Hello, and thank you for joining us today for this investor event regarding our ADPKD program. We will be making forward-looking statements today. Actual results may differ materially from those discussed, and we encourage you to review our SEC filings, including our most recent 10-Q regarding risk factors related to our business. Our objectives today are to provide a background on Regulus and our ADPKD program. We also have assembled a number of leaders in the field of ADPKD to discuss the disease, its genetic drivers, the role of miR-17, and new ways to characterize the disease's evolution. We also will be presenting our next gen antisense oligo, RGLS8429, its role to discovery and preclinical validation data. And then our colleagues at Regulus will discuss our clinical development plans going forward and longer-term strategy, after which we'll have time for questions and answers. A bit of a brief background on Regulus. Regulus was established in 2007 as a joint venture between two leaders in oligonucleotides, Ionis and Alnylam, where we were their exclusive licensee to their collective technology and intellectual property to develop oligonucleotides directed at microRNA. We have since evolved the technology to preferentially distribute to the kidney. Now, microRNAs are short, highly conserved, non-coding RNA molecules involved in the regulation of gene expression. However, their aberrant expression of this microRNA is implicated in numerous diseases. Regulus is focused on developing oligos that sterically bind to target microRNA that are over or under expressed to address rare diseases of the kidney and CNS. Today's agenda will include an overview of the disease and the standard of care by Dr. Alan Yu, followed by a discussion around the genetics of the disease, the biology of polycystins by Dr. Peter Harris. Then Vishal Patel will discuss the role of miR-17 in ADPKD and its discovery around the implications in the disease. Finally, Dr. Tim Klein will present the novel imaging biomarkers we're employing in this study longitudinally to characterize the changing cystic architecture in patients with ADPKD. Then my colleagues at Regulus, Dr. Ed Lee, in charge of Translational Medicine, will walk through the discovery of RGLS8429 and its preclinical characteristics. Then my colleague, Dr. Preston Klassen, will provide an overview of the clinical program, and we'll wrap up with a Q&A session. With that, I'll turn it over to Dr. Alan Yu to provide an overview of ADPKD. Alan? Hello, I'm Alan Yu from the University of Kansas Medical Center, and I'm going to talk about the overview of ADPKD and the standard of care. ADPKD is a rare disease, but as rare diseases go, it is by far one of the most common, meaning that the diagnosed prevalence is about four in 10,000, which means that in the U.S., there are approximately 166,000 patients. It's caused predominantly by mutations in two main genes, PKD 1 and PKD 2, of which you'll hear more later. Mutations in these genes cause the renal tubule epithelium to dilate and enlarge, and then pinch off to form fluid-filled cysts in the kidney that enlarge progressively over time, cause the entire kidneys to enlarge, so that even though a normal kidney is about the size of your fist, PKD kidneys get to the size of a football or even larger, and fill the abdominal cavity and cause considerable complications, which include hypertension, hematuria, urinary tract infections, and kidney stones. Eventually, it causes a decline in kidney function and chronic kidney disease, leading to end-stage renal disease or kidney failure, which occurs on average at age 55. ADPKD is the fourth most common cause of kidney failure in this country and accounts for 5%-10% of patients requiring dialysis or a transplant. The cysts can also occur outside the kidney, most notably in the liver, which can become massively enlarged and cause considerable discomfort. In fact, chronic back, flank, and abdominal pain are extremely common, and about 12% of patients are on prescription pain meds. PKD patients also develop cerebral artery aneurysms, about 10% of cases, and these have the propensity to rupture and cause subarachnoid hemorrhage, which is a devastating form of stroke. So this is a severe and devastating disease. The cysts, we believe, start in utero and grow throughout life. So here, what we're showing is that the cysts are growing exponentially throughout life, and as they do so, the entire total kidney volume also enlarges and tracks really well with cyst volume. And that's extremely useful because total kidney volume can be quite easily measure. From an MRI or CT scan, and so that is a really good biomarker of disease progression. In contrast, glomerular filtration rate, which is commonly used as a measure of kidney function, is a very late biomarker. So this is a theoretical diagram showing that total kidney volume, or as is shown here, often adjusted for height. So height-adjusted total kidney volume increases exponentially throughout life, but GFR stays normal for several decades of life due to compensatory mechanisms, and then only fairly late on, starts to drop quite rapidly. So height-adjusted total kidney volume is a much better early prognostic biomarker, and the corollary of this is that children and young adults often have normal GFR and may remain undiagnosed for many years, or may be diagnosed, but because their kidney function is normal, might not get referred to a nephrologist for appropriate treatment. So height-adjusted total kidney volume is currently, the strongest prognostic biomarker we have for ADPKD. This is one of many studies that have looked at this, showing the association between height-adjusted total kidney volume and measures of decline in kidney function, a decrease in GFR, stage 5 CKD or end-stage renal disease. And to put this in context, what this study showed was that in patients in the highest as compared to the lowest quintile of height-adjusted total kidney volume, they had a 19 times increased odds of developing kidney failure 13 years later. So total kidney volume is now qualified by the FDA for use as a prognostic enrichment biomarker, meaning that its measurements at baseline can be used to select patients with ADPKD at high risk for progressive decline in renal function for inclusion in interventional clinical trials. But TKV is not the whole answer because it just provides a snapshot of the disease progression at one point in time, and doesn't account for the fact that patients of different ages might have faster or slower progression. To account for that, the Mayo Group developed an imaging classification system in which they took accounts of one kidney volume and the age at which that was measured, as shown on the left, to extrapolate a trajectory for the rate at which those kidneys have progressed, and used that to sub-classify patients with typical imaging findings into five groups, ranging from 1A, over here on the right, which are representing patients with the slowest rate of kidney growth, to 1E, the patients with the fastest rate of kidney growth. As shown in this study, Mayo Imaging Class predicts kidney function decline later in life very well. So over on the left, you can see patients with class 1A have the slowest rate of decline. The eGFR hardly changes as they get older, and at the other extreme, on the right, class 1E patients' GFR is dropping even when they're in young adulthood and reaches end-stage renal disease on average in their forties. So switching gears, let's talk about current therapy for ADPKD. So currently, there is only one approved therapy specific for ADPKD, and that's tolvaptan, a vasopressin V2 receptor antagonist that inhibits the action of vasopressin in the collecting duct and reduces cyclic AMP levels, which are thought to drive cell proliferation and cyst fluid secretion. The only other things we have are supportive care, antihypertensive therapy, drugs that inhibit the renin-angiotensin system, pain control, and treatments of complications. This shows the evidence for the efficacy of tolvaptan in ADPKD in the two pivotal trials, TEMPO 3:4 on the left and REPRISE on the right. So in TEMPO 3:4, tolvaptan, as compared to placebo, reduced the annual rates of total kidney volume growth by about a half. And on the right, it shows that tolvaptan reduced the rate of decline in GFR over 12 months by about a third. So clear evidence that tolvaptan is efficacious, but this efficacy is somewhat limited, and we don't know whether it's sustained over a long period of time, but sufficient to project that if it were sustained, that dialysis would be delayed or transplant would be delayed by 2.5-three years. But this comes at a cost. tolvaptan has significant side effects, most notably aquaresis side effects. So because it's blocking the effects of vasopressin, which is a hormone that concentrates the urine, it causes polyuria, nocturia, and thirst, and that occurs in at least a third of patients. There is a rare incidence of elevation in liver function tests, transaminases, that in one study caused two patients to reach a level where they were at risk of liver failure. And so even in the trials, 10%-15% of patients have discontinued due to adverse effects. So tolvaptan is approved to slow kidney function decline in adults at risk of rapidly progressing ADPKD. In practice, in the US, many of us use the Mayo Imaging Classification system and consider classes 1C to 1E as being indicated for treatment. tolvaptan distribution is restricted due to the REMS program that's required by the FDA, Risk Evaluation and Mitigation Strategy, which essentially means that patients have to have monthly liver function tests for 18 months and every three months thereafter, which can be a significant imposition for the patients and for nephrology practices. Use of tolvaptan around the country is highly variable. In Kansas, we have a high uptake of tolvaptan. Other centers, on the other hand, some of them have low use of tolvaptan, and many nephrology practices don't use it at all. Tolerability has been a major limitation. Average fluid intake that's required is six liters a day, so that's a lot. That's like 12 twenty-ounce bottles of water a day. Patients have to have access to a bathroom to go to void every half an hour to one hour. So this is not a medication that can be taken by airline pilots, long-distance truck drivers, teachers, surgeons, and so on. And we often find also that our young patients that have active social lives don't want to take this at all. And so it's clear that there is still a huge unmet need in ADPKD for an intervention that has improved or added efficacy to tolvaptan and/or is better tolerated than tolvaptan. Finally, I would just comment that there is a fairly nice path to approval in ADPKD. So there are the usual clinical endpoints that we use for kidney disease medications, which are end-stage renal disease and death. The accepted surrogate endpoint for full approval, which is worsening kidney function, generally measured by the eGFR slope. But unique to ADPKD, total kidney volume has now been accepted as a reasonably likely surrogate endpoint, which means that it can be used as a basis for accelerated approval, with the proviso that a post-marketing study with clinical endpoints can be conducted. So with that, I think I'll conclude here, and I'm looking forward to the questions later. Thanks, Alan. And now I'd like to hand it over to Dr. Peter Harris to discuss the genetics of ADPKD and the biology of the disease. So, thank you for inviting me here to discuss the genetics of ADPKD and the cell biology of the polycystins to this Regulus Therapeutics Research and Development Day. So ADPKD is associated with a lot of changes at the cellular level, including dedifferentiation resulting in turning on of fetally expressed proteins like c-Myc. Polarization defects, where molecules that are normally on the basal or lateral surface land up on the apical surface. There's a moderate rate of increase in cell division. The epithelium changes its characteristics, so that it's secretory, so that it continues to increase the cyst size, even if they're separated from the tubule, and the extracellular matrix becomes disorganized. So one of the questions in ADPKD is: How do we make a cyst if we have just a single change? The classical view is that a second somatic mutation is required within the somatic tissue, and only then, when there's no polycystin present, does a cyst develop. I'll show you a piece of data in a minute, supportive of that idea. The other view is that the haploinsufficiency of the polycystin itself can be a trigger or increasing the likelihood of cyst development, and then other changes such as kidney injury damage, stochastic changes of expression, or other events can lead to a cystic transformation, and I will show you data from hypomorphic models, which is consistent with that idea, and both models may be in play in ADPKD. This is data about somatic mutations, published by Hanna Renner, Renner's group, and you can see here that mutations, somatic mutations were found throughout the gene and interestingly, in up to 93% of the cysts which were analyzed. So similar to the level that we find from testing of germline variants. The question here, I think, is whether these cysts are required for cyst initiation or if they occur in larger cysts as part of the growth and survival of those cysts. And I think of importance, these types of studies are done in larger cysts. So there's two major genes for ADPKD, PKD1 and PKD2. PKD1 accounts for about 78% of cases, and PKD2, 15%. PKD1 is a rather complex gene, with the first 33 exons in a complex region of the genome that's duplicated, making it a little more difficult to study. As well as these two genes, there's a number of other loci I will highlight. Still some cases that are unresolved and other, genomic complexity. So if we look at the breakdown of the, of the PK types of variants, that we have, causing ADPKD, and this data is from the ADPKD variant database, we can see in the 63% of, PKD1 cases, we have a variant which is predicted to truncate the protein, and so 37%, including a large number of missense, but some in-frame changes and a small number of atypical splicing, which are classed as, non-truncating. In PKD2, we have a larger majority that are truncating changes, around 86%, and then with a smaller number of missense, and then a significant number of non-canonical splicing changes. The variants, and we're looking here at PKD1, are found throughout the gene. They're of all the different types that we have causing genetic diseases, so splicing, frame shifting, in-frame, missense, and then larger rearrangements. Any single change that causes an inactivating lethal allele is likely to result in PKD. Of importance, there's no single change that accounts for more than 2% of the ADPKD population worldwide, and in our database, we have over 1,650 variants, but there's more than that now. So a lot of different variants cause this disease. We've classified variants into ones that are maybe missense changes that may be fully penetrant, like the one on the left with a large chemical difference and at a well-conserved site in orthologs and in the domain. Whereas on the right, this is one that may be more likely to be hypomorphic, a more conservative change, and although it's conserved in orthologs, it's not conserved in the domain structure. As you've heard from Alan, there we use imaging also to classify patients. And so patients with a larger kidney volume at a particular age here classified into the five Mayo imaging classes are more likely to progress to renal failure more rapidly than the ones with smaller kidneys, and you can see that on the right side of the slide here. When we looked at by Kaplan-Meier analysis at age, renal failure, but judging by the genotypic groups on the left here, you can see that the truncating variants are associated with most severe disease, around 55 years on average, PKD2 around 74 years, and then the two missense groups are in there at around 60 and 64 years of age. If we look at the Mayo imaging classes, we can see even greater difference with the largest kidneys. The 1E group are reaching renal failure about 25 years more earlier than the 1B group, with the 1A group, with only few patients reaching renal failure. So this is showing you that the imaging class is collecting other information apart from just the genetic cause of the disease, which is probably other genetic modifiers as well as environmental factors. We can also see that there's a change in trajectory also associated with the severity of the disease, and the bottom left probably best shows this, with the patients with the milder disease having a region of a period of conserved function before renal failure occurs. Whereas for patients with the largest kidneys or with PKD1 truncating changes, we see a decline at an early stage. If we look at the relationship between the genotype and the Mayo imaging class and the size of the kidneys, we can see that the genotype they progress at about the same level, but we can see the PKD1 truncating have the higher or larger kidneys at 20 years of age. And likewise, for the size groups, the 1E have a much larger kidneys at 20 years of age. So this is showing you that there's a lot going on before 20 years of age, which is determining the severity of the disease. Unusually, we find biallelic variants that are causing an ADPKD-like disease. Here, we can see a missense change that is well conserved and causing renal failure at about the same time as a normal PKD1 change. But in this case, we have two copies of that variant. They're homozygous. Individuals that are heterozygous in this pedigree just have a small number of cysts. To try and prove that this was a hypomorphic change, we made a mouse model, and we can see in this mouse model that we have slowly progressive disease up to 12 months of age. If this was a fully inactivating mutation, then the animal would not be viable, and obviously, if it was neutral, we wouldn't have cyst development. These hypomorphic alleles are important in so-called biallelic cases, causing very early onset disease, as illustrated here. Whereas a normal truncating nonsense mutation to PKD1 is the disease-causing change, but in the infant with a very severe disease, we have the addition of the hypomorphic variant I showed you on the last slide. If we look, if we then mimic this in a mouse model, we can see that we have very rapidly progressive disease, with 25% of the body weight of the animal made up by kidneys at 25 days of age, and many of the animals dying at this early period. So that shows that this hypomorphic change is having a modifying effect along with the nonsense mutation. So this gives us, brings us to the idea that the dosage of the polycystin is important to the severity of the disease. If we don't have any polycystin, this is not viable in mice or in humans. If we have a 50% reduction, we have the adult-onset ADPKD that we're familiar with. If we add in a hypomorphic allele, we can have the early onset, and if we have a hypomorphic allele, which it can mimic the presentation of PKD2 or with a very just a few cysts if it's a weak allele. That doesn't mean that there's not other chance factors going on, like chance factors, somatic variants, other germline variants, and kidney injury, which are modifying how the disease presents in each individual patient. As I mentioned, there's a number of other genes now which are associated with an ADPKD phenotype. Most of these, apart from IFT140, are involved in the process of folding and trafficking of proteins, and I'll come on to show a little bit more about this. And you can see that each of them has a rather individual characteristic. DNAJB11, for instance, the kidneys stay small, but because of fibrosis, we have renal failure later in life. IFT140 is a cilia-related protein, and in this case, we get a few large cysts that result in enlarged kidneys, but not usually renal failure. If we look at the whole PKD, ADPKD population overall, and this is illustrated with the CRISP and the HALT populations, we can see here in this group, there's about 80% are PKD1, either truncating or non-truncating. About 15% are PKD2, truncating or non-truncating. About 3.7% remain unresolved, but you can see the other 1% is made up of a number of other genes, some of the ones that I just showed you, as well as unusual presentations of collagen genes. So the PKD1 and PKD2 encode polycystin one and polycystin two. These two proteins form a complex, and they're membrane-bound proteins. Polycystin-2 is like a TRP or a type of TRP channel, calcium channel. Polycystin-1 has a very large extracellular region. It's thought that these proteins reside on the primary cilium, as well as other sites in the cell. But the site on the primary cilium is likely important for PKD development, and there's plenty of pieces of evidence associated with that, including that there's a number of syndromic diseases which affect the growth or the development or transport of cilia into cilia, which also have PKD as part of their phenotype. We can see here from cryo-EM images, the structure here with three Polycystin-2s associated with one Polycystin-1, and a channel or a pore here, which is somewhat different in structure than we see in the heterotetrameric Polycystin-2 channel. As I mentioned, cilia is thought to be important in PKD, but the precise role of the polycystins of complex on cilia is somewhat controversial and unresolved. There's suggestions of a flow receptor particular signaling pathways such as calcium, cyclic AMP, and Wnt, a regulator of cis-dependent, cis-activating pathway. It's been suggested by the Somlo group. Since there's vesicles that contain high levels of these polycystins, and at least in other species, cilia can bud off vesicles, whether they may also be important as part of the mix. If we breed together a Polycystin-1 and Polycystin-2 model, we can see on the top right-hand corner or in the top right-hand image, that we get much more severe disease than with the animal models by themselves. And we think this is because Polycystin-1 and Polycystin-2 are important for trafficking the polycystins to the cell surface and also to the cilium. If you look at the Polycystin-2 negative cells at the bottom, you don't see the mature form of the polycystin one. And so that's why we think that there's an interrelationship between Polycystin-1, Polycystin-2, and each is important for the other's trafficking. If we take a mutation to another of these genes, like GANAB, which is important for glycosylation and quality control of the protein, you can see also that this mature form of the polycystins is absent, and you can see on the right-hand side that Polycystin-2 is not transported to the cilium. As I mentioned, there's quite a large number of these proteins that are involved in glycosylation, folding, quality control, and transport of the of Polycystin-1 and Polycystin-2, as well as a wide range of other membrane and secreted proteins that are involved in polycystin. The polycystins seem particularly sensitive to reductions of proteins in this pathway. We've used this relationship between Polycystin-1 and Polycystin-2 to assay the significance of some missense changes, and this is measuring the surface localization of co-expressed Polycystin-1 and Polycystin-2. And only when they're co-expressed, do we see the surface localization of the protein. If we use this assay, we can see that many of the Polycystin-1s that are predicted to be missense changes that are predicted to be pathogenic are not seen on the surface. We're looking at the top diagram here with no surface signal, whereas some other high weaker alleles, we see more surface localization. At the bottom, we can see what we see with some of these hypomorphic alleles, including the 3277C, whereas we get an intermediate value. Again, suggesting that dosage of the polycystin is important for the severity of the disease. Recently, there's been exciting work from the Somlo group that's shown that reexpression of Polycystin-1 using a transgene in animals that already have PKD development can reverse the disease. And you can see the difference between the third and the fourth is to do with reexpression of the polycystin. So this is optimistic for the idea that we might be able to reexpress the polycystins or maybe increase the level of the polycystins that may be advantageous. There's a lot of different signaling pathways that have been suggested. This is a somewhat old diagram, so I've put some other pathways here at the bottom, but there's still quite a lot of uncertainty of which is important. Cyclic AMP is the one with the only approved treatment right now, tolvaptan here, as a vasopressin receptor antagonist. So I think this trying to target downstream pathways in the PKD still may be problematic. So just to summarize, there's two major genes, PKD1 and PKD2, with a truncating PKD1 variants more severe than PKD2, and PKD1, more than PKD, more severe than PKD2. We can see this just looking at kidney size. There's a number of other genes now that have been associated with an ADPKD-like phenotype. Many of these involved in folding and trafficking of Polycystin-1. We feel that a dosage model best explains what we see in patients and in animal models, although somatic mutations may also be important for cyst growth and survival. It's multiple signal pathways that have been involved in ADPKD, but this is not completely sorted out yet. I think proximal treatments targeting the level of the primarily defective proteins, the polycystins, is attractive in ADPKD for this reason. So thank you. This is the conclusion of my presentation now. I look forward to answering questions in the Q&A session. Thank you. Thanks, Peter. And now we'd like to hand it over to Dr. Vishal Patel to talk about the role of miR-17 in ADPKD. Hello, my name is Vishal Patel, and I'm a professor of internal medicine in the division of nephrology at the University of Texas Southwestern Medical Center. Our long-standing interest has been understanding ADPKD pathogenesis, and particularly, in particular, developing therapeutic approaches for ADPKD. As you've already heard, ADPKD is caused primarily due to heterozygous mutations in one of two genes, PKD-one or PKD-two. So as far as therapeutic development is concerned, we focused our energies towards these genetic root causes. So let me start by taking a step back and introducing key genetic mechanisms by which PKD-one or PKD-two mutations produce ADPKD. So everyone here knows that each one of us inherit two copies of the same gene, one coming from the father, shown here in blue, and the other one coming from the mother shown here in pink. In individuals with ADPKD, one copy is defective that comes from the affected parent, in this example, the father. This lowers the dosage by 50%. However, it turns out this 50% reduction is not sufficient to phenotypically express the disease in target tissues. In many cells in the cystic kidney, the second normal copy undergoes a somatic second hit mutation, which then lowers the dosage to 0%. Of course, at 0% dosage, that is sufficient to cause ADPKD. However, we now know that many cells in the kidney of individuals with polycystic kidney disease continue to express the remaining normal copy. So in that setting, where the normal copy is still around, how does the disease progress? We now know, after lots of modeling in mice and other preclinical models, that the dosage needs to fall by an additional 20%-30%. In other words, the remaining normal copy gets suppressed by 20%-30%, lowering the dosage now -70% to -80%, which falls below a critical threshold and now expresses the disease in the kidney. We've been very interested in understanding where does this delta arise from? So more formally stated, these are our research question: How is the remaining PKD1 copy regulated in the context of ADPKD? Any insights into this question, perhaps, will allow us to harness the remaining copy for what essentially amounts to endogenous PKD1 or PKD2 gene therapy. So now that I'm gonna segue into microRNAs by showing you this beautiful painting by Guillaume Courtois. He, of course, is depicting the biblical battle between David and Goliath. I use this as a way to introduce microRNA biology. MicroRNAs are, in essence, David, and in this setting, mRNAs are Goliaths. MicroRNAs are a fraction of the size of mRNAs, yet when sequence alignment is proper and Watson-Crick base pairing happens, this interaction leads to microRNA-mediated repression of mRNAs. So when I started my independent laboratory at UT Southwestern in 2010, we were extremely interested in understanding how this genetic warfare plays out in the context of kidney diseases, in particular, polycystic kidney disease. So we began really by doing very simple experiment, and we simply took PKD1 and PKD2 mutant mouse kidneys and human ADPKD samples and screened for aberrant microRNA expression in these three datasets. We overlapped the three datasets to identify an aberrant microRNAs gene signature in mice and human samples of polycystic kidney disease. In it, we identified a microRNA called miR-17. We've worked on this microRNA for many years now. We've published quite a bit on it. And I'm not gonna go over all of the data, but it presents some of the key highlights of what we have learned over the intervening 10 years. So the first thing is, if miR-17 is the David, what is its Goliath? A simple bioinformatic search revealed that some of the top miR-17 mRNA targets were PKD genes, in this example, PKD1. So I'm showing you a PKD1 3' UTR nucleotide sequence aligned by various species, and the red box is the miR-17 binding motif. This in silico prediction, indeed, we verified using independent experiments and show that miR-17 physically interacts at this red box locus on PKD1 mRNA in cells, both from mice and humans, and also tissues of human, of mouse kidneys. This interaction, physical interaction, actually leads to PKD1 repression. So I'm not showing you all of that data, but in essence, with this new information, the idea then became that in individuals with polycystic kidney disease, where one copy of the PKD1 gene is already inactivated, mRNAs that are made by the remaining copy are further repressed by epi factors, in our case, miR-17, which reduces microRNA function, further lowers gene dosage, and aggravates disease progression. We've spent quite a bit of time to flesh out this biological pathway, and I'm gonna show you some of the key experiments that we've done using in vivo gene modeling. So one of the first experiments we did was to take normal kidneys from mice and transgenically express miR-17 in them. What we've noticed, as you can see here, is in miR-17 overexpressing kidneys, we saw microcysts. These cysts actually came from collecting ducts, which is the target tubule type in ADPKD. In addition to this cystic phenotype in mice, what we found that at the molecular level, high levels of miR-17 was associated with reduced PKD1 gene dosage. So now we have two independent observation. One, miR-17 expression is at least partly sufficient to phenotypically present with cystic like disease. And second, at the molecular level, it seems to repress PKD1 gene dosage. So, overexpression of miR-17, perhaps partly, is sufficient to produce a PKD-like phenotype. Next, we turn to mice that already had polycystic kidney disease. So let me briefly explain the genetic makeup of these mice. These mice carry two PKD1 copies. The first copy is a germline mutation, called the RC mutation, shown here in red dot. All cells of the body of these mice have this RC mutation. The other copy is normal. So in essence, these mice are RC plus. The plus copy, however, has loxP sites in critical introns of the PKD1 gene, shown here as yellow and purple triangles. So now when you expose Cre to this setting, a kidney-specific Cre, you remove the floxed copy, and in essence, turn the RC plus mouse now into an RC minus mouse, lowering PKD1 dosage, activating miR-17 expression, and consistently expressing, producing a reliably aggressive polycystic kidney phenotype, shown here. In this setting, when you delete miR-17, we see a marked phenotypic regression. The kidneys are smaller and shown here in graphs, KW or BW stands for kidney or body weight. Red dots are single mutants, green dots are double mutants, and as you can see, deleting miR-17 lowers the kidney weight over body weight ratio, suggesting that the kidneys are smaller. Moreover, multiple studies assessing kidney function, shown here, is blood urea nitrogen, also showed that miR-17 deletion improves kidney function. More importantly, in addition to this phenotypic regression, what we found was that in the double mutants, PKD1 dosage was increased. So very simplistically thinking about it, if you go back to the genetic makeup of this mouse, one of the major ways by which PKD1 dosage goes up in the double mutant is if the second non-inactivated allele is, in fact, producing more PKD1. This was perhaps an indirect indication that removing miR-17 out of the system allows the remaining copy to function at a higher level compared to when miR-17 is still present in the system. So that leads to a very simple model. We all know that PKD1 is required to suppress this growth. We propose, based on the studies that I've just shown you, that miR-17 represses PKD1, and in that way, modifies the ADPKD phenotype. So life's simple, miR-17 represses PKD1, which we know represses ADPKD. Well, nothing is this straightforward in biology, unfortunately. There's a couple issues with this model. First is, microRNAs, unlike transcription factors, are not on and off switches. They act as rheostats. So they dial down gene dosage of their target mRNA by 10%, 20%. So the question at the time was: Is miR-17 mediated repression of PKD1, which perhaps is 10%-20% reduction in dosage, is that even biologically relevant, especially, when faced with such, such an aggressive phenotype? The second question is, miR-17, like all other microRNAs, are is pleiotropic. It targets many other mRNAs besides PKD1. And so... How can we be certain that majority of the miR-17 biological effect is in fact being derived through PKD1, and not from some of its other mRNA targets? So in essence, the question is, are the beneficial effects of miR-17 truly via PKD1 de-repression? To address this question, we really need to disentangle and decouple this pleiotropy. In essence, we need to design an experiment where we retain miR-17 in the system, maintain its repressive activity on its mRNA targets, except the PKD1 gene. If preventing miR-17 interaction with PKD1 is sufficient to phenocopy the effects of deleting miR-17 itself, that would imply that miR-17 biological activity in the setting of ADPKD is derived through PKD1 de-repression. To model this experiment, we devise this approach. I'm showing you the PKD1 gene in a cartoonish way, obviously not drawn to scale, but the blue box is the three prime UTR region. Within this blue box is the red box, where the miR-17 motif resides. We use CRISPR-Cas9 gene editing to remove the miR-17 motif. So the idea here is that the endogenous mouse PKD1 gene now lacks the ability to produce an mRNA that contains the miR-17 motif. And therefore, this mRNA will evade miR-17-mediated repression and perhaps will be more translationally stable, and may function better. So here's the evidence that indeed our editing approach works well. So I'm showing you the miR-17 motif shown in green letters, and that's shown here in caps. The scissors are basically the guides that we use to chop off the miR-17 motif. The free ends at which the motif is cut off, then reconnect, and we produce a contiguous mRNA that lacks the miR-17 motif. We model this editing approach in the context of ADPKD. So here's the experimental setup. I've already shown you this mouse model, the RC minus mouse, that reliably produces polycystic kidney disease. What we're saying is that the remaining copy of the PKD1 gene has an Achilles heel, and that Achilles heel is its miR-17 binding motif. Removing this binding motif, using an editing approach that I just described, could perhaps stabilize the mRNAs that are made by the remaining copy and could have a beneficial effect in the setting of polycystic kidney disease. So here are the data. RC plus, where one copy of the PKD1 gene has a germline RC mutation, the other copy is wild type, has normal kidney. Compare that to RC minus. I've already shown you this data. RC minus kidneys have aggressive polycystic kidney disease. When we mutate the miR-17 binding site from the RC allele and the other copy of the PKD1 gene is still around, the kidneys continue to remain normal. However, when we remove the miR-17 binding motif from the RC copy and the other copy is now deleted, we see a marked phenotypic regression. Moreover, from a molecular standpoint, we find that PKD1 protein level is increased in the RC delta 17 minus mice compared to RC minus mice. For the aficionados in the audience, PKD1 biochemistry truly is very challenging and we've tried to make sure that this finding is actually real and tested two different antibodies that go after two different epitopes, one at the C terminus and one at the N terminus, and we find consistent reproducible results. I want to point out that the four different genotypes that we've derived all came from the same founder, so there isn't a simplistic explanation of different genetic backgrounds to explain for these findings. Nevertheless, we've confirmed these findings by making a second founder of delta 17 mutations and a third founder. In each case, we see that removing miR-17 binding site from the RC allele increases PKD1 protein expression and leads to phenotypic regression of polycystic kidney disease. This visual data are shown here in a graphical presentation. Again, KW or BW means kidney or body weight ratio. Y-axis here, please note, is broken to accommodate all data points. And horizontally, I'm showing a dashed line that represents normal value. Compared to RC plus, RC minus mice have dramatically enlarged kidney, representing severe ADPKD. When you remove the miR-17 binding motif from the RC allele, that brings the kidneys back closer to normal. Moreover, again, the BUN blood urea nitrogen levels also improve or normalize. In vivo modeling of miR-17 motif deletion shows that it phenocopies the effect of deleting miR-17 microRNA globally. The question next was whether this is a phenomenon of the RC mutation and of mouse modeling, or is this a more generalizable finding? And so for this, we moved towards primary human ADPKD cell lines. Our collaborator at University of Kansas Medical Center, Dr. Darren Wallace, provided us with cyst epithelial cells that are derived from ADPKD kidneys. We got four donor cell lines, and we did a couple of things with it. First, we genotyped these cells and found pretty severe PKD1 or PKD2 mutations. Please note that even though these cells came from the cysts of people with ADPKD, the second copy of the PKD1 gene was normal, shown here with a plus sign. So these are indeed heterozygous mutations present in cyst epithelial cells. Next, what we did was we edited the miR-17 motif from these cell lines and compared these edited to unedited isogenic controls. So each donor cell line serves as a control for the edited pair. Western blotting revealed that PC1 or PKD1 protein levels go up every time we removed the miR-17 binding motif from the remaining allele, suggesting that even in human primary ADPKD cell lines, taking miR-17 away from the remaining PKD1 copy leads to higher PKD1 protein expression level. We did phenotypic analyses of these cell lines. So growing the edited and unedited cells in 3D Matrigel cultures, we noted that compared to the unedited cells, isogenic control, isogenic edited mutants, in which miR-17 binding motif had been removed, grew at a slower pace, suggesting that increasing PKD1 protein dosage leads to smaller cyst formation in this 3D culture model. Moreover, using immunofluorescence, we found that p-CREB levels were reduced in the edited compared to the unedited isogenic controls. p-CREB, by the way, is a readout for the cyclic AMP pathway that is targeted by tolvaptan, the only FDA-approved medication for polycystic disease. This and other assays to us suggested that harnessing the remaining copy and raising PKD1 protein levels leads to the resolution of downstream pathogenic signaling events that involve proliferation, 3D cyst growth, cyclic AMP pathway activation. So this brings me to the conclusion of this presentation. The conclusion is that what we thought was a David and Goliath story is, in fact, an Achilles heel story. Our conclusion is that PKD1, especially in the context of ADPKD, has an Achilles heel, and that Achilles heel is its miR-17 binding motif. This leads to cis-repression of the PKD1 mRNA in the setting of ADPKD. An anti-miR-17 drug that harnesses the remaining PKD1 allele and, in essence, acts as an endogenous gene therapy agent, is in clinical development. At this point, I have to mention, I didn't go over the data, but miR-17 indeed is a pleiotropic factor, and it targets many other mRNAs that may be ADPKD relevant in addition to PKD1. Some examples are PKD2, PPAR alpha. Some of the biological effects of miR-17 overactivity in ADPKD is likely derived also through repression of PKD1, PPAR alpha. The beneficial effects of an anti-miR-17 therapy, it probably is also derived through derepression of PKD2, PPAR alpha, et cetera. With that, I conclude. Thank you for your attention, and I look forward to taking questions during the Q&A session. Thank you. Thanks, Vishal. Now I'd like to hand it over to Dr. Tim Klein to discuss novel imaging biomarkers in ADPKD. Hello, and thanks so much for joining today. My name's Timothy Klein. I'm an AI scientist and assistant professor of Radiology at Mayo Clinic, and today I'm gonna be talking about novel MR imaging biomarkers for ADPKD. So I think we all know the importance of measurements of total kidney volume. This is FDA-approved prognostic biomarker for ADPKD research as well as clinical trials. It's used heavily to track disease progression, as well as evaluate therapy effectiveness. And really, the imaging modality of choice here is MRI, and this is due to a number of reasons, including its high soft tissue contrast. It also utilizes non-ionizing radiation, as well as provides very high accuracy for the measurements. And therefore, we sought to develop automated tools in order to make this process efficient, accurate, as well as reproducible. However, the phenotypes associated with ADPKD are highly variable, so patients with fairly similar total kidney volumes can have very different disease presentations. So highlighting here, if we take the two images on the left, two patients with fairly similar total kidney volumes, but very different phenotypes in terms of their cystic burden. So in the case on the left, we have a case with lower number of cysts as well as larger cyst sizes, whereas on the right we have a greater number of cysts as well as smaller cyst sizes. So doing something that can quantify these differences is really of interest to our group. So our objective really was to develop an automated segmentation method to quantify these different ADPKD phenotypes. And we can imagine, starting with our MR image, we can do things like individually segment all of the cysts, which gives us a number of in-depth metrics in terms of total cyst number, cyst size distribution, as well as cyst parenchyma surface area as well. And with these measurements, we can also do things like look at overall total cyst volume, as well as calculate things like cystic index. So wanted to highlight here an overview of the AI model that we built to do this. I won't go into too much detail, although just to highlight that what the model essentially does is take the MR image and output a segmentation or mask that separately labels all of the cysts in the images. And this is what you can see in the bottom right. We did a fair amount of validation work, comparing the output of the AI with human readers. So in this case, we're showing a number of different comparisons, where in Panel A, we have the inter-rater measurements, so the comparison between Reader one and Reader two. As you can see, there's some bias where one of the readers tends to measure more cysts than the second reader. So this really highlights the variation that can happen with two different readers performing the measurements. As we move to Panel B, we have the comparison of Reader one to the AI model, Panel C, Reader two to the AI model. And finally, what was particularly interesting was in Panel D, when we compared the average of the two readers to that of the AI model, we saw a very close match. Which gave us confidence that the AI model was a really nice, accurate, and reproducible method to do things like measure cyst number and some of the additional imaging metrics that we discussed previously. Shown here is a visual comparison of the human readers versus the AI. So you can see the output or the tracings that were performed by the two different readers, as well as the automated approach and a 3D rendering which really highlights the differences between these three different cases as we move down the columns. You can see that in this case, the automated method is allowing us to measure anything from 10 to thousands of individual cysts. After developing the tool, then we sought to evaluate its utility for predicting progression of PKD. So here we used what is known as the CRISP study data, which is the Consortium for Radiologic Imaging Studies of Polycystic Kidney Disease. This was a study that began in 2001, and the purpose of this study was to use MRI to track kidney volume in ADPKD patients, really aiming to correlate these measurements with renal function decline. I also wanted to highlight what the findings were from this study, which was that the MRI measurements of kidney volume reliably indicated ADPKD progression, and this study really formed the basis for linking increased volume with reduced renal function. So what we were able to do then was go back and look at these, baseline images, extract, the additional imaging, metrics, and then assess how informative they were, for predicting subsequent progression. Wanted to highlight that, to do something like this, instance-level segmentation of the cysts, we really require a high, quality, regarding the images that are acquired. So in this case, there were actually 24 images, from this study that weren't ultimately utilized, based on a few different, reasons highlighted on the right side, where we saw some cases that had really low cyst contrast. So essentially, even a person couldn't identify the cyst within some of the slices within the image. In the second example, we have motion that would occur oftentimes from respiratory motion that would essentially cause sort of a jigsaw pattern, where then we wouldn't be able to measure properties like the cyst surface area very well. And finally, in C, we have large slice thicknesses. So some cases only had images with something like nine-millimeter or greater slice thickness, and then we really can't measure some of the small cysts accurately. So just wanted to highlight that then in this study, we're really starting to look at now incorporating these new imaging metrics within the clinical trials. So showing some examples of the rich data that we can pull from just a single MR image. So here we, in Panel A, have a T2 weighted MRI. Then we're doing things like kidney and liver segmentation to get organ volumes. But then applying this instance level cyst segmentation allows us to get properties like total cyst volume, total cyst number, and other metrics like the renal parenchyma volume and cyst parenchyma surface area. So what was particularly interesting was applying these new imaging metrics then to the CRISP study, where we could extract these imaging features that from images that were acquired in 2001, and then look at subsequent changes in kidney function, in this case the eight-year slope of eGFR. And what we're showing in the middle is the scatter plots comparing the eight-year slope eGFR to various image-based metrics. So for example, in the first panel, we have a height-adjusted TKV, where we see an R squared relationship of about 0.19. Then moving through the newer imaging metrics and basically seeing that we can stratify much better and predict much better how these patients are gonna progress. So something like total cyst number or cyst parenchyma surface area, there we're getting a relationship on the order of an R squared of about 0.26. I also wanted to just highlight what these visual representations mean. So essentially, each of the color codings, so if we start on the left side, we have the yellow diamond image, and below it the yellow circle image. These are two patients with very similar total kidney volumes, but very different presentations in terms of their cystic burden. So these are really the metrics that, that are now able to be provided by the instance segmentation. We're essentially comparing, various, kidney volume sizes, where the bottom are cases that, progressed, much more rapidly than the cases on the top. I think you can qualitatively, see these differences. This is something that a clinician is doing all the time, incorporating some of the visual phenotype information. However, we're trying to put, hard quantitative, measurements, in, in their place. So I wanted to highlight, that there's obviously a lot of, potential opportunities for future research as well. So something that has been worked on by my group is to go beyond, just quantification of cyst number to actually do classification of the cyst. So for example, differentiating simple cysts from proteinaceous or hemorrhagic cysts. And there we can utilize information from multiple pulse sequences. So for example, a T2, T2-weighted, combined with a T1-weighted image, and use the intensity information to drive a classification to differentiate those simple cysts from maybe some of the more complex cysts as well. Another potential opportunity is looking at individual cyst tracking. So our current method allows us to calculate cyst number and track this over time, but to actually track individual cysts and see their growth changes is a particularly difficult image processing task. But our lab has been working on this for some time, and I think this will be something that we can incorporate relatively quickly. So in summary, we've worked on developing a lot of new imaging metrics that can be extracted from routine MR images. Some of the ones that we highlighted included total cyst number as well as cyst parenchyma surface area. I think excitingly is that this study now is the first interventional trial to start using and measuring a number of these parameters. And there's also exciting opportunities for future investigation, for doing things like cyst classification and cyst tracking as well. And with that, I'd like to acknowledge those who've worked on this project, particularly Adriana Gregory and Dr. Torres, Dr. Harris, and Dr. Erickson, as well as our funding sources. And that concludes my talk today. I look forward to answering questions during the Q&A session. Thanks, Tim. Now I'd like to hand it over to my colleague, Dr. Ed Lee, to discuss RGLS8429's discovery and preclinical characteristics. Thank you, Jay. Hello, my name is Edmund Lee. I'm the Vice President of Translational Medicine here at Regulus. Today I'm gonna talk about the discovery of our next generation, anti-miR-17 oligonucleotide, called RGLS8429, for the treatment of ADPKD. Before we start, I would like to provide some background regarding Regulus' focus on anti-miR-17 oligonucleotide. Regulus has a long-standing interest in miR-17 biology and its role in different disease areas, including oncology and ADPKD. As you've heard from Dr. Vishal Patel earlier, genetic knockdown of miR-17 has been shown to attenuate cyst growth in different kinds of mouse models of ADPKD, shown here, including PKD1 knockout and PKD1 flox RC mouse model of ADPKD. In collaboration with the Patel lab, together, we have shown that the first generation anti-miR-17 oligonucleotide, called RGLS4326, have demonstrated efficacy also in multiple mouse model of ADPKD. including shown here, PKD2 knockout and also PKD1 flox RC mouse model. It is with this data, and together with other non-clinical safety profiles from our IND enabling study, that prompt us to initiate the clinical development of RGLS4326. In particular, in phase I-B clinical trials, the first-generation molecule have demonstrated statistically significant increase in urinary PC1 and PC2 levels in patients who have ADPKD. However, during further development of the first-generation molecule, we've observed dose-limiting CNS effects in animals who received repeat high doses of RGLS4326, including in the chronic toxicity study. Although no CNS effects were observed ever in humans for the first-generation molecule, with the anticipated limitation on dosing in patients, we've made the decision to pursue our next generation molecule, called RGLS8429 instead, in order to eliminate the dose-limiting CNS effect. So what is RGLS4326, or what I would call 8429 simply in this, the rest of the presentation? 8429 was designed to specifically avoid the dose-limiting CNS effects while still maintaining all the beneficial attributes of our first-generation molecule. In the next several minutes, I'm gonna discuss how we characterize, how we discovered the molecule, and most importantly, show that 8429 have shown similar potency, PK, PD, and efficacy profiles as compared to our first-generation molecule. Most importantly, no CNS effects were observed in the IND enabling studies for 8429, including the recently completed mouse chronic toxicity study. Right now, 8429, the clinical development of 8429 is currently underway. As a reminder, RGLS8429 was granted orphan drug designation by the FDA last year. We've recently completed the phase 1 clinical trials of RGLS8429 in healthy volunteers, and we're now right in the middle of phase I-B trial, evaluating RGLS8429 in ADPKD patients. So how do we get here? Well, in order to get to RGLS8429, we asked a question: What is actually causing the CNS effect for the first-generation molecule to begin with? Given that we know inducible knockout of miR-17 in adult mice did not cause any CNS effects, we hypothesized that the CNS effects was not related to the on-target inhibition of miR-17 by our first generation molecule but instead was related to its off-target activity. In order to find what off-target activity we're talking about, we turned to an in vitro safety pharmacology panel, namely the Panlab SpectrumScreen, which includes over 170 different molecular targets, such as transporters, ion channels, and GPCR receptors. Using this panel, we have identified 39 molecular targets that were able to displace the natural ligand of the respective receptors. Further characterization led us to narrow down to three targets with binding IC50 of less than 10 micromolar. Because we know that binding doesn't necessarily lead to modulation of receptor functions, we turn to the functional assay of those specific three different receptors and try to find out whether or not 4326 has actual functional effects on those receptors. The data from those study allow us to identify a receptor called the AMPA receptors, which upon treatment with 4326, its activity would be inhibited. In particular, 4326 behave as a competitive antagonist to inhibit AMPA receptor function, with the IC50 values of around 0.4 micromolar. This is important because AMPA receptors are glutamate receptors on the excitatory synapses that mediate fast excitatory neurotransmission in the CNS. And more importantly, Perampanel, an FDA-approved AMPA receptor antagonist, which is indicated for the treatment of epileptic seizures, also cause similar CNS effect in the non-clinical studies. Putting this together, we hypothesized that AMPA receptors, or rather, the off-target inhibition of AMPA receptors from the first generation molecule, is causing the CNS effects. We now know that through our collaboration with Dr. Sobolevsky from Columbia University, that our first-generation molecule indeed directly interacts with the AMPA receptor via the ligand binding domain. On the left-hand side, you can see the cryo-EM structure, showing that 4326, or the three-prime end of 4326, is inserted directly into each of the ligand binding domain clamshells of the AMPA receptor, with the five-prime end of the molecule protruding outwards. If you look closer, you can see that 4326, when binding to the ligand binding domain, is positioned in the way that otherwise would be occupied by the natural ligand glutamate. Even closer, you can see that the way the 4326 is interacting with the AMPA receptor. We've been able to identify specific chemical moiety of the RGLS4326 that interact with the same amino acid residues of the AMPA receptor that would otherwise be interacting with the glutamate. The structural information was incredibly helpful. As you see on the left-hand side, when you overlay the cryo-EM structure of the glutamate bind, ligand binding domain structure to the RGLS4326 binding structure, you can see that again, RGLS4326 is occupying the same space that was otherwise being occupied by the glutamate. More importantly, upon binding to the ligand binding domain, it actually further opened up the clamshells to a more rigid structure, which in turn caused the closure of the channel pore at the transmembrane domain. Putting all this together, it allows to have a structural understanding of how RGLS4326 is inhibiting AMPA receptors. Under normal circumstances, you see the natural ligand glutamate binding to the AMPA receptors, allowing the proper function of the receptors. In the presence of RGLS4326, RGLS4326 competes with glutamate, displaces the natural ligand, binds to ligand binding domain, leads into a more open structure, leading to the closure of the channel pore at the transmembrane domain, and ultimately causing CNS effect that we've observed. Because we know that AMPA receptor interaction is causing a CNS effect, we design a focused library based on the structure of our first-generation molecule and use it to screen against binding with the AMPA receptor. This allow us to identify a specific chemical moiety within RGLS4326, that's responsible for the off-target inhibition. On the left-hand side, you can see anti-miR-17 oligos that contain the specific moiety, interact with the receptor, and was able to displace a radiolabeled AMPA from the receptor, just like 4326. In contrast, for the anti-miR-17 oligos that do not have the specific moiety, did not interact with the AMPA receptor. And most importantly, we've been able to shown that that specific part of 4326 is not important for its potency against miR-17s. In the middle panel, you can see that regardless of whether or not an anti-miR-17 oligo contain this moiety, it still have very similar activity against miR-17. This is further showcased by the data you can see on the right-hand side, where a head-to-head comparison between RGLS8429 and the first generation molecule, that RGLS8429 show very similar potency against the intended target, miR-17, based on the luciferase assay, both based on miR-17 or its downstream target PKD1. Because RGLS8429 does not interact with AMPA receptor, when we further test RGLS8429, no CNS effects were observed in mice receiving RGLS8429. Even when we test it at the dose level and duration, while exceeding dose that we've tested for the first-generation molecule, including both acute, subchronic, and the recently completed chronic six months toxicity study in mice. On the left-hand side, you can see that after daily, weekly, or biweekly dosing of our first generation molecule, as soon as you reach a certain level of total dose, around 2000 milligrams per kilogram, CNS effect started to occur. In contrast, for RGLS8429, because it doesn't interact with the AMPA receptors, in similar study design frequency, including daily, weekly, and biweekly dosing, even at a much higher total dose level, no CNS effects were occurred. Similarly, no CNS effect were observed in non-human primate study receiving RGLS8429. We anticipate to initiate a chronic toxicity study in non-human primate by the end of this year. Now, before we move on to further characterize RGLS8429, I just want to mention one thing. Recall I mentioned earlier on that Perampanel, which is an AMPA receptor antagonist indicated for the treatment of epileptic seizures. Perampanel show efficacy in a mouse model of severe tonic seizures. And when we test RGLS4326 in this, in this model, as we expect, because RGLS4326 also antagonize or inhibit the function of the AMPA receptors, RGLS4326 also show efficacy in this mouse model. In contrast, using another anti-miR-17 oligos that have very similar potency compared to our first-generation molecule, but did not interact with the AMPA receptor, no efficacy was observed in this model. This is particularly important because putting all this data together, it further support the notion that CNS effects indeed was caused by off-target inhibition of the AMPA receptor, but not related to on-target inhibition of miR-17. Moving on further characterization of RGLS8429. Previously, we have shown that following a single subcutaneous administration, our first-generation molecule have a preferential distribution profile to the kidney, as shown on the left-hand side in the subpanel A. When we test 8429 head-to-head with the first generation molecule, the two molecules show very similar PK or pharmacokinetic profiles, where 8429, just like 4326, also show preferential distribution to the kidney with a kidney to liver ratio of around at least 20-fold. On the right-hand side, you can see that as you would expect, 8429 also shows very similar PD profile compared to our first generation molecule. This is measured by assessing the amount of miR-17 that following treatment, how much miR-17 were being displaced from the high molecular weight polysome in the kidney, including both tested in the normal mouse model or a PKD mouse model. Finally, in collaboration with Dr. Patel's lab, we have shown that RGLS8429 also have very similar efficacy in both the human primary ADPKD 3D cyst culture model and also in vivo in the PKD1 flox RC mouse model of ADPKD. On the left-hand side, you can see that RGLS8429, just like our first-generation molecule, was able to increase PC1, PC2 proteins, reduce the cyst growth, and overall decreasing the cyst growth of the 3D culture, just like the first-generation molecule. On the right-hand side, you can see the 8429 after treating it in the PKD1 flox RC mice, can reduce the kidney weight over body weight ratio, cyst index, as assessed by histology, proliferation, and also kidney function, as shown by serum BUN. In summary, we've shown the inhibition of miR-17 has shown very promising results in both mouse models, in preclinical settings, and also in ADPKD patients in the clinical setting. We've shown that inhibition of miR-17 can slow disease progression, preserve kidney function, and extend survival in multiple preclinical mouse models of ADPKD. In the clinic, we've also shown that using a first-generation molecule can increase key PD markers, urine PC1 and PC2, in ADPKD patients. Altogether, this is particularly important because you've just heard from the early speakers that by other independent groups, increasing PC1 and PC2 level has been shown to show efficacy in ADPKD mouse models. So in closing, I would just like to mention that the clinical development of RGLS8429 is currently underway. I've just shown you that RGLS8429 retain all the favorable potency, PK, PD, and efficacy profiles compared to RGLS4326. More importantly, we've identified the specific chemical modality within RGLS4326 that is responsible for its off-target inhibition to the AMPA receptor. And we've been able to engineer out the specific chemical modality of the molecule for our next generation molecule, RGLS8429. Because eight four two nine does not interact with the AMPA receptor, no CNS effects were observed even when tested at dose level and duration, while exceeding dose that we've done for 4326, including the acute, subchronic, and chronic mouse toxicity study. We anticipated to also start the chronic toxicity study in non-human primates by the end of this year. That concludes my presentation. Thank you for your attention. I would love to answer any questions you might have during the Q&A session. With that, my colleague, Preston, will discuss in a little bit more detail regarding clinical development of RGLS 8429. Thank you. Thanks, Ed, and hello, everyone. My name is Preston Klassen. I'm President and Head of R&D at Regulus Therapeutics, and I'm going to briefly cover the clinical development program for RGLS8429, with specific emphasis on the ongoing phase I-B multiple ascending dose study, including the study design, as well as our expectations regarding the upcoming data, and then also our plans for later stage clinical development, including the opportunity for accelerated approval. There are four key components to the clinical development program: a phase I single ascending dose in healthy volunteers, which is already completed, and in addition to safety and tolerability, it demonstrated a PK profile that is comparable to our first generation, RGLS4326. phase I-B multiple ascending dose in patients with ADPKD, which is ongoing, and it measures Polycystin one and Polycystin two as the key pharmacodynamic and mechanistic readouts. We'll talk more about that study in a bit, including our expectations for that polycystin readout and also the use of exploratory renal imaging that Dr. Klein has highlighted today in his presentation. Importantly, this study will be used to select the dose that we take forward into phase II. Now, the phase II trial will be randomized, placebo-controlled, with a 12-month treatment duration and a primary endpoint of total kidney volume measured by magnetic resonance imaging. As I will cover in a moment, we think that there's significant precedence for FDA's acceptance of this study design as a single pivotal trial supportive of accelerated approval in ADPKD. Following accelerated approval, confirmation of clinical benefit will be required in the form of a phase III post-approval trial. This study will again be randomized, placebo-controlled, and will examine the rate of change in kidney function over two years, as measured by estimated glomerular filtration rate, or eGFR. I want to emphasize that this planning for phase II and phase III, including the use of an accelerated approval pathway, does require confirmation with FDA, which we will be obtaining. I'll talk about all of this further in a moment... Last year, we completed the single-ascending dose study in healthy volunteers with 8429. It was weight-based dosing of 0.5, one two, and four milligram per kilogram, and the study focused on safety tolerability and the pharmacokinetic profile. RGLS8429 was well tolerated with no serious adverse events, and overall, AEs were more frequent in the placebo arm than inactive, and the profile of AEs did not demonstrate any patterns of concern. The plasma exposure was approximately linear across these 4 doses, and importantly, was similar to the PK data from the first-generation compound, RGLS4326. Now, the ongoing phase I-B multiple-ascending dose study is enrolling patients with a confirmed diagnosis of ADPKD, a Mayo classification of 1C, 1D, or 1E, and an eGFR greater than 30 mL/min. We plan to test weight-based doses of one, two, and 3 mg/kg, dosed every other week over three months. On the basis of these weight-based dosing results, we will test one or two additional fixed doses with the ultimate goal of moving to a fixed dose in a prefilled syringe for later stage development and, of course, commercialization. Each cohort targets 12 subjects in a three-to-one randomization. In addition to safety and tolerability, our primary focus is on the pharmacodynamic and mechanistic marker, polycystin, both polycystin one and polycystin two. We will be performing MRI at baseline and after three months to evaluate the kidney and a variety of PKD-specific measurements. These are the measures that Dr. Klein spoke to today, and we have particular interest among these in total cyst number and cyst parenchyma surface area. This study is on track. We've completed enrollment in the one milligram per kilogram cohort, and we anticipate those data will be available later this month or in early October. The two milligram per kilogram cohort is near completion in terms of enrollment, and we intend to announce key data from each cohort as it becomes available. So let's talk just a bit about what we expect to see and learn coming out of this study. Again, the key focus for this trial is that PKD and mechanistic marker in polycystin. As you've just heard from Doctors Harris and Patel, polycystin helps regulate renal tubular cell growth and function, and when polycystin is suppressed, you see abnormal proliferation and the transformation of renal tubular epithelial cells into cystic epithelial cells. And a key part of the pathobiology of ADPKD is that miR-17 is upregulated in disease, and that suppresses polycystin. And because RGLS8429 blocks the action of miR-17, we should see subsequent increases in polycystin, and we will therefore use polycystin as that key PD mechanistic readout to determine the dose that we do take forward into phase II. Now, we do expect to see significant increases in both PC1 and PC2 with this first dosing cohort at 1 mg/kg, and I'll demonstrate why we are confident in that on the next slide. We will also be examining pre- and post-renal MRI in the trial, in part to get experience with executing MRIs in the clinical trial setting, because the logistics of that kind of imaging can be challenging. And I'll also be clear on our expectations in terms of the imaging results after I talk more about polycystin. As I mentioned, we do have confidence that we'll see increases in polycystin with this first 1 mg/kg cohort with RGLS8429, and that is specifically because we have already demonstrated significant increases in both PC1 and PC2 with our first-generation product, RGLS4326, when dosed over six weeks. This slide shows phase I-B polycystin results with 4326 at 1 mg/kg, dosed out to 44 days or six weeks, and polycystin continues to increase even after the dosing is stopped. By week 10 here, we saw a 60% increase in Polycystin-1 and a 40% increase in polycystin-2. As I just covered with you, we know that our current compound, 8429, has identical binding to miR-17, with similar potency against miR-17 and the same level of anti-miR-17 activity as seen with the first-generation 4326. We also know that 8429 has similar pharmacokinetic profile as 4326 does. So the impact on polycystin at one milligram per kilogram with 8429 should be similar to what we have already seen with 4326. In fact, it may be greater since we're dosing for twice as long, 12 weeks instead of six weeks. So we do have a high degree of confidence in the ability of RGLS8429 to demonstrate those significant increases in polycystin with just this first cohort, and then we'll explore the dose range by moving up to two milligrams per kilogram and finally three milligrams pe phase III r kilogram in terms of weight-based dosing. We also are excited to examine some of the novel imaging parameters or markers that Dr. Klein just talked about. This will really be the first time that these kinds of imaging techniques are measured in a prospective PKD clinical trial. These measures will, of course, be exploratory, and we don't really have significant expectations here since they haven't really been examined this way before in a clinical trial setting. In terms of the more standard total kidney volume, we also don't have significant expectations that we will see reductions in TKV over three months, as that usually takes at least 12 months of treatment to be able to see. However, if some of these newer measures, as we believe they may be, are more sensitive, perhaps we'll see some directional improvements. And so it really makes sense to get experience with both the MRI procedure itself, because that's, of course, an important part of the next phase II trial, and also get some experience with these novel and perhaps more sensitive imaging measures. I'll turn briefly to our plans and expectations for phase II and phase III development. Importantly, ADPKD is a serious disease with significant unmet need and where surrogate endpoint, total kidney volume, is considered to be reasonably likely to predict clinical benefit. As such, ADPKD does meet the requirements for when accelerated approval pathways from a regulatory perspective are considered. FDA has really already set precedent for allowing an accelerated approval pathway for other sponsors in ADPKD, and two of them are listed here. So therefore, we believe it is possible to achieve accelerated approval on the basis of a single pivotal phase II trial, if we can demonstrate statistically significant reduction in total kidney volume over 12 months compared to placebo. As with all accelerated approvals, confirmation of clinical benefit is required after approval. In the case of ADPKD, this would mean a phase III trial demonstrating significant improvement in eGFR over time compared to placebo, and this is typically a two-year trial on treatment duration. So our current plans for the pivotal phase II trial for accelerated approval will include approximately 300 patients, with either one dose randomized two to one against placebo, or two doses randomized one to one to one against placebo again. There will be a 12-month treatment period, and the primary endpoint is total kidney volume. Decisions on the final dose selection for this study will be made once the results of the ongoing phase I-B trial are complete. And we're also considering having patients in this phase II trial continue blinded therapy for an additional 12 months in order to contribute to an eventual phase III eGFR endpoint, and those plans will be finalized next year as we engage with the FDA and consider the capital structure of the company. So in summary, we're very excited about the clinical program for RGLS8429. Our operational execution of the phase I-B trial is on track. We believe the probability of success is high based on the knowledge we've gained from our first-gen program. The primary goals of this phase I-B trial, again, are to provide the key mechanistic readout in polycystic, to gain experience with conducting renal MRI in a clinical trial setting while exploring novel imaging measures, and finally, to enable selection of a fixed dose to take into phase II. The planned next study will be at phase II trial, examining total kidney volume, and that does have the potential to serve as a single pivotal trial for accelerated approval, with, of course, a post-approval requirement to confirm clinical benefit in phase III study utilizing eGFR. We will have data from our first cohort in the phase I-B coming up in the next few weeks, and we do look forward to sharing that. Data from future cohorts are expected to be shared as they complete. Additional information regarding the design of phase II, the timeline for execution of that trial, will be shared as we see the complete phase I-B dataset and interact with regulatory authorities to finalize our plans. So in closing, in addition to our strong commitment to bringing a novel therapeutic to patients with ADPKD, we do have additional research efforts with oligonucleotides targeting microRNAs across both CNS and renal indications. These include work in rare pediatric epilepsies and ALS, among others, and we will talk more about this and our advancing renal programs, hopefully in a dedicated investor event next year. Before turning to Q&A, I want to specifically thank our academic partners, including Doctors Yu, Harris, Patel, and Klein, who graciously joined us today, and all of our participating investigators and their staff in our clinical trials. Most importantly, the patients who give their time and their commitment to our clinical programs. Thank you. Now, I think we'll turn over to the Q&A. This is going to include everyone that you've heard from today, and we will also be joined by Dr. Rekha Garg, our Head of Clinical and Regulatory, as well as Cris Calsada, our Chief Financial Officer. And Jay's going to moderate the Q&A session, so I will turn it over to him now. Great. Thanks very much for your attention during these presentations. We now are moving into our live Q&A session. We've received several questions along different lines of discussion. We're going to start off with this one question from Yanan from Wells Fargo: What level of polycystin change is expected to have meaningful clinical consequences? I'll turn that over to Vishal to discuss his views, and Preston and Ed can chime in as well. Thank you, Jay. So, any level is good as far as I'm concerned, when you're down at the bottom by 70%, 80%, or sometimes 0%. Any increase will be good. Is there a golden number that you will hit? Based on mouse studies, I mean, we think that, you know, getting the dosage above 50% would be something. And talking about dosage in the tissue, in the cells, getting it above 50% would have tremendous benefit. And that is based on a lot of data already out there, which says that if you reduce PKD1 dosage by 50%, you actually don't get the disease. So, if you bring it up close to 50%, you should get some amelioration there. Whether going above 50%, more is better? Perhaps not sure, but a recent study showed that a transgenic PKD1 reexpression in the setting of PKD1 deletion, where the dosage probably was higher than 100%, still showed some beneficial effects. So there may be some safety built in, in terms of how high you could go. The fact that the first gen showed an 80% increase in the urine, which is, please realize, an indirect measure of what's actually happening in the kidney, that to me is a very promising indication. My views are that bringing the dosage level to 50% or higher would be good. Yeah, I'll just chime in very briefly. Our opinion is, you know, very similar to that of Dr. Patel. Preclinical models, as you mentioned, you know, 50% have shown improvements with the transgenic that was recently published, pushing it up to even 150% actually helped revert kidney tissue to look normal again, or at least some of it to look normal again. So we definitely think it's a dosage effect. Really pleased that the first, RGLS4326. At 1 mg per kg, we're able to drive up to 60% improvements in Polycystin-1. And so now we're going to repeat that at 1 mg per kg with the current compound, RGLS8429, and then go to twofold and threefold higher that as we go across the dosing range. So we're excited to see the results. Great. Thanks very much. We have another question here from Whitney. This is directed to Dr. Klein. Which of the more exploratory imaging endpoints would you think would be more likely to show changes earlier in this three-month study versus total kidney volume, which we understand typically requires around a year to see an impact? Yeah, I think that's a great question. I think probably the parameter for cyst parenchyma surface area has the most potential for showing early changes. And essentially, this metric is related to where we think the cysts are having the most significant burden for the kidneys. So it's essentially removing a lot of the impact of the exophytic cysts and just focusing on the burden of the cysts in adjacent to healthy parenchyma tissue. So I think probably CPSA or cyst parenchyma surface area, as well as looking at potentially changes in cyst number as well, will be particularly interesting to investigate. We're obviously still measuring all of the typical parameters like total kidney volume, total liver volume as well. But I think these new metrics, at least from our initial research, has shown, you know, a lot of value in terms of predicting progression for these patients. So really excited to be involved and see how this looks in this study for sure. In a related question, have you looked at or thought about taking all of these imaging biomarkers collectively to predict impact on disease progression versus each one individually? Yes, we definitely have. We're actually collaborating also with Dr. Yu, who could maybe comment as well. So we've looked at applying these metrics in a few different studies like CRISP, like HALT, as well as individual institution datasets like we have here at Mayo Clinic. But Dr. Yu, maybe you want to comment on the work that we're doing, looking at essentially establishing a new sort of classification system based on these new imaging biomarkers. Yeah. Thanks, Tim. I don't have a lot to add to that exactly. We're using the CRISP and HALT database to combine total kidney volume with the cyst parameters and try and come up with a combined biomarker that would have prognostic value and potentially also, I guess, be a monitoring or response biomarker. Great. Thanks very much. We got one question, which is directed to all the experts in the field in terms of, you know, a required product profile for reasonable adoption and strong adoption relative to the approved therapeutic tolvaptan. You know, what aim should regulators take in thinking about what to try to demonstrate in a registrational trial? Maybe, Preston, do you want to triage that with each of the experts? Sure. Alan, why don't you start off? Sure. So I think that if you could develop a product that has equivalent efficacy to tolvaptan, but be better tolerated, that would already be a huge win, and I think that would be readily adopted. And so equivalent efficacy would be a 50% reduction in total kidney volume or a 30% reduction in GFR slope, which I think many of us believe is clinically meaningful, but without the significant intolerability and without the need for a REMS program for liver toxicity. Vishal or Peter, anything to add? Nothing more than what Alan said. I think hitting the benchmark gold standard efficacy, that's current gold standard set by tolvaptan. And then a safety profile, obviously that's better than tolvaptan in increasing the adaptability. That will help with increasing adaptability. So some of the few minor things, for example, you know, not having to take the medication on a daily basis, twice a day, and things like that, which clearly this drug hits, right? It's once, it's certainly not on a daily basis. So some of those profiles are, you know, probably also key benefits of this medication that could help with the increased uptake. But, tolvaptan-like efficacy and cleaner safety. Yeah, I don't really have anything to add to this. I think I agree with what's been said. I think I'll just the only thing I'll chime in on is that, you know, as we move forward, for example, in phase two, we do plan to be including some portion of patients who are on tolvaptan, because we want to get a sense of both monotherapy and with tolvaptan in combination. The two mechanisms should be different, and so, presumably, that would be additive or perhaps even synergistic, but we... You know, that remains to be seen. And that's very much in line with how FDA urges sponsors to obviously be testing, you know, any new potential assets in development as monotherapy, but also in combination with existing therapy. Thank you, all. And a related question in terms of the current market, couple questions. One is, what percentage of patients do you believe are being treated with tolvaptan today? I know that we have discussed in the past that dose reductions have been required to manage tolerability, and there's been some cycling. But there's a question related to that, and then from there, how large the undiagnosed population is and what are ways that we can address that? So we will start with you, Dr. Yu. Sure, I'll tackle that. So, I don't think we really know how frequently tolvaptan is being used, but I think it's very variable. Our center probably has one of the highest adoption rates. I think the Mayo and KU does, too, and about a third of our patients are on tolvaptan. But of the ones that get treated with tolvaptan, about a third of them drop out, in at least half of the cases because of side effects. But across the country, there's a huge variability, including places that don't prescribe tolvaptan to their patients, and it's related both to patient tolerability, and also to the administrative burden of the REMS program. There was a very interesting study that was done in Japan by Inui and colleagues, where they actually looked at the adoption of tolvaptan. There they have a national drug monitoring database across all of the prefectures of Japan. And they found a 10- to 20-fold variation in the rate of tolvaptan prescription per 1,000 of the population. So clearly, it's hugely variable. To the other question, the... I think it was the percentage. Undiagnosed. Diagnosed versus undiagnosed. Yeah. Maybe Peter can help me a little bit here. But there was one study by Lanktree, where they looked at population-based whole genome sequencing, and they found a rate of ADPKD mutations of about 1 in 1,000, whereas the diagnosed prevalence, we think, is about 1 in 2,500, let's say. For every one diagnosed patient, there is at least one to two undiagnosed. I would say in addition to that, we've done some patient engagement studies, and we find that there's a significant number of patients that what we call unactivated, meaning that they know they have the diagnosis, but they either don't want to reveal it, or they don't go to a primary care physician, or they go to the primary care physician, but they're told they don't need a nephrologist. I think there's a large hidden population there that's really not come to light yet. Yeah, I could just add, you know, I think there was a, a Geisinger study, population study recently that suggested, you know, a fairly high level of ADPKD patients, even higher than the one in 1,000 with a, with a clinical diagnosis. So I think there are patients that are, that are not necessarily coming to the notice of the, of the nephrologist. I think more genetic analysis of, of populations can be helpful to help identify these, these patients. Thank you. Related to the utilization of these novel imaging biomarkers and total kidney volume, there's a question about how the two relate, and is there a sense that there will be a lag in seeing an impact, for instance, on total kidney volume, or you might see something in imaging biomarkers earlier, and how does that tie together with respect to an overall regulatory strategy? Preston, you want to take that? Sure. I'll start, and then we can... We have others chime in. So, just to maybe take the last part first, at the regulatory strategy. It is very clear that there are... That the accelerated approval pathway utilizes total kidney volume. FDA has gone on record, it's on the list of acceptable surrogate endpoints in ADPKD specifically, and of course, other sponsors have received formal designation of accelerated approval pathway using total kidney volume. And then the other, you know, approval pathway for standard approval or once you have accelerated approval in the post-approval setting, you still need to perform a confirmatory study to demonstrate clinical benefit, and that would be through eGFR.... In terms of the relationship between, and maybe I think there was a question about lag. So our general expectations are for polycystic, that you start to see those changes relatively quickly over the course of, you know, the first couple of months. We're going to learn much more about that in the phase I-B study. But as you saw with first-gen four, three two, six, at one milligram per kilogram, we saw the curve started to come up with 6 weeks of dosing, but then actually increases continued on through 10 weeks of observation total. And so, we're just going to learn a lot more about that with this current study. And then total kidney volume itself, and Dr. Klein can certainly comment on that, tends to take longer. And then eGFR, of course, would, you know, presumably take a bit longer as well to demonstrate a clear impact. That also has to do with, you know, population, sample size, et cetera. But generally speaking, it's around 12 months or maybe 18, for total kidney volume, and then about two years of study for eGFR. And then in terms of the novel markers that Dr. Klein has developed and discussed today, I think as we're all, you know, have stated, we don't know yet necessarily. This is an important aspect of this trial to get some sense of these biomarkers, use them in a clinical trial setting. This will be the first prospective clinical trial in ADPKD that utilizes these novel measures. And so our belief is that they're more sensitive, and we'll see changes, you know, earlier, but we need to gather information to start to understand that. Dr. Klein, any additional comment on that? No, I think that covered it really well. You know, I think the big point there is that we expect more sensitivity from some of these more advanced imaging biomarkers. So seeing changes, for example, impacts on cysts of different sizes is one thing that we're interested in looking at. So, you know, obviously, kidney volume gives us a picture of sort of an overarching picture, but I think a lot of these newer imaging biomarkers are going to help us refine, you know, what changes we can impact and how those will potentially then impact patients' progression of the disease. Great. Thanks very much. We have another question from Joe Schwartz, asking whether we're expecting to see equal upregulation of both PC1 and PC2, or does this depend on the mutational status of the patients? And are we enrolling patients with both mutations? I'll start just briefly with the enrollment criterion, and ask Rekha to maybe chime in and give a better sense of what we're targeting here, and how we're characterizing the patients, and then we could turn it to Preston to triage the remaining piece. So we are enrolling patients with ADPKD, we are taking baseline mutational status. And so as we've seen before, we've seen increases in both Polycystin-1 and polycystin-2, because they're both regulated by miR-17. In terms of additional stratification or otherwise, Rekha, do you want to comment on that? Sure. So in terms of the subjects, we're also enrolling based on two criteria. One is the Mayo classification, so patients with 1C, D, and E, as well as the eGFR. So the eGFR is up to 90 in terms of the eGFR for the patient's enrollment. Patients are getting, we are getting the PKD1 or PKD2 in terms of the genetic aspects, which we're not... There's no criteria for which subjects will, you know, how many subjects to enroll in terms of that. It's more related to getting the genetic mutations for these biomarkers. So when we have the data from cohort 1, we'll know the distribution of those genetic biomarkers. Ed, maybe you could comment on the question about do you expect equal upregulation, PC1, PC2, and/or does that depend on mutational status? Yeah. I mean, historically, both from a preclinical study and also clinical study, we tend to see more increase of PC1 than PC2 in terms of magnitude or percentage increases. Now, recall that all three anti-miR-17 is expected to repress both PC1 and PC2, and we've seen that again and again in preclinical models and also in the first-generation study. So I think this upcoming study from the 1B study for RGLS8429 will be very informative in terms of further understanding what is the association between background mutation type and mutation you get versus the level of changes, what you see. Vishal, do you have any thoughts on that? Yeah. I just wanted to comment briefly about the mechanism of action, and people often ask which mutation are you trying to fix? Very simplistically speaking, you are harnessing the non-mutated remaining copy of PKD1 and/or PKD2 gene. So, it really doesn't matter which mutation the germline mutation is, is it in the PKD1 or the PKD2 gene? As long as the germline mutation is an inactivating mutation, that copy cannot be harnessed. That copy does not make an mRNA to which miR-17 can bind, and therefore, miR-17 can be repressed. That copy is gone. The other one that's around is the one that this drug would harness, right? So, a priori, deciding which mutant patients to enroll, it, you know, is not that important, frankly, as long as you have one copy or some residual PKD1 and PKD2. So that's just adding a little bit more color to what Ed and Preston were mentioning. Great. Thanks, Vishal. We're almost out of time. We're gonna take one last question here. I think it highlights the very limited options available for patients with the disease. And this comes from a patient asking: "If all goes well phase I-B, when may we begin recruitment in a phase II study?" Preston, do you wanna take that? Yeah, sure. Just at a high level. Obviously, we've got to get through phase I-B. we do plan to roll out data as each cohort does become available. We're not giving, you know, specific guidance yet in terms of the, you know, particular timeline for enrollment beginning, you know, in phase II at this point. We need to wrap up, in particular, the weight-based dose-ranging portion of phase I-B, one, two, and three milligrams per kilogram. We'll then likely do at least one additional fixed-dose cohort, 'cause we want to prepare and be ready to roll into phase II with a fixed-dose, prefilled syringe. That would be the ideal scenario because that's where we need to be for commercialization, for patients to be able to self-administer at home. Then we also do need, of course, to confirm our phase II and phase III plans, including the use of accelerated approval pathway with FDA. So all of that will take some time. We'll be providing more updates around the overall timeline for phase II and phase III as we get into the larger bulk and kinda near completion of phase I-B trial itself. But just again, in terms of specific upcoming data, as I mentioned, we do have a Cohort 1 at the one milligram per kilogram dose level, is expected to be available within a few weeks, essentially. So we do look forward to talking about that at that time. Then finally, with respect to Cohort 2, we know we started that earlier this summer. Preston, when might we anticipate completing enrollment of that and initiating Cohort 3? Yeah. So each cohort is we're targeting 12 patients in a three-to-one r andomization against placebo. And we've just been informed that we are targeting enrolling the last patient in the second cohort early next week. And so, we're already now turning our attention from the enrollment machine, so to speak, to Cohort 3. So that's great. We're on track. Things are going very well from an operational perspective, and kudos to the team for being able to execute that. Great. Well, I'll echo Preston's comments earlier. Thank you so much to our academic partners and all the investigators working in this study, and most importantly, the patients who are participating in our clinical programs, without which we wouldn't be able to be working to advance this novel new therapy for ADPKD. And with that, we'll conclude the call. Thank you, everyone.