Wave Life Sciences Ltd. (WVE) Earnings Call Transcript & Summary

[Audio Gap] [Operator Instructions] It is now my pleasure to turn today's program over to Kate Rausch, Head of Investor Relations at Wave Life Sciences. Please go ahead.

Thank you, operator. Good morning, and welcome to Wave's 2021 Analyst and Investor Research Webcast. The slides that accompany today's presentation will be available in the Investors section of our website at www.wavelifesciences.com following the webcast. Before we begin, I would like to remind you that management may make forward-looking statements during today's presentation. These statements are subject to a number of risks and uncertainties that could cause our actual results to differ materially from those described in these forward-looking statements. The factors that could cause actual results to differ are discussed in our SEC filings, including our annual report on Form 10-K for the year ended December 31, 2020, and our quarterly report on Form 10-Q for the quarter ended June 30, 2021. We undertake no obligation to update or revise any forward-looking statements for any reason. In the room with me today are Dr. Paul Bolno, President and CEO of Wave Life Sciences, who will begin the presentation this morning with opening remarks. We also have Dr. Chandra Vargeese, Wave's Chief Technology Officer, who will be discussing Wave'S PRISM platform and progress towards building a best-in-class ADAR editing capability, followed by Dr. Ken Rhodes, Senior Vice President, Therapeutics Discovery, who will discuss ADAR editing in the CNS; and finally, Dr. Paloma Giangrande, VP of Platform Discovery Sciences in Biology will present an update on our alpha-1 antitrypsin deficiency program. Following the presentations, we will open the call up for Q&A. I'd now like to turn the call over to Paul. Paul?

Thanks, Kate. Good morning, and welcome to Wave's 2021 Analyst and Investor Research Webcast. We're excited to share with you today the recent progress we have made on our PRISM platform and how we are applying that to new targets with the ultimate goal of advancing life-changing treatments for patients diagnosed with genetically defined diseases. Today, we are a participant in a genetic revolution where breakthroughs and understanding the human genome and how mutations drive disease pathology have provided enormous opportunity to make an impact from millions of people and their families. There are more than 6,000 known genetically defined diseases that are driven by mutations in single genes and new genetic links to both rare and common diseases will continue to emerge. Advances in genetic testing have enabled more rapid identification of patients with specific mutations who would be eligible for genetically targeted treatments. So this takes the question. If we know the genetic cause of thousands of diseases and how to rapidly identify likely responders and what is the bottleneck to bringing more disease-modifying solutions forward? The answer is up to 85% of genetic diseases are beyond the reach of traditional treatments. At Wave, we are building a leading genetic medicines company focused on harnessing the endogenous machinery in our bodies to drug the transcriptome, thereby modulating the expression of faulty genes. By intervening at the RNA level, we can design therapeutics, tailored to address underlying genetic drivers of disease with an opportunity to achieve potent and durable effects within frequent adjustable dosing. We can take advantage of simplified delivery strategies, leveraging free uptake and endogenous enzymes to avoid delivery vehicles. RNA-targeting therapeutics have an ideal balance of precision, durability, potency and safety, as well as a clear line of sight to commercialization with established regulatory access and reimbursement pathways. Our focus on the transcriptome goes hand-in-hand with a fundamental truth. As a biological machinery to address genetic diseases already exist in ourselves, it just needs to be controlled with the right tools. We are able to achieve this through the rational design of highly specific oligonucleotides. We have built a genetic toolkit comprised of 3 modalities, all with the capability of modulating gene expression by efficiently and precisely harnessing existing cellular machinery. Most recently, we have been able to utilize ADAR enzymes to establish what we believe is a best-in-class RNA editing modality. You'll hear a lot about this capability today, including how we are applying editing in the CNS and other tissue types. Using ADAR to edit single bases in RNA gives us the dexterity to restore functional protein or even up-regulate expression for control how it functions. The ability to access all these mechanisms with 1 platform enables us to develop the tools to optimally address disease biology. Our PRISM platform can therefore be summed up in 1 simple concept, a unique ability to unlock the body's own ability to treat genetic disease using oligonucleotides. Through the lens of stereochemistry and the ability to control the 3 structural features of oligonucleotides, we can design precise tools that work in unison with the body's biological machinery. Again, these are short and fully chemically modified genetic medicines that are either freely taken up by cells or compatible with GalNAc delivery. Through PRISM, we identify, aggregate, leverage and deploy pharmacologic insights with a deep understanding of the interplay among sequence, chemistry and stereochemistry. We have unique backbone chemistry, including our PN modifications, which positively impact the pharmacology of all of our modalities, as Chandra will review today. Naturally occurring nucleic acids are not suitable for use of therapeutics due to a lack of stability. In addition to sugar modifications, backbone chemical modifications, including phosphorothioate, they are used to stabilize different types of therapeutics, including antisense, RNAi and guide strength. With each chiral chemical modification, diversity has introduced into these oligonucleotides. Why is this important? Because chirality matters. Stereochemistry affects pharmacology of oligonucleotides in vitro and in vivo. Recently, others in the nucleic acid field, including Alnylam, have been following our lead with scientific publications, recognizing the positive impact of chirality. We're excited to share more on the impact of stereochemistry shortly. Controlling chirality is core to how we design, optimize and manufacture our therapeutic oligonucleotide candidates. With PRISM, we have the ability to apply this control to every step of the drug discovery and development process, and we have built a dominant IP position in this space. While others may assume this could add cost and complexity, we are able to do this efficiently and at a comparable cost to stereorandom methods. For nearly a decade, we have been investing in and iterating on our PRISM platform. Our vision is to take genetically defined targets, predict a sequence and apply the therapeutic modality best suited for the biology. We use our platform engine to screen candidate and optimize the pharmacologic profile based on predefined design principles. This process is powered by predictive modeling and our machine learning capability. With the learnings from each program, we refine and improve design principles and deploy them to subsequent programs. For this reason, the platform today is distinct from the understanding and technology used in our initial programs. The evolution of our platform is reflected in the significant improvements of our hit rates from our screens of silencing therapeutic candidates. Years ago, our hit rates were approximately 10% to 15% with our stereorandom screens on par with industry standards. With the application of chiral control to every screen, chemistry enhancements, including PN chemistry and machine learning, we are sustaining hit rates of 80%. While these reflect our screens for our silencing oligonucleotide, we are on track to achieve similar progression with our other modalities. Additionally, we have been able to open up new sequence space that has been ignored or overlooked, allowing ways to pursue new potential target sites and accelerate our discovery pipeline. We are also applying data sciences to identify and explore new targets with our splicing modality. An example of this recent work that will be presented at the OTS conference this week, our data science team used large genomic data sets to develop a model that identifies exon amenable to exon skipping oligonucleotide. They predicted more than 2,500 potential targets, most of which were previously undiscovered. From this work, we have identified multiple targets that could fuel new exploratory programs. Our portfolio of stereopure PN modified oligonucleotide spans multiple modalities, including our 3 clinical programs, WVE-004, our variant selective silencing candidate targeting C9orf72 for ALS and FTD; WVE-003, our allele selective silencing candidate targeting SNP3 for HD; and WVE-N531, our exon 53, exon skipping candidate for DMD. We expect to generate data from the clinical trials for these candidates through 2022, which represents the first clinical data with PN modified compounds. In a moment, Chandra will review our PRISM platform and why we are so excited about these modifications. Additionally, today, you'll hear more about our quickly evolving editing capability, including an update on our AATD program from Paloma and the application of ADAR editing in the CNS from Ken. It's an exciting time for nucleic acid therapeutics, and we are only just beginning to realize the full potential of our PRISM platform and its ability to redefine oligonucleotide therapeutics. I'd now like to turn the call over to Chandra Vargeese, Chief Technology Officer, to review our PRISM platform and ADAR editing modality. Chandra?

Thanks, Paul. Good morning to everyone on the webcast, and thank you for joining. Our PRISM platform is built on the reality that there exists enormous opportunity to tune the pharmacological properties of oligonucleotide therapeutics by leveraging 3 key features of the molecule: sequence, chemistry and [indiscernible]. Each of these features can be modulated. We can change the sequence or incorporate a modified base. We can change and innovate on the chemistry of drivers of the backbone, and we can control chirality of all sterocenters, including those in the backbone, deleverage this rich diversity to design molecules with desirable profiles. Wave is uniquely positioned to rationally design our molecules because of our deep understanding of the interplay between these tunable features and our ability to control them. With PRISM, we modulate these features to design therapeutic candidates with desirable activity profiles and compressing their potency, tissue exposure and duration of activity. To optimize activity profiles, some of our design principles are universal. For example, the stability and tissue exposure across modalities. On the contrary, potency is fine-tuned based on the interaction of oligonucleotide and the endogenous enzymes requiring our unique ability to control sequence, chemistry and stereochemistry. Since Wave's inception, we have immediately generated stereopure molecules, those in which the chirality at every stereocenter in the phosphorothioate or PS modified backbone is controlled. Without this control, oligonucleotides are a mixture of potentially millions of different isomers. Last year, we unveiled our proprietary phosphoryl guanidine-based modification, or PN chemistry, which like the PS modification introduces chiral center into the backbone. We also developed the capabilities to control the chirality of these PN linkages in order to fully integrate PN chemistry into our platform. The incorporation of PN chemistry conveys several desirable properties to our stereopure molecules. Because it is neutral, it even breaks up the negative charge on the backbone, and we believe that this property drives tissue exposure benefits. Oligonucleotides with PN backbones retain complementary base pairing properties. And consequently, it retains the resulting specificity inherent to these molecules. In the last year, we have shown you many examples of how the addition of few judiciously placed PN linkages into our oligonucleotide backbone has positively impacted the pharmacology of our molecules. What have been truly remarkable and differentiates this modification is the magnitude and consistency of benefits we see, regardless of the modality, slicing, silencing or editing. I'll review several data sets that demonstrates this impact on potency, tissue exposure and durability of our oligonucleotides on the coming slides. On Slide 20, we show how the interaction of just a few rationally placed PN linkages provides a notable potency benefit across modalities in vitro. On the left, we compare a series of stereopure silencing molecules that add to an RNA stage dependent silencing mechanism with PS modification, shown in light blue, or with the few PN modifications shown in dark blue. Across multiple sequences, we observe a downward shift, indicating a potency gain for this modality. In the middle, we compare a series of splicing molecules with PN modifications or with the few PN -- with PS modification or with the few PN modifications. Across multiple sequences, we observe an upward shift, indicating a potency gain for this modality. On the right, we illustrate the dramatic improvement in both potency and maximum editing, driven by the application of stereopure PS and PN backbone modification. I will provide more detailed updates on our editing capabilities a little later. Importantly, these potency shifts with PN chemistry in vitro translate in vivo. Here we show PK/PD profiles for 2 molecules from our C9orf72 discovery efforts. These PS/PO and PN containing silencing molecules engage RNA stage to decrease pathological C9orf72 transcripts, and they illustrate the potency and tissue exposure benefits that we observed with PN chemistry. The graph shows exposure on the X-axis and mRNA expression on Y-axis. The PN molecule shown in dark blue shifts rightward, indicating increased exposure and downward indicating increased RNA silencing compared with the PS/PO molecule of the same sequence and same chemical modification pattern. Now for the splicing modality, we highlight 2 examples from our DMD discovery efforts that illustrate that meaningful biological benefits can be realized with the addition of a few PN backbone linkages. The graph shows data from a survival study we performed in a double knockout or dKO mouse model for DMD with an aggressive and lethal genotype. In this dKO mice, treatment with 1 of our PN continuing exon skipping compounds shown in green led to 100% survival of the duration of a 40-week study where administered at a 75% lower dose than the PS modified molecule shown in light blue. In a separate 6-week experiment, in the same detailed mouse model, the PN containing molecules accumulated to a higher levels in all muscle types compared with the PS/PO oligo, as shown on the top of the slide. The PN molecule also led to more exon skipping and more dystrophin restoration in all muscles evaluated. These differences are most dramatic in the diaphragm and heart, which likely explains the profound survival benefit observed in these mice. We have also applied our stereopure PS/PN modification in another silencing modality RNAi. To engage the RNAi pathway, we use double standard siRNA. These molecules engage with AGO2 with the guide strand loading onto the enzyme and interacting with the RNA to form a substrate that is cleared, ultimately leading to the degradation of the target RNA. To study our siRNAs, we benchmarked our molecules to reference molecules from the literature that are well studied in ESC and advanced ESC format. We then generated double-stranded GalNAc-conjugated siRNAs targeting transparent in mRNA or TTR. Because TTR protein is segregated into serum, we can track levels of TTR protein in serum over time in the same mice. Both control of stereochemistry and application of PN chemistry drive a threefold increase in potency, twofold increase in durability of effect and an unprecedented 35-fold increase in AGO2 loading as compared to reference ESC. Then we apply PRISM to the state-of-the-art RNAi design based on an advanced ESC chemistry. We again observed a significant benefit from the molecules containing PN chemistry shown in dark blue with a 50% increase in duration of activity after a single dose of 1 mg per kg compared to the reference molecule shown in black at the 2-month time points. We're excited about these data because they highlight that stereochemistry and PN chemistry enhanced pharmacology of another oligonucleotide modality, this time, one relying on double-stranded RNA. Now with PRISM, we leverage the full diversity of nucleic acid chemistry, including diversity resulting from stereochemistry to optimize molecules. This enabled us to understand structural activity relationship, or SAR, which is the standard practice in the industry for all types of modalities, including small molecules and antibodies. The oligonucleotide space has ignored this diversity for decades, missing out on an entire dimension of therapeutic optimization. Thus far, I've highlighted how we use PRISM to optimize potency and duration of activity. Through PRISM, we are also able to identify and adjust for the impact of stereochemistry on the tolerability of our molecules in vivo. I'll walk you through 2 examples where we showed a dramatically different profile for closely related stereoisomers, enabling us to tune tolerability while maintaining potency by selecting the right isomer. In the first example, we highlight 2 GalNAc-conjugated molecules that promote RNA stage dependent silencing of a liver target. At the top of the slide, we illustrate 2 stereo isomers that are identical in design, except at a single stereocenter. As seen in the graph on the left, these 2 isomers have comparable activity against the target mRNA in vivo. By contract, the graph in the middle and to the right show how they differ dramatically in tolerability with isomer 1 leading to a significant expression of multiple tolerability markers, including liver enzymes, AST and ALT and TNF receptors that induces apoptosis. By contrast, isomer 2 does not induce significant expression of any of these markers. In the second example, we highlight molecules that promote RNA stage dependent silencing for an undisclosed CNS target. At the top of the slide, we illustrate 2 stereoisomers that only differ at pre-stereocenters in the backbone. Once again, as seen on the left, these isomers have comparable and potent activity against the target mRNA in vivo. They again differ in how well they are tolerated with the isomer 2 leading to a dramatic and significant body weight loss, beginning almost immediately after treatment. By contrast, mice treated with isomer 1 gained weight comparably to those treated with PBS. For these few examples, if you had ignored stereochemistry, we would have either selected for a mixture containing isomers with unfavorable profiles with potential risk for tolerability issues emerging over time, or these isomers may have driven us to avoid this target sequence and potentially sacrifice potents. By controlling stereochemistry, we can design our own tolerability signals without compromising potency. With our PRISM platform, we apply novel chemistry and unique design principles across modalities to truly innovate on oligonucleotide discovery and development. Our stereo PS and PN backbone modifications represent a breakthrough in the nucleic acid field. One of the best ways to demonstrate this is to discuss how we leverage PRISM to develop our RNA editing modality. We were inspired by early academic publications in the field, demonstrating that oligonucleotides can lead to RNA editing with endogenous ADAR, getting back to Tod Woolf's publication in 1995. We took early structural insights into ADAR RNA substrate interactions and initiated our work on this modality with the aim of expanding our capabilities to include editing all through free uptake of chemically-modified oligonucleotide. We evaluated more than 1,000 ADAR oligonucleotides to produce new insight into the relationship between the molecule structure and its ability to elicit robust ADAR editing activity. What was most exciting, it was not until we applied chiral control and PN modifications to our oligonucleotide that we saw the strong potential for this capability, enabling us to rapidly progress to demonstrating efficient in vivo editing. Today, we are introducing our AIMers, short chemically modified oligonucleotide to record endogenous ADAR enzymes to change a specific adenosine or A to inosine or I, which self-read as a guanosine or G. For the rest of the presentation, we will refer to our ADAR editing oligonucleotide as AIMers. A-to-I editing is a common post transitional RNA modification that generates transcriptomic diversity. And the ADAR enzymes are ubiquitously expressed across tissues. We are at the forefront of this modality, defining new levels of editing that can be achieved as well as the tissues and cell types amenable to this approach. I'm excited to have the opportunity to share more with you today. I'll start off with how ADAR expands our toolkit to adverse diseases with RNA therapeutics by enabling protein restoration, upgradation and even modification of protein function. Next, I'll describe how we design, optimize our AIMers, which differentiates us from other editing technologies. Last, I'll discuss how we have built our structures without GalNAc-conjugated AIMers and to use our unique chemistry to expand beyond editing in the liver. Nearly half of non-pathogenic SNPs are GTA mutation. Many of these are loss of function mutations. The unique ability of ADAR to correct these mutations and restore protein expression creates significant opportunities for us to potentially treat a broad spectrum of human diseases. Later in the presentation, Ken and Paloma will walk through our data supporting how we use this application on the therapeutic targets, including restoration of functional protein to address alpha-1 antitrypsin deficiency. The application of PRISM to RNA editing opens the door to therapeutic applications, extending beyond the restoration of protein function. We can use ADAR to upgrade protein expression to modify protein function by offering post-translational modifications or protein-protein interactions or to also protein stability. Ken will share data on an example of this later on. Now, I'll highlight how we apply our unique chemistry capabilities to design and optimize AIMers. Each of our AIMers have several consistent features. They're single stranded oligonucleotides that are typically 30 basis or fewer. They are fully chemically modified, including modified nuclear basis and stereopure PS and PN backbone modifications. They're able to be conjugated to ligands to aid in tissue targeting. By utilizing our unique chemistry platform, we can rationally design AIMers for each target to efficiently recruit ADAR to maximize editing. Here, we show the activity of GalNAc-conjugated beta-actin stereopure AIMer with and without PN linkages and compared to the matched stereorandom control shown in black in primary hepatocytes. The addition of PN chemistry substantially improves both potency and editing efficiency. We have confidence that the endogenous ADAR editing capacity of a cell is sufficient to support therapeutic ADAR editing without disrupting homeostatic ADAR activity. In the graph, we highlight editing levels observed in pre-transcript when we evaluated editing for each transcript in isolation shown in light blue, or when all the pre-transcripts were targeted in the same experiment in the same cells in the same time shown in dark blue. Under both conditions, editing levels of the each transcript are almost the same, suggesting there is ample reservoir of ADAR editing capacity for us to tap into. A challenge for ADAR is the defined sequence phase of the target. To navigate this, we have evaluated sequences complementary to the edit side as one dimension of our rational approach to designing AIMers. For this investigation, we changed the sequence motive on the AIMer independently of the sequence motive in the target in over 300 AIMers. This heat map, shown on the right of the slide, reveals a clear pattern in the design that helps us define the sequences in our AIMers that support the most robust editing. And data sets like this example, enable us to predict sequence design in our AIMers to optimize our editing. We have developed AIMers, the direct editing of multiple endogenous transcripts. This graph highlights everything at 8 different targets. These observations helped build our confidence that we can develop in editing technology that with some optimization to the target can be applied more broadly. Next, I'll show how we translated RNA editing in vivo. In this section, I'll highlight the work we have done to establish proof-of-concept in vivo in multiple tissues, starting with GalNAc-conjugated AIMers in liver and expanding to CNS, ophthalmology and beyond. It is important our AIMers are optimized for human ADAR. While we have demonstrated editing in nonhuman primates, we know some diseases targets will be unavailable in nonhuman primates to evaluate in vivo target engagement. As such, we developed a transonic human mouse ADAR, mouse with human ADAR which expresses human ADAR in liver and neurons at levels that are equivalent with human tissues, as shown by the western blots on the right-hand side of the slide. This model enables us for the assessment of in vivo target engagement with human ADAR and is especially important for disease targets that are unavailable in nonhuman primates. Now I'll start by reviewing the success that we have had with GalNAc conjugated AIMers to achieve highly specific RNA editing in liver with subcutaneous administration. As we have previously described, we achieved efficient RNA editing in vitro that GalNAc conjugated AIMers across a variety of cell lines, including nonhuman primate and human primary hepatocytes. See they moved in vivo dosing nonhuman primates subcutaneously once-a-day for 5 days with 3 chemically distinct AIMers. We detected up to 50% editing as compared to baseline at an initial time point, which was remarkably durable up to 44 days post last dose, as shown in the middle of the slide. We also determined that these AIMers are highly specific. We performed full transcriptome RNAC in primary human hepatocytes. We only detected editing at the target sequence in the acting transcript and absorbed nominal off-target editing across the transcriptome, as shown on the right. These results drove our decision to initiate our first therapeutic program with GalNAc conjugated AIMers for AATD. Now building on our progress with GalNAc conjugated AIMers, we then explored unconjugated AIMers to CNS and eye. As Kendell will discuss later on, these data have led us to our MECP2 discovery program and undisclosed exploratory programs in neurology. In a recent study, we observed dose-dependent editing of the UGP2 in the posterior compartment of the eye, including the retina at multiple time points post injection. We used intravitreal injections of clinically relevant route of administration to deliver these AIMers to mouse eyes. At 1 week post dose, we observed up to 50% mean editing and these high levels of editing pushed us at least 1 month post dose. We also observed wide distribution of AIMers at the 1-month time point in the sections from these eyes, as shown by the pink staining on the right side of the slide. These data suggest our platform may support development of AIMers for ophthalmology diseases. Turning to CNS. Last year, we demonstrated editing in multiple CNS cell types, as shown on the left and in the CNS of mice at single time point with UGP2 targeting aim. The initial result was exciting, and we sought to replicate it and explore durability of editing in CNS. We performed a follow-up study where mice with the same single 100 microgram dose, but we were followed over a much longer window of time. As in the previous experiment, we observed editing throughout the brain, with peak heading observed approximately 1 month post dose in all the tissues and the robust editing persisting for at least 4 months post dose. We're very excited by the durability of this editing as this result is consistent with what we have demonstrated with GalNAc conjugated AIMers and stereopure PN modified constructs across other modalities. We also evaluated tissues from 3-month time point and absorbed wide distribution of AIMers throughout the CNS, as shown by the pink caning in the top 2 rows of the slide. As the target landscape for ADAR editing outside liver, CNS and ophthalmology is vast, we also evaluated the potential of unconjugated AIMers using systemic delivery. In preliminary studies in NHP, we absorbed in several tissues of interest, including kidney, liver, lung and heart after a single subcutaneous dose. While desired editing levels will vary by target. These results are compelling, particularly in hard to reach issues such as heart. We are also evaluating the potential of unconjugated AIMers for delivery to the immune system, our self found in human peripheral blood. Cells of the immune system are known to be difficult to edit. But editing in these cells calls great promise for the treatment of many diseases, including cancer and autoimmune diseases. After AIMer treatment in vitro using 3 uptakes, we observed robust editing in variety of immune cells in PBMCs, including CD4 positive and CD8 positive T cells, monocytes, B cells, NK cell and regulatory T cells. The work we are doing to expand the capacity of our ADAR editing modality in its infancy. Many developments, I've shared so far today, we like our preliminary lines. Our work to improve our designs and to explore new application for this technology is ongoing. As we have demonstrated with our silence components as learning from our platform inform our design principles, we can significantly improve the activity of these editing molecules. On this final slide, I show you a premier of what is coming. Much of the data I shared today were up with an AIMer with an initial chemistry that edited UGP2. The activity of this AIMer in cultured neuron is illustrated in light green on the left of the slide. On the right, in dark blue, you can see that you're seeing 75% more editing in the same cells under the same conditions with our optimized chemistry. We are currently applying optimized AIMers to therapeutic targets as you're about to hear from Ken and Paloma. Our ADAR editing modality has come a long ways in a short period of time. We have an elegant approach to RNA editing, enabled by chemistry and biology expertise. We'll continue to establish our leadership position as we advance the therapeutic programs and scientific publications. I look forward to updating you again soon. With that, I will now turn the call over to Ken. Ken?

Thanks, Sandra, and good morning, everyone. Over the last several years, the evolution of our PRISM platform has shaped our neurology pipeline by 3 ongoing clinical programs, WVE-004 for ALS and FTD, 003 for HD and N531 for DMD to incorporate PN backbone modifications and leverage translational in vivo models, providing greater insight into dose response relationships and target engagement. As these programs advance, we continue to apply lessons learned to characterize and bring forward new therapeutic candidates. Today, I'm excited to describe how we're applying ADAR within our neurology discovery pipeline to access targets previously thought to be undruggable. Following on Chandra's update on the progress of our ADAR platform, we can now expand our neurology portfolio to include diseases where restorational protein function is required and where we would like to access pharmacology that is currently beyond the reach of other oligonucleotide based approaches. RNA editing provides a unique approach to treat a wide array of neurological diseases that cannot be targeted through silencing or exon skipping mechanisms. As Chandra presented, we've demonstrated that we can achieve robust RNA editing in CNS tissues. Using UGP2 as an example and using first generation AIMer molecules, we've demonstrated potent, widespread and sustained editing throughout the mouse CNS with the effect persisting out to at least 16 weeks following a single ICV dose. These data indicate that PN chemistry is driving widespread and durable effect when applied to AIMers. As described on Slide 57, there are several ways in which we could use ADAR to address disease mechanisms or disease-causing mutations across a diverse spectrum of neurologic diseases from epilepsies to dementia, apple insufficiencies to gain a function mutation. To provide one example of how we're using RNA editing in our neurology portfolio, I'll focus on correction of a specific nonsense mutation that leads to reduced expression of MECP2, a protein found in the nucleus of neurons and glial cells that is required for normal brain development. Mutations in MECP2 are the cause of Rett syndrome, a devastating neurological disorder, mainly affecting girls and characterized by profound neurodevelopmental delays and seizures. MECP2 regulates expression of a variety of genes involved in neuronal development and synaptic plasticity. Nonsense mutations in MECP2 lead to truncation of MECP2 protein, a loss of the ability of MECP2 protein to localize to the cell nucleus and the loss of the ability of MECP2 to recruit other proteins required to carry out its normal function as a regulator of gene expression during brand development. Using ADAR, we can edit nonsense mutations in MECP2 to restore expression and activity of MECP2 protein. Although individual patients can have distinct mutations in MECP2, our initial focus is on a common and prevalent mutation, R168X, where x represents a signal to stop mRNA translation. Based on success in editing R168X, we can apply this approach to expand our Rett program to develop aimers that address additional disease-causing MECP2 nonsense mutations. In the example on Slide 59, we use ADAR to edit the R168X nonsense mutation. As you can see in this figure, using our AIMer constructs, we obtained concentration dependent editing of MECP2 up to about 70%. Incorporation of PN chemistry and the AIMer molecules greatly improves editing activity. As shown in the bottom right-hand figure, we see that ADAR editing restores for length MECP2, expressed here as a GFP fusion protein in this in vitro system. ADAR editing of the R168X mutation replaces the missense stop codon with an mRNA triplet in coding the amino acid tryptophan as tryptophan is not normally present at that location in the wild-type MECP2 protein, it became important for us to show that the edited MECP2 retains its normal cellular localization and is able to associate with other nuclear proteins that are required for MECP2 to carry out its role as a modulator of gene expression. As I mentioned earlier, MECP2 is a nuclear protein. In Slide 60, we show that edited MECP2 is present in the nucleus of transfected cells. We see this using immunofluorescent staining of epitope tagged MECP2, shown in blue, in the left-hand figure. We also see this using subcellular fractionation shown in the figure on the right, where we demonstrate that MECP2 is found in the nuclear fraction and not in the cytoplasmic fraction of transfected cells. To further demonstrate that edited MECP2 retains its normal function, we used a co-immunoprecipitation strategy to show that edited MECP2 associates with members of a protein complex that work in concert with MECP 2 to regulate the expression of target genes. As shown on the right side of Slide 61, immunoprecipitation of MECP2 from transfected cells also pulls down other members of the protein complex that are critical for MECP2 to regulate gene expression. Together, these data strongly indicate that edited MECP2 is present in the nucleus and associates with binding partners that are important for its normal function. Ongoing studies are looking to demonstrate that edited MECP2 restores regulation of known target genes. Changing gears now to an often discussed use of ADAR editing, which is to modulate protein-protein interactions. Protein-protein interactions have been notoriously difficult to drug using small molecules, particularly when the interaction interface is shallow, large and hydrophobic or very deep within a protein fold. In addition, small molecules targeting these interactions often have properties that give them poor specificity and poor selectivity for the desired target. RNA editing to change amino acids within a protein interaction interface can overcome the limitations of small molecules and provides a novel approach to modulating this class of targets. To exemplify our ability to modulate protein-protein interactions using Adar, we turn to the well-characterized KEAP1 Nrf2 system. The transcription factor Nrf2 is the master regulator of the cellular antioxidant response is involved in the metabolism of xenobiotics and is involved in immune cell activation. Nrf2 also plays an important role in the pharmacological activity of immunomodulatory drugs used to treat CNS indications, such as multiple sclerosis. Under normal cellular conditions, Nrf2 is bound to the regulatory protein, KEAP1. Finding the KEAP1 retains Nrf2 in the cytoplasm and targets it for degradation by the proteasome. However, under conditions of oxidative stress, the interaction between Nrf2 and KEAP1 is disrupted, stabilizing Nrf2 and allowing it to translocate to the nucleus where it activates expression of a host of genes involved in a cellular antioxidant response. As a proof-of-concept experiment, we look to see if we could limit the cellular stress response by using ADAR to edit individual amino acids at the protein-protein interaction interface between Nrf2 and KEAP1. If these edits work as designed, we would expect to see downstream activation of the Nrf2 gene expression program even in the absence of cellular stressors. On Slide 65, we demonstrate successful mRNA editing of KEAP1 and Nrf2 at 8 unique sites within each protein. Each site was edited using a unique AIMer. As you can see on the right, editing ranged between 20% to 60% across these targeted sites. To confirm that the edits were biologically meaningful, we next looked at the expression of genes dependent on the stabilization and nuclear translocation of Nrf2. The right side of this slide shows the effects of individual AIMer treatments. As expected, treatment with each AIMer resulted in increased expression of known downstream Nrf2 target genes involved in the antioxidant response. Control treatment did not result in increased expression of any of the Nrf2 target genes, indicating that AIMer treatment did not lead to Nrf2 dependent gene expression changes through nonspecific mechanisms such as increased cellular stress. In summary, we continue to advance our neurology portfolio. In our silencing programs, we've previously shared potent and extremely durable target engagement and widespread drug distribution throughout rodent and nonhuman primate range. Although relatively new and in early stages of discovery and preclinical characterization, our use of ADAR editing in the CNS provides additional opportunities to address unmet needs in patients suffering from neurodegenerative disease, allowing access to targets and mechanisms that cannot be approached through modalities such as silencing or exon skipping or using small molecules. This is truly an exciting new frontier for Wave and for neuroscience, opening an array of new opportunities in neurologic disease. With that, I'll turn the call over to Paloma to discuss our Alpha-1 antitrypsin program. Paloma?

Thanks, Ken. Hi, everyone. I'm very excited to be speaking with you all today. As Chandra described, we've demonstrated our AIMers are GalNAc compatible and can achieve significant RNA editing in vivo and liver, including in the liver of NHP. These results have led us to rapidly advance our first ADAR editing program for a hepatic indication, Alpha-1 antitrypsin deficiency, or AATD. AATD is an inherited genetic disorder that is most commonly caused by mutation in the SERPINA1 gene, commonly known as Z allele. This mutation leads to misfolding and aggregation of Alpha-1 antitrypsin protein or Z-AAT in hepatocyte and the lack of functional AAT in circulation, which results in progressive lung damage, liver damage or both. With ADAR editing, we aim to correct the SERPINA1 mRNA to restore circulating functional wild-type Alpha-1 antitrypsin protein, or M-AAT, to protect the lung and reduce Z-AAT protein aggregation in liver, all while retaining the innate physiological regulation of M-AAT. With our GalNAc conjugated stereopure AIMers, we anticipate replacing chronic weekly IV-AAT augmentation therapy with a subcutaneously administered therapy that addresses all goals of treatment. This is superior to a silencing modality, which would only impact the liver. Of note, approximately 200,000 people in the US and EU are homozygous for disease in mutation with the highest risk of lung and liver disease. Earlier this year, we presented data demonstrating restoration of wild-type M-AAT in vivo with ADAR editing with our initial AIMer design. Today, I will start by reviewing those data, and we'll provide an update on the specificity of editing and durability of protein restoration from that study. Using Prism based chemistry optimization, we continue to enhance the editing potency of our GalNAc AIMers. I will share data demonstrating how this optimization translates to protein restoration in vivo. Finally, I'll speak to our path to selecting a development candidate. We started with the in vitro evaluation of multiple GalNAc AIMers with unique chemistry. Shown here on Slide 71, or SA1-1 and SA1-2. We successfully demonstrated upwards of 60% editing of the SERPINA1 redial transcript to wild-type in hepatitis in vitro, which led to a threefold increase in functional wild type AAT protein. Supported by these results, we developed a proprietary transgenic mouse model containing both human SERPINA1 and human ADAR that enables pharmacokinetic and pharmacodynamic assessment of human specific GalNAc AIMers in vivo. This human ADAR mouse enables us to optimize AIMers to human ADAR, which is expected to improve translation into the clinic. Of note, aggregation of human AAT protein in liver, which is observed in AATD patients is recapitulated in this mouse model, as shown on the right of Slide 71. We next advanced 2 distinct AIMers, SA1-3 and SA1-4 into an initial in vivo study to assess editing and protein restoration after 1 week. Following 3 subcutaneous doses, we achieved up to 40% editing at day seven. We were encouraged by these initial results as they approached the level of correction representative of a heterozygous NV patients with very low-risk of disease. To evaluate the specificity of the SA1-4 GalNAc AIMer, we performed RNA seek. On the left, you can see total sequence coverage across the entire SERPINA1 transcript for the AIMer treated samples. The percentage of unedited T and edited series are indicated for each group. Editing is only detected at the intended on target sequence in the SERPINA1 transcript. Thus, the protein being produced using this approach is truly wild-type M-AAT protein. This also confirms there is no editing of high standard residues as has been seen with DNA targeting approaches. Furthermore, to assess off-target editing for the whole transcriptome, we applied a mutation calling software to search edit sites. From this analysis, we observed nominal off-target editing across the transcriptome. Sites where potential off-target editing occur at either low read coverage in the analysis or occurred at low percentage of less than 10%, indicating that these are rare events. In both analyses, shown on the left and right, we find a high percentage of editing that is specific for the target site in the SERPINA1 transcript. Next, we look at how this level of editing impacted the circulating human AAT protein. We saw a threefold increase in circulating AAT as compared to PBS control at this initial time point. This magnitude of increase is promising as it is representative of, one, the fold increase that may achieve phenotypes with lower rates of disease; and two, total circulating AAT concentration approaching 570 micrograms per mil or 11 micromolar in these mice. This result established a floor from which to optimize potency, which I'll discuss later on. Using mass spectrometry, we investigated the isoforms of the circulating AAT protein and confirmed that the majority was restored wild-type M-AAT. It was very exciting to see such levels of wild-type protein being generated post editing at this first time point. As you can see on the right of Slide 75, we also observed that there was a significant increase in neutrophil elastase inhibition post editing, confirming the functionality of this restored wild-type M-AAT protein. Our next step was to understand how durable this protein restoration was over time. We repeated this experiment with the SA1-4 GalNAc AIMer out to a 35-day time point. As shown on the left-hand side of Slide 76, following 3 initial doses, we observed a significant and durable increase of greater than or equal to threefold over PBS control out to 35 days. Notably, with our initial AIMer chemistry, we maintained 11 micromolar threshold for approximately 3 weeks in the disease model. As shown on the right side of the slide, we confirm the presence of wild-type M-AAT protein at the day 35 time point. Also, at this time point, we observed elevated Z-AAT protein levels in the AIMer treated group as compared to PBS. This suggests clearance of intracellular D-AAT aggregates with AIMer treatment. So next, we wanted to understand if chemistry optimization could further improve potency and durability. In vitro, we demonstrated that we could design more potent editing AIMers. With sequences held constant, we adjusted chemistry and stereochemistry to optimize the potency of the SA1-4 GalNAc AIMer. With 2 optimized GalNAc AIMers, SA1-5 and SA1-6, we demonstrated at ADAR editing of approximately 50% with half the dose used in our first study in mice at the same day 7 time point. Next, we wanted to understand how optimization impacted protein restoration. We repeated the initial in-vivo study with an optimized version of the SA1-4 AIMer named SA1-5 and looked at circulating AAT protein at an initial day seven time point. Consistent with an increase in percent editing, as shown on the right of Slide 78, the total serum AAT relative to PBS was increased from threefold to fourfold or greater than 15 micromolar with the SA1-5 AIMer. When evaluated with mass spectrometry, the proportion of wild-type M-AAT was increased from 75% to approximately 85% of total AAP. Importantly, the optimization and chemistry establishes a range within which to advance dose range studies over longer duration. In summary, our initial in vivo study has demonstrated that our AIMers restore therapeutically meaningful levels of functional wild-type M-AAT protein at an initial time point. These AIMers are highly specific with no notable off-target or by standard editing. This protein restoration is durable, with significant levels of wild-type M-AAT detected in serum at over 30 days post last dose. We continue to optimize our AIMers and we have demonstrated that we can improve editing efficiency, drive AAT protein above 15 micromolar and also increase the overall percentage of circulating wild-type protein to 85% of total. Our focus today is advancing ongoing and planned studies that will assess durability, dose response and PK/PD of optimized AIMers. We'll also be looking at the reduction in V-AAT protein aggregates in liver and changes in liver pathology. We expect to have an AATD development candidate in 2022. The next, I'll turn back to Paul. Paul?

Thank you, Paloma, and thank you, Chandra and Ken, for speaking in detail about these exciting updates. With dosing underway in WVE-004 and 003 clinical trials and in 531 soon behind, we are entering a potentially transformational period of data generation for Wave. We believe clinical data will unlock value in multiple ways, including insights into the clinical effects of PN chemistry by way of biomarker results, and implications for dosing intervals. Through 2022, we will also gain insight into the ability of PN chemistry to improve cellular uptake and distribution with our exon-skipping modality. Our advances in chemistry set us apart from others in the field and are enabling us to lead the way in developing ADAR editing therapeutics. Starting with our Alpha-1 antitrypsin program, we are on a path to generate proof of principle that we can harness human biological machinery to treat genetic diseases of the liver and beyond. And with that, I'd like to turn the call back to the operator to begin the Q&A portion of our program.

[Operator Instructions] Our first question comes from Salim Syed with Mizuho.

Thanks so much for all the detailed color guys. Looks great. Just a few for me, if I can. Paul, just a high level of strategy question here. Obviously, the presentation they focused a lot on ADAR. And when I look at your pipeline slide right now, it's just the Alpha-1 antitrypsin program. So just curious how you're thinking about the longer-term strategy here with your ADAR platform? How much of your pipeline do you think will be ADAR, if we were to think about this in a three-year time frame or a five-year time frame? And then second question, maybe just -- I don't know, I guess, anybody can take this one. On Slide 20, this is a slide that you guys had presented prior regarding the silencing and the splicing of PS/PO chemistry versus PS/PN chemistry, and I'm curious why the darker blue dots seem to have a little bit more variability than the -- dots of -- the PNB in the darker blue here. Is there some natural variability or more variability on the PN chemistry side versus the PS/PO?

Great question. So we'll do -- I'll take the strategy question first, and then we'll cover the PS/PO. And I really appreciate really trying to think about the advancement of ADAR. Obviously, we're rapidly pushing the AATD program forward. And it's because it does two things. One, obviously, it's an important and meaningful disease with a substantial number of patients who could be potentially amenable to the therapy. So very important for what we're developing it for. Equally important is the human proof of principle data that comes from that in demonstrating our ability to measure on target engagement of editing by measuring the Alpha-1 antitrypsin program. So this program really allows us to do both features. One, adding a meaningful medicine but also to be able to validate ADAR in humans. Secondly is, as you point out, and that was really -- I'm glad you picked up on it today, is really the breadth of where ADAR could potentially go. And I think there's really two areas of focus, and I think you heard that today. One is, obviously, based on the exploration of AATD, continuing to explore GalNAc conjugated medicines that have the potential to edit. And we do see the opportunity to expand the pipeline there. I think the other that's all equally important is the work that Ken was speaking to about opening up an area of biology that we have extensive experience in, which is neurology. And so across the path around four therapeutic areas, we do see a substantial build-out of ADAR editing capability. I think we'll continue to also view the opportunity to collaborate with others as we expand into other therapeutic areas as well. But I think the key is really, as we said at the very beginning of the call that it really opens up the ability to utilize multiple endogenous machineries and cells of silencing, splicing and now editing, so that as we think about important meaningful diseases, we have this unique position now to be able to pick the right tool to do the right job. And so there is a vast universe of therapeutic potential that over the next three to five years, we can open up. So we're pretty excited about that. But I think what we've also learned from the past is a data driven approach. And I think that's why also today was really important for us is, as we talked about new ways of utilizing ADAR like nonsense mutations or drugging and modulating protein ProTmune interactions. It was important us to go beyond the statement and demonstrate that indeed we could build labors that have that capability. So we're excited about what's to come, and we're marching through with the portfolio to do that. Your second question that relates to the rank order, ask Chandra, if she wants to add to that, because they are ranked ordered by a sequence. But Chandra, do you want to add to that?

Yes, sure. So the graph is actually plotted in a way that we rank out of using the PS/PO compound as a reference. So everything is towards that. So we didn't rank order the PN. So that's why you see a uniform line for the PS/PO.

Our next question comes from Mani Foroohar with SVB Leerink.

I guess, tightening the look a little bit more narrowly at Alpha-1 antitrypsin. You talked about being in the clinic, perhaps next year. Obviously, there's a little -- there's quite a bit of optimization that goes into before you go into humans. Can you give us a sense of what are kind of the key metrics or hurdles you're looking to clear before getting into first in human? And then I have a follow-up on regulatory pathway.

Sure. And it's a great question. I hope -- and I think what you pointed out was really our path to candidate selection in 2022. And that will really inform the clinical time line. So to that point, I think Paloma did a really nice job of assessing where we are. I think we were able to achieve substantial chemical optimization on potency. I think we're there. I think the next piece that we're evaluating is that PK/PD dose response and looking at durability. So that we want to go into the clinic as quickly as possible. We also want to go there with a molecule that we think can really go the distance. So I think there's a piece around durability, so is response. And then one of the other features that's pretty interesting that Palomo pointed out as well, is we do see the ability potentially to actually see a removal of the aggregates for the liver. So being able to treat both hepatic by clearance of those aggregates as well as long as something we'll evaluate as we build the total package together to the clinic. But I think the key is we need to establish that candidate and along the way, ample opportunities for us to provide updates both on the data that we're generating and as well as how that informs the clinical time line. But it's an exciting place to be starting at now that we see substantial editing.

That makes sense to me. And then I guess also on AAT, we've seen a little bit of regulatory clarity from Arrowhead. Obviously, it's a very different olodaterol technology. What do you draw from what a potential regulatory pathway would be towards the pivotal indication? At what point do you think you need to have a study with a biopsy endpoint? And how quickly would you be able to move into that?

I think we have two ways to think about this. And it's really, first and foremost, a different approach to the silencing approach is to your point right now, which is heavily focused on aggregate removal from the liver of what the impact to the liver, which is what's unique about editing is restoration of the functional protein and protection of the one. So I do think we have opportunities as part of those studies to be able to assess both. I think what we'll be starting with in the proof of principal study is answering the first most important question, which is do we edit and do we create measurable functional protein that's in this serum. So that will be a critical inflection as we think about benchmarks from some of the protein fusion therapies. And instead of kind of having to go for weekly IV infusions, bring patients to the potential of actually restoring functional balance of wild-type protein that is there when the patient needs it, not overexpressed, but there when it needs to be. I think the unique opportunity, as we were just saying around the potential to also see a clearance of the aggregates, lets us really think about the opportunity on both ends of the spectrum, liver and lung. And in that, obviously, to your point, we will require potential evaluation of the liver. But I think we'll also be able to assess measurements in the serum of what's happening to the Z protein that could be leaving liver. So we could have a biomarker assessment that's at least enabling us in the clinic to look at whether or not we're having an impact on both liver clearance of aggregate as well as restoration or protein function. So that's part one. And then obviously, first step is getting into the clinic, and then that's clearly the assessment that we do as part of that proof of concept -- proof of mechanism study. I think what's important on a third piece is not only liver and lung, but also what that opens up to the entire ADAR editing capability in general. So there's an ability to assess this particular therapeutic program. But by doing that, really unlocking our capability to expand much more broadly across other editing modalities.

Great. That's really helpful. I think we have another -- a lot of other questions waiting, so I'll hop in the queue.

Our next question comes from Joon Lee with Truist.

In addition to optimizing your AIMers for editing efficiency, is there a way to screen for immunogenicity? Do certain stereopure oligonucleotides have more or less immunogenicity compared to other [ therapure ] ones and do AIMers form secondary structures? Just curious because a company developing antisense oligo for Angelman has impressive efficacy, but has some inflammation issues that the FDA may be concerned about. So any thoughts there would be helpful. And I have a follow-up.

Yes. No. And we have a lot of thoughts there, because I think when we think about programs like you brought up before, I mean, I think that brings up just the challenges that we -- Chandra spoke to at the very beginning of the call, which is when you're dealing with mixtures of hundreds of thousands and millions of dollar goes depending on the length, you have all different isomers, right? So isomers are not a wave issue. Isomers are a mixture issue. What we have is the ability to do what you're saying, which is how do we create, and that's really the work that we've done now over the last years, which is being able to build characterization of individual isomers before we go into studies. So to your point, the work that goes into is testing a molecule that we're going to potentially bring to the clinic would involve that assessment around tolerability and immunogenicity. That's part of the characterization process that goes into identifying the individual algorithm. The benefit of that and really getting back to the whole thesis when we started around really rationally designing and why you move forward with a single drug is, once we do characterize it that will be the isomer that goes into a patient. So you can have fluctuations of what dominant isomers are and what aggregates. And I think that can cause some of those changes you're alluding to. So as part of the buildup to selecting the program, yes, we do that analysis, not just in how we pick the sequence. But actually on the design of that individual oligo before it goes into the clinic. In terms of structure, and I'll just see if I don't complete the answer, I will -- for Chandra. But I think one of the other pieces that's important for us using short therapeutically relevant size, oligonucleotide is to avoid really along oligo that could form secondary structures. So by staying in the size range we are and the modifications we are, we're not concerned about that. That's obviously a bigger concern as people build potentially bigger, longer molecules. I think you had a second question.

How about the reach of ADAR editing system compared to Cas13 and now Cas73 like an editing capabilities, if you can compare and contrast some of the pros and cons there would be great.

Yes. I'm happy also to kind of keep opening up as we look back. But I think stepping back again on why we approached RNA editing and RNA editing via Adar is, first and foremost, to use an endogenous machinery inside the cell, so that we could drug it, attack it, right? So to be able to utilize an endogenous base edit or inside a cell to do the work we need to do. We like RNA for the period of pieces that we brought up at the beginning, which is when we can control redosing, so we can give the drug more frequently. It's reversible. We can characterize and make sure that by targeting RNA, we're not permanently editing the human genome, right? So we're not interfering with DNA. So there's a whole bunch of pieces that really drive our piece of building a re-dosable, titratable, reversible RNA editing capability that we see as a core feature as we think about editing differently than, to your point, some of the other tools that are available for editing.

Our next question comes from Paul Matteis with Stifel.

This is Alex on for Paul. On the AAT program, the 15 micromolar is great. I'm just curious if you're hoping to get more towards the normal serum concentration in the 20-plus range as you move forward? And would that need to be all wild-type? How are you thinking about sort of the target for achieving that serum level for your lead candidate?

Yes. No, and we appreciate it. And when we think about the range, and this is why that threefold number becomes really important is, what we are trying to do in an attempt for these patients who are homozygous. So if we think about the homozygous patient population, about 200,000 patients in the U.S. and Europe, the premise for the treatment for these patients is to restore a heterozygous phenotype. Therefore, by achieving -- even that these patients who go on to get protein replacement therapy at 11 micromolar weekly, our view is the ability to give them a restorative protein where they can consistently stay above that threshold in a way that they can self-administer subcutaneously and hopefully do that infrequently. So I think as we get to your point on 15 micromolar, we think we're above that range. Obviously, the steps that will go forward as we kind of move to the clinic is really assessing how durable that is, where they stay in that range and what the redosing intervals are. But our view is, on a potency threshold, we're there, we'll have to continue to evaluate that as we move toward the clinic. But again, the ultimate goal of therapy for these patients is getting back to it and the phenotype [indiscernible]. And that, we believe we can achieve based on where we are now. So that's the current plan, but we want to be able to evaluate. Again, how frequently they need to administer. I mean we think, as we talk to physicians and patients, the ability to have subcutaneous administration that doesn't require patients to go in frequently, whether it's one or even emerges to three, having to go into hospitals for IV infusion, that's important. I think the second piece, just not to forget that right now, we're talking about a threshold and protein, which, again, provides pulmonary protection. But I think what we're also interested in evaluating as we think about the characterization for looking at the liver clearance of the aggregate, opens up the second piece where we see the advantages of an editing approach now, which is both the protein production that you're referring to, but also the removal of the liver aggregates, right, which ultimately provides hepatic protection. So if you go back, I mean, we're already achieving with that dose. Paloma was sharing 50% editing, which is kind of where you theoretically want to be for a heterozygous phenotype. So I think that was really -- the important update for us today was moving even from where we were to where we are and now really setting us on the course to bring a program forward. So we're excited.

And then one follow-up question to you. Beyond sort of standard safety studies, what sort of work can you do, if any, in nonhuman primates to kind of get a better understanding of those properties before getting into the clinic in humans?

Great question. So what's unique is they're not -- they're in transgenic HD. So for us, it really is about really evaluating the PK/PD from the SA1 level. And then ultimately, like we would do for any program, evaluate distribution, safety and all of the other work that you do in NHP. The support we have, and I think this is why we're excited about this program, really becoming a defining program for ADAR and editing for us is, there is a very good correlation between PK/PD from a mouse with GalNAc to an NHP with GalNAc to a human. And so leveraging on the GalNAc distribution work, we can lean on that in terms of looking at correlations of concentration to continue to support what we said across the whole portfolio is one of the core drivers as we think about C9, as we think about SNP3, as we think about N531, is being able to have PK/PD modeling to support dosing and then animal models to support therapeutic index and ultimately, have that translate to an efficient starting dose in the clinic. And I think we're on the path to evaluating that with the Alpha-1 antitrypsin program. I don't know, Ken, anything you want to add, but...

No, I would just follow-up. It becomes just like any other development program, right, where we want to learn about dose response relationships and tolerability in a highly relevant pharmacological species as we prepare to go into the clinic.

Our next question comes from Luca Issi with RBC.

Congrats on all the progress. Great CD editing efficiency going up for higher doses. Maybe one question on business development and the other one on the clinical path here. So on business development, I think, we'd love to hear thoughts on how you're thinking about BD in the context of ADAR. It looks to me that the ProQR and Eli Lilly deal as well the shape deal with Roche, were really designed to share the long-term upside, more than optimizing for the upfront cash. I'm wondering if a similar deal will be of interest to you? And then maybe on clinical, I'll just double-click on a prior conversation. I know early days, but how are you thinking about the ultimate endpoint that is required for approval for A1AT. Will A1AT serum level be sufficient? Or do you need to show either an [ FPV1 ] benefit, maybe fibrosis benefit or possibly both? Any thoughts on that would be great.

Yes. No, and I'm excited to talk about both of those topics. I'll just say at the beginning, one of the things that was exciting as it relates to the 50% editing we shared is we actually saw the 50% editing at cap the dose. So I think the improvement of potency is not just about being able to push doses higher to see more editing, which is obviously exciting in and of itself. But I think what was critical, at least from the take-home is that 50% number was achieved at half the dose from those initial data sets we shared previously. So it's an important note. As it relates to BD, I mean, I think you summed it up really nicely for us, which is there's a tremendous amount of value that we can create with the current program on its pathway. And I think we want to make sure that as it relates, at least now in Alpha-1 antitrypsin, that we can make sure that, that program, as I said, it goes to distance to get to clinical evaluation and assessment of protein and really validates the underlying principle of editing. Now as it relates to being able to expand the universe of things we can do and as we said, there's a substantial number of different indications as we share today from everything, from immunology and what we can do in immune cells through CNS. I think we're very disciplined around spending the resources and the capital that we have in areas that are aligned with our core strategy. And I think as we think about the ability in CNS and other tissues to drive it there. So I think there are opportunities as we just have discussions with a number of ways of expanding our portfolio. BD is a core driver. But it's important, another piece that we've used in the past is that we also won deals that not just fully fund the work that we're going to do, but really bring additional capital into Wave that our runway spending. So deals that are runway extending, deals that expand the portfolio are critical to us. So as we look around, it's great to see interest shaping up in this space. But BD has always been part of our strategy, and will continue to be. As it relates to clinical endpoints, as we shared earlier as well, the initial clinical endpoints will be really focused on really that assessment of propane, because it's going to tell us a lot, right? The ability, one, to see and characterize wild-type protein is going to give us the ability to characterize our editing efficiency in humans. And the ability to look at whether or not there's any mutant protein, protein will allow us to assess is that -- is there now protein coming out of the liver that we can assess. And so as we think about that core initial data set that's going to provide critical information to us, that's really important. And then once we have that, there's advances on two fronts. One is, obviously, how the program progresses around threshold levels of protein and what happens to the liver in terms of demonstrating protection of both and how we think about studies that can build the label appropriately. But then there's also the critical feature, which is, again, unlocking the full potential of ADAR, where we will be driving ADR programs across the portfolio that gets supported by that. So a lot of activity in this space, and that's why we're excited to share that today with you.

[Operator Instructions] Our next question comes from Eun Yang with Jefferies.

So for the optimized AAT formulation, you mentioned the clinical candidate selection next year. Once you selected a candidate, I assume that you're going to be running an IND-enabling study, which is going to take about a year. So is it reasonable to assume the clinical studies start sometime in 2023?

So before we guide to when the study is going to start, I think you're right. So the good news, they don't take a lot of years. And we view it as a lot of work goes into that candidate initially that defined as pharmacology. And then the next thing happens, as you pointed out, are one of the IND-enabling studies that need to be done. Now the benefit, again, of being an RNA therapeutic, and in a lot of ways, is similar to what we've done before, is with three clinical programs already underway and having a pretty precedented path and characterization, we would take this through the standard IND-enabling studies. I don't know, Ken, if there's anything you think we should add to that?

No, I think that's right, Paul. It's going to be a fairly standard and straightforward IND-enabling study package, really only, wrinkle is the duration of action and how long those studies need to run, but we'll have more information about that as a lead emerge, our candidate emerges. And we get ready for those studies.

Okay. And then one of the gating factors for choosing the clinical candidate is obviously durability. So can you talk about what's the kind of durability you are looking for, number one? And the second question is on the second quarter conference call, you guided that there would be additional durability in those response data in second half of this year. So does that apply to the optimized AAT formulation?

So I'll answer both those questions. So we provided the durability on the first construct, and I think that's what we were sharing in terms of where we were in [indiscernible]. That got us as -- it almost showed out for three weeks. I think our view is, given that right now on the protein at a weekly avenue we know there's more work. We want to go as long as possible. I mean, the advantage of this therapy is, one, the potential for self-administered subcu therapy. So the key for us is to push that dosing interval out as long as possible. Step one in that was could we achieve higher potency. So pushed that curve up. And as you saw in the data today, we could achieve that. And the next is to see how that characterizes itself out in terms of durability. But I think we've learned across the PN platform that, that is a core characteristic of our medicines. If we go back to the C9 preclinical data, we had data remember out at six months still with complete knockdown of the poly (GP). So I think to Ken's point earlier on, the work that gets done, we're going to be characterizing that durability work. But the first core threshold that we needed to achieve was substantial potent editing that puts us well within the range that we would like to be in with achieving that. I think now is about the optimization of the rest of the pharmacology.

I see. So are we expecting additional data before year-end? Or is it going to be more in the 2022?

So I think there's always a potential with a lot of work underway to deliver data. I think guiding on that is always difficult, but the team here is constantly generating data as we move towards the clinic, and it is appropriate. We'll continue to provide updates on key drivers that kind of build to the program. So I assume we'll have ample opportunities to continue to provide updates on the program. You did mention one thing at the beginning, and I think it's important because we do know that others are using like liposomal formulations. And so obviously, you put a formulation into a vial. But what's important to remember is that these are GalNAc conjugated all of those. So we're not formulating them in [indiscernible].

This concludes the Q&A portion of today's webcast. I will now turn the call back to Paul Bolno, President and CEO of Wave Life Sciences, for closing remarks.

Thank you, everyone, for joining the webcast, and thank you, everyone at Wave for their hard work and especially in preparing the data and presentation today. We look forward to connecting with many of you after this call. Have a great day. Thank you.

Thank you to all our participants for joining us today. This concludes our webcast. You may now disconnect. Have a good day.