PTC Therapeutics, Inc. (PTCT) Earnings Call Transcript & Summary

Good day, and thank you for standing by. Welcome to the PTC Huntington's Disease Deep Dive Conference Call. [Operator Instructions]. Please be advised that today's conference is being recorded. I would now like to hand the conference over to your host today, Kylie O'Keefe, Senior Vice President, Commercial and Corporate Strategy. Please go ahead.

Good morning, everyone. Thank you for joining us today for the first addition to our 2021 deep dive series. Today, we will be focused on our PTC518 program in Huntington's disease. Joining me on today's call is our Chief Executive Officer, Dr. Stuart Peltz; our Chief Development Officer, Dr. Matthew Klein; and our Executive Director of Biology, Dr. Anu Bhattacharyya. Slides are available on the Investors section of the website for you to follow along during the presentation today. I will now pass the call over to our Chief Executive Officer, Stuart Peltz. Stu?

Thanks, Kylie, and thanks for joining us today. I'm very excited to discuss with you our Huntington's disease program and our development candidate, PTC518, an orally bioavailable small molecule that we are developing to treat HD, or Huntington's disease. While there's been recent trials and tribulations on this space, we are confident in our differentiated molecule, PTC518, has the pharmaceutical properties associated with it that sets it apart. Before we start, let me remind you that today's presentation will include forward-looking statements. I refer you to our SEC filings. As you know, we have built a robust and diversified set of platforms and programs that range from discovery, development to commercialization. We believe that our pipeline will continue to drive both short- and long-term value for all of our stakeholders. Today, we'll focus on the splicing platform and, in particular, the Huntington's program and PTC518. During the presentation today, I'll cover an overview of our splicing platform and the path that we pioneered that led to the success of Evrysdi. I will also review Huntington's disease. Anu Bhattacharyya will review our preclinical program for PTC518. Lastly, Matt Klein is very excited to share our preliminary Phase I results. Let's start with the splicing platform. We have developed this platform over the last 2 decades and have built a proven track record. When we began this journey, there were many skeptics that thought it was impossible. However, with the success of Evrysdi, we have demonstrated that this is not the case. Our research has helped us establish a sophisticated set of databases to be able to identify splicing targets, and we have built a robust set of assays that allow us to look at hundreds of specific splice sites on a single assay. We also utilized our RNA splicing technology to be able to rapidly sequence and identify all splice sites within the genome. In addition, we've developed a proprietary library of compounds that modulates splicing biology. Lastly, we've assembled a world-class team of scientists who understand how to modulate splicing and have acquired the experience and expertise to chemically optimize these molecules for the appropriate biological and pharmaceutical properties. In essence, we have built the science, the infrastructure and the chemical libraries that allow us to continue as the leaders in utilizing splicing as a target to develop new therapies. Now I'll go a little deeper on the splicing platform. Here's an example of an RNA that has multiple exons and neutrons, and there are specific regions of the RNA that define the intron and exon borders. There's a complex called U1, shown here on the right, that binds to the region that defines the 5 primary ends of the exon-neutron border. while the U2 conflict shown on the left, lies specifically to 3 splice site junction. Both complexes are made up of both proteins and RNA. We have currently based our platform and identified molecules that modulate splicing due to the U1 splice sites. To remind you, in SMA, people have a mutation or lack the SMN1 gene survived because of SMN2. The SMN2 transcript differs from SMN1 in that most of the time exon 7 gets skipped, leading to a lower concentration of full-like SMN protein that leads to this disorder. Risdiplam strengthens the splice site and pushes the reactions to make a full-length protein and, therefore, is able to treat the disease. On the next slide, it shows you why. On the bottom, you can see a cartoon of the U1 SMN1 RNA. On the left, you can see a perfect match for the U1 binding site on the exon-neutron junction. In the case of SMN2, it's not a perfect binding site, allowing the U1 to be easily to stabilize or fall off. If you look at the top part of the slide, you can see what the structure looks like. On the left is the perfect structure. However, on the right, you can see the SMN2 is a noncanonical element that has mismatches that forms a pocket. Risdiplam binds to the open pocket stabilizing the interaction and allows inclusion of exon 7. As you can see on the left, SMN mice treated with isoprene show increased levels of SMN protein in all tissues and organs. Risdiplam upon treatment resulted in efficacious SMN protein levels in the brain and reach levels in which the mice would have no disease phenotypes. On the right side, you see in the analogous level of SMN protein shown in the blood. We know that the SMN is an intracellular protein. The results here were key to us. That gave us high confidence that a given dose would lead to an exposure level in the blood that we can measure. We demonstrated in multiple clinical trials that Risdiplam treated patients showed SMN levels found in healthy volunteers. On the graph on the left, you'll see baseline. And on the right, you see post treatment. These powerful results ultimately led to the demonstration of clinical benefit in SMA patients of all types of disease severities. In addition, we have created an app, a road map that led to the success of Risdiplam and also validated our splicing platform. A key point that I cannot emphasize enough is that we can measure the amount of drug in the blood, and we've shown that this reflects the drug concentration observed in all parts of the brain. That's the beauty of this approach. The 1:1 ratio between blood and brain allowed us to choose a dose that results in the drug exposure that caused a desired change of SMN2 slightly that was efficacious. We'll show you how today we use this and have applied this road map to Huntington's disease. Now let's focus on our Huntington's disease program. Today, we're going to spend time to provide an overview of HD and describe the different modalities that are being considered for potential treatments. We'll also address some of the key questions surrounding HD development that we've been hearing. Analogous to the SMA program, we'll address why measuring HTT lowering in the blood is the appropriate biomarker and why this is a more appropriate marker than measuring HTT levels in the CSF. Lastly, We will review and highlight the evidence that supports the benefits of lowering mutant HTT and the evidence that supports the tolerability of lowering wild-type HTT. I'll start with a high-level overview of Huntington's disease. Huntington's disease, or HD, is a progressive debilitating neurogenerative disorder that is caused by a genetic defect in the HTT gene that results in CAG trinucleotide expansion. HD has a broad impact on the person's functional abilities and usually results in a movement disorder and cognitive loss. HD has an estimated global prevalence of around 135,000 patients. I find this disease particularly insidious as onset commonly occur in a person's late 30s or early 40s, which is often considered the prime of their life as you're just getting your feet under yourself with a career, marriage or children. HD takes its time, but you know it's the inevitable consequences. There are unfortunately no disease-modifying therapy to treat the underlying cause of the disease. HD is caused by a mutation of at least 1 copy of the gene for huntingtin protein or HTT. As you can see in the figure on the left, the CAG repeat expansion, shown in yellow and the HTT gene reduces an mRNA encoding the mutant HTT protein, which forms aggregates within the cell. On the right, We see that the number of repeats of CAG is closely correlated with disease with full penetrant and individuals with greater than 40 repeats. This means that with the genetic screening and individual, we know their likelihood of developing their disease far before onset of symptoms. HD is a whole brain disease. On the right-hand side of the picture of the cartoon showing the key regions of the effective rate, including the cortex and the striatum. The aggregate fits that form for me from mutant HTT in the brain becomes toxic cause and cell death. The brain MRI images on the left clearly demonstrate that in an HD patient, there is extensive neuronal law in virtually all parts of their brain when compared to a healthy brain. HTT is a large, primarily intracellular protein. Wild-type HTT is essential in embryonic development but does not seem to be required post development. As shown in the right figure, HTT is shown to be ubiquitously expressed throughout the body and that the mutant HTT may also affect other systems like cardiac function and digestion as an example. With that, I'll now hand it over to Anu, who is our Executive Director of Biology and has championed and been the leader of our HD preclinical program since it began almost a decade ago. We're excited to share results that support exactly why we believe PTC518 is such a promising therapeutic for HD patients. Anu?

Thank you, Stu. I'm very excited to be able to share with you the story of the identification and development of PTC518. However, before I dive into the story, I want to address some of the key questions surrounding the development of a disease-modifying huntingtin-lowering therapeutics. First, let me start with how the huntingtin-lowering modalities differ from each other. As Stu discussed, HD is a whole brain disease and the route pathogenic cause of HD as the huntingtin or HTT gene. The huntingtin protein with the expanded polyglutamine stretch formed mutant huntingtin aggregate, and this aggregation is a major cause of the clinical symptoms associated with HD. Therefore, reducing this disease burden by targeting huntingtin expression and making less huntingtin is the basis of many disease-modifying approaches. So how do these different huntingtin-lowering modalities compare? As Stu discussed, we have developed oral small molecules that provide multiple advantages in ease of use, equal and brought distribution across the brain, lowering in the periphery, type durability and reversibility. As shown here on the right, the brain is a highly vascularized tissue, which enables broad tissue distribution to all key affected regions of the brain. We are confident in our small molecule huntingtin-lowering approach based on the ability of PTC518 to distribute not only in the whole brain but also in HD-affected peripheral tissue. Let's compare this to an antisense oligonucleotide, or an ASO, which is delivered intrathecally. This approach provides little reduction in the deep region of the brain, the striatum, as compared to the cortex. Lowering huntingtin in the striatum is critical for a therapeutic, as loss of striatum, medium spine neuron is an important hallmark of degeneration in HD. In the case of a gene therapy lowering approach, a vector is delivered directly into the striatum. While this increases huntingtin-lowering in the striatum, there is less lowering in the cortex and other distill areas of the brain. Also note that a gene therapy approach is not reversible and can't be redosed. A key objective in any drug development program is to establish a strong PK/PD relationship. This enables you to accurately predict the dose that will lead to the desired exposure, which will, in turn, translate to targeted huntingtin-lowering in the affected tissue. Here, we link the huntingtin-lowering potency in cells to systemic exposure of the molecule and target engagement our huntingtin-lowering in animals. Using this model, we can then accurately predict for any given compound what dose and exposure will be required to lower huntingtin in animal tissues as well as humans. The next key question involves the correlation between blood and CNS exposure. To understand that, let's turn to some background on the brain vasculature and CNS drug discovery. Proof of target engagement is a key step in CNS drug discovery, and drugs targeting the CNS need to penetrate into the brain through blood circulation. As I mentioned earlier, the brain is composed of an extensive vascular network. This high vascularization in the brain ensures that all the cells have their own blood supply. Therefore, a small molecule can gain a direct entry into these target cells through the blood supply and target all affected sections uniformly. If a small molecule has a strong correlation between systemic and CNS exposure, its target engagement, or PD effect, in the blood and brain should be comparable. Therefore, evaluating huntingtin-lowering activity of PTC518 in blood will serve as a surrogate for huntingtin lowering in the brain. Now let me spend some time on why we think measuring mutant huntingtin levels in the CSF may not be an appropriate target for brain huntingtin-lowering. To dig into this a little deeper, let's review the functions of the CSF. The cerebrospinal fluid, or CSF, is an acellular membrane-bound compartment that surrounds and cushions the brain and removes metabolic waste and cellular debris. However, CSF does not contain a significant amount of protein. We also know that mutant huntingtin levels in the brain are significantly higher than that in the CSF. Because mutant huntingtin levels are so low in the CSF, one would need an ultrasensitive assay to measure it. This leads to assay inconsistency and variability between different batches and assay run. Also studies with multiple huntingtin-lowering modalities have shown that there is a weak correlation between huntingtin-lowering in different brain regions and that in the CSF. In addition, the source of huntingtin in the CSF and contributions of specific regions in the brain to CSF mutant huntingtin levels are 2 big unknowns. All these technical and biological challenges when taken together question the interpretation of meaningful treatment-related changes in mutant huntingtin levels in the CSF. As I mentioned earlier, decrease in mutant huntingtin expression is expected to mitigate HD pathology and improve symptoms. So next, I will talk about the effects of huntingtin-lowering, both in the context of wild-type and mutant huntingtin. First, let me start with the effects of lowering wild-type huntingtin. This question has been addressed by reviewing human as well as animal data. One example was shown in humans. There are published case studies of people with rare genetic variation in huntingtin allele, In these subjects, the loss of 1 wild-type allele was not associated with HD. This suggests that a heterozygous reduction in huntingtin levels is not associated with the onset of HD, and that 50% of the normal levels of huntingtin are compatible with normal development and brain function. In addition, several studies in adult rodents and nonhuman primates indicate that 50% lowering of wild-type huntingtin could be well tolerated. Now let me turn to the effects of lowering mutant huntingtin. Since the mapping of the huntingtin gene and generation of different HD animal models, several reports have been published that support benefits the partial lowering of mutant huntingtin levels to treat HD. Here, we're going to highlight 2 such reports. The first one is a human case study. The study revealed that a transcription lowering variant located on the mutant huntingtin allele was associated with the delay in the age of onset by almost 10 years. This variant, which is located in the Huntington promoter is associated with reducing HTT levels by approximately 50%. Multiple animal studies have evaluated the effect of mutant huntingtin-lowering on phenotypic benefits. The study shown on this slide used the aggressive HD fragment model, R62. As you see here on the left, there was a 30% to 40% reduction in stratium Huntington. These R62 mice were used to study HD phenotypic deficit. As shown here on the right, the phenotype worsens over time in the untreated groups. Whereas in the treated mice with 30% to 40% reduction in Huntington expression, as shown here in black, there was a significant reduction of HD phenotype. These data indicate that 30% to 50% mutant huntingtin-lowering could be beneficial in HD patients. In the next section, we will discuss the preclinical development of PTC518, and we will emphasize why we're confident in the promise of a small molecule splicing approach to treat this devastating disease. To introduce our preclinical program, let's first lay out some key characteristics off a promising oral huntingtin-lowering therapeutic. So let's start with how PTC518 lowers HTT levels. The way PTC518 reduces huntingtin levels is quite remarkable. As you see here, this is the huntingtin pre-mRNA, or at least a part of it with 2 exons fighting on intron. In the absence of the molecule, the huntingtin pre-mRNA with all its exons and introns goes through its usual slicing, generating a full-length huntingtin mRNA. In the presence of PTC518, a pseudoexon is activated within this intron. This pseudoexon contains a premature stop codon. As a result, a stop codon pseudoexon containing huntingtin mRNA is made, which is subject to degradation. Thus, huntingtin mRNA levels are reduced but prevents the production of the huntingtin protein. One important stage in the drug discovery process is the lead optimization stage. It's an iterative process where we refine the chemical structures of the lead molecules in an attempt to improve their drug-like property. The goal here is to find a drug candidate that could be advanced to development. This multi-component process includes various key optimization parameters that are laid out on this slide. These were all implemented to identify PTC518. So we will take a few moments now to highlight our DMPK huntingtin-lowering activity in cells as well as pharmacology optimization processes. Before we dive into the data, first, let's look at the most appropriate animal models used to establish a PK/PD relationship. One of the challenges in HD research is the lack of robust full length animal models of the disease. The pseudoexon activated by PTC518 is not conserved in mice, therefore, we needed to evaluate the effect of PTC518 in a mouse model that has the full-length human huntingtin gene with all its exons and introns, like the [indiscernible]. For PK, we use wild-type mice. They give us a pretty good read on systemic drug exposure, our concentrations of drug in the plasma upon oral dosing. Although they lack the human huntingtin gene, we have used nonhuman primates to demonstrate drug distribution, especially in the brain as they have a similar anatomical structure to the human. In summary, we utilized BACHD for PD, and wild-type mice and nonhuman primate for PK. The first important characteristic of PTC518 is its oral bioavailability, which has biological advantages while also reducing patient burden. A tablet, when taken orally, breaks down in the stomach, small intestine and liver circulated in the bloodstream to be distributed throughout the body, including the brain. However, the brain has an extra layer of protection known as the blood-brain barrier. This is a highly selective barrier between the blood, supply and the brain. The blood-brain barrier regulates the passage of selective molecules necessary for brain function while keeping out potential toxins. PTC518 was designed to penetrate the blood-brain barrier so that it can be evenly distributed across the whole brain. We showed that PTC518 is only bioavailable and crossed the blood bin barrier because we were able to demonstrate brain huntingtin-lowering and that too in a dose-dependent manner. This brings us to titratability, which is shown on the next slide. In this study, BACHD mice are orally dosed for 21 days. After oral dosing, brain tissue was extracted, and huntingtin protein lowering was measured using an electrochemiluminescence, or ECL-based sandwich immunoassay on the MSP platform. The Y-axis is the HTT protein level and the x-axis is increasing dose level. Here, 100 person means no change in HTT protein levels. The 2 dashed lines here show the range between 30% to 50% huntingtin-lowering. As discussed earlier, human and animal data indicate that 30% to 50% lowering of mutant huntingtin could be beneficial in HD patients. As you see here, compared to the vehicle group, which is shown here in gray, we observed a dose-dependent decrease in huntingtin protein levels in BACHD brain after oral dosing. To reiterate, upon oral dosing, PTC518 can lower huntingtin protein levels in BACHD brain in a dose-dependent manner. We evaluated the effect of huntingtin mRNA and protein lowering in human cells expressing endogenous huntingtin. The graph on the left shows mRNA levels, and the one on the right shows protein levels. The Y-axis shows HTT mRNA or protein levels, and the X-axis is drug concentration. As you see here, we see a dose-dependent decrease in huntingtin mRNA levels on the left, and the concomitant decrease in huntingtin protein levels on the right. Therefore, PTC518 lowers huntingtin mRNA and protein levels at low nanomolar concentrations in these cells. Following demonstration of activity in human cells, the next step was to show proof of splicing and protein lowering invivo using BACHD mice. We dosed PTC518 in BACHD and tracked huntingtin splicing and drug concentration over a period of 24 hours. The plot on the left monitors splicing in whole blood after a single dose of PTC518. The plot on the right monitors systemic drug concentration over the course of 24 hours at the same dose. As you see here, there is a time-dependent splicing effect. We see a significant drop down to 30% between 2 to 4 hours post dosing. This slightly affect returns to pre-dose levels by 16 hours as the drug concentration dropped below the in vitro IC50 levels, which is the concentration at which we see 50% huntingtin-lowering in cells. So we demonstrated 2 important things here: we showed that PTC518 is a reversible splicing modifier and we were able to achieve a strong PK/PD relationship with PTC518. What I just demonstrated was RNA in whole blood after a single dose. Now how does it translate to huntingtin's protein lowering at the same dose but now dosed for multiple days in the same tissue from the same mouse model. Here, we show approximately 50% reduction in huntingtin protein levels in BACHD white blood cells after 21 days of oral dosing with PTC518. Therefore, we were able to show a very strong correlation between splicing and protein lowering in the same target tissue after single as well as multiple days of dosing. So how about the brain at the same dose? As shown in the next slide, here, we compare huntingtin-lowering between the blood and brain in BACHD mice. As you can appreciate, we see very similar levels of huntingtin-lowering between the brain and blood from the same animal. So the takeaway from this slide is that huntingtin-lowering in blood matches that in the brain. Now we can check off reduction of HTT in the CNS as well as in the periphery and turn our discussion to uniform lowering across key areas of the brain. PTC518 demonstrates robust huntingtin reduction in BACHD mouse brain. As we have emphasized, HD is a whole brain disease and PTC518 is equally distributed across the whole brain. This is supported by the data here showing uniform lowering of huntingtin protein levels across all discrete sections of the brain. The striatum cortex and cerebellum are shown here. As a reminder, we also show similar levels of lowering in the periphery. Now adding on to this equal distribution story, we show here that in nonhuman primates, PTC518 can cross the blood-brain barrier. As you can see, the X-axis is pre-plasma drug levels. The unbound drug concentration in systemic circulation determines drug concentrations at the active site and its effect. This unbound or free PTC518 will effectively lower huntingtin in all target tissues, including the brain. The Y-axis shows drug concentrations in the CSF, which is widely used as a surrogate for unbound brain concentration. As you see, there is a strong linear relationship between plasma and CSF levels of PTC518. The data here shows that total drug concentrations in CSF or the unbound concentration in the plasma in humans will be a good estimate of pharmacologically active PTC518 available to lower huntingtin levels in the CNS. As you can see, we have demonstrated uniform lowering in key regions of the brain. Next, optimizing for selectivity was also very important so we could maximize huntingtin-lowering effect and minimize off-target effects. To assess selectivity, we performed transcriptome-wide RNA sequencing in cells expressing endogenous huntingtin. We saw very few gene expression and splicing changes at 10x the concentration at which we see 50% huntingtin-lowering in these cells. Therefore, PTC518 is a highly selective huntingtin-lowering splicing modifier. Just to remind everyone, elsewhere in the body, drugs can diffuse freely between the bloodstream and surrounding tissue, but our brain cells must be protected from harmful agents in the bloodstream. Even if a molecule can passively penetrate the blood-brain barrier, it can be actively pumped back out to the bloodstream. This phenomenon is called efflux. Therefore, it is very important to address efflux in brain targeting therapeutics so that they are not actively pumped out of the brain and achieved improved efficacy in the CNS. Now let's compare PTC518 to another splicing modifier, but one that is effluxed. Here, you see a clear difference in protein lowering between the brain and the periphery. With PTC518 at a given dose, which is shown here on the left, we'll see similar reductions in HTT levels between central and peripheral compartment. However, with an efflux molecule, which is shown here in the middle, we see that 50% reduction in huntingtin protein levels in the brain resulted in greater than 90% reduction in the periphery. Now if this effluxed molecule were to be dosed such that we wanted to achieve 30% to 50% reduction in the blood, it would amount to less than 15% huntingtin reduction in the brain. In summary, the totality of the preclinical data support that PTC518 is a promising potential HD therapeutic. As you can see, we have demonstrated that PTC518 fulfills the key characteristics we outlined for a successful small molecule. PTC518 is orally bioavailable, blood-brain barrier, penetrable, selective, titratable and not efluxxed. It's not only uniformly lowers huntingtin across all key sections of the brain but also reduces mRNA and protein levels uniformly in the periphery. I will now turn it over to Matthew Klein, our Chief Development Officer, who will discuss our preliminary data for Phase I healthy volunteer clinical trial. Matt?

Thanks, Anu. We are excited to be able to share initial data from our PTC518 Phase I healthy volunteer study. While the study is still ongoing, we wanted to take this opportunity to share some key and exciting data that we have already collected from the SAD treatment cohorts as well as from the first 2 MAD cohorts. I would like to begin with a review of the Phase I study design. In addition to capturing critical safety and pharmacology data, the Phase I study was designed to demonstrate proof of spiking methods, just as we did in the Phase I Evrysdi program, and to inform dose levels for future efficacy studies. There are 4 parts to a Phase I study: single ascending dose cohorts; multiple ascending dose cohorts; a food effect cohort; and a CSF sampling analysis cohort. The SAD portion of the study included 5 cohorts, each with 8 subjects, 6 of whom received PTC518 and 2 of whom received placebo. The MAD portion of the study will include up to 5 cohorts, each also with each subjects, 6 receiving PTC518 and 2 receiving placebo. All MAD subjects will be dosed for 14 days and then followed for an additional 14 days. In addition, we will include 1 food effect cohort and 1 cohort that will include both peripheral and CSF pharmacology analyses to confirm PTC518 blood-brain barrier penetration and lack of eflux, the critical PTC518 design components Anu described. At this time, we have completed dosing in all 5 SAD cohorts and 2 MAD cohorts, and these are the data we will be sharing today. Before sharing the Basil data, I wanted to again show one of the key pieces of preclinical data we sought to replicate in the Phase I study: dose-dependent lowering of HTT mRNA as observed in the BACHD mouse. As we have said many times, Phase I success would be defined as confirming safety, predictable pharmacology and dose-dependent evidence of HTT lowering in peripheral blood cells as observed in the mouse. So with that as our clearly defined target, let me share data from the SAD cohorts. As shown on Slide 58, there was a dose-dependent lowering of HTT mRNA after a single dose of PTC518. On the Y-axis of this graph is present baseline HTT mRNA level, measured following PTC518 administration. As you can see, following a single dose of PTC518, there is a dose-dependent reduction of HTT mRNA across the 45-milligram, 90-milligram and 135-milligram doses, providing strong evidence of proof of splicing mechanism. Furthermore, we were able to achieve the desired target lowering of 30% to 50% of HTT mRNA with just a single dose of PTC518. In addition to the PTC518 pharmacodynamic splicing effects, we also observed dose-dependent predictable pharmacology. This graph on Slide 59 demonstrates the linear relationship between PTC518 dose level and plasma concentration. These levels of plasma exposure were consistent with those predicted by our preclinical pharmacology studies. In addition, the half-life at each dose level appears to be over 70 hours, which is a bit longer than observed in the animal models. Finally, and importantly, PTC518 was found to be well tolerated in all SAD cohorts with no safety findings. In summary, we successfully dosed 5 cohorts in the SAD study and achieved all study objectives. PTC518 was found to be well tolerated at all dose levels with no observed safety findings. There was predictable dose-dependent pharmacology and, most significantly, there was dose-dependent splicing of HTT mRNA. And even with a single dose, we were able to achieve the desired 30% to 50% lowering of HTT mRNA levels. Next, I will share results in the first 2 MAD cohorts, in which we dosed subjects with 15-milligrams and 30-milligrams of PTC518 for 14 days. Here are the splicing data from these 2 cohorts. Again, as you can see, we have dose-dependent reduction in HTT mRNA levels, with the 15-milligram daily dose, resulting in a 40% to 50% reduction in HTT mRNA levels and the 30-milligram daily dose resulting in a 50% to 60% reduction in HTT mRNA levels. Another important observation from the first 2 MAD cohorts was the maintenance of splicing activity for up to 72 hours following treatment completion. As I mentioned, MAD study subjects are treated for 14 days with PTC518 and then followed for 2 weeks after dosing completion. Here, we see that while we observed peak reduction in HTT mRNA levels in approximately 6 hours following the final dose. There are still persistent end-market splicing occurring for 24, 48 and even up to 72 hours. We do, of course, see a return to baseline over the subsequent days, highlighting the reversibility of PTC518. This prolonged splicing effect is likely related to the long half-life of PTC518 and suggests that in subsequent MAD dosing cohorts, we consider exploring less frequent dosing regimens. In summary, in the 2 completed MAD cohorts, PTC518 was found to be well tolerated with no observed safety findings, and we achieved the desired dose-dependent lowering of HTT mRNA beyond the targeted 30% to 50% reduction. In addition, we see that splicing activity persists for up to 72 hours following the final dose, which supports exploring alternative dosing frequencies in the subsequent MAD cohorts. We are, of course, very excited about these results and now look forward to completing the remaining parts of the Phase I study, including the additional MAD cohorts, HTT protein analysis, the CSF cohort to confirm the expected CNS high distribution and the [ Flu Detect ] study. In parallel, we are continuing our planning for the next steps in the PTC518 development plan as we look forward to establishing the safety and efficacy of PTC518 in Huntington's disease patients. I wanted to briefly highlight some of our ongoing clinical development planning activities. As we have discussed many times, the mechanism of action of PTC518 makes it an attractive therapy for all HD populations, including manifest, pre-manifest and juvenile onset HD. However, we are, of course, well aware of the challenges in HD drug development. Therefore, our team is in the process of conducting extensive natural history work to identify the optimal development strategy including clinical endpoints that capture meaningful treatment effect, biomarkers that may provide reliable and more rapid confirmation of treatment benefit and, importantly, appropriate study populations in whom we can practically establish evidence of treatment effect in the course of a clinical trial, given what is understood about heterogeneity of disease symptoms and disease progression. We look forward to sharing more details on the development plan as it is finalized. I will now hand the presentation back over to Stu for concluding remarks. Stu?

Thanks, Matt. I just want to make a few concluding points. As you heard from Anu, we've demonstrated that PTC518 fulfills the key characteristics we outlined for a successful small molecule. PTC518 is orally bioavailable, penetrates the blood-brain barrier, selective, titratable and not efluxxed, and not only uniformly lowers HTT across all sections of the brain but also reduces mRNA and protein levels uniformly in the periphery. As Anu noted, a key objective in any drug development program is to establish a strong PK/PD relationship. As demonstrated in the SMA program, this enables us to accurately utilize a dose that will lead to a desired drug exposure, which in turn will translate to targeted HTT lowering in all tissues within the brain. The HTT program has now been able to follow the same path and replicate these results in our preclinical studies as well as in our preliminary results from our Phase I study. Our next step, once the study is completed, will be to go on to demonstrate clinical benefit in the HD patient population. As you've hopefully seen from the presentation today, the splicing platform has proven to be a robust engine to identify therapeutic candidates for the SMN2, HTT as well as the next set of targets we're currently working on. We look forward to continuing to be the leader in the splicing field. I'll now turn the -- I'll now turn it over to the operator for Q&A. Operator?

[Operator Instructions] Please standby while we compile the Q&A roster. Our first question comes from Eric Joseph with JPMorgan.

Thanks for hosting this deep dive session. Very informative. I guess with the Phase I data so far, to what extent have you looked at protein expression -- or changes in protein expression So far? From the mouse model data, it seems like maybe 14 days of treatment might be a little brief to realize or see the one-to-one relationship between mRNA and protein levels. But -- and what you've looked at so far and insofar as protein expression is kind of the key biomarker here. Are you seeing indications that would support a similar dose-responsive relationship? And I guess given the tolerability as you good tolerability, as you continue with the remaining cohorts, do you have any flexibility with the ratio of treatment that you might be able to look at protein expression and optimize dose selection?

Yes. Thanks, Eric. Thanks for that question. So yes, as you said, that we -- in the Phase I trial, we demonstrated that obviously a orally available small molecule that reduced the HTT mRNA in a dose dependent manner. And of course, there was no -- it was well tolerated with no safety concerns. And what we've shown in the past is that we would expect that similar to the messenger RNA that we saw protein lowering, we would see analogous levels of protein lowering. And we've seen that not only in cells, in multiple cells, but also in the BACHD now So we're able to be able to show that in our preclinical data and that we saw really the reduction between RNA protein were the same at the dose that we were giving. In terms of the flexibility, in terms of the dose, and I think that's a good point is that -- and I think this sort of exemplifies the -- what we -- the advantage of being able to have an orally bioavailable molecule that distributes throughout the body and gets to all places within the doses that we can actually measure that level. So we know what dose gives us a particular exposure on that exposure. We know what gives a particular reduction in terms of the HTT mRNA level. So we certainly can of focus in on to get the precise dosing within that as well. So I think you're also right that given that we have a wide therapeutic window where there's an effective dose even at the lowest level that we gave in the MAD with the 15 mg. I think it gives us a lot of opportunity and flexibility to optimize the -- what the dose that we ultimately choose. And so I think your 14 days, as you could see, is getting pretty close to what steady-state. It may take a little bit longer, obviously, but we're getting to a pretty good idea of what -- in terms of the mRNA level, what that would be.

Okay. Got it. And from a PK standpoint, I guess, any additional comments or insight with respect to the metabolism of the 518? You're looking at it, a food effect study, I'm just wondering whether there are any drug-drug interaction considerations with 518, how you're exploring those given the potential food polypharmacy in the Huntington patient population?

Yes. I think -- again, there's -- yes, I think there's -- obviously, we've looked at this in the half life, we said, is on the order of [ 270 hours ] or so. So we have pretty good flexibility on that. And that over time in terms of drug-drug interactions well -- the food effect is we'll be doing just to see if it changes the -- any in terms of the levels that we see within the blood, so that -- that's always a good thing for us to do. In terms of drug-drug interaction, that's something we probably will think about as we move forward longer on, on what we might need to be part of that. And that will obviously, of course, do the formal DDI studies as part of our normal development program.

Our next question comes from Gena Wang with Barclays.

A few, first, maybe the question regarding the nonhuman primates data on Slide 49. Can you hear me okay?

Yes. Now I can.

Okay. Sorry. So I have a few questions. The first one is regarding the nonhuman primates data on Slide 49. Just want to make sure, because the CSF unit is different from the plasma unit, I just want to make sure those are the same unit? We do use that nanogram per gram. Is that equivalent to similar to millimeter? And also another question is here, we only see the plasma in a CSF correlation. Do you have data on different parts of the brain, and particular different components of the [indiscernible], how the drug distribution look like? This is the first question. And I have -- maybe I will take my other questions after that.

Sure. In terms of the nonhuman primates that we shared on the details, we showed that there was a linear relationship between the levels in blood and the levels of the CSF. So that's true. And that the nonhuman primate studies, they were in terminal. So we didn't do brain tissue analysis that we're not carrying out on these. But I want to just also remind you that we did do this in mice and did show that it got to all tissues within the brain, and we showed that obviously in our presentation here. But I think also the nice thing in terms of the nonhuman primate study, assuming that we saw the 1:1 levels that we saw elsewhere, also gives you a pretty good indication that the drug isn't being efluxxed as well because we would have anticipated that there would be lower levels in the CSF. Therefore, I think we're pretty confident based on that, that there's going to be distribution across the whole brain, right? It's a only by ratable small molecule. And the fact that you see through the CSF, you'll be able to see it through that. So that's in terms of that. And that's also -- I should point out that -- and we did a lot of work on this to ensure these compounds didn't get efluxxed, and so we showed that also in cell-based models and where you monitor eflux as well.

Thank you, Stu. But I do think the nonhuman primates is a larger animal, and that's a very, very important, especially for the deeper brain -- of the brain of mouse. I think there is a criticism data from mouse the CNS system, we cannot translate just far away from human because of the size, particularly because of size. I think it will be important for investors to understand the brain distribution in nonhuman primateses.

My next question is regarding the Slide 43. So you did show mRNA knockdown versus -- I'm sorry, not the lowering, versus approaching...

Yes. Can you keep closer to the phone? It's hard to -- you're muffled.

Can you hear me okay now?

Yes.

Okay. So my next question is regarding the RNA, the Slide 43, RNA and the protein correlation. Here, this is a cell line, and I think the Roche actually did a study in nonhuman primates in both cortex in -- for the RNA and the protein correlation. And it is not actually 3:1:1, so that translates back to, I think, the first -- Eric's question, regarding the protein level, I think it is important to show protein level in human regarding the actual final product percentage of protein knockdown may not be 1:1. Just wanted to hear your thoughts regarding mRNA knockdown correlation between and versus protein.

Sure. So let's -- I think there's a couple of things. I think we're mixing a little bit here apples and oranges in terms of -- because we got 2 different types of molecules. We have in 1 case, we have a small molecule that is -- that actually [indiscernible] passes the blood brain-barrier goes into -- and is able to go into all tissue types, right? That's the major advantage of being able to be able to measure -- use a small molecule versus other molecules that are either put in intrathecally. And so the nice thing about a small molecule is that it's brought to the very tissues for the blood, every tissue -- every cell and tissues needs blood, and as a consequence of that, you're able to get small molecules very much distributed very well. So -- and that's not the case with drugs that are intrathecally given to the deep recesses of the brain. You can have very different differences. So we would expect -- so I don't think one -- first of all, point number 1 is I would not actually compare one to the other as a consequence of that, because they're very different ways of distribution, very different ways, and we're just being put in. So -- and that's a relation. So we would expect, in our view that we'll be able to get every cells of similar levels of mRNA and protein level lowering. And that's in a subset of preclinical data was able to show us. And so I think that -- in that we saw a pretty good -- a very good correlation between RNA protein within that. And that's also because we know that the molecule gets throughout the cell, and that's why you could see a very good distribution of it. I think that -- and I think in that sense, looking at the CSF and in comparing it in animals, we have a 1:1 correlation between blood and the CSF. That should give you a pretty good feel as well in that. So it's very different in that. And that has a lot to do with very different differential bio distribution of an anti [ sense ] like a nucleotide versus a small molecule, which could be much easier by -- distribute it.

Okay. My last question is regarding the knockdown mRNA level, just wondering if you can share in terms of a total mRNA versus a mutant mRNA, what is the difference you see regarding the knockdown level?

Right. So in the case that we're doing is we're measuring total knockdown of both wild type and mutant when we do these measurements. So we're showing an overall level so that they're the same.

Okay. Will you be planning to show mutant knockdown level?

Go ahead, Matt.

I was going to say, this is in the healthy volunteers, right? So they only have wild type. As they move the program forward, obviously, the next step will be to move into HD patients where we'll have the ability then to demonstrate a differential knockdown. But we expect it to be 1:1 given the non-real selectivity of the approach.

Okay. But this is also the splicing form, the one you are checking in this splicing form, the modified version?

Yes. So maybe just to remind you that we said that the pseudoexon that gets integrated into the RNA that causes the rapid degradation is in both mutant wild-type form. So we would anticipate that the levels of reduction are the same, and that's what we see. We're looking for loss of the full length on messenger RNA .

Our next question comes from Alethia Young with Cantor Fitzgerald.

Congrats on the early progress here. A lot of questions out of the gate, so let me try to get just 2 quick ones. And I guess just in thinking about dose, I know there's a trade-off between doing too much on the wild type, but I mean do you feel like you should push the dose a little bit more to try to get in the mutant lowering beyond? It seems like you have a very clean, safe profile. So it seems like you've potentially pushed there. And then I guess, just to ask like the elephant in the room question, specifically, when you think about -- you referred to this a little bit with the antisense and gene therapy, but do you think it's really the 1:1 ratio between the brain and the blood and then obviously some activity to clean up the periphery as well. Is that what gives you confidence that you'll have a greater translatability than maybe the antisense gene therapy?

Yes. That's 2 good questions. I think in terms of dosing, I mean, so we -- I think we're getting up to near 70% reduction at doses. So we can't -- I think the nice aspect of the drug is it's shown a very nice titratable, able to do this. So we have the luxury of being able to really look at different levels within the loss of RNA in the healthy volunteer study. So we're -- obviously, we know it's titratable and reversible. And those are again 2 nice aspects of that. And in terms of the -- like what you said in terms of the 1:1 ratio, yes, that's -- I think that's a really excellent point. And that didn't come by just happenstance. We went through chemical optimization, and the chemical optimization really worked on, not only passing the blood-brain barrier to get in, but also, okay -- and also that has been passed the blood-brain barrier but also doesn't get kicked out. So these are really good questions in terms of -- to really differentiate our molecule through versus other molecules, because it's in the -- the brain has, as -- and you talked about of highly protective with eflux with pumps that move kick things out, and therefore, you can see in the blood levels in the blood that are much lower than what we see -- than in the blood, but they're lower levels seen in the brain, and that's because they're being fluxxed out. We work to make sure that they didn't get efluxxed so that there's a very good ratio so that what we see in the brain reflects what we see -- what we see in the blood reflects what's seen in the brain, and we found that to be the case. And so that's exactly what we did initially with getting to Evrysdi. So -- and that we also showed that it reaches all parts of the brain so it's not efluxxed out. So we feel pretty confident. We showed that at Evrysdi. We showed that for PTC518. So the degree of HTT lowered it really is nicely controlled there. So I think your point is quite right. And since it's a whole brain disease, this was, to us, a critical point in particular because cells, the brain cells die in the case of Huntington. So being able to get to all cells we thought was very important.

Our next question comes from Robyn Karnauskas with Truist.

Really good presentation today, very helpful. I know you're doing some work on what populations you should go after, but maybe you could opine a little bit on what types of patients might have a phenotypic benefit from lowering? And specifically, if you think about some patients who've had longer course of disease or they have longer repeats, do you have any understanding of, even in mouse models, what have happened to this aggregates after you're lowering mutant protein and how -- and what phenomic differences you would see and how long it might take to see it? That's my first question. Then I have a follow-up.

Yes. Again, another really good question in terms of -- I'll set it up, and then I'll pass it to Matt. I mean I think you're -- so we know, right, it's a monogenetic disease as a consequence of the mutant. We approach -- we know and as a consequence of the toxic mutant huntingtin protein and, therefore, lowering it should give benefit and then the benefit, the question then comes how do you show benefit. And so we've been thinking, obviously, that's the key question. So we believe now we have a drug that's had that's efficacious, what's the next step? And so let me pass that to Matt, who's been thinking a lot about it from a clinical development perspective as well. Matt?

Yes. Thanks, and Robyn, thanks for the question. Obviously, the clinical trial strategy is a key component in any neurodegenerative disease program, and then obviously clearly the case in Huntington's disease. As Stu mentioned, our approach of targeting mutant HTT toxic protein, we believe would have benefited all HD populations and in patients in all stages of disease. The question of where do you go first to show the benefit really becomes one of understanding, rate of change in different populations, what's the right symptomology that you can capture to really show that effect, right? And this is really the challenge to say within the time course of a clinical trial, what's that sweet spot in terms of a population that's progressing at a fast enough rate that you can capture a meaningful clinical benefit? And so a lot of the work we're doing now is looking not only at the scales, which obviously have been used before and had all the pluses and minuses of composite disease scales, but really trying to hone in on specific symptoms that are the drivers of the changes in these scales and symptoms that are easy to measure and I think gives us a clear indication of treatment effect in a relatively shorter period of time. And so we have a few clues on that, and our teams are working really hard on that with external experts, a lot of the national district databases and really thinking about this in the context of what's been done and what we can learn from what's been done so that we can ensure we have the greatest probability of success in showing this molecule's effect. And then from there, of course, maybe reach out to the other population. So it starts with what's that sweet spot, hit that, and then extends to the broader aging population.

Got it. So then maybe just to set us up into the next data release and update from your program, specifically in the patients who have Huntington's disease, can you just help us understand give us some expectations so our expectations are not too high and what exactly you're looking for. And I know you're collecting the -- you made a case for why CSF isn't going to be as reliable, but do you expect to see a correlation in these patients with Huntington's, with the CSF and the blood? Just some -- just help us get good setup in the data.

Sure. Maybe -- so I'll start in terms of the CSF. The CSF in our view, and actually, I think there was a recent paper that was put out is the CSF, we already know it in terms of when people were looking at the striatum and cortex in animal models and then looking within the CSF. So you know there's a disproportionate change within the regions within the brain and yet you get a level of within the CSF. So we don't think it's really a very good or predictive model. And I think it's -- I know people have focused on this. But I think we tried today to explain very much why there are some real issues with using this, not only does it really talk about the proportionality of what's going on within the brain and, therefore, it's not really telling you what's occurring, there's also some real issues. Like we don't even know where the where it's coming from in terms of measuring that. And second of all, the assays because they're so low, they're at temp to lower levels that you see, you have the limit of quantitation with that, which gives high variability. So even the measurements are probably relatively questionable. That's why, from our point of view, we use, at the end of the day, levels of blood, because we think that's a real -- as we've shown in animal models, we've shown that -- we used the CF to measure our drug levels so that we know it passes the blood-brain barrier and then below it gets there. So I think the blood in the CSF is important for that. So then for next step, Matt, do you want to go through your thinking on that?

Yes. So obviously, the immediate next steps, Robyn, are to get some of the final pieces -- important pieces of data from this Phase I study, including the CSF analysis to confirm the pharmacology that Stu discussed and showing that important expected similar ratio between plasma exposure and CSF exposure. Then obviously, the next major important milestone in the program is being able to recapitulate these findings in patients with Huntington's disease, being able to show that we are able to decrease the levels of mutant huntingtin, mRNA and protein. And that's really for us then the key next step. Obviously, at that point, we'll also start wanting to look at other potential signals of clinical benefit in the shorter term, some biomarkers that we're looking at to do that. And then, of course, after that, will be the final step, which is establishing clinical efficacy. In terms of setting expectations, we're not going to tell you that we have a way to establish clinical efficacy in a 1-month trial. No one would believe that. It's still -- we still know [indiscernible] generates is going to be a longer period of time in a longer trial to be able to show that definitive evidence of clinical benefit that's going to get you -- get us our approval for PTC518. But what we realize along the way is that we need to show and will next show the effect of PTC518 on a mutant, huntingtin-lowering protein levels and explore some other signals that can give us confirmation that we're moving things relevant to the disease.

Our next question comes from Brian Abrahams with RBC Capital Markets.

Congrats on the early data, and thanks for hosting this. Two questions from me. I guess, first off, maybe just a clarification. Can you just clarify, have you looked at the protein knockdown data yet at this point? Is that something you're going to be looking at? And if so, when should we expect that? And then secondarily, maybe on the PK front, how does the steady-state PK with the multiple doses at 15 to 30 map to the single doses that you looked at? I guess I noticed that steady state, the knockdown is relatively modest for right after dosing for -- I guess, at the steady state for the 15 milligrams and much more robust for 30 milligrams. Just sort of wondering does 15 sort of map to your 5 milligram single dose? And just, I guess, trying to understand how the exposure levels and splicing activity will correlate.

Yes. Thanks for that. Yes, let me clarify. Yes, we will be planning to look at protein levels. And we know we haven't done that yet, but we're going to be doing that. And then at an appropriate meeting, we'll be presenting that data as well. On your PK front, I think, actually that's a good question. That's a subtle question. It's -- in a sense it's predictive in terms of the half-life But what you're pointing out is interesting in terms of looking at the RNA level after day 1 versus day 14. So maybe this will help you. If you think about a day 1 pretreatment, there's x amount of Messenger huntingtin already there, right? So it's already spliced and is at that level. So that RNA is not already going to be affected as a consequence of the drug, because you've already made that level. That steady-state Messenger RNA is already there. So it's not surprising that when you look only at a day that you got -- that RNA is going to be there for that measurement that we're doing. And so -- so it looks different as a consequence. However, by day 14, the -- you're treating and you're reducing the level of an RNA that's being made, and you're also degrading the RNA that was there on day 1 so that you get to a steady state -- a new steady-state level of RNA that's present, that's more a consequence of the titration of what the drug is doing. And so that's what -- I think that disconnect that you're seeing is not really a consequence of the effect of a change in the drug. It's a consequence of getting to the steady-state level of the RNA. And day 14 lets you get that versus day 1, and that's what you see the difference of. Does that make sense to you?

Yes. That's really helpful.

Our next question comes from Raju Prasad with William Blair.

This is Sami on for Raj. I was curious what other splice types you're monitoring and maybe how the prolonged half-life this compound is affecting that? And also, are you monitoring any other additional biomarkers, in particular, the neurofilament light chain protein, either in preclinical studies or in the Phase I or Phase II studies?

Sure. So yes, so in terms of the preclinical markers, I think obviously, you've got to remember that it's in healthy volunteers so that there's no anticipation yet in Phase I. So we're planning that to be a phase. And we'll have more of a chance to look at that, of course, within in Phase II. We can, of course, look, but these are HD patients, and the value of this, I'm not sure of. Does that answer your question?

Yes. I guess would you be looking at those in both CSF and blood or just in blood?

Well, on -- you know, in preclinical models like the BACHD, I don't think you see any changes. Yes, but I think we would probably be looking. In looking at it, I think we would. Matt, do you want to comment on that?

Yes. Sami, as we had said, I think what we -- in Phase I, obviously, there really wasn't any benefit to looking at preliminary change early markers of HD because these are healthy volunteers. But certainly, as we move into Phase II and exploring the -- as we talked about the effects of 518 on the HD mRNA in protein levels, we'll start exploring biomarkers like NSL and look in the compartments that has been shown to be relative, which is the blood and the CSF. So that would be our thoughts in terms of doing that moving forward.

Got you. And if you could just comment quickly on any kind of alternative splicing activity you're monitoring?

Yes. So that's a good point. I think when you might -- you might have noticed within the -- or we talked about there was a -- when we looked at the overall effects when we -- in terms of 518, it was an incredibly small number of factors that we saw that ultimately changed. And so there was -- so -- and I think we talked a little bit this during that -- during the call. So there's not a lot of changes that occur as a consequence of this. And we pretty much made it pretty selective as we made it. So there were probably in the small range of around 20 or so that changed, I believe, is my recollection. And so -- but when we look at SMA or other things, we see a very good selectivity as a consequence at the treatment level that we're giving. So we feel pretty confident at the end of the day that we -- it's a pretty molecule. There's very few changes that we saw even at concentrations of 10x the amount of the doses that we're giving.

Our next question comes from Danielle Brill with Raymond James.

I have a couple clarifying questions. For Stu, exactly how similar is 518's PK/PD profile to Evrysdi? Just trying to understand how appropriate it is to use those data for benchmarking. And then can you provide a bit more color on how you're assessing safety in the Phase I trial? Are you measuring off-target gene engagement?

Sure. So in terms of the PK/PD, I think it's interesting in the case, it's a little bit of -- and it's a good question in a sense of -- in one case, in the case of SMA, you're looking at there were 2 different genes with 2 different transcripts. In 1 case, it included all the exons and the other one excluded exon 7 on the whole, which led to right -- as a consequence of a leaky U1 exon that we -- in the case of Evrysdi, it stabilized that interaction so that even on SMN2 transcript, you would get the RNA that would include exon 7. So that's how that works quite nicely. In the case of the Huntington's disease, it's a pseudoexon that's normally not included. So -- and again, it's due to -- in this case -- and in both cases, a non-canonically use site that -- U1 site that doesn't work in this case. And so the molecule is able to now make this much more effective in being able to move it in. And you're seeing actually very good changes. Probably the PK/PD, from that point of view, is pretty similar from the point of view is to know they're both highly potent efficacious molecules that cause, in one case, the switching to an SMN1-like transcript. And in this case, the inclusion of the pseudoexon that causes reduction of the RNA. But in terms of PK/PD, it's highly -- it's -- they both were shown to be highly titratable, effective and produce lowering. Again, with Evrysdi, we saw it across the brain as well as within the CNF and other -- in the periphery and other tissues. And similarly, with 518, we see, in a sense, really an analogous change in terms of reduction of the RNA throughout tissues in the periphery as well as, and this is really critical, in all parts of the brain. And that's really -- again, I can't emphasize enough the fact that it gets delivered to the brain and therefore, even in deep recesses, tissues of the brain, we're able to get 518 to it to lower the HTT there. So that's, I think, really the exciting aspect of this is that it's highly effective, highly selective and get to the -- and can get to tissues even deep within the brain. And I think that ultimately makes it potentially one of the best-in-class as we continue to move forward.

Okay. And then can you provide some more color on how you're looking at safety in the Phase I study?

Matt, do you want to talk a little bit about this?

Yes. Sure, Danielle. So in the Phase I study, is very standard approach in terms of close safety monitoring. So obviously, we're looking at a whole comprehensive battery of laboratory values, physical examination, screening for [ AEs ], screening for SAEs. And as we mentioned that thus far, we've had no safety issues reported, and 5 appears to be well tolerated. We're not doing any gene screening, alternative slicing screening in the Phase I study. But as Stu mentioned previously, this is something we did look at in the preclinical program in terms of demonstrating and verifying that we had the desire selectivity and specificity that were key components of the 518 development program. And also, as Stu mentioned, we saw very few changes at even 10x the concentration of that we're going to be using in terms of in the clinic. So basically, at multiples of the effective dose level, we saw minimal changes, so we'd expect even less in the clinic. And I also want to emphasize the other important point that we mentioned earlier, we're -- obviously, when we went into Phase I, we were guided on a dose escalation strategy based on our therapeutic window, and we were really pleased to see the targeted 40% to 50% splicing reduction we wanted to achieve at the lowest dose range, at the lowest dose, 11. So that gives us a lot of comfort, and we have a healthy window as we move forward.

Our next question comes from Colin Bristow with UBS.

This is [ Ping ] on for Colin. So we have question on the chart of using the co-blood HTT over the CSF HTT assay indication of the brain HTT levels, since there -- we know there is a linear correlation between this 2. You do mention setup is unreliable, and one rationale that brain HTT level is like larger than 100-fold of the CSF HTT level. So what's the difference of HTT mRNA protein levels in the white blood cells versus in neurons? Another question is you also mentioned 518 distributes equally well in striatum and in the cortex. So is this purely based on how we think of the blood vessels that are distributed in the brain? Or do you have any microscopic or imaging evidence suggesting this? So our final question is what -- would you be able to show any biodistribution of 518 in different sections of the brain from this Phase I trial?

Yes. Okay. So -- okay, yes, thanks for the question. So I think what you're asking is when we looked at the CSF, we're doing it simply to look when we do this, and we'll be -- so just as we've demonstrated and others that it gets to the brain, passes the blood-brain barrier, and we'll get into the CSF. And we showed in the nonhuman primates that the level of free drug in the CSF equals the level of that drug in the blood. So we were actually able to demonstrate that it's -- and it gets into the CSF in that level, and therefore, we would anticipate it to get into all aspects of the brain as a consequence of that. And we've shown that within the BACHD mouse model as well. So I want to make sure that we separate the notion of doing PK and looking at that versus the level of HTT RNA protein in the CSF, which is different. And because that's really sort of a consequence of the HTT levels, right, in terms of cells breaking, and you don't really know where it comes from. The levels -- the RNA levels in the blood and brain are pretty similar, and you're looking -- where is HTT RNA and protein. The vast predominant amount of that of RNA and protein are within cells, and the level of -- within the blood and brain cell, it looks like they're pretty similar as well. So It makes sense to us that, that is a much better marker than looking into the CSF where you -- where we already know there's not proportionality to what you see in striatum and the cortex, the changes that would count at least when people were using like antisense oligonucleotides versus what you see in the CSF. So you don't have a very good correlation, and there's a huge difference in the level of the amount of protein in the CSF, which -- and you don't know where it comes from. So it's hard really to say is that good or bad, and let me remind you that the Huntington's disease is a whole brain disease. So it's incredibly important to be able to get it to a whole brain. And therefore, the levels of protein levels that we see in the blood as we see changes in the brain are really important. And we showed that you see that in the striatum and cortex in the cerebellum in the BACHD mouse model. You can't make such claims in the CSF nor -- because they're so low in terms of levels, you are near the limits of quantification, and that's not a good place to be, because the variability at that spot is very high. So you can get very different levels, and you can get fooled, because if the variability is that high, you can get different baseline measures that make -- that can actually cause you a variable level for what you think you have. So we think at the end of the day, you're much -- it's much better if you can to measure what you can see in the blood. And I think we have pretty good data reflecting that a small molecule like this will reflect what's happening within the brain.

Yes. Thank you so much for the explanation. It's really helpful. Okay. So maybe just the last question, like I mentioned. So will you show any biodistribution of 518 in different sections of the brain, like from the Phase I trial? Like is there -- is that something you'll be able to show?

Yes. We'd love to be able to do that. But these are healthy volunteers. And so I don't think we're sort of able to do that for this type of study.

Our next question comes from Joseph Thome with Cowen & Company.

One question in sort of the preclinical BACHD mouse model, do these mice show striatum atrophy over time? And have you done any work to show maybe a delay of atrophy with PTC518 treatment? And then second, if you could just walk us through sort of the next stage of disclosures from the program, when can we see additional data from the remainder MAD cohorts? Will that be this quarter or later in the year? And then you mentioned the natural history data. When do you think you'll be in a place to kind of establish that Phase II study?

Yes. Joe, thanks for the question. Yes. The -- in terms of the preclinical, the BACHD mouse model, it doesn't have strong phenotypes in that perspective, so we're not able to do that. While we're able to show -- use it really as a PK/PD predominantly, so we weren't able to do that. In terms of the -- we anticipate finishing up most of what, Matt talked about in terms of the planned studies. I mean most of what was planned at the moment, probably by the end of this quarter. But if we decide to do a few more cohorts that would go into that. But -- and our goal will be to be able to wrap this up and then present it at an appropriate meeting in terms of that. In terms of the natural history and the clinical plans, Matt, do you want to talk a little bit about that?

Yes. Sure, Joe. So we're obviously working in parallel to get that design set up. And I think probably by the time we've got the Phase I study wrapped up, we'll be able to provide you more detail on the design of that trial. Obviously, we're working in the natural history work will continue as we move the program forward. And to achieve all of those goals in terms of identifying the ideal population and efficacy endpoints. So we'll continue to provide updates as we finalize the subsequent studies.

Awesome. Maybe just one more quick one. In terms of sort of the recent antisense landscape data we've seen recently, has this changed at all your opinion on how long you want to follow patients in a Phase II to gain confidence in the molecule before making that investment to go to a pivotal trial?

Yes. Thanks for that question Yes. I mean, the -- there's a couple -- I mean it's a complicated question from the point of view is, in a way, The question is what -- it's hard for me to know exactly what the dose and exposure was with the antisense oligonucleotides to be capable of saying you have to see -- I think, again, what we often do is mix together the notion that, well, they didn't -- what was the efficacy? Where part of the real question is -- I think there's 2 questions: was there an exposure that was -- that did this that would have given clinical benefit that would have led to clinical benefit and, therefore, you did measures. If answer A is true that you got good exposure, than you didn't see, I would agree with you that you might have to think about that. However, in my view, in this case, I don't know, and I don't know that the results show you that there was enough exposure within the brain that people can actually show you this that it actually demonstrates -- that there was enough exposure to be able to say that you offered clinical benefit. And to me, that's the crux of the issue. And so I mean, we'll think about it. Obviously, the team is incredibly experienced an orphan disease and know the trials and tribulations of pioneering a new field. But I also don't -- so we'll think hard about it, but I don't want to -- I don't know if we know enough to say that there was enough exposure that you can say we should- -- that you know that it was a -- it could have been efficacious if it went longer. But as you said, I mean, in terms of thinking about it, and Matt, maybe you want to talk a little bit about how we're trying to think about defining what is the right clinical endpoint for this disease.

So Just so to reiterate on your point, so I don't think we can extrapolate too much from the ASO experience given that I don't think it's a true trial of the approach of HTT lowering due to the biodistribution issue. I mean that's a significant issue given the whole brain nature of the disease and the progression. At the same time, we want to be thoughtful about having the appropriate duration to inform future studies. But we are 100% committed to bringing this drug forward to patients who desperately need a drug. And while we're aware of the incredible need, obviously, also aware of the disappointment with the recent news from the other programs here we remain committed to this approach. We believe it's the right approach. We believe it's the right molecule, the right attributes so that we can make an important impact on this disease for people who need a drug. And so while we obviously want to move quickly in delivering this drug to patients, we also are going to move carefully and thoughtfully so that we're sure that we design these studies and design this program that gives us the highest likelihood of success. So that's really how we think about everything we're doing in this program.

That concludes today's question-and-answer session. I'd like to turn the call back to Stu Peltz for closing remarks.

Well, thank you, everyone, for joining us today. I hope you see why we're so excited about this molecule and that I think now we're very confident that we're able to reduce huntingtin protein, and now the next step is the clinical benefit. So we're planning our clinical trial. So stay tuned for that, and thanks for joining today.

This concludes today's conference call. Thank you for participating. You may now disconnect.