding effect, where if you have that A in a specific codon, deamination can convert that codon to something else. One of the more well-studied substrates for human ADAR2 is a glutamate receptor, which is a calcium ion channel, and it's known that the two forms of the protein, have different ion conductance through the channel. It's also true that in the case of the secondary structure, substrate for ADARs, the A is often in an AU pair.
The deamination of that A creates an inosine U mismatch, which destabilizes the secondary structure. The reaction of double helical RNA in the cytoplasm and weakening the secondary structure and preventing that RNA from binding to cytoplasmic receptors for duplex RNAs is one of the, probably the most important biological function for human ADAR1, preventing an immune response because those receptors are there for recognizing viral RNA. There are other consequences of A-to-I editing as well. A-to-I editing can affect splicing. The reaction within microRNA precursors or microRNA binding sites also can affect expression levels of specific target genes. One thing that's gotten, right, the community very excited, and again, why we're here today, is this idea that you can do directed editing, right, with a guiding oligonucleotide.
Recall I said that the, a requirement for ADAR recognition is that the RNA is in a duplex structure. We can create that duplex with a synthetic oligonucleotide, right, binding to a particular transcript, which will then recruit the ADAR to do the reaction at a specific site. You can imagine we may want to convert an A that is there because of a G to A mutation that causes some genetic disease, perhaps by creating a premature termination codon. Right? We can recruit the ADAR to that site, convert that A to inosine, which will now be translated as if it were a G, which could allow for read-through or some other therapeutic effect, repairing the defect or making a corrective edit. Okay? The guiding oligonucleotide then would be the therapeutic in this case.
In this slide, I show you some key advances that I think are key advances in ADAR research over the roughly the last 30 years. Starting with the discovery of ADAR1 in 1988. Here we have in 1995, the first report of directed RNA editing. Todd Woolf's lab showed in Xenopus eggs he could direct the ADAR reaction to make a corrective edit. In 1996, we see the discovery of ADAR2. Right? In 2000, we see the role of ADAR2 in neurobiology. Recall, I indicated the importance of that glutamate receptor editing. That's a very important role for ADAR2, controlling the property of neurotransmitter receptors. We see in 2012 now, others doing directed RNA editing with ADAR fusion proteins.
This is not recruiting endogenous ADAR. This is actually generating a fusion protein where the ADAR deaminase domain is connected to some other targeting element and directing it that way. In 2012, we see the role of ADAR1 established in the innate immunity. Okay? As early as 2014, ProQR has shown, right, that you can do directed editing with endogenous ADAR. This is a very important advance, right? The idea that we could recruit endogenous ADAR to do a therapeutic edit. In 2016, our lab actually reported the crystal structures of human ADAR2 RNA complexes.
This is important for the field because now this allows for rational design, rational optimization of the strand that's going to be the guiding strand for ADAR, the therapeutic in the directed editing application. In 2019, the field was really invigorated by the discovery of a role for ADAR1 in cancer. It's now known that certain cancers are addicted to ADAR1's activity, and loss of ADAR1 activity is particularly lethal to those cancers. ADAR1 is now a target for cancer chemotherapeutic development. In 2022, we see directed editing in primates with short oligonucleotides. I believe you're going to hear more information from ProQR in this event about their work directing editing in primates with chemically synthesized short oligonucleotides.
This idea of recruiting a human effector protein to alter properties of RNA, as outlined in this part of the slide here, where we could have an editing oligonucleotide recruiting an ADAR to do a corrective edit. I see this as very analogous to two other pathways that use synthetic oligonucleotides to recruit human proteins to effect a change in RNA. Of course, I'm talking about antisense, where the use of ASOs and the antisense pathway will recruit RNase H to cleave particular target mRNAs, causing a loss of function of that, for that target. Also, RNA interference pathway, where an siRNA's guide strand recruits the human AGO2 protein to direct it to an mRNA for cleavage and loss of function. Okay?
Each of these pathways now have multiple FDA-approved therapeutics, and I see this idea, this pathway for directed RNA editing in a similar vein, and excited about the opportunity to participate in the advancement of these EONs to the point where we can show that there's approved therapeutics in this pathway as well. Importantly, the antisense and RNAi pathways lead to a loss of function, right? We're cleaving targets, we're losing function. An exciting application of directed ADAR editing is we could have a gain of function. Okay? For instance, I made reference to this repair of a premature termination codon, right?
We could actually restore the activity of a protein in a cellular context where that activity is lost with directed editing, which I think is a very exciting opportunity in this space of oligonucleotide therapeutics. If you're thinking about directed RNA editing, the natural tendency is to compare this to genome editing. I think there are some interesting advantages to using directed RNA editing. The effects are reversible in the case of RNA editing, whereas in the case of genome editing, right, that's going to be a permanent change to the genome. In the case of RNA editing, there'll be no need to deliver the enzyme because we're going to use ADAR as the effector protein, right?
It's, it's endogenously expressed in human cells, whereas in the case of genome editing, one needs to deliver the Cas protein. Of course, ADAR is a human protein and not one of bacterial origin, and of course, we're not going to have the immune responses that we would have with a bacterial protein. Challenges do exist for the use of RNA editing. We would expect that we would have to have continuous administration for the therapeutic effect, whereas in the case of genome editing, it's likely it would be a one-and-done sort of therapeutic. For certain sequences, we see low editing efficiencies for directed RNA editing. Okay? Now, I see this actually as an opportunity for innovation. Okay?
What we are very interested in doing, and we are working with ProQR to do this, to optimize the guiding oligonucleotides to improve editing efficiency and using structure and biochemistry to guide that optimization process. For the rest of my presentation, I want to talk to you about some examples of where we've been doing that with ProQR. In 2016, our lab, working with my colleague here at UC Davis, Andy Fisher, solved the first crystal structure, right, of an ADAR deaminase domain bound to RNA at a relevant point in the reaction pathway. We did that by using a nucleoside analog that we had developed that mimics the deamination transition state, the adenosine deamination transition state. We could trap this complex at a point along the reaction pathway that was catalytically relevant.
By solving the crystal structures, then we could identify the key contacts between the ADAR protein and the strands of the double helical substrate. We see multiple contacts to the strand that is edited, the strand that I have indicated in red here. We also see multiple contacts to the strand that I've indicated in blue here. Right? This is the guiding strand. This is the EON, right? This would be our therapeutic ultimately. We are now using this structural information, right, to optimize the individual components, right, of this guiding strand. Right? I'm going to talk about how we're doing that. Of course, we think about this guide strand, right, as the drug, right? We're trying to optimize the components of this guide strand.
I'm going to draw your attention to this nucleotide I've indicated as the orphan, right? This nucleotide right here is the nucleotide that was base paired to the reactive A, but in the process of the reaction, that A flips out of the helix and occupies the active site of the enzyme. After that base flipping step, this orphan position is left alone. We refer to it as the orphan position. I'm going to talk about how we've optimized that orphan nucleotide to maximize the amount of editing that we can see. Okay? I'm also going to talk about how we've optimized the minus one nucleotide position, right, for a sequence that's been a challenging sequence to target by using ADARs. All right. Let's start with the orphan position.
Again, we have multiple crystal structures of the deaminase domain of ADAR2 bound to RNA, right? What I'm showing you here on the right-hand side of this slide is the orphan C in contact to glutamate 488. This is the residue that's found in the endogenous ADAR, the wild type enzyme. Okay? We also saw the crystal structure of a hyperactive mutant of ADAR, okay, that has a glutamine at that position. We see a very similar contact. Okay? Glutamine, of course, can be a hydrogen bond donor and make a hydrogen bond to this nitrogen at the orphan position. Glutamate would be a hydrogen bond acceptor, and for that to actually make the same contact, that glutamate would have to be protonated to be a glutamic acid, and that's shown on this slide.
In this case, if you lose the proton, right, this proton right here, this would be a mismatched hydrogen bonding pair. Okay. With the glutamine, the glutamine always has a hydrogen bond donor. That would be a matched pair, hydrogen bond acceptor, hydrogen bond donor. This suggested to us, of course, because this is a hyperactive mutant, that if we had an analog of the C that could donate a hydrogen bond, right, that could interact with the glutamate form of the protein, we might mimic the effect of the hyperactive mutant. Okay. This led us then. First, I should say, we tested this idea of a pro-protonation-dependent interaction by studying the enzyme kinetics of the wild type protein and the mutant at different pHs.
Those studies supported this hypothesis that for the wild type protein, there was a protonation-dependent effect consistent with this idea that you needed to have this hydrogen bond donor at this position. This led us, this combination of crystal structure and biochemical analysis then led us to explore analogs for the orphan position that had a hydrogen bond donor at this position. In other ways was like a normal cytidine. This led us to the testing of this analog that's referred to in the literature as Benner's base, because Steve Benner developed this analog in his work with orthogonal base pairing systems for DNA. This was in the literature, but not for this application, right? We're using it in a new way here.
It is a cytidine analog with an N 3 hydrogen bond donor. We showed in data that I'm not showing you here, the using enzyme kinetic analysis that Z-base accelerated the rate of the reaction for both ADAR1 and ADAR2 in vitro. What I'm showing you on this slide now is the use of that Z-base in collaboration now with scientists at ProQR. This is important collaborative effort here where our lab is doing the structural work and the biochemical analysis, and they're taking this information, and they're using it in targets that are of interest to them. What I'm showing you here is two different oligonucleotides that are chemically modified and stabilized for metabolic stability, so improved metabolic stability.
The orphan position is varied to either a normal C or the Z-base. We see for two different targets, one target in mouse liver fibroblasts, another target in retinal pigment epithelial cells. These are human cells. This is mouse cells. We see that the Z-base modification, this one nucleotide modification at the orphan position, causes leads to a substantial improvement in the amount of editing that is seen. In about a 2- to 3-fold enhancement, and we know that our other targets lead to even greater enhancement of editing that is observed. This one nucleotide is improving the editing yields. Okay?
I wanna just emphasize here the discovery of this nucleoside analog was enabled by the structural analysis identifying the contact, the biochemical analysis exploring the nature of that contact, and a chemical intuition about what it was needed to mimic the effect of a hyperactive mutant that then led us to this, okay? This is ongoing work. This approach, I should say, is continuing in our collaboration with ProQR. I want to tell you one other short story and a similar sort of idea. Okay? Here, what we're gonna try to do is we're gonna try to overcome the natural tendency of ADARs to react at sites that have certain nucleotides nearby poorly. Okay? It's known that ADARs have nearest neighbor preferences.
It's well known in the literature that both ADAR1 and ADAR2 prefer to react at A's that have either a U or an A on the five prime side. The natural substrates for these enzymes are depleted for those sequences that have either a C or particularly a G on the five prime side. There are therapeutically relevant target sites, for instance, premature termination codons, that we may want to repair, right, that have G on the five prime side. How are we going to enable editing at those what would be naturally not preferred sites? I'll tell you about that story.
Our structural studies explain this nearest neighbor preference, because If we look at our crystal structures, where the optimal 5-prime nearest neighbor sequence has been studied, that has a 5-prime U in a pair with A, there is a loop of the protein that has a glycine residue, position 489, that comes very close to the minor groove edge of that base pair. If we have a 5-prime G, the sequence that's not preferred by the enzyme, and we model a GC pair at this position, right? Mimicking what we would see for a 5-prime G. We see that there would be a clash with the protein at that position where the 2-amino group is in the minor groove. How can we potentially overcome this?
Well, a very talented graduate student in my lab, Erin Dougherty, who's now a post-doc with Jennifer Doudna at Berkeley. What she did when she was in my lab is she actually explored different base pairing partners opposite that 5-prime G for both ADAR2 and ADAR1-P110. Found that for ADAR2, there was clearly a preference to have a G opposite the 5-prime G. For ADAR1, both A and G enhanced editing. Okay? Clearly better than the GC pair that's shown here in green. There's something going on with the purine mismatches, if you will. Okay? This is also shown in a different substrate for ADAR2, that the GG pair seemed to be beneficial. What's going on there? Well, we turn again, to crystallography, right?
We generated a substrate that had a GG pair next to an editing site, solved the crystal structure and found that that GG pair was in a very specific hydrogen bonding pair geometry. Okay? The five-prime G, the G of the substrate strand is now no longer in its normal anti-conformation, but it has flipped into what's referred to as a syn conformation. It's presenting what's referred to its Hoogsteen edge to the base in the guiding strand, which is a G, to form a specific pairing partner. Okay? How does that help? Well, that takes that amino group that was clashing in the minor groove, right, and puts it into the major groove.
That's actually shown, I think, more clearly in this bottom chemical structure formula here, where you see the two amino group in the minor groove in a normal GC pair. When you have a G syn, G anti-pairing interaction, that two amino group is moved into the major groove and no longer clashes. However, we still have a two amino group from the G that is in our guiding strand, and we see that it's very close to G 489. Okay? Is there a way that we could potentially overcome that effect or improve it even further, I would say. Okay. I told you about this G anti, G syn pair, but recall that for ADAR1, the A also enhanced editing, right? What could be going on when there's an A paired with the G?
It turns out that there are structures of RNAs that have AG pairs, and the G is in the syn conformation, but the A is protonated to form the same sort of pairing interaction. Right? The pKa of A is around 3.7. Protonation at that site, right, will be unfavorable at physiological pH. If you dig into the literature for nucleoside analogs, it's well established that 3-deaza-A has a pKa of around 6.8. It's much easier to protonate that position than at physiological pH. This suggested to us that if we used 3-deaza-A, we may be able to form the same sort of pairing interaction and potentially be able to improve editing even further. Okay?
We see that, in fact, when we compare A and G to 3-deaza-A, this is in vitro kinetic analysis. This is with ADAR1. We see an improvement in the catalytic rate of editing when we have 3-deaza-A paired with G. Okay? We then went and solved the crystal structure of 3-deaza-A paired with a 5-prime G, and it pairs as we had predicted, right? Where the 5-prime G is in a syn conformation. The 3-deaza-A is base paired with that in a very nice pairing interaction. We also see additional stabilization of the 2-amino group hydrogen bonded to its own phosphate.
Importantly, when we overlay now the structure with 3-deaza-A paired with G to the optimal five prime nearest neighbor, the one that has the five prime U, we see nearly perfect overlap. Okay? Using 3-deaza-A to pair with G has created a structure that now to ADAR looks almost like the optimal five prime nearest neighbor, and we see this in improved rates of deamination. Okay. Again, what we've done there is we've used the structural studies, right, to generate hypotheses about how the guiding strand could be optimized by using chemical changes to the structure of the components of the guiding strand. This, I think, is an exciting opportunity to continue to develop these guide strands for directed RNA editing, and we're continuing to do this in collaboration with ProQR.
To summarize, I told you that ADAR-mediated RNA editing is capable of rewriting genetic information, importantly at the RNA level in a reversible manner. Human ADARs have important roles in neurobiology, innate immunity, and cancer. Okay. Synthetic oligonucleotides, we refer to those as EONs or editing oligonucleotides, can be used to direct ADARs to make corrective edits. And we're using high-resolution structures of ADAR RNA complexes along with rigorous biochemical analysis, where we're actually measuring rates of deamination with purified proteins and substrates of varying structure and sequence. This has enabled optimization of EONs by rational design, and I showed you 2 examples of that where we worked together with ProQR. Z at the orphan position, and then 3-deaza-A at the -1 position. Okay.
We continue to do this at other positions along the guiding strand, different positions along the guiding strand, and also for different target sequences. Okay. I do want to acknowledge the people in my lab who do this work. I have a fantastic group of graduate students and postdocs here at UC Davis. have benefited from collaborations around the world, including with scientists at ProQR. I wanna acknowledge the funding sources for work that we do on ADAR. Thank you again for the opportunity to present this work.
Thanks, Dr. Beal, for a really excellent overview on ADAR science and ADAR-mediated RNA editing. I will now go into more details on our Axiomer platform, including some of the data that we've generated over the last months. This section does contain data generated in-house but also in collaboration, explaining the undisclosed targets in some of the following slides. As mentioned by Peter during his presentation and on the video you watched earlier, ADAR enzymes act on duplex RNAs, so referenced as double-stranded RNAs. By learning from this natural process, we have developed our editing oligonucleotides, EONs, with the objective to mimic that. We looked very carefully what nature has given us.
Editing oligonucleotides consist of short, single-stranded RNA, and I will get back to that a little bit later, that are actually complementary to target RNA, and the target RNA is indeed also a single-stranded molecule. By binding it, the editing oligonucleotides create a double-stranded structure which is resembling what you find in nature, and then that will attract ADARs and allow the A-to-I editing to be performed. Our editing oligonucleotides are optimized to obtain very precise editing on target RNA, and I'm going to show you how in the next few slides. On this slide, you can see that the EON sequence defines the target RNA binding and provides stability. Our molecules are around 25 to 30 nucleotides, and this size makes the molecules a developable entity, so to speak. The two specific regions we can recognize in the EONs are described as follows.
We have the ADAR binding domain, or ABR, and that backbone modifications enable ADAR binding and actually disable off-target editing. The second region we have is the editing-enabling region, or EER, and that actually is a very short region of about three nucleotides, and this is where ADAR is to achieve very specific A-to-I editing. This region actually is designed to push the target adenosine into the catalytic domain of the enzyme and allowing the enzyme to perform its task. Over the years, we've developed our expertise and now very much understand how it's needed and what is needed to optimize the sequences that we are now using. On this slide, you can actually see the advancements that was made over the last 8 years in trying to understand and delineate the ground rules that are needed to design these molecules. Together with Dr.
Beal at UC Davis, we worked hard to understand the specific components that are needed and that are allowed by ADAR to be did, put into these molecules and to optimize the performance of such molecules. We changed the bases, the ribose modifications, and the linkages, as you can see over here, and put in a lot of modifications in order to improve the pharmacodynamics, but also the kinetics of the molecules that we are using right now. This work led to our deep knowledge of the molecules that we are now designing and using for our development programs, but also led to a portfolio of our 10 foundational patent families, as you can see over here.
In short, in my presentation, I will actually show that we have consistent RNA editing over all the models that we've tested in the nervous system and liver, including non-human primate studies in vivo. I can show that we have used GalNAc to actually target the EONs to the liver, specifically to the hepatocytes, which leads to increased editing, that will be used in our subsequent development as well. By showing different applications of the technology, I'm going to show you the broad applicability of our Axiomer platform as well. In order to establish a strong platform, we have established multiple cell systems and test systems in which we can test the performance of oligonucleotides. We have been focusing to set the system up in the nervous system and the liver-originated cells.
In as such, we've developed different cell models, more complex models like organoids, and then the in vivo models as well. The cell models as such we can use for screening a large number of molecules to optimize the performance. The organoid systems, which are more complex, we are going to be using to actually test the effect of such variants in the actual situation of the disease. We are going to be using the mice or non-human primates to study the performance in vivo. Uptake, effectivity, and safety will be part of those studies as well. Let's dive into the CNS first. Here you can see that looking at the neural models of neuronal origin, we've tested several of these EONs for their performance of editing.
You can see on the left, we by genetic uptake, tested a EON targeting ACTB, and you can see that we have a nice editing efficacy of over 50%. We took those learnings and we targeted a specific gene that is expressed in neuronal cells, APP or amyloid precursor protein, and we were able to get over 20% editing as well. It actually shows that in these two systems, or in the neuronal cells, the editing machinery is present and available and capable of performing the editing reaction. Next, we looked into co-more complex molecule systems, the cerebral organoids, and that's shown on the next slide over here.
Here you can actually see that we were able to create and culture three-dimensional structures containing a lot of different cells of the brain, and mimicking thereby the real conditions and cell-to-cell interactions. In fact, you can actually distinguish certain cells from the brain regions that they were derived from. The experiment actually called for a genetic uptake, as we've done with cultured cells. What you can actually see on the right panel is that we were able to get very high level of editing using ACTB, but also the APP gave us 65% editing in the cerebral organoids. This allows us to study the processes in, let's say, ex vivo circumstances much better. With that knowledge, we started to study the reaction in vivo. We scaled up to vivo models at first in mice.
For this experiment in mice, we were able to dose these animals with a single dose through ICV injection. After four weeks, we were able to show that we have extremely high editing in certain brain regions of up to 40%. Also, moreover, we could show that this 40% editing resulted in a 26-fold change in protein functionality, which is actually a very nice proof that we can actually correct the G to A mutation. Following results in the mice, we took the study into the non-human primates, as you can see over here. What we've done is we dosed these animals with an integral injection, and we tested the editing after seven days using dPCR.
As you can see on the right, we have in the spine region up to 50% editing, as you can see over here. If you analyze the brains of these animals, we can actually see up to 30% editing, which is extremely promising for applications in the nervous system as we move forward. In summary, I think we've shown consistent editing of around 50% in all the cell models, organoid models, and the vivo studies that we've done, which has a huge impact on our way of thinking of using this technology and moving forward into the development. As we've indicated before, we are going to be focusing on the liver for our internal pipeline development. Being at the origin of multiple diseases, the liver is really a target organ with a high potential to target with these editing therapies.
In these following slides, I will take you through some of the data that we've generated in the different cell models, organoids, in vivo, as we've done for the nervous system approach as well. We selected several cell models for the first step in the selection and the optimization of the Axiomer, but mainly focusing on primary human hepatocytes, as you can see over here. On the left, you can see the experiment where we took the housekeeping gene ACTB, genetically with our cells, and you can see that we have over 60%-70% editing, and that's something that really shows that the editing machinery is really present in the hepatocytes and able to perform the reaction.
We took those learnings into a cell line that harbors the mutation that causes alpha-1 antitrypsin, as you can see on the right, and we were able to generate an EON which resulted in a very robust, more than 50% editing. That allows you to actually translate the learnings from the housekeeping gene into a clinically relevant target, and you can see that we have in both situations, high level editing obtained. A next step, we turn to Liver Microtissues, because those are more complex organoids that allow us to show effect of certain treatments in more relevant models. You can see that the LMTs, as we call them, are a co-culture of primary hepatocytes, Kupffer cells, and liver endothelial cells in a three-dimensional form, which is then forming a spheroid.
On the left, you see a very nice live image of an LMT stained, and you can actually see some of the more complex structures starting to form after a certain point in time, like the presence of bile channels, using specific stainings. That actually shows you, and we'll use that in our later development program, that certain essential processes can be mimicked in these very complex cultures, as you can see over here. We tested the editing potential of such LMTs on the right and incubating them with the housekeeping gene for actin B. We are showing very nice editing that increases over time, showing the robustness of editing up to 40%, but also showing the endurance and the duration of the reaction as well.
This model will allow us to really study processes that are important for product development later down the line. Having learned all of this, we took our lessons and then created data in vivo using wild type mice, as you can see over here. In this particular case, we dosed these animals subQ with a five times 10 mg/kg dose and tested the editing performance at day seven after the last dose. As you can see here in wild type mice, so no additional ADAR expression, we find over 40% and even up to 50% editing, showing that our EONs actually can achieve high double-digit editing in vivo. In summary, we are again showing very consistent editing in the liver in these three different models with approximately 50% editing, including in vivo results.
That is a very nice basis to take our developments and learnings into the next step into development. As the liver will be our target organ for us in our product development, we will use a proven modality called GalNAc to deliver our EONs to the hepatocytes in the liver. GalNAc is really a g-conjugate which is attached to the EONs and allow us to increase cell uptake, specifically hepatocytes, by specifically targeting a receptor which is called the asialoglycoprotein receptor, which is found on the surface of hepatocytes. I'm going to show you a little bit more about the effect of such conjugation in our models. On the left, you can see the biochemical editing assay in which we test the capability of the EONs to edit at a certain position by incubating the EON with the target RNA in combination with the ADAR molecule.
We can test the A to I performance of such molecules. You can see also that the addition of GalNAc to such EON does not interfere with the reaction at all, as you can see on the left. We concluded that ADAR does allow for GalNAc to be present on the molecules. We took those lessons into a vivo study, as you can see on the right, and we tested a GalNAc EON combination and compared to a EON without a GalNAc, as you can see over here. We dosed these animals four times with 10 milligrams per kilogram, and then after 14 days, we tested the editing potential. As you can see, we observed a 10-fold increase after using the GalNAc conjugate, as you can see over here.
This reiterates and shows that the GalNAc can be actually used to enhance the uptake and efficacy of the EONs that we are using right now, and we will use that going forward in our development. On this slide, I would really like to summarize what we've just seen in the slides that I've presented. Overall, we can report consistent RNA editing in all the models that we've tested. As you can see on the left, we have up to 40% editing in the CNS of mouse in vivo. We have achieved 50% editing in the liver as well in vivo. Of course, last but certainly not least, we have about 50% editing reported in the nervous system of non-human primates, which bodes well for the next steps in our development.
In addition to that, we've tested and shown that the use of GalNAc can really help us to efficiently target our EONs to the hepatocytes, which are the appropriate cells for our further development into liver targets. Now that we've looked at the Axiomer proof of concept data in general and the promising results that we've obtained there, we form the basis to look further now to focus on the broad therapeutic potential of our platform. On this slide, you can actually see how we vision the use of Axiomer to develop a novel class of medicines in different therapeutic areas. On the left side, you can actually see that we can actually correct mutations. There are many, many G to A mutations in the literature that we can actually use the Axiomer to develop a therapeutic to correct those.
Maybe even more excitingly, on the right you can see protein modulation, because that's a very broad field that we can change the properties of specific proteins with the purpose of developing therapeutics. For instance, we can alter protein function to, or to include protective variants to actually show and achieve loss or gain of function proteins helping to address or prevent disease. On the middle, you can see that we can actually interfere with post-translational modifications like glycosylation, phosphorylation, what have you. Those are important to regulate protein activity, preventing immune escape or slowing down degradation as well. On the right, we can choose to use the technology to change protein interaction, and that actually has a bearing to changing localization, folding, protein function, and what have you.
Specifically, I'm going to give you some examples of all of these applications on the next slide. Showing this would actually allow us to think about using the technology into broad therapeutic areas like common diseases, also rare diseases, but also to target a wide variety of target organs, so to speak. Here I would like to show you two examples of G to A mutation correction, which were done in our labs. On the left, you can see actually we picked a gene which is mutated in the brain, and as I can remember, I can remind you of our previous result, that we were able that after a single injections, we're able to get a 40% editing level in the brain.
On this particular slide, you can see that after that, so four weeks after the single dose, we were able to correct the protein's function, up to 26-fold, and that is something that is very impressive. On the right panel, you see another example of a G to A correction, and that is in the SEPT9 gene. In organoids that we cultured, we were able that after two weeks of culture, after a single treatment, we were able to get a very significant correction of the protein localization and function again, as you can see over here. We have twfour examples of G to A mutation corrections that we can actually show protein correction, as I, as you can see.
The second example that I would like to give you that we are able to include a protective variant into the target messenger RNA that alters the function of a protein. That's something that is shown over here, where we use PCSK9 as our model. Pro-PCSK9 use autocleavage at position 152 to mature into PCSK9. In patients, PCSK9 increases LDL cholesterol, so inhibition of autocleavage would lower PCSK9, resulting in a lower cholesterol load. We turned to genetics and found a family with a very low level of PCSK9 and a very low level of cholesterol. Very healthy otherwise. They have this very specific mutation at the autocleavage site, which actually result in a lower PCSK9 load in their bloods.
Thinking about that, we designed some of these EON that could take care of a specific amino acid change in this autocleavage site, and the result is shown on this particular site. On the left, you can actually see that after the editing reaction in these cells, we cannot go up to 25% editing in this particular experiment using our dPCR. That in turn resulted on the right panel in about 80% reduction of PCSK9 secretion, and also led to a shift in ratio from pro to uncleaved PCSK9, as you can see over here. It changed from 70/30 to up to 25%-75% in treated samples. The inability to undergo autocleavage likely retained the proenzyme in the endoplasmic reticulum, where it acts as a dominant negative protein, preventing the exit of mature PCSK9.
This actually shows that with a change in an amino acid, you can interfere with the progression and maturation of a specific protein. That is one of the second bucket. The third bucket is really to show that we can interfere with post-translational modifications. This particular example that I'll show you here is that post-translational modifications are really very much important for function of the protein, such as stability, localization, and certain ion channels depend on it, whether they can open or close, depending on the editing or phosphorylation state. Protein phosphorylation or dephosphorylation is a PTM consisting of the addition or removal of phosphate groups to specific amino acids which reside on proteins. We selected a non-disclosed protein which is phosphorylated for its function, and we designed some of the EONs to actually alter the phosphorylation site.
After the editing, we could show, this example shows you the ratio between phosphorylation and non-phosphorylation, that we had a very significant reduction of over 25% of phosphorylation, which really opens up the way to manipulate the function of these proteins after treatment. We will study that procedure much more. We are going to be updating you as we go forward. Actually, this is the second example of a protein modulation as we can perform on these particular proteins. The last example that I'm going to give you is really to change a protein interaction with another ligand. In this particular example, I'm going to tell you about protein interaction. Protein-protein interaction or protein other ligand interactions are very common and important for function of certain processes.
We select an example of such interaction. ANGPTL3 is an inhibitor of lipoprotein lipase and requires heparin binding for its function. We found in genetics, in human genetics, specific variants that have certain SNPs or variants in the heparin binding domain of ANGPTL3. ANGPTL3 with such variants actually leads to activation of LPL because the fact that heparin binding is impaired in those variants. We designed EONs to mimic this change in the heparin binding domain and tested that. On this slide, you can see the result. After editing the ANGPTL3 with a change in the heparin binding domain, we could actually show an 80% decrease in heparin binding.
Moreover, it seems that lower level editing in this particular case did not result in dominant decrease in heparin binding, but a higher degree of editing could actually show an 80% decrease in heparin binding. Very promising for the next step. With this example, I could show that by changing a specific, very important interaction, we could change the functionality of the protein as well. In summary, we think that the platform has broad applicability. We show that the potential is there to correct G to A mutations, and there are many monogenic diseases that are caused by this particular mutation, allowing just to use a G to A correction to restore the wild type protein again.
We have discovered the potential to modulate protein function, acting to include a protective variant or to change the post-translational modification, also being able to change protein interactions with other ligands to create a phenotype in these proteins. We feel that all of these examples lead to a broad therapeutic potential of the platform and will allow us to develop the products for rare and more common diseases. Okay, with this presentation, I hope that we've shown the consistent editing of all the platforms that we tested in the nervous system and in the liver, including non-human primate data. We were able to show that addition of GalNAc allows us to efficiently target the EONs to the hepatocyte and increase the editing efficiency.
By showing the different examples of the application, we feel that we have a very broad applicability of our technology, and that will provide us with a means to develop new products into rare and more common diseases as well. Thank you very much for your attention. With that, I'm going to hand over to René for the IP overview and partnering strategy.
Thank you, Gerard, for the deep dive on the progress that we're making with our Axiomer platform. I will give you a short break and a breather. Steve Jobs was famous for many reasons, one being his practice of one more thing. He would often wait until the end of a presentation to share the one more thing, which was often a big announcement. Today, the one more thing I would like to share with you, I will not save for the end of my presentation, but say it now. If you want to use an oligo to recruit ADAR, ProQR is the place to be. We have the science, and importantly, as I will highlight, we have the IP and a partnership strategy where we are open for business as points of proof for our leadership in RNA editing.
We have been building an IP portfolio since we began working on Axiomer in 2014. Therefore, we are confident that we have a leading IP position to support our proprietary ADAR-mediating RNA editing platform technology. Back in 2014, I shared an office with Gerard and Bart Klein, who is today our Senior Vice President of Axiomer Strategy. Bart is on top of being a seasoned patent attorney, also a brilliant scientist, and wrote our first patent application in 2014 in my presence. This was the basis of our IP estate, which protects all the foundational elements of the platform using endogenous ADAR for therapeutic purposes beyond 2040. Initially, Bart and Gerard came up with the idea to use natural known ADAR attractors to attach to oligonucleotides in the form of stem loop structures. This appeared to work just fine and could indeed recruit endogenous ADAR.
However, these oligos were still rather large. In 2016, Bart worked together with a few other ProQR inventors to achieve the same recruitment of endogenous ADAR, without the need of a stem loop structure. This allowed an even better opportunity to use the oligos for clinical purposes, simply because they were much smaller. These are the oligos that we are developing today and have a solid IP position on to back up this early work. Patents have been granted in the major jurisdictions such as the United States of America and Europe for both types of technologies as indicated in the last column. The others are in various stages of examination and are expected to be granted in due course. In summary, we have 10 published patent families currently comprising of 22 patents. We are working hard, very hard.
We continue to invest in and expand our IP estate to maintain our leading position. We recently announced the successful defense of a key Axiomer patent protecting ADAR-mediated RNA editing. The oppositions were filed in February 2021 with the European Patent Office by two separate straw men against our granted patents, which is related to targeted RNA editing using endogenous ADARs. The opposition division of the EPO held a public hearing on March 7 and 8 and ruled in favor of our position after minor amendments of the main claim and one dependent claim. These non-material modifications didn't weaken the claim in the patent at all. These claims were amended such that the oligos were limited to being chemically modified. In general terms, not specific to chemistries, to be completely novel over the prior art. This successful defense confirms that our IP is likely seen as problematic to others.
Why otherwise would multiple parties try to oppose a patent position spending valuable time and money? Given the importance and value of our intellectual property estate, we intend to continue to defend against such challenges and remain confident in our leading position. One more thing, ProQR's Axiomer-leading IP portfolio is robust, broad, and provides coverage for the fundamental features of the technologies beyond 2040. We now turn to our partnership strategy. We believe our IP, along with our deep expertise in RNA editing and oligotherapeutic approaches, is a source of great value to potential partners. For ProQR, a key part of our strategy is the ability to selectively form partnerships. They bring important resources, capabilities, and funding to our to further enhance our programs.
They also advance applications of the platform beyond the targets we intend to develop independently, creating a new class of medicine based on Axiomer. We closed our first Axiomer partnership with Eli Lilly in September 2021, expanding it last December, bringing the total potential value to approximately $3.9 billion plus royalties. The expanded collaboration is important validation of our leadership in ADAR-mediated RNA editing, our robust IP, and the potential of a broad applicability of the Axiomer platform. Lilly has the option to expand the partnership further with another five targets for an additional $50 million opt-in fee to reach a total 15 targets. I would say this is a very, very pleasant sort of Damocles dangling above our heads. We look forward continuing to progress the productive partnership with Eli Lilly, improving the lives of patients together.
Now a few words on our ophthalmology programs, which do not utilize Axiomer. We are encouraged by the progress that we're making together with Lazard to identify a partner who can continue to advance these programs. I have great expectations finding new homes for these assets. Of course, when there is an update, we will certainly share this with you. Having said all that, beyond our partnership with Eli Lilly, we do have appetite and capacity to further selectively partner our Axiomer RNA editing platform technology. With regard to that, we are seeking research collaboration partnership that could take the form of single or multiple asset licensing agreements or broader therapeutic or organ-specific transactions. We are right now focused on dialogues around liver, extrahepatic, and metabolic disorders, as well as rare neurological diseases with an unmet medical need.
The potential is broad, extremely broad, as highlighted on this slide. We continue to be motivated to execute value-creating transactions, harnessing ProQR's leadership position in precise RNA editing and endogenous ADAR recruitment. Given the vast potential of Axiomer to target a broad range of diseases, we look forward keeping you abreast and updated on developments with our partnership strategy. Before I turn the presentation back to Gerard, who will present our pipeline overview, I would like to highlight that the shop was open for business during renovation last year. With Sarah, Daniel, and myself, we have had since I rejoined in June last year, almost weekly investor calls and visits. I promise we intend to continue to do so.
I realize that there's a lot to take in from today's presentations, and we are happy, very happy and honored to take continuous one-on-one meetings, whether virtual or face-to-face, to further elaborate and interact with you and present ProQR as an investment case. As a matter of fact, we are in New York in the first week of April. If you're interested, please reach out to Sarah. With that, and without further ado, I would like to give the floor back to Gerard.
Okay. Thank you, René. I think we will now focus on the in-house pipeline development for ProQR. In the next slide, I will take you through the first targets that we will be developing going forward. As can be seen in the previous sections, ADAR editing is a very active mechanism which is widely present in the whole body. As such, Axiomer platform really takes advantage of the natural activity and has a broad applicability basically to, with the possibility to target multiple organ systems, as you can see over here. For our initial pipeline products, we are going to be focusing on the liver, and we feel that the liver is very attractive as a target organ for Axiomer for really multiple reasons, which are shown on this slide.
A very obvious one is the high editing potential in the liver, and I hope I've convinced you of that in my platform slides previously. We can actually enhance the liver targeting by using the GalNAc technology, which has been proven in previous developments for using oligonucleotides as well. 85% in the liver consists of hepatocytes, and they have the asialoglycoprotein receptor, which is targeted by using GalNAc. In addition to that, we've shown that GalNAc does not interfere with the function of ADAR itself. In drug development itself, biomarkers are essential to understand target engagement, and also to assess efficacy and safety as well.
The presence of biomarkers as such allows us to monitor and see the target engagement, that we will have with our Axiomer technology, and that's what we'll be using moving forward into development as well. The liver is at the center of a high metabolic activities, and it does influence other target organs. The high metabolic activities, as you can see, involve bile acid production and excretion, lipid production, and plasma protein synthesis, to name a few. There are many diseases that originate from the liver, and by delivering medicine to the liver, we can target multiple organs, as you can see over here. The numerous liver-originated diseases are listed over here, and to name a few are cholestatic diseases, metabolic disorders, or cardiovascular diseases or CVDs.
As Daniel mentioned before, our rationale is to go after new targets guided by human genetics. I'm going to show you details of that in the two therapeutic areas that we've decided to pursue. I'm going to walk you through the rationale of the initial targets and give you some more information on the disabling diseases that we plan to investigate. Let me turn to AX-0810 , which is our first pipeline program. It's designed to target cholestatic diseases. In short, when there is an excessive accumulation of bile acids in the liver, that can lead to cell stress and damage. We call those diseases cholestatic diseases. On the left, you can see the liver cells that are actually at the center of bile acid production. Bile acids are derived from cholesterol metabolism.
Once the bile acids are produced, they are stored in the gallbladder, but excreted from the liver to the intestines. There, the bile acids facilitate the digestion of fat, nutrients for growth, fat-soluble vitamins, and carry away waste. Most of the bile acid, some 95%, return into the liver via reuptake from the portal vein. On the right, you can see that individuals with cholestatic diseases actually have a dysfunctional bile duct that causes bile acid accumulation in the liver. The buildup of bile acid is really toxic and leads to inflammation and liver damage. On this particular slide, you can see that without treatment, the accumulation of bile acid in cholestatic disease leads to progressive liver degeneration, starting with liver inflammation, progressing to fibrosis, and then unfortunately leading to liver failure, malignancies, and poor life prognosis.
Out of all the known cholestatic diseases, two of them are leading conditions for liver transplantation in adult and pediatric, well, population. Primary sclerosing cholangitis or PSC and biliary atresia or BA. Here you can see the high unmet medical need, which still remains in PSC and BA. PSC is a condition that is being diagnosed at the age of 30-40 years with a higher prevalence in men. There are about 80,000 patients in the North Americas, Europe, and Japan, and it's a autoimmune liver disorder leading to bile duct stricture. BA, on the other hand, is a pediatric condition affecting about 24,000 individuals, and it's very rapidly diagnosed in the first weeks of life and caused by congenital absence or defective bile ducts.
Both of the conditions are associated with symptoms that have a huge impact on quality of life, such as pruritus, fatigue, poor weight gain or weight loss. Of course, they have a rapid progression into cirrhosis and liver failure. Both PSC and BA have a high unmet medical need since there are no approved therapies available at this time. For PSC, the only approach currently available is liver transplant. As such, PSC is known as the leading autoimmune indication for liver transplant in the adult population. Unfortunately, 20%-40% of the patients present PSC recurrence after liver transplant.
For BA, newborn babies can undergo a very complex surgery, which is called hepatoportoenterostomy or HPE, for which a surgeon removes the extrahepatic duct bile duct, excuse me, for the infant, and links the liver directly to the intestine to remove the bile's acids. Post-operation procedure, a majority of children will eventually develop portal hypertension and require liver transplant before reaching adulthood. On this particular slide, I'd like to present with you the solution that we have in mind. As I've mentioned previously, we learned from human genetics to inform our pipeline development. I turn to sodium taurocholate co-transporting polypeptide, also known as NTCP, which is actually the main transporter involved in bile acid reuptake into the liver.
As cholestatic diseases are caused by accumulation of bile acids in the liver, altering the reuptake could have a positive impact on the course of such condition. Interestingly, it has been reported that loss of function variants of NTCP naturally occur in some people with only a mild phenotype without any comorbidities. Furthermore, pharmacological modulation of NTCP has been shown to improve outcomes in a mouse model of cholestasis, reducing inflammation and liver damage. We have consolidated our learnings from scientific data and human genetics into a novel therapeutic. AX-0810 targeting NTCP is designed to reduce bile acid reuptake into the liver.
It's an editing oligonucleotide that will attach or bind to the NTCP messenger RNA and will create, after editing, a loss of function variant that will be no longer be able to perform the task of reuptake the bile acids into the hepatocytes. On the right, you can see that as such, AX-0810 aims to reduce the bile acid reuptake and reduce the bile acid load into the liver. With this program, our objective is to alleviate symptoms in PSC and BA, and to prevent or delay the progression of liver damage and need for liver transplant. To ensure success of this program, we have a rigorous and pragmatic approach that we think will increase our chances of success to commercialization. On the left, you can see that we have a robust preclinical model package available with the possibility of assessing proof of mechanism early in models.
We have the complex organoid model that actually has all the machinery available to study bile acid synthesis and uptake and reuptake in our in culture, as well as animal models that have bile duct structure conditions as well. The program does allow to a early start clinical trial in healthy volunteers. Such, we can study that timely insight in safety and target engagement with validated biomarkers that we'll use. For instance, we will be looking at bile acids in serum, liver enzymes like ALP, amongst others. Of course, we have a plan moving forward into later stage clinical trial with disease-specific endpoints and a clear regulatory pathway.
As mentioned on the previous slide, there is the possibility of generating early clinical insights via phase 1 in healthy volunteers with the objective to assess safety, tolerability, and pharmacokinetics and dynamics of our molecule without actually having being bothered by the concomitant pathological conditions. We also plan to measure RNA editing when circulating exosomes in plasma that will allow us to get an early insight in the target engagement of our EONs. The aim is to develop 1 clinical trial to ensure timely recruitment and data generation. The trial design will include single and multiple dose ascending cohorts and targeted entry into the clinic by late 2024 or early 2025. In summary, AX-0810 is our first pipeline program that's being developed to high unmet medical need in cholestatic diseases, with PSC and BA being the lead indications.
It's a novel and on-target approach originating from human genetics and with the objective to reduce bile acid reuptake into the liver through a NTCP loss of function variant. We have a rigorous approach to increase the probability of success from preclinical to clinical stage, including use of biomarkers which are validated and a clear regulatory pathway. The next steps will be the generation and selection of our lead candidates, for which the data will be presented in due course at scientific conferences. We anticipate entry of clinical trials late 2024 or early 2025. Like I said, Axiomer can address numerous liver-originated diseases, and we'll be looking now into our second pipeline program, as you can see over here.
This example is a clear example of Axiomer's broad development potential by targeting the B4GALT1 protein, which is highly present in the liver and can have a positive impact on risk related to cardiovascular diseases. Okay, we have selected AX-1412 as our second pipeline product to address the remaining needs to reduce residual risk factors for cardiovascular diseases. We aim to develop this program up to the early clinical stage, and after the first clinical data has been obtained, we will partner with another company for further development. Cardiovascular diseases or CVDs are really a group of health conditions that affect the heart and blood vessels, such as atherosclerosis, which can lead to severe problems like heart attacks, heart failure and stroke and are the leading cause of disability and death globally, becoming a significant health issue.
Shockingly, there are around 80 million people dying from CVDs each year, making up 30% of all deaths globally, according to a report by the WHO in 2021. In the United States alone, the American Heart Association estimates that by 2035, more than 130 million adults will have some form of CVD. Despite the standards of care, the medical need remains. Even with existing therapies, less than 35% of Americans with high LDL cholesterol levels reach their target levels recommended by guidelines. CVD events still occur even when LDL cholesterol levels meet clinical goals. Many patients also struggle to continue taking their meds 2 years after a CVD event. In addition to that, about 5%-10% patients cannot tolerate high doses of, for instance, statins, primarily due to muscle aches.
LDL, there are certain independent risk factors involved in CVDs that may have a negative synergistic effect. LDL cholesterol is reported as a major risk factor for CVD, increasing the risk of arterial plaque formation, as you can see on the left of this cartoon. Another risk factor reported in literature is fibrinogen, and it's increased risk for blood clotting. Furthermore, LDL cholesterol and fibrinogen are two independent risk factors involved in CVD that may have negative synergistic effects on the risk of cardiovascular diseases. LDL cholesterol risk factors is being addressed by current standard of care, but none of the treatments on the market specifically focus to address the both, as you can see over here.
Interestingly, it has been reported in literature, so human genetics, that an old-order Amish-enriched missense variant in a functional B4GALT1 or the beta-galactosyltransferase protein, was associated with lower LDL cholesterol and a lower plasma fibrinogen. B4GALT1 is an enzyme that transfer galactose from uridine diphosphate galactose to specific glycoprotein substrates like apolipoprotein B and fibrinogen. Furthermore, a specific loss-of-function variant in B4GALT1 has been reported to be associated with decreased coronary artery disease in a gene-based analysis. It's learning from that protective variant, which was reported in human genetics that made us decide to develop AX-1412 as an Axiomer, which leads to a loss of function of B4GALT1 as described in literature. The B4GALT1 loss-of-function variant induces a hypogalactosylation of the apolipoprotein B 100 fibrinogen and limit their negative effects.
It is a novel and unique approach which can address simultaneously two cardiovascular risk factors. That's really not suitable for knockdown technologies as shown in literature, leads to semi-lethality and a severe developmental abnormality in mouse studies. AX- 1412 can lower LDL cholesterol and fibrinogen levels to reduce residual risk in cardiovascular diseases and prevent or delay the development of cardiovascular events. As presented with our first pipeline project, our development approach for the second program also meets all the thresholds we've put to increase probability of success, including a very strong preclinical model package available with good translatability to clinic, a timely early-stage clinical phase 1 that will give us some initial insights into the clinical effect. We're going to go after conditions with disease-specific endpoints and a good guidance from regulatory agencies.
Okay, as mentioned in the previous slide, there is the possibility, like for our first pipeline project, to generate early clinical insights via a phase one trial in healthy volunteers. This with the object to assess safety, tolerability, pharmacodynamics, and kinetics of our program, AX-1412 . This without interference of co-committed pathological conditions and to investigate target engagement and PD effects through specific biomarkers. We will also plan to measure RNA editing through exosome analysis. That will give us a very early insight into target engagement at the RNA level as well. We aim to have one clinical trial to ensure timely recruitment and data generation. The trial design will also include single and multiple dose ascending cohorts and target an entry into the clinic in late 2024 or early 2025.
In this summary slide, we're in the next steps for our program. I can reiterate that although there are several approaches to lower the risk of cardiovascular disease, including reducing LDL cholesterol and ApoB-100 levels. Reducing fibrinogen levels may offer additional benefits to patients with unmet medical needs in this large population. AX -1412 is a novel and unique approach originated from human genetics that could have pleiotropic effect for cardiovascular protection and is really not suitable for knockdown technologies. As our second pipeline program, we have a rigorous approach to increase probability of success from preclinical to the clinical stage, including the use of our validated biomarkers, and next steps will also include the generation and selection of a lead molecule for which data will be presented in due course at scientific conferences.
We aim to enter the clinic by late 2024 or early 25 with this program. In summary, I'm very excited to have announced our two development programs, and we will pursue to develop liver-originated disorders for these programs as we move forward. We feel that the broad applicability of our Axiomer platform does not limit us to target one organ only, and we aim over the next years to broaden our pipeline, targeting metabolic conditions, rare neuro disorders, and potentially other conditions that we will not disclose today. We do have a very dynamic, thorough, and robust capacity to find new targets suitable for Axiomer, and we will continue to push the boundaries to develop new therapies using our technology as we are just scratching the surface of our Axiomer potential and are at the beginning of a new era, the Axiomer era.
I will now hand it over to Daniel for the summary and milestone section. Thank you very much for your attention.
Thank you, Gerard. We're confident that we have put ProQR in a strong position to execute on the business plan to capture the full value of the Axiomer opportunity. We believe this approach holds great promise, and we see RNA editing as the next evolution in RNA therapies. Axiomer can contribute by creating a new class of medicines with broad therapeutic potential. Today, we announced AX-0810 for cholestatic diseases targeting the NTCP channel and AX-1412 for cardiovascular diseases, targeting the B4GALT1 gene as our initial pipeline programs.
These programs, along with the other targets in our pipeline, share several key features: a deep rooting in human genetics, the potential to have a major impact in a high unmet medical need, the ability to leverage the existing proven delivery technology, the opportunity to establish target engagement via bar-biomarkers in early clinical testing, as well as well-defined clinical endpoints.
We plan to advance these programs into the clinic in late 2024 or early 2025. The platform has shown consistent RNA editing in models in the nervous system and liver, including in vivo data demonstrating up to 40% editing in the nervous system of mice, leading to a 26-fold change in protein function. We've also seen up to 50% editing in the liver of mice and up to 50% editing in the nervous system of non-human primates. These effects were durable, with editing and protein activity reported four weeks following a single injection. We also showed that GalNAc does not interfere with A-to-I editing and that it increased the editing efficiency by 10-fold in vivo.
Axiomer has shown broad applicability and proof of concept by mutation correction with pro-protein recovery, by generation of variants in wild type sequences, also by changing protein function with dominant negative effect, by regulating translational modifications, and by impacting protein-protein interactions. This opens the door to many more applications beyond what we have discussed today. We believe that Axiomer can potentially lead to hundreds of new medicines. As we have seen today, a lot of exciting progress lays ahead in the platform and pipeline. After this update today, we will continue to share more platform data as the time progresses. This will include additional data presentations at scientific conferences and publications in peer-reviewed journals. Furthermore, we plan to share additional platform data over the next 12 months, including further non-human primate data with delivery to hepatocytes in liver.
As we progress, we will also provide further insights in the pipeline and discovery progress to populate rich pipeline that captures the broad Axiomer opportunity. On the pipeline. We will, over the course of the next 18 months, share several updates with translational data on our initial two clinical programs. These will include non-clinical proof-of-concept data updates with supporting these programs, as well as several translational data updates that will help you to understand the translational de-risking strategy into the clinic. Subsequently, we plan to advance our two programs, AX-0810 for cholestatic diseases targeting the NTCP channel and AX-1412 for cardiovascular disease targeting the B4GALT1 gene in late 2024 or early 2025.
We will continue to execute with high priority on our Lilly partnership and work towards the opt-in from Lilly that we anticipate this calendar year, which would come with a $50 million payment to ProQR. We expect to enter into another one or two multi-discovery partnerships as we progress. We will continue to further enforce our IP estate as we plan to preserve and expand our leading IP position. With the cash on hand, we are well-funded to execute on all that we have discussed today with a runway into mid-2026. Today, we have highlighted the progress that we have made to advance our proprietary Axiomer RNA editing platform technology. We also demonstrated its broad applicability and how we plan to develop treatments for patients that suffer from diseases with high unmet medical needs.
We have the science, a platform approach with broad applicability and preclinical proof of concept across multiple models and targets. We're now advancing our initial pipeline programs. Our leading IP estate and our recently expanded partnership with Eli Lilly are important validating points. We have an experienced team executing on this strategy. As a company, we are well-funded with a cash position that funds us into mid-2026 to execute on the Axiomer business plan we have laid out today. ProQR is leading the advancement of the RNA editing field as a new class of therapies for patients. I want to thank you all for joining today to learn about our Axiomer platform and pipeline, and I will now hand the call back to the operator for questions.
Thank you, Daniel. At this time, we'll begin conducting our Q&A session. As a reminder to the audience, you may submit questions through the Q&A text box at the bottom of your webcast player. To our analysts, we remind you to raise your hand to indicate you would like to join the queue. With that, the first question will come from Steven Seedhouse at Raymond James. Steve, please go ahead.
Yeah. Hi, everyone. Thanks so much for hosting the event and for the detailed update, and thanks for taking the questions. I had a few that I'll work through some science ones and then some strategy ones. Just first on the science. I'm curious about the turnover of an exogenous editing oligonucleotide and how you would compare or contrast that with the turnover of RNAi or RNase H, you know, for siRNAs or ASOs. Are there any implications, if there are differences, for the dose that you would need to use, you know, in humans? Is it basically gonna be in the same ballpark as an oligo that would be used for knockdown or splice correction? Thanks.
Hey, Steve, thank you for the question. Thanks for attending the event. We're gonna have the science questions be addressed by Gerard.
Thanks, Steve, for the very good question. I would say that the oligonucleotides that we are developing for our editing reaction, the EONs, are quite similar to the ones that we developed with the full modification for the earlier, let's say, splice modulation. Somewhat different, but still similar. I would say that the actual dosing frequency, we have now data that goes out to, let's say, four weeks and beyond. And the initial experiments that are ongoing are focusing on to see how long the effect lasts. We anticipate a, let's say, a quarterly or maybe a twice yearly dosing that will be envisaged for these, the EONs. I hope that answers your question.
Yeah, thanks. maybe just on cholestatic program, a really interesting approach. You know, there's a few of these diseases that via IBAT inhibitors, you know, reducing bile acids has proven to be effective, some pediatric indications and arguably PBC as well. You've kinda highlighted biliary atresia and PSC, in fact, as maybe the first two indications that you would pursue for that mechanism. I'm curious if that's more of a sort of commercial unmet need decision, and you're sort of anticipating, you know, proof of concept coming pharmacologically in those indications in the coming year or so before you're in the clinic. Or do you think the mechanism is actually better suited for those diseases on a scientific or medical basis?
Right. Again, a good question. I think the focus of our program that will tackle the PSC or BA lies in has a basis that we are going to be targeting the disease on target in the hepatocyte where the primary problem lies. The let's say the faulty bile duct excretion of bile acids and the high influx of bile acids through NTCP is actually causing the problem and then the progression from the liver into let's say the late stage progression into towards the liver transplantation. We feel the IBAT inhibitors target the enterocytes transporters, whereas we are going to be focusing to limit the re-uptake of bile acids in the hepatocytes.
That is a more direct way of dealing with the problem. Looking in human genetics, we found that there are people out there with a loss of function variant of the NTCP that actually gives rise to a lower uptake into the hepatocyte. You know, learning what nature has given us, we are going to apply that to the problem, which we find in the cholestatic diseases.
Could I ask just on this program, if, like, should Lilly, for example, opt in on an additional five targets, could this be among those targets that are partnered with Lilly? Or is this carved out now and going forward as ProQR's wholly owned?
Yeah, Steve, I'll take that question. This is indeed carved out and will be ProQR's program going forward. We are in control about the future of these programs that we have announced.
Great. Last question from me. Just, I was hoping you could comment on just the path that the clinic over the next two years. My guess is the market today is sort of expressing some ambivalence to the timeline of late 2024, early 2025. It might be helpful if you could just provide some context for sort of why that is and, you know, what boxes you need to check and sort of, yeah, maybe when these programs became defined by you as lead programs, and maybe that'll help, you know, just provide some context and, comfort on the timeline. Thanks so much.
Absolutely. Happy to. Obviously we are working on really exciting novel technology that allows us potentially to make dozens, if not hundreds or thousands of different medicines. We're very mindful of the fact that the first few programs carry the burden of successful translation of the wider platform and unlocking that value into the clinic. We're really mindful of not cutting unnecessary corners by really building a solid translational data package that de-risks our steps into the clinic. Which allows us to go into the clinic with high probability of success, as the importance of the validation of this platform in the clinic with these first molecules is so significant.
We are in the next 18-24 months, going through a number of different steps to move these programs towards that first clinical step. That does include additional proof of concept work that we will announce over the next 12 months that will allow us at scientific meetings to share this and provide further context around these programs. Over the next 18 months, we will provide additional translational data that will give further insights in the de-risking translational strategy. So how these molecules are preemptively de-risked to translate into the clinic. And subsequently, we move into the clinic.
I think as Gerard laid out in his slides today, one of the key criteria in our selection of these targets was that we can test these targets in healthy volunteers, which will make the clinical execution much more straightforward in terms of timelines, in terms of execution, in terms of cost. Will give us a data set that will early on in the clinical development, give us high confidence in terms of target engagement of these molecules in human. All designed really to maximize the probability of technical success of translation into the clinic. There's quite a bit of news flow you can expect from that.
Great. Well, thanks for hosting a great event and for the detailed overview.
Thanks.
Thank you for the question, Steve. The next question comes from Jennifer Kim at Cantor. Jennifer, you may unmute your line.
Hey, guys. Thanks for taking my questions. This is very helpful. I have a few questions. Maybe to start off, to follow up on what the last analyst asked. Could you elaborate how soon we could get early clinical data in humans? Also on the path to translational and proof of concept work, I know you said over the next 12 months, but is there a way to think could this come more latter half or could that trickle more into starting into in early 2024? Maybe we could start there.
Jen, thank you for your questions and for attending the event. We're guiding through the next 12 months, but in the next 12 months we're gonna have multiple data disclosures through scientific presentations at scientific conferences. We're gonna do peer-reviewed journal publications. We're gonna be talking a lot about the programs and providing additional insights, both proof of concept data as well as translational data. Yeah, we're not guiding through this year or next year, but it's gonna come over the next 12 months, spread out across multiple different disclosures. Sorry, your other question was?
A follow-up question. I think you talked about being interested in one to two sort of multi-target discovery partnerships as you progress. Do you believe that's a 2023 focus? From your conversations that you've had with potential partners, do you believe that there's anything specific that those potential partners are looking for before pulling the trigger?
Yeah, that's a great question. We are looking at, potentially doing one or two additional business development multi-target discovery deals over the course of this year or next year. We think that, we have the bandwidth to execute on that. Certainly, the platform is broad enough to not cannibalize on the opportunity on the platform, by doing, one or two additional of these transactions. We actually, through the partnership with Lilly, have gained, some really important insights. We've seen how important the partnership has been for us as a company. Yeah, the Lilly team has been great to work with. We, we plan to expand that and to go, further onto the strategic side as well. That could happen over the course of this year.
We obviously, you know, with these things can never firmly guide to these events, but it's certainly our aspiration.
Okay. Then maybe one last question. You've been pretty consistent highlighting GalNAc. Can you remind us of whether or how you're leveraging your partnership with Lilly on that front? Have you made any decisions there?
Yeah, great question, Jennifer Kim. As part of the partnership that we have with Eli Lilly, we are also granted access to their delivery technology. We can, on a case-by-case basis, use specific delivery technology that Eli Lilly developed and validated for programs that we develop in our wholly owned pipelines. That, for example, does include the GalNAc that Eli Lilly uses. We have the possibility to use that selectively for the targets that we bring into the pipeline as we progress forward. We do believe that GalNAc is the way to go. We think GalNAc has been tested and proven and is certainly the most straightforward way to get preferential delivery to the hepatocytes. Likely we will use GalNAc going forwards for the programs translating those into the clinic.
Maybe the Eli Lilly GalNAc, we'll come back to that at a later time point.
Okay, great. Thanks again, guys.
Thank you, Jennifer.
Thank you for the question, Jennifer. The next question comes from Alexandra Van Harten at Kempen. You may unmute your line.
Hey, guys, thank you so much for taking my question. I have some questions, actually. Firstly, you already elaborated a bit on the news flow for the coming year. I was curious, are you also planning on the announcement of additional programs, or do you desire to have a certain cadence of new programs?
Thanks, Alexandra, for the question. Yes, we do have a rich news flow for the next 12 months and beyond, both on the strategic BD side as well as on data, translational data and the pipeline. We may decide over time to advance additional programs going forward, but we're not guiding to that right now. There's two programs that we now plan to use to validate the platform in the clinic, and those will be prioritized, and those will be the first data points to show. In the meantime, we will obviously progress further on the pipeline as it will help us to further build on the internal pipeline as well as on the potential business development front. There is additional progress to be made there.
How that will fit into the news flow, remains to be seen.
All right. Thank you.
Thank you. A second question, right?
Yes. No, thank you very much. It's very clear. I have another question that's more on the scientific side, about the very impressive editing efficiency that you showed for ACTB across in vitro and in vivo models. I was curious, how does this efficiency compare across different genes and specifically your announced target genes? Is there a lot of variability there?
Thanks, Alexandra. That's a great question. I think that we started earlier in during the time that we started to understand what the ground rules were, and we now understand how we design these EONs for for differential editing. The editing of the ACTB, which is a housekeeping gene, is very robust over the different systems, as you've seen, from neuronal cells, hepatocytes, the complex organoid systems that allow us to really study the actual editing in the, let's say, the surroundings that we want to see it in, but also in the vivo situation. When we take these learnings into other genes that I've shown before, we do see that there's a portability in the concept.
We take the learnings that we have from the housekeeping genes, develop our EONs, for instance, for the Alpha-1 or the ANGPTL3, we see robust editing as we move forward. Every target has its own challenges and optimizations needed. All in all, the ground rules that we've taken in are portable from one gene to another. That helps a lot. Hope that answers your question.
Yes, very clear. Thanks a lot. Appreciate it.
Thank you for the questions, Alexandra. The next question comes from Jon Wolleben at JMP.
Hey, guys. Thanks for all the information today. A couple from me. Just wondering about the data you presented pre-clinically from the rodents and the non-human primates. Do you think that these are at, you know, potentially relevant human doses? How should we think about the translation from 40%-50% editing to, you know, a potential benefit? Is that, you know, enough editing efficiency or do you need more? You know, just thinking about how we should translate this from the pre-clinical to the potential clinical data.
Yeah. Thank you, Jon, for the question. Thank you for attending the event. I'll let Pieter address your questions on the science.
We think that the robust 50% editing really is a very good basis to address the, let's say, clinical relevance in the other models that we have. We feel that, you know, we are looking at human genetics to give us guidance on what we need to achieve. We see in the models that we chose, the data shows that heterozygotes, people, you know, which equals, let's say 50% editing already have a clinical benefit. We do also have data showing that, in our models, that less than 50% would actually do the job and give us a good result that we would like to achieve. Anywhere between 25 and 50% would be my target to go after in a relevant program.
That's the current thinking.
That's helpful. One bigger picture, how are you thinking about prioritizing internal programs? I guess the big difference between the liver program and the cardiovascular would be the size of the market and potential patients. What other kind of attributes are you using to strategically pick what stays in and what goes out?
Yeah. Jon, let me comment on that. We've obviously done a lot of work to triage all the opportunity that lays ahead of us with Axiomer. We see so many opportunities how Axiomer can be applied to generate new therapeutics that we developed a very comprehensive system to essentially prioritize these targets, which led to the pipeline that we have announced today.
I think important elements in that systematic approach to us are to prioritize targets that have a deep root in human genetics so that we completely know that the, let's say, post-editing RNA product will have beneficial effects on disease state in patients. Second, for this new technology, we have decided that we wanna really study this mechanism of action isolation of any other variables.
We have the kind of luxury position to be able to use a existing therapeutic modality, and therefore we can use a lot of known moieties for delivery, for example, that allow us to with high probability of confidence, with high confidence, deliver these molecules to the target organ and not have to worry about actual delivery of these molecules, such that we can really look at the therapeutic effect. For that we pick the liver as a target organ, as delivery to liver is pretty much the risk. In the liver we have, I think, the advantage that we can look at a lot of biomarkers in plasma.
That allows us to, at very early on, study target engagement of our molecules and the downstream effect of that in biomarkers. That, I think, provides us a quick path to building confidence in the broader platform and the individual programs. For these two programs, we actually prioritize targets that can be studied in healthy volunteers, because we all know that studies in patients can be very lengthy, can be complex to recruit, can be complex to have a consistent pattern in the data. In healthy controls, there's obviously much less variability and much better to control and execute such study in an efficient way. Those are some of the important matters.
I think for what we will long-term keep, for the in-house pipeline and what will be potentially used for individual product partnerships, largely has to do with the time and cost to get a product developed to market and the ultimate commercial infrastructure that's required to commercialize this product. Looking at the two assets that we today have announced and that will move forward into the clinic, the product for cholestatic diseases, AX-0810 will be prioritized for our in-house pipeline to develop. For AX-1412 , we will likely seek a partner after we have established clinical proof of concept in the first clinical trials, such that we can continue development there with a partner for the future.
I think that pattern you will see throughout the rest of our pipeline as we, you know, as we progress.
Very helpful context. Thanks, Daniel, and congrats on the progress.
Super. Thank you, Jon.
Thanks for the question, Jon. The next question comes from Yigal Nochomovitz from Citi.
Hi, team. This is Ashik on for Yigal Nochomovitz on Citi. Appreciate you taking my questions and very interesting presentation. I just have a few, maybe first a more practical question. It sounds like you're expecting that you'll be able to generate some initial clinical data, at least within your guided cash runway. I'm curious if you can talk a little more about what specifically you mean by proof of concept, meaning data within healthy volunteers, or do you think that timeframe is enough to show POC in natural patients?
Yeah. Ashik, thank you for attending the event today, thank you for this question. We have indeed a pretty robust cash runway. As we have announced with our year-end cash today, we have a runway that is well into 2026, into mid-2026. There's quite a bit of upside to the runway with potentially the $50 million opt-in from Lilly with potentially additional BD deals on the Axiomer platform, with potentially divestment from the ophthalmology portfolio. All non-dilutive ways to help us extend our runway significantly beyond that potential. From a cash runway perspective, we feel pretty comfortable. Currently, we're guiding to the start of these trials in late 2024, early 2025.
With that, you can count on having clinical data from these initial trials, certainly well within a runway. If that includes patient data in a follow-up study, that we are not guiding towards yet because it also depends on, you know, the learnings from the trials. What is most important, I think, for us, Ashik, in validating this technology into the clinic is that we get reliable data that is answering the question on how efficiently we can edit in a human setting, in vivo editing, that is. The trials that we're executing on are designed to answer those questions while in parallel advancing two potentially very important treatments forwards on a development trajectory.
Got it. That's very clear. Maybe I'll ask one on ADAR itself. I guess as we think about this from a broader perspective, what's the rate limiting step within this editing process? Is it the recruitment and availability of ADAR, or is it more to do with the dosing and persistence of your oligonucleotide? What's the rate limiting step? And maybe how efficient or consistent is the recruitment of ADAR, especially across your various experiments?
Those are great questions. I think the recruitment rate of ADAR and the oligo, we studied that in detail, and I think it's a very efficient process, to be honest. I think that, having learned a lot from our experiments, rate limiting in the beginning was really to understand the chemical modifications that we could enter into these oligos that could actually be compatible with what ADAR could do. Basically, the chemical modifications that we introduced, not to enhance delivery, but also stability. We very carefully titrated those together with Dr. Beal to see what the balance would be between effectivity and delivery. Now we have learned a lot more.
Including the use of GalNAc, I think the limitations start to disappear, and the actual dosing and frequency starts to become more oligo-like, if you will. That has been in the field of oligos. We're understanding more about the chemical modifications, about the behavior of these molecules, and what ADAR can actually tolerate in that. I hope that answers your question.
Yeah, that's very helpful. Last one from me. I think we have a good understanding of what you're looking for on the efficacy front, especially from a biomarker perspective, but what are you looking for on the safety side? I assume you'll maybe talk about potential liver AEs or something to that effect, but you also alluded to ADAR's role as an antiviral player. I'm wondering if there are potential immune AEs you might be keeping an eye on or something to that effect.
Great questions again. We would look for toxicities that would be related to oligonucleotides. Our EONs basically are oligonucleotides, as you would see in other mode of actions. We would be looking at systemic events, so to speak, that would focus on liver and others that are known to the field. I don't think that we will see off-target effects in activating the immune system because that's a slightly different mode of action. Mainly we will be looking at the oligo associated toxicities that are out there. Having, let's say, 4 decades of experience from the field in using oligonucleotides, we know fairly well in what to look for and what to select against.
That's something that it will be, yeah, front and center in our development programs moving forward.
Great. Thanks for taking all my questions.
Thank you, Ashik.
Thank you for the questions. The next question comes from Dae Gon Ha at Stifel.
Great. Good morning. Hope you guys can hear me. Thanks for hosting this call, and it's extremely helpful. Maybe I'll just go one by one. Daniel, at a high level, just to follow up on an earlier, I think it was Steve's question. I guess when you think about prioritizing programs, can you maybe tell us a little bit about how you're thinking the insights from these initial two programs, what kind of de-risking they would provide for your subsequent programs which remain undisclosed at this time? Is it just the similarities in the organ targeting, the cell types that you're targeting, the nature of the edit? What would be sort of the main sort of take home from a read-through perspective?
Yeah, Dae Gon, thank you for the question. Look, I think the first organ we wanna unlock on the Axiomer platform is the liver. We think in the liver, there's a lot of opportunity to target proteins that are excreted from the liver and, you know, can cause or prevent disease throughout the entire human body. We think that with these first programs, we can really start to build a correlation model to understand how our molecules translate through the non-clinical models into the clinic. Based on that correlation, we can extrapolate for future targets, how they will behave, which largely has to do with delivery, durability, dosing, editing efficiency, translation of that editing efficiency into biomarkers.
There are some disease specificity, of course, around the biomarker aspect, but we think we will learn enough by changing few variables from target to target, to really be able to increase the probability of success for each following molecule on the basis of the de-risking we're doing with the first few. I think the, we're trying to keep as many of the variables, the same. Targeting hepatocytes with GalNAc-conjugated molecules will allow us to really study the different targets in isolation of any other variables.
Great. Great. Thanks so much. Maybe some science questions since I'm a little bit naive on the ADAR front. Maybe, Gerard, on the ADAR expression profile, if I could ask a two-part question. Can you maybe walk us through the differentiated or differential expression profile in between cell types, in between ages, in between, I guess, organs, or maybe under disease conditions, do ADARs get expressed differently? Then on the other side, can you also maybe talk a little bit about ADAR subtypes, whether they are constitutively expressed at a similar level, or how you could go about designing EONs appropriately so that you are recruiting the type that you want for the output?
Right. Excellent question. I think that in general, we spend a lot of time in understanding expression of ADAR as a class of editing molecules. As you may know, there are three ADARs, of which two ADARs are interesting to us because they can help us to edit. You have the ADAR1, which has two isoforms, which is actually the 150 isoform and the 110, which the 150 isoform is basically interferon inducible and is basically cytoplasmic, but also shuttles into the nucleus where the smaller form of ADAR1 is mainly nuclear. Then you have ADAR2, which is nuclear as well. We have an abundance of ADAR molecules that are in the cell.
We found that comparing different cell lines, there are some editing efficiencies that differ from cell to cell. We find that if you look at the hepatocyte, we have a very fairly well and consistent pattern of editing taking place. That's, that are learnings from our systems that we took from testing the system in hepatocytes, primary hepatocytes, and then we took them into more complex systems, which we call LMTs or liver microtissues. We see a very nice portability of editing efficiencies that we can take from those learnings. I think what we've seen is also taking that into vivo now we can actually see that the... Well, there's a robust editing taking place in every system that we now test.
We do feel that there's a consistency around the editing taking place in the different or cell systems that we're testing. So that helps a lot in the development and taking the next steps.
Gotcha. Just two clarification questions. On one of the, I think it was the spinal cord data NHP, there was an N=3, but with the two asterisks saying two were excluded due to some human error. Can you maybe clarify what was that and would that kind of induce any kind of concern when you think about sort of CNS target down the line? second clarification is for your, the B4GALT1 program, what's the intended target cell type? Is it hepatocytes? Because I saw myeloid cells on one of your slides. If it's a naked EON that you're injecting, is it intended to go into your bone marrow, or is it just in the circulation that it would induce the change?
No. Let me start with the second part of your question. I think the intended target cells are hepatocytes, in which B4GALT1 is expressed. It's not solely dedicated to hepatocytes, but the intended target proteins that we would like to affect are expressed in hepatocytes. fibrinogen and ApoB-100 are expressed there and through the, let's say in local inactivation or loss of function variant induction, we are going to be affecting the glycosylation patterns of those proteins and thereby limiting their negative effect. That's the answer to your second part. The first part is we've been working since seven months to get the proof of concept data ready.
We've been gathering information in animal models like mice, wild type mice, but also now human primates. The first experiment in non-human primates that we show actually was intended to dose three animals through intrathecal delivery. Unfortunately, due to human error, two of the animals got a faulty dosing, so we didn't deliver the dose. The dose wasn't delivered in appropriate, let's say a functional compartment for us to measure the editing. Since the data of the 3rd animal was very robust, we decided to show it. In the near future, we'll be going out with additional points of data in non-human primates. You can rest assured that will be a continued effort to show editing in large animals as well.
Great. Thanks very much. Very helpful.
Thank you, Dae Gon.
Thank you, Dae Gon. The next question comes from Keay Nakae at Chardan.
Hi, can you hear me?
Yes. Hi, Keay. We can hear you.
Great. A couple of questions, first for Gerard. In this application for your synthetic oligo where you're trying to recruit endogenous ADAR, what factors contribute to the length being, 10-20 nucleotides, as you stated earlier?
You mean why can we go so low, so small?
Yeah. How do you know in this application that's the optimal length?
Well, the EONs as we develop them are between 25 and 30 nucleotides long, and that's the, let's say, the intended length. We will do a length optimization in each and every target, so there will be some variation to the length. In general, what we see in our development and our EON designs, we are around the 25 to 30 nucleotides, which is a length that we can develop as we would be doing for an exon skip or another moiety or mode of action.
I think the what dictates the length because as René referred to at the start, we did start with all the nucleotides that had hairpin structures to intend to actually recruit ADAR from the beginning. By understanding more what ADAR can and cannot tolerate, we were able to, by also chemical modifications, to decrease the length. We're finding that the current length that we are using, which is again between 25 and 30 nucleotides, is most optimal to provide a double-stranded, yeah, let's call it a docking moiety for ADAR to bind to and then to exert the deamination reaction on the intended adenosine.
Great. You know, Dr. Beal talked about how beneficial having the crystal structures has been in the understanding and facilitating a rational design. What other technologies or characterizations do you think are needed for you to develop even more e-efficient, EONs?
Indeed the crystal structures as being discovered by Dr. Beal were very, very helpful in modeling the binding of the oligonucleotide to both the target RNA and the ADAR itself. It allowed us to model introduction of chemical modifications at different locations and then the interaction with the amino acids of the ADAR. In addition to that, as he explained, we have optimized the editing-enabling region or EER to mimic something that we feel that is most optimal by using a chemical modification. Modeling and understanding the chemical interaction is actually key to develop these programs further. We are actually expanding the collaboration with Dr. Beal that we also understand that certain sequence contexts, which are less amenable to editing, we can now resolve.
There's some really nice scientific data also being published that shows that we can overcome certain limitations as well.
Okay. Maybe a question on the IP. Obviously you have a nice portfolio here that you've built. Maybe back at the 10,000 foot level, what is it do you think you own? Obviously recruiting endogenous has proven itself to be the way to go here. You're using your EONs, and the EER being a critical design point. Maybe just recharacterize what do you think you own on the IP front here for ADAR?
We own the world. As mentioned, one more thing, if you want to use an ADAR, an EON to attract ADAR, it's us. We have the patents, we have also confirmation. It's not us that telling you that we have a leading IP position, but it's now confirmed also in Europe by winning a very, very important opposition. That's the long and the short. You can look it up in our patent slide. Happy to take you through in detail. But the short and the long story is, if you want to use an EON to attract ADAR, it's us.
Okay. Very good. Thanks.
Thank you, Keay.
Thank you for the questions, Keay Nakae. I'll turn it back to Daniel for concluding remarks.
Yes. Thank you, Sarah. Thank you all for attending the session today. We're pleased to have shared with you the progress we've made on the RNA editing front and outline to plan going forward. We look forward to keeping you up to date. Thank you all. Have a good day.