PTC Therapeutics (PTCT) Earnings Call Transcript & Summary

Okay. Good morning, everyone. Thank you for joining us at PTC's R&D Day. Thank you for those who are joining us live here in New York at the Yale Club as well as those who are joining us virtually online. I'm joined today by our Chief Scientific Officer; [ Neal Olmsted ]; our Chief Medical Officer, Lee Golden, members of our R&D leadership team, who will be presenting today as well as members of our IR team. For the past 2 years, a lot of the focus on PTC has been on our late-stage programs and commercial products which makes a lot of sense. However, over that time, we brought the same emphasis on focus and execution to our earlier-stage R&D programs. When we did a restructuring a couple of years ago, we made the decision that we were going to focus on small molecule therapies as well as focus on science that PTC can uniquely leverage to deliver transformative therapies for patients with diseases of high unmet need. Specifically, this meant we were going to focus on our small molecule splicing platform as well as our ferroptosis platform. And we look forward today to sharing a lot of the progress that we have made over the past couple of years. PTC currently has 2 laboratory facilities, 1 in New Jersey and 1 in Northern California. The majority of our splicing work occurs in New Jersey. And our inflammation of ferroptosis program work, most of that occurs in our Mountain View, California location. However, there is a lot of collaboration between the teams. In fact, we have certain specific areas of expertise at each facility that's leveraged by the entire PTC research organization. So today, in terms of agenda, the first part of the morning, we'll be focused on our splicing platform. We'll give you a sort of deep dive into the science around splicing and highlight some of our recent programs. Then we'll have a deep dive into our inflammation and ferroptosis programs. And we'll have time for question and answer after each part of the presentation. I'll mention that we'll be limiting the Q&A to just topics that are covered today, not the Sephience launch, which is going great and other things, but this is really a day focused on PTC's early stage R&D platform. As we will be making forward-looking statements, we -- I refer you all to our statements regarding forward-looking statements, risks and uncertainties that's on this slide as well as our SEC filings. Before I get started and turn the platform over to Dr. Trotta and Bhattacharyya, I'm going to just give a sort of simple readers digest overview of RNA and splicing to provide a foundation for some of the deeper science that will be covered this morning. This central dogma that many of us learned in high school biology that DNA is transcribed into RNA, which gets translated into protein was described almost 7 decades ago by [ Watson and Krick ]. It's now quite clear that this is an oversimplified if not in complete description of how protein is generated. A critical insight into understanding how protein is generated from DNA came with the concept that DNA is first transcribed into a pre-mRNA that contains coding sequences called exons and noncoding sequences called introns. And then a process known as slicing occurs to splice out those introns to form the messenger RNA that will ultimately get translated into protein. Splicing is really the key of how we're able to get 1 million proteins from just 20,000 genes. This important work on splicing earned [ Roberts and Sharp ], the Nobel Prize in 1993. Now splicing is an incredibly complex entity. There are many important participants. This spliceosome, as it's called, contains a number of protein or mRNA and enzyme interactions. There's also enhancers, inhibitors and modifiers that all contribute to the splicing process that, again, takes us from that pre-mRNA to the mRNA. And of course, all of these different aspects of the spliceosome represent potential drug targets. Now the earliest attempts at targeting splicing were done with oligonucleotides, which makes sense because it was believed to affect these splicing processes, you needed the sequence specificity that was believed could only be provided by oligos. However, we know very well that oligos have drawbacks. They're not orally bioavailable. They are not easy to get biodistribute in the brain because if you want to treat a CNS disease, you often rely on intrathecal or intraparenchymal administration, and that really is not an optimal solution if you're trying to treat a whole brain disease, where whole brain biodistribution is necessary. We were interested in trying to use oral therapies to target splicing. This made sense. I mean after all, in small molecule drug development, we often need specificity for enzymes in order to have an effective drug. And of course, we know very well that oral molecules have a number of advantages over oligos. They're easily administered, much easier to manufacture. You're able to get full body, systemic biodistribution. And particularly, if you want to treat CNS diseases you're able to get full brain biodistribution. And finally, titration is a lot easier to achieve with an oral small molecule than it is within oligo. PTC pioneered the field of oral small molecules splicing. I think everyone is aware of our first 2 programs, one to treat spinal muscular atrophy and another to treat Huntington's disease. Evrysdi, which is commercialized by Roche is now the global leading SMA therapy and votoplam, which we partnered with Novartis is the leading oral therapy for Huntington's disease. We're incredibly proud to have pioneered splicing. And these 2 programs not only serve as the incredible innovative compounds that prove that you can get meaningful therapies with oral splicers, but they also provided us a number of important learnings that we'll be discussing today that have allowed us to greatly expand our efforts as well as the efforts of others in trying to develop oral small molecule splicing agents. So how do these molecules work? Well, as I mentioned, there are these coding sequences of exons and noncoding sequences of introns. And our initial efforts focused on the elements of the spliceosome that defined the 5 prime splice sites on the exons. In nature, more than half of these U1 5 prime slice site interactions are known as canonical, which means that there's a perfect complementary fit between the U1 spliceosome component and the 5 prime splice site on the pre-mRNA. This strong interaction leads to the inclusion of exons in the final mRNA. About 45% of these interactions are noncanonical were weak. And when you have a weak U1 5 prime slice site interaction, you do not efficiently get exon inclusion. So you typically will not have that weak interaction results in the inclusion of a sequence in the final mRNA. So our strategy was to develop small molecules that can specifically target these weak interactions, turn these weak interactions into strong interactions and affect the inclusion of exons in the final mRNA. And this is what was done with both of Evrysdi and votoplam. In the case of Evrysdi, we forced exon inclusion to increase the SMN protein, which is deficient in patients with SMA. In the case of Huntington's disease, we forced the inclusion of essentially a stop codon that prevents the translation of the full-length Huntington protein to decrease that toxic protein that causes the disease. So let me describe a little bit further how these therapies work. Let me start with SMA. So SMA is a disease caused by a deficiency in spinal motor neuron protein or SMN. The majority of SMN protein comes from the SMN1 gene. Now there's a second gene that's quite similar, the SMN2 gene which differs, as you can see, in just a single nucleotide here in exon 7. That single nucleotide change leads to a weak U1 interaction and as we discussed, that weak interaction does not lead to efficient inclusion of that exon. And as a result, you get very little functional SMN protein from the SMN2 gene the majority of which you get is a truncated unstable protein. Now in the disease states, there's mutations in the SMN1 gene that results in a significant decrease in the availability of SMN protein. So our therapeutic strategy was to target the U1 binding site on that exon 7 with a small molecule that could again -- sorry, that could turn that weak interaction into a strong interaction, force the inclusion of exon 7 into the mRNA that would then get translated in increase the amount of the deficient SMN protein. I think we all know the punchline here, this worked. These data are important -- were important proof-of-concept in vivo data, which demonstrates dose-dependent increase in SMN protein in the SMA mouse model. You can see that we were able to achieve levels of the SMN protein that were equivalent to those that would result in being phenotypically normal. Also importantly, in these early in vivo studies, we were able to demonstrate that we were getting uniform biodistribution, uniform increase in SMN protein in both the CNS and the blood. And we'll be coming back to these types of figures time and time again throughout the presentation today because these are really important proof-of-concept data that substantiate that we're having the desired target engagement and dose-dependent splicing activity and dose-dependent change in the target protein. From these in vivo data, again, we've -- development was completed. And as I mentioned, Evrysdi is now the leading global SMA therapy. So this program not only brought a transformative therapy to individuals with SMA, but also provided us with a blueprint of how to successfully discover and develop an impactful therapy -- splicing therapy, particularly for CNS disease. A number of critical elements in turn, including the need for sequence specificity, selectivity adequate blood-brain barrier crossing not being [ a flux ], all those critical elements that we learned from the SMA program were brought forward to our next program, which was the votoplam program for Huntington's disease. Now this is the pre-mRNA of the HTT protein. And as you can see, there is a pseudoexon that lives between exon 49 and exon 50 of the HTT transcript. And then without compound in the normal state, the pseudoexon is not included in the final mRNA and therefore, you get production of the full length HTT protein. And with votoplam, what we do is we force the inclusion of that pseudoexon, which then acts as -- or basically induces a nonsense mutation or act as a stop codon to stop the translation and formation of HTT protein. So again, what we've done here is the other side of the corner, what we did with SMA. And in SMA, we used splicing to increase the levels of a deficient protein. And in the case of Huntington's disease, we're using splicing to decrease the production of a toxic disease-causing protein. Again, these are the data that -- those important proof of concept data showing that we're getting dose-dependent decreases in Huntington protein production, and we also have demonstrated and we've shared these data before that we get full brain biodistribution, equal amounts of HTT lowering in each region of the brain and consistent lowering both centrally in the brain tissue as well as peripherally in the blood. We are incredibly proud to have pioneered field of oral small molecule splicing. What seemed like a heretical concept 2 decades ago, that we could use small molecules to drug RNA has now been validated as a powerful source of drug development. Over the past 2 decades, you'll see -- and you'll hear this from Dr. Trotta shortly, our teams have made a number of important learnings that have allowed us to significantly expand the universe of potential druggable splicing targets. We've augmented our small molecule library and we have advanced our screening tools to build something that we call PTSeek which is a powerful platform engine that we believe has the potential to lead to the development of a number of innovative and impactful therapies for a variety of different diseases. So with that brief introduction and overview of RNA biology and splicing, I'll now turn the platform over to Dr. Chris Trotta. Chris?

Okay. Thanks, Matt. That was a very thorough overview of our splicing efforts at PTC. And as thorough as you might feel that was, I'm going to take us a little bit deeper and get really into the science. And I think one of the themes here is that we've learned a lot from these first 2 programs, and I really want to give you a sense of what we learned at the time and how we brought that forward to PTC, and I'll go through that in detail. So I don't know if I introduce myself. I'm Chris Trotta, and I'm the Senior Vice President of the discovery splicing team at PTC Therapeutics, with over 2 decades of experience in drug discovery my multidisciplinary team of scientists has established PTC as a pioneer in targeting RNA and splicing to discover novel small molecules. It's been appreciated that targeting RNA and specifically RNA splicing has opened the door to important targets that lay outside of the traditional areas of drug discovery that target proteins and enzymes. While much of the validation, as Matt said, it was done by antisense oligos or siRNA, PTC has taken a different approach, and we focus on discovery of small molecules as a modality with a proven track record in drug discovery. In my talk this morning, I will walk you to through these learnings from our efforts in targeting splicing. And I will -- and how we leverage these insights to develop the PTC platform. This has opened the door to take aim at the vast amount of targets that are found throughout the genome with our efforts to be able to develop novel small molecule therapies to target genetic disorders, cancers and other diseases, a few of which we will showcase today for you. I will begin by taking you deeper into the science behind the discovery of small molecule splicing modulators. So we start with splicing. So what is splicing? Matt laid this out, but I wanted to give a better picture of this. This is really the -- captures the complexity of splicing. And so basically, splicing is the process that a cell utilizes to take DNA and turn it into a functional protein. And so this is a pre-messenger RNA. In this case, we're looking at the Huntington gene. And you can see -- this covers about 180,000 nucleotides within the genome and there are 67 exons and not introduced exons, which in this depiction, are the dark purple segments of the gene of the pre-mRNA. In light purple, are the introns. And so splicing is the -- the challenge of splicing is to remove the intronic sequences and join these exons to produce the mature mRNA. And you can see that there's a lot of sequence needs to be removed. So this process is required for a majority of genes throughout the human genome, where there can be anywhere from 20 to over 200 exons in a pre-mRNA transcript. Processing of the pre-mRNA with such a vast amount of real estate within the transcript into a mature RNA requires precision. And I think Matt alluded to this precision. Recognition of the boundary between the exons and the introns occurs at locations known as the 5 prime and the 3 prime splice sites. The first step of splicing is accomplished by the U1 ribonucleoprotein particle called here the U1 snRNP or as I will refer to it throughout the talk -- sorry, the U1 snRNP or as I will refer to it throughout the talk is the U1 snRNP. Once the U1 snRNP is bound to the 5 prime splice site, this complex with pre-mRNA is then recognized by the spliceosome, which will subsequently identify the 3 prime splice site that is found at the end of the intron. The splices all functions as a single turnover enzyme as it assembles onto the complex and catalyzes the reactions that will produce, remove the introns and join the exons to produce the mature mRNA product. This process occurs for each exon in a pre-mRNA. The recognition step serves as a gatekeeper and represents a crucial step in splicing. And this comes from literature and splicing as well as our appreciation of just the dynamic nature of how U1 is interacting at the 5 prime slice site. Identifying the correct site among hundreds of thousands of nucleotides found within introns requires a high degree of specificity. If we zoom in to U1, so going from the cartoon picture on the left here, to yet another illustration of U1 interaction. However, we represent it as a zipper. And so the interaction between the U1 snRNP and the RNA -- pre-mRNA at the 5 prime splice site intron, exon boundary is basically accomplished through base pairing and base stacking interactions between the snRNA part of the U1 snRNP and the pre-mRNA. So when there is a perfect match between these 2, the complex is stable. The interaction is stable and will lead to engagement by the spliceosome, allowing the reaction to proceed to intron removal and joining of exons. As Matt mentioned, there are -- however, for more than 50% of splice sites throughout the genome, this is not the case. These are called noncanonical 5 prime exons because they do not have a perfect match at the 5-prime slice site. And in particular, at the last several nucleotides of the exon here shown in dark purple. And so this results in a very weak association of U1 with the 5 prime splice site and often leads to dissociation of the U1 snRNP as the spliceosome is trying to recognize. And as such, it will be -- this exon will be failure to splice. So for these cases, splicing has evolved several strategies to enhance the interaction between the U1 snRNP and the weak noncomplementary 5-prime splice sites. One of the main strategies is the use of cellular proteins called exonic splicing enhancers. These bind to specific sequences found within the exon. Once they are bound to their target sequence, in some cases, they will interact directly with the U1 snRNP, effectively clamping it into place at the 5 prime splice site. This will inhibit the dissociation of U1 and allows the recognition complex by the spliceosome, which leads to splicing and inclusion of exon into an mRNA. This is a critical mechanism that the cell uses so it can sample 5 prime splice sites throughout the large number of exons that are found in any given gene. Doing this accurately and correctly is critical to being able to splice properly. A well-understood example of this comes from SMN. -- specifically the SMN1 gene which encodes the -- as we heard, encodes the SMN protein. Exon 7 of this gene carries a noncomplementary 5 prime splice site. So its interaction with U1 is already somewhat compromised. And so the way the cell has dealt with this is there is an exonic splicing enhancer binding site in exon 7. Once bound, this exonic splicing enhancer protein can stabilize U1 binding at the 5 prime splice site and activate inclusion of exon 7. This leads to constitutive production of SMN protein from the SMN1 gene, as we know, SMN is a critical protein required for motor neuron function. Mutations or deletions in SM-1 gene drastically reduced the production of the SMN protein in patients with the disease SMA, the production of SMN becomes predominantly dependent on SMN2 gene, which is nearly an identical copy, as Matt discussed, of the SMN1 gene. This arose -- millions of years ago when human and chimpanzee lineages diverged. So SMN2 carries this critical sequence change. It's a cytosine to [ eurocell ], this is the binding site for the exonic splicing enhancer, and as such, SMN2 exon 7 is poorly included into mRNA. And this is the reason for the disease is that functional protein coming from SMN2 is very low. So with this clear understanding of the splicing and how an enhancer can regulate the slicing in exon 7, PTC embarked on a molecular based screen. So we took this into account when we developed an HTS screen to screen for small molecules that could target this activity and reverse the week interaction between U1 and the 5 prime slice site. We're basically asking the question, could a small molecule function as a 5-prime splice site enhancer? Our efforts led to the discovery of a small molecule splicing enhancer that could modulate splicing of exon 7 and SMN2, allowing SMN protein expression to near wild-type levels from the SMN2 gene. And through optimization of this hit molecule, it led to Evrysdi, which is a first-in-class, first in mechanism, small molecule therapeutic that is the leading treatment of spinal muscular atrophy globally. Evrysdi functions as a sequence-specific splicing enhancer, it represents a groundbreaking finding in splicing as it's the first demonstration that a small molecule could function as a splicing enhancer by directly targeting the interaction between the U1 snRNP and a week 5 prime splice site, and it was a breakthrough in drug discovery, where a small molecule splicing modulator could be advanced to a therapeutic treatment for patients with a devastating genetic disease. As these findings were emerging from our SMA program, we applied these learnings to accelerate our efforts to target the expression of Huntington gene to lower repeat expanded mutant Huntington protein as a treatment for Huntington's disease. Our strategy to target expression of the Huntington protein utilized our compound library, which at this point now contains a splicing centric chemical scaffolds to identify a brain-penetrant small molecule that could knock down the Huntington mRNA, thereby lowering mutant Huntington protein. These screening efforts led to discovery of the second small molecule splicing modulator, which we demonstrated promotes the inclusion of a previously unknown exon in intron 49 of the Huntington pre-mRNA. This event is very rare -- this exon is very rarely incorporated into the Huntington mature mRNA. In fact, a survey of hundreds of publicly available RNA seq data sets reveals that the presence of this exon within the mature Huntington mRNA is ultra rare. It's almost never seen, but it actually is seen. So this is a biological process that the cell can use to regulate gene expression. And this is basically due to a combination -- and Matt highlighted this, a combination between the week 5 prime splice site, where U1 will not productively interact and the presence of a stop codon. So even when this is sliced in occasionally, this stop codon will cause translation termination, which will lead to decay of the mRNA through a cellular process known as nonsense mediated decay. votoplam functions as a 5 prime splice site enhancer to target the conclusion of exon 49 A, which shifts the splicing to an isoform that will be subject to nonsense mediated decay and with the end result being a lowering of the protein production. This was a foundational discovery by PTC and established a new modality for the regulation of gene expression through splicing. The use of a small molecule modulator to tap into a splicing regulatory mechanism to lower protein expression. When we compare the activity of PTC's 2 first-in-class splicing modulators for SMA and Huntington's disease, a second fundamental insight jumped out immediately. There is a high degree of selectivity of each molecule for their target sequence. This was achieved through lead optimization of the initial hit molecules that led to elicit sequence specificity. We refer to these exons as compound induced exons or i exons. These 2 programs demonstrate the ability to target the inclusion of exon in mRNA to either promote protein production as in the case for SMN protein with Evrysdi or to decrease protein production as in the case of votoplam for HD. Several published studies in our own internal work suggest that these small molecule splicing enhancers function through direct binding of the interface between U1 and the pre-mRNA highlighted here in orange. If we look at the interface, this interaction occurs between the U1 snRNA and the last 4 nucleotides of the 5 prime splice site of the targeted i exon. As this mechanism became clear to us we got very excited and realized we were just at the tip of the iceberg. When you think about a splicing modifier as targeting the last 4 nucleotides of an exon, if you do the math, there's 4 to the fourth potential targetable sequences or 256 potential targeted sequences. Each of these sequences can be considered a distinct class of 5 prime splice sites with i exon throughout the transcriptome. As demonstrated on the wheel to the right, which represents with each sliver of the pie representing a different sequence, SMN2, exon 7 and HTT exon 49 represent the first 2 sequences that can be specifically targeted by small molecule. Taking our learnings from these 2 programs, we asked 2 key questions. First, how broad is i exon regulation of gene expression? In other words, how many and in what genes do i exons exist? And second, can we systematically discover distinct small molecules for each i exon sequence class found at the 5 Prime splice site. We reason that success within these 2 efforts combined would open up the ability to target a wide range of therapeutically relevant i exons in targets of high interest to human disease. To uncover the extent of i exon regulation, I'm showing here one type of experiment where we did this. We've also, as I mentioned, looked at RNA seq data sets, but we directly interrogated the transcriptome to understand the true scope of i exon inclusion. And the way we did this was we genetically modified the snRNA from U1 to each of the potential 256 sequences. So as depicted here as the ends, basically, we had a U1 variant that would match each of those 256 sequences. When we did this, we then treated cells and we look to see what exons were affected. We started, we looked at SMN2 and HTT, the genetically modified U1 variance for those 2 indeed induced in i exon class that matched the 5 prime splice site that we see Evrysdi and votoplam respectively, affecting. Upon expanding this analysis to the remaining 254 sequences, we found evidence for i exons for all 256 sequences. Furthermore, amongst all the i exons observed, we estimate that there are i exons covering a majority of genes throughout the genome, including numerous genes of therapeutic interest. This insight into i exon regulation was pointing us directly at numerous druggable targets and led us to tackle the next and probably more important question, can we actually find molecules that can do this? So can we systematically discover small molecules that target the novel 5 prime splice site sequences as hit molecules for i exons that we could develop in exactly the same way we achieved with SMN2 and the Huntington pre-mRNA but within new therapeutic targets. So to accomplish this, we built off the foundational discovery, some of which Matt and I have outlined today as well as we just saw a moment ago, and we developed what we call PTSeek. This is a novel platform that utilizes transcriptome wide interrogation. So directly looking at splicing, transcriptome wide and we use a innovative screening tools and some and informatics insights that we have gained from understanding the 5-prime splice site enhancers and work that we've -- of targeting splicing over the last several years. We couple this with PTCs custom design library of novel small molecules that cover unique splicing centric chemical space, which has been built over the last 2 decades. To be clear, our efforts to this point led to the discovery of an entire universe of targetable i exons. And with PTSeek, we now had a small molecule strategy to uncover novel splicing modulators. Employing PTSeek for several years has led to just that, the discovery of dozen and counting novel chemical scaffolds that target novel 5 prime splice site sequences within i exons. As with Huntington and SMN splicing modulators, these novel scaffolds have a distinct structure sequence relationship. So each novel molecule targets a very specific sliver of the pie here. And this -- and as such, they all -- from the beginning, they demonstrate a high specificity for a novel target sequences. Equally important and many of these targetable i exons, many of these i exons are found in therapeutically relevant targets, some of which my colleague Anu will be presenting shortly. So connecting these 2 -- connecting the discovery of a dozen novel molecules that target distinct sequences that are found in i exons of therapeutically relevant targets is unprecedented, and it really positions PTC to advance numerous splice modulators into the clinic over the next several years. As a sneak peek to some of the data Anu will discuss, I want to highlight another significant advantage of the PTSeek approach. The molecules that we have discovered are high-quality hits. Since these molecules are derived from splicing centric scaffolds, we have found that progression from a hit through early lead and lead optimization has been greatly accelerated relative to our initial efforts targeting splicing exemplified here by comparison to the SMA drug development time line. This has allowed us access to more targets and a quicker path to a development candidate. So with that, I will sum up. PTC's efforts to target splicing have led to tremendous success in building a sustainable drug discovery platform, first, through a targeted high-throughput approach for SMA and HD, where fundamental insights into the innovative mechanism of targeting splicing with small molecules [ mods ] splicing modulators has led to the establishment of a platform approach, which focuses on transcriptome wide discovery of small molecules from PTC's proprietary splicing library. PTSeek is a best-in-class discovery engine. And over the past 3 years, has and continues to deliver novel small molecules that target a wide range of therapeutically relevant targets implicated in diseases of high unmet medical need. I would now like to turn the podium over to Anu to take us through some of the exciting emerging programs from our efforts to target RNA splicing.

You can all hear me, right? Thanks, Chris. Good morning, everyone. My name is Anu Bhattacharyya. I have been at PTC for more than 18 years. I lead our splicing programs team. Chris just did an excellent job in describing how we have ushered in the next generation of splicing programs. I'm going to -- I'm sorry, these programs have this opportunity to deliver novel splice modulators like Evrysdi and votoplam across a wide array of indications. So I'm going to talk to you about a few of these programs really highlighting the great work that our fantastic splicing cross-functional team has done. The first 3 are late stage and the last 2 are early-stage programs. And in some cases, you will note that we're not going to disclose specific targets to maintain competitive advantage. So the first program targets a key driver of disease pathology across a group of indications called nucleotide repeat disorders or NRDs. In nucleotide repeat disorders, individuals have a segment of DNA repeats in a given gene that is usually longer than what is typically seen in unaffected individuals. Right after birth, these repeats cause no symptoms. As the individual ages, the repeats expand further, especially in affected tissues. This expansion phenomenon is called somatic expansion and somatic expansion has been documented in almost all nucleotide repeat disorders. Our initial target indication for this program are Huntington's disease and myotonic dystrophy 1, or DM1, and we will focus on these 2 indications first. For Huntington's disease, this is a differentiating approach from what we're targeting in the votoplam program. And this approach of targeting somatic expansion is complementary to Huntington lowering. So let's dig a little bit deeper into this phenomenon of somatic expansion. The rate of somatic expansion is key here. If the rate of expansion is faster, that will result in earlier disease onset and faster disease progression. On the other hand, when the rate of expansion is slower, patients will have delayed onset and slower disease progression. So there is a strong correlation between the expansion rate and disease onset and progression. And this expansion rate is primarily driven by a family of proteins called mismatch repair or MMR proteins. Therefore, lowering the levels of these MMR proteins will slow down somatic expansion, which was significantly delayed disease onset and progression. Our program objective is to use PTC splicing technology to modulate the levels of these MMR proteins and target the rate of somatic expansion. Now of all the MMR proteins, MSH3 has really emerged as the most promising target in the past 5 years and the data is specifically strong in Huntington's disease and DM1. As I said, MSH3 has very strong human genetics evidence. Here, mother nature has done the work to validate a correlation between MSH3 levels and disease onset and progression. As you see here in the table, there are HD patients who have these different variations in their MSH3 gene. As a result of these variations, these patients have reduced levels of MSH3 protein. Now when this is plotted out in there on the right, you see a clear linear correlation between the extent of MSH3 reduction in these HD patients and their disease onset. Now not only does this human data validate the therapeutic potential of MSH3 reduction, but it also values MSH3 protein as a potential surrogate biomarker. A similar trend in correlation between somatic expansion and disease onset was also observed in DM1 patients. So our program objective was to use our splicing technology defined novel molecules that will lower MSH3 levels. And I'm very happy to report that we have identified novel splicing modulators that lower MSH3. The mechanism, as laid out on the slide is similar to what Chris described earlier for Huntington lowering. In this case, the molecules target a novel sequence distinct from the SMA and HD molecules, and they target a different gene, MSH3. These novel molecules promote inclusion of an exon bringing in a premature stop codon in the MSH3 mRNA. This induces degradation of the mRNA and reduction in MSH3 protein levels. So we have been using our in-house splicing expertise to significantly expedite both drug discovery and optimization processes. Here is a snapshot of the results of that work. Now the data shown on this slide represents a lot of hard work from our very experienced splicing cross-functional team. We have identified novel compounds that are highly potent and selective. Here are some key pieces of data from one such compound. As you see here on the left, the compound shows a robust and correlative dose-dependent lowering of MSH3 mRNA and protein levels in cells. And this data is validated in animals as shown here in the middle. You see a nice dose-dependent reduction of MSH3 protein levels. Just as we have done in SMA and HD, we optimize these small molecules to cross the blood bring barrier not be e-fluxed and show uniform distribution throughout the CNS. We established a robust PK/PD relationship in this program. From our previous experience with SMA and HD, strong PK/PD data generated in animals transits very well to the clinic. Next, as proof of concept, we tested our compounds in a specific expansion model. This was to see if MSH3 lowering has an effect on somatic expansion. As you see here on the right, we were able to show that 30% to 50% lowering of MSH3 levels significantly stalled expansion in this model system. This is in line with what's been reported and it represents what will be our target lowering in the clinic. This is a very promising program because it's built on a strong scientific foundation. By targeting MSH3 lowering, we can tackle what's been shown to be a key driver of disease pathogenesis in repeat disorders like Huntington's disease and DM1. Again, in HD, this approach of MSH3 lowering will be complementary to HTT lowering. And this could be particularly appealing for juvenile onset HD because this population is linked to higher levels of somatic expansion. Our molecules target a novel splicing event that results in reductions of MSH3 mRNA and protein levels. They have favorable drug-like properties for CNS indications, such as Huntington's disease, our plan is to select a clinical candidate in the first part of 2026 and be Phase I ready by the end of the year. Next, we're going to provide an update on our SCA3 program. SCA3 or spinocerebellar ataxia 3, is a monogenetic neurodegenerative disorder. It primarily affects the cerebellum, brainstem and spinal cord. Balance problems and in coordination are typically the first visible symptoms of SCA3. And currently, there are no disease-modifying therapies. Now in SCA3 patients, the mutant ataxin-3 protein is the primary toxicity driver that leads to neurodegeneration. Our therapeutic strategy is to lower ataxin 3 level, utilizing our splicing technology. We have identified novel splicing modulators that skip or exclude exon 4 in ataxin-3 pre-mRNA. This is a different mechanism from what I described earlier for MSH3. Now this is a great example of how different splicing mechanisms can be leveraged to modulate protein production. Here, the molecules skip or exclude exon 4 and that breaks the frame of the ataxin 3 mRNA, which creates an mRNA with a premature stop codon that then gets degraded. This results in reduction in ataxin-3 protein levels. Here are some key results of our lead compound. As you see here, the compound shows a robust dose-dependent reduction in ataxin-3 mRNA and protein levels in cells and animals. Just as we have done in other CNS programs, the molecules in this program were optimized to cross the blood brain barrier and not be e-flexed and the extent of ataxin 3 lowering, we observed in animals is in the range believed to be needed for functional benefit. So for the SCA3 program, we identified a novel splicing event that results in ataxin mRNA and protein lowering. We demonstrated robust in vitro and in vivo activity and optimal drug-like properties. Similar to our MSH3 program, the plan is to select a clinical candidate in the first part of 2026 and be Phase I ready by the end of the year. Next, I will talk about a program that targets brain tumors and metastasis. This is another great example of the utility of PTSeek in identifying novel splicing modulators across different indications, including oncology, which can be a partnership opportunity for us. This oncology target is an interesting one. It has a dual role. It promotes cancer cells to grow, and it suppresses T cells from being activated. So lowering it will have a dual effect. It will block the cancer cells from proliferating and it will enhance T cell function to kill those cancer cells. Also, it's been shown that the lowering of this target will synergistically enhance the anticancer effects of immune checkpoint inhibitors. The molecules we've identified in this program have excellent CNS exposure, which makes them suitable for targeting brain tumors. As I mentioned, we have leveraged PTSeek again to identify this novel class of splicing modulators. These molecules modulate splicing of the target, promoting inclusion of an exon. And this induces mRNA decay and lowers protein levels. Again, similar to what I described for MSH3, but now with a different class of novel molecules. Here, I'm showing you an example of a lead compound that is very potent in reducing mRNA and protein levels in cells. This reduction in mRNA and protein levels leads to T cell activation as evidenced by a significant increase in cytokine levels, which is shown here in the middle. This brain penetrant compound also shows a robust lowering of the target in animals shown here on the right. In summary, we made excellent progress on this program. We have identified a novel class of splicing compounds that have great activity and selectivity. By lowering the target the compound activate T cells, they have great brain penetration. Next, we're going to assess the compound synergy with checkpoint inhibitors. Our plan is to select a clinical candidate early 2026. Now I will shift to some of our earlier stage programs. First, I will start with the program that targets sickle cell disease and beta thalassemia. Now sickle cell disease and beta-thalassemia patients share a common pathology. They make defective adult hemoglobin. One therapeutic strategy that has emerged to be the most promising is induction of fetal hemoglobin levels in these patients. Again, mother nature has validated our therapeutic strategy. Sickle cell disease and beta-thalassemia patients with increased levels of fetal hemoglobin show significantly reduced symptoms. And this process of fetal hemoglobin induction is regulated by multiple proteins that we believe are great splicing targets. So we have leveraged the PTSeek platform to look for novel molecules that will target these key proteins. PTSeek, again, has enabled us to identify novel splicing molecules that target one of these key inhibitors of fetal hemoglobin. These molecules modulate splicing of the target again, by promoting inclusion up in exon, as shown here on the slide. This induces mRNA decay and lowers protein levels. Now reduction of this key inhibitor will lead to an increase in fetal hemoglobin, and that will be disease modifying for sickle cell disease and beta thalassemia patients. As you see here on the slide, one of these early molecules shows a nice dose-dependent reduction of the target inhibitor which in turn induces fetal hemoglobin levels in adult blood cells in the therapeutic range. This is an early stage program. And as you see here on the right, the extent of fetal hemoglobin induction is already very encouraging. We will continue to identify and optimize these small molecules for potency, efficacy and drug-like properties and move this program as quickly as possible. Next, I will talk about another early-stage program that has application in multiple neurodegenerative diseases. Now with SMA and HD, we targeted a single gene to address disease pathology using splicing. This is another example like MSH3, where splicing can be applied to target specific protein and targeting this specific protein will have a huge impact on a key pathway implicated in disease pathology across multiple indications. And this is a new powerful way we can use our splicing program. Here, our therapeutic strategy is to mitigate the neuronal dysfunction that results from protein aggregation. And this is a common pathology of many neurodegenerative diseases. Again, with PTSeek, we can identify novel splicing modulators that can specifically target key proteins in this pathway. Our program's objective is to have a single therapy that has broad utility across multiple disorders. We have been able to identify novel splicing modulators that target one of these key proteins. This program is early but already our fantastic splicing team has done a tremendous job in identifying compounds with good activity and selectivity. As you see here from the green arrow, we are already seeing target modulation in the range that is deemed therapeutic, and it's just the beginning. We will continue to identify and optimize molecules for potency, efficacy and drug-like properties and move this program also as quickly as possible. So as Chris and I showed you today, we have truly pioneered splicing drug discovery and development. It started with our groundbreaking work in SMA that led to the development of Evrysdi, which was then followed by our HD program and development of votoplam. We use the learnings from these programs to develop our innovative PTSeek platform. This platform is built upon decades of experience that allows us to expedite the discovery and development of novel splicing molecules. Now what's really exciting about PTSeek is that it unlocks a whole new therapeutic space by selectively targeting RNA and it gives us the advantage of delivering therapies that can't be addressed with traditional small molecule approaches. It also positions us to broaden the scope and continue to deliver best-in-class splice modulators like Evrysdi and votoplam in a wide array of indications. Thank you so much for your attention.

Great. Thank you very much, Chris and Anu. Fabulous presentations, and I hope everyone sees how with our focused efforts over the past few years, we've been able to significantly advance and evolve our splicing platform that we said we believe can be a significant source of innovative therapies, both for PTC development and commercialization as well as strategic partnerships for noncore therapeutic areas.

We'd like to open the floor now to questions in the room looks like there's a lot of questions. So we'll have microphones that will circulate around and see as many as we can. It looks like Brian, you're up first.

That was really insightful presentation. Just a couple of questions on the MSH3 program, if I could, because that seems like it's one of the ones that's furthest along. How do you think about who would be the most optimal patients, Huntington's patients, for instance, for a splice modulator like that? Would they need to be kind of early stage or even presymptomatic just given the potential to stem the repeat expansion. It also looks like in the chart, you we're looking at about 30% to 50% reductions, which I think was similar to what you were targeting with votoplam and HTT. So just wondering if there's a reason for that window and if there's any concern about going too high in potential cancer risk. And then lastly, curious if you could maybe compare and contrast the concept of using splice modulation here versus just allosteric direct inhibition of MSH3 with a small molecule which I think is around the same stage of development.

Terrific. Thanks for the questions, Brian. Let me tackle the first and then I'll have Anu handle the 2. So we -- so in terms of the relative role or the role of MSH3 lowering, we certainly, as Anu said, see this as complementary to HTT lowering. And there's a number of papers emerging and a lot of work emerging, showing me the power of putting these 2 therapies together. It's also possible that something like MSH3 lowering may be more applicable in, say, juvenile HD, where there is known to be a more aggressive disease, larger repeat expansion that's driving that accelerated clinical presentation and disease course. So I think we see a complementary HTT in all cases and maybe particularly alone suited for juvenile HD. Anu, do you want to tackle the other 2 questions?

Can you hear me? Yes. So starting off with 30% to 50%. Why? So I'm going to go back to the slide where I showed the very strong genetic -- human genetics data. And what we saw was that between 30% and 50%, you're talking about up to 10.5 years delay in disease onset and up to 50% reduction of MSH3, systemic reduction of MSH3 is deemed safe in terms of any potential cancer risk. So that's really what we're shooting for. That's the first question. And then -- sorry, second point. And the third was, you were talking about allosteric inhibitors. I think the way I think about it is, right now, we're not really sure about what path that will take because this is going to be the first as opposed to on our end, we have tremendous experience in already lowering HTT, and we know the range we can do that selectively with a small molecule. So we feel like there is very strong evidence that modulating the levels of MSH3 is going to be therapeutically beneficial as you have seen with other modalities. So we're definitely very confident about protein modulation.

And I'd also say that we've had programs and looked at other difference modulators, including BMS1, and we've obviously selected to focus on MSH3 for a couple of reasons. One, MSH3 has upstream to BMS1; and second, to really affect or to get an effect on expansion with BMS1, you tend to have need protein reductions in excess of 70% to 80%, which is getting very, very high. So for both those reasons, being upstream and also just being able to have more achievable lowering window we're moving forward to MSH3. And I think the literature supports that choice as well.

Great. This is Brian Cheng from JPMorgan. Maybe just first one is more of a broader question. As we think about the scaffold that's underlying across all your new pipeline. Can you talk about whether the scaffold -- if part of the scaffold is the same across the new programs that you disclosed today and also compared to the older program, Evrysdi and votoplam. And given that there is still going to be that interface with U1. Just how much read-through is there between the approved late stage and also the earlier stage pipeline? And then the second question, going back to MSH3 in Huntington, how should we think about the target population here? And do you have a sense of how the patient target should differ or may not differ compared to votoplam's program? Is it going to be based on the level of somatic extension that you see?

Sure. So the first question, Chris or Anu either if you want to take that and talk about how different these molecules are.

Yes. I'm not the chemist on the program, but I can't answer the question. Basically, structured sequence relationship is the term we use to describe this, and it really is the case that these new target sequences that are being affected -- it's new chemistry. It's a new molecule. It's reminiscent in terms of general principles from the Huntington and HTT, but it's a novel chemical scaffold. And it's true for most of the programs we've showed today as well as some of the other slivers on the wheel that we didn't discuss. So I think just the accumulating evidence we're gathering from doing this is we can get distinct novel chemical scaffolds.

Just to emphasize the point, both the SMA program and HD program, shared and common targeting the same dinucleotide sequence, right, that GA sequence what we presented today, which is truly novel and incredibly exciting is we've expanded this universe beyond just a simple dinucleotide repeat to the full spectrum of 4 nucleotide sequences, which did the quick math on a 4x4x4. And what you're seeing in some of these new programs is an entirely different splicing target sequence and the fact that we had the slicing centric chemical library that using PTSeek allows us to more rapidly and reliably pair the novel chemical matter with this target or the target of interest. So that's what's really -- there's a lot of new things here that have really not been talked about before that are incredibly exciting in terms of really significantly expanding the universe splicing targets and then being able to leverage those new targets with a splicing centric chemical library to develop potential therapies. Anu, if you want to add a comment on the second question, which is around -- we had said that we see the MSH3 targeting as complementary to HTT lowering, but are there other specific cases where you would think about trying to identify specific somatic expansion rates that may make an individual particularly more amenable to the approach.

I mean you touched on juvenile onset HD, I think that's a population that has been shown to have greater somatic expansion. So in terms of adult onset HD to be complementary to HTT lowering from the literature and from just talking to everyone in the field, it sounds like even if you -- well, a, you go in early, like we're doing with HTT lowering where majority of the neurons are spared and the expansion hasn't happened in majority of the population. So that's a great place to start. But even you're hearing from the field that you can go in a little bit later with MSH3 reduction because you still have -- yes, you still have a big portion of the neurons and like in the [ striatum ] that are still alive, and you can spare them from expanding. So I think first is probably HTT similar to the HTT lowering population.

I need to introduce myself. Tazeen Ahmad from Bank of America. Couple of questions from me on MSH3, if I could. So you've talked about having several compounds being ready to move into the clinic next year. Based on what you've seen on the impact of somatic expression in the model system, how confident are you that you'll be able to replicate that, let's say, and I don't know healthy volunteers and then going on to actual patients. Related to that, in the past, you might have had to go through more than 1 molecule in order to find the right fit. Based on previous experience, what have you learned that could maybe make that process a little bit more efficient? And then I'm going to throw in a quick commercial question, sorry. On DM1, in particular, it's a topic near and dear to my heart. We look at several companies that are trying to develop therapies there. How do you make a decision about wanting to go into a space which might have a lot of competition, especially early on, where you may not be fully aware of what your product profile might look like?

Great. So let me make sure we get all of those orders. I think and the first question really thinking about the reliability and how we've set up our preclinical screens to be reliable predictors of what we can see in the clinic.

Yes. I mean I'm just going to say that I'm going to use the HD playbook. And as I said, very early on, we think about in vitro to in vivo translation, and in establishing PK/PD relationship, we are very confident what a Phase I, Phase II proof-of-concept trial will look like. And just taking 1 step further the efficacy model where we show 30% to 50% reduction showed significant delay in somatic expansion. That's a widely used model and that's been published in the literature, and we see exactly what other people have shown. So we have a lot of confidence in that model that the extent of reduction we're seeing, a, we'll be able to translate that into the clinic because we have a very strong understanding of PK/PD. And b, if we are shooting for that extent of reduction, we should be able to significantly install somatic expansion.

And the second question was about how do we think about the role of backup molecules as we do this development. And clearly, we talked a lot about how PTC gives us a faster track to get through those early hit to lead the lead up and the development candidate, but how are we thinking about backups as we move programs forward?

Yes. I mean for every program, you have multiple scaffolds that you're looking at, so you can derisk and then also you have multiple lead molecules. And then obviously, the goal will be to look at safety, efficacy and everything and then advance the compounds and then have backup. And like we have done again with HD, we're not just focused on 1 chemical scaffold. We're actually working on multiple scaffolds at the same time.

And the other part is once we get a development candidate and even start Phase I, the work doesn't stop for exactly the reasons you highlighted is we continue to do work and leverage the learnings of what we brought into the clinic to continue to see to have a backup ready and in some cases, maybe that backup could be take the poll position at some point in time. And I think that's common to a lot of different drug development programs. I think when you look at all the work that goes into getting these molecules forward. It's incredibly important to make sure that all the exon in 1 basket, but that we're continuing to make sure that we have backup plans as we take a compound through the key preclinical and early-stage clinical programs. In terms of your question regarding competition in commercial landscape, look, it's something we think a lot about. I mean, whenever we decide, and this will come up a bit as we start talking about the inflammation of ferroptosis programs, how do you select an indication and some of it is a scientifically rational target. Of course, in the case of splicing, we know we're hitting something incredibly relevant to the disease in a highly specific and selective way. We also look at developability. Is there a clinical path forward? Can you actually do the clinical trial? Are there endpoints? And as Anu alluded to, as one of the things we're looking at with MSH3, is this idea that when you have such good literature-based evidence now establishing the association between MSH3 and somatic expansion, you started thinking about the use of [ sericite ] markers and translational biomarkers early on. And then we think about the commercial landscape. So we do pay a lot of attention to what does the competitive landscape look like today? What do we think that competitive landscape looks like tomorrow? And then, of course, as you also mentioned, Tazeen, sometimes it's hard to know because there's things out there that aren't on the radar screen. I would say this program is one of them that it's hard to assess. And so this is something we continually go back to. I think we've all learned in the case both of rare disease and particularly in neuromuscular neurodegenerative disorders, something that looks good in Phase I into Phase II may not get there all the way. So we have to have a balanced approach as we look at the competitive landscape today and what may be tomorrow and not necessarily always assume just because something is in the lead, it's going to get approved. And then also, we, of course, look at if it is approved, what is the relative benefit of our [ TPP ]. What does our molecule offer that perhaps the approved one doesn't. So it's a continuous process that we look at every step of the way.

Joe Thome from TD Cowen. Maybe can you talk a little bit about how the Novartis partnership on [ 518 ] plays into the Huntington's compound today? Does it make it potentially easier to combine with [ 518 ] or limit that in any way? And then second, in regards to SCA, I guess, how easy is it once you maybe establish proof of concept with SCA3 to go to other members of family like SCA1 and some others. And then just quickly 1 more. Obviously, some other programs in SCA and Huntington's have had some hiccups with FDA. I guess what learnings have you taken from this in terms of how you want to design your initial studies for both these programs?

Thanks for the question, Joe. The first one, again, as we mentioned, we see as the field does the potential synergy of using Huntington lowering along with targeting somatic expansion. We're incredibly excited about the Novartis partnership. It's a terrific partnership for us, we're excited on the progress that we're making with votoplam. This program sits outside that partnership. But I think nonetheless, that there is a clear path that one could think of developing first and establishing first the necessary things we would need to do for somatic expansion and then down the line, thinking about how those therapies can be looked at together. Anu, do you want to touch on the specificity or the issues of SCA3.

Just for the SCA3 program, we're targeting ataxin 3 splicing. So it will only work on SCA3 patients. So that's different. And you asked about some of the earlier oligos, for example, especially, I think the SCA3 oligo, the issue was lack of efficacy and some preclinical safety signals. So obviously, we have a small molecule. We are going to target the entire brain and also -- and I mentioned this on my slide when I was introducing the indication for SCA3 you have to target the cerebellum and brainstem. And that, we believe, we can do with a small molecule as we will see very nice uniform reduction and even from distribution throughout the CNS.

And in terms of your third question about thinking across all of those programs on the regulatory pathway learnings that we've made, one of the things that the SMA program taught us, and again, I talked about a blueprint or a path, which we could follow with other programs, is the ability to get evidence of target engagement and proof of splicing even in healthy volunteers, given the fact that these molecules that have systemic exposure. We're able to readily avail ourselves of the peripheral compartment and get PK/PD early on, which is incredibly valuable. And the other part of this, and Anu alluded to it, in our presentation, when we look at the patients who have single nuclear polymorphisms that are affecting MSH3 and that effect on disease onset and disease progression. I think we're now thinking a little bit more ahead of time of how we can start establishing the body of evidence that we could take the FDA very early in a development program and start having discussions around what could be the basis of a surrogate end point so that when we start the clinical program, we already have some alignment with the agency on how they would be thinking about the necessary data package to support an accelerated approval. I think this is a learning that we have made. I think we certainly learned, we've had very productive discussions with FDA and [indiscernible] on the Huntington program. And I think we know it would be in our best interest get on board with the agency sooner. And they're actually doing -- they're introducing programs as well to engage earlier responses around biomarkers. So I think we have a convergence of interest, and we're certainly doing our part thinking about ways that as quickly as possible. We could have the steps or the details of what could be an accelerated approval development program.

This is Clara Dong from Jefferies. So maybe a quick one on MSH3. First, because it's part of the mismatch repair system. So when it's lower to the level you showed in his slide, do you see any compensatory MNR components up gradually? Or it's completely an independent mechanism and then also on SCA3, do you have a sense of magnitude of RNA and protein reduction correlates to meaningful disease modification and based on the level you are seeing in vivo and in vitro, how should we think about this the threshold in humans for this modification? And maybe a quick one on the market opportunity for SCA3 as well. So in the slide, you've highlighted the problems in Japan, 2,000 patients there. How are you planning to integrate the disease prevalence across the world into your clinical development and to reflect the future of commercial opportunity as well?

Thanks, Clara. Anu do you want to -- the first question, Clara, was just about was regarding MSH3. And can you just repeat it one more time?

Any update?

Off-target?

Compensatory. So it's interesting as Matt was talking about the MMR proteins that are involved in that pathway, MSH3 is most upstream, the most proximal. And MSH3 and that pathway actually is very specific for repeat microsatellite repeats. So in terms of compensatory, if you are targeting MSH3 and you're lowering MSH3 from all the preclinical data, it looks like you're not going to upregulate another parallel pathway that will compensate for the loss of MSH3. So we're very confident that lowering MSH3 will do its job. What was the...

And the second question was...

SCA3. Yes. I mean in terms of target engagement and the range that we're shooting for is very much in line with what we saw in our SCA3 animal model, the CSA3 mouse. We're shooting for anywhere, I think 40% is what we're looking for and that we believe will give the therapeutic benefit that you need in the SCA3. And that's based on a lot of preclinical data.

So I'll just quickly answer the question on Japan. We have global development capabilities. We've done clinical studies in Japan and also have the full infrastructure in Japan for commercial services as well. Gena, I think you're next.

Gena Wang from Barclays. I think I spent many, many years when I was at back at school, doing my PhD post-COVID doing splicing and so it's very exciting to see the advancement of where the field is now. So I wanted to ask a few scientific questions. So I think the first one, regarding your U1 GGUC, the complex, right? So how do you testing the noncanonical 5 splice donor sites that when you try to target a specific, say, Huntington or MSH3, what is the -- like the whole transcriptome sequencing that makes you feel confident that whatever the target, the small molecule, you identify will be very specific to that particular gene? So that's the first question. And the second question is regarding the MSH3. I know -- we know that this is truly is the housekeeping genes, right? So there are some other studies suggesting will be involved in the Huntington path. And when we look at it is the [ locking ] mouse model. So -- but on the other hand, it is a housekeeping essential proteins for DNA mismatch repair. So any concern, and I think others also question about the potential cancers. So how much have you done when you do this and especially this oral delivery and you basically penetrate everywhere, the key focus is the brain and how do you test in the other parts of the body, make sure there is truly, there's no upper regulating of the DNA breakdown or other consequences?

Terry. Thanks, Gena. So Chris, the first question is regarding how do we assess specific...

So I think Matt used the word heretical at one point. I think Gena, I know you've done splicing for a long time. And I think it was really the case that as you outlined, could this be done? And the short answer is empirically by optimizing the molecule, we select for that potency against the target of interest. And by doing this to the exclusion of other targets, the molecules that win are the ones that can accomplish this, how that's actually happening at the molecular level is not completely understood as the interaction is very dynamic at the 5-prime splice site. That structure is very dynamic. But for Huntington, for SMA for MSH3, these are the top hits for these molecule classes and we move forward with that. And that's one of the reasons why it's nice to expand out to other sequences because there are going to be some targets that are just preferentially affected by a molecule like this at that 5 prime splice site. So what we can't gain for in one particular chemical class, we can grab it in another by finding a molecule that's optimal for a different 5 prime splice site sequence. And actually, sometimes, this plays in the same target. There can be more than one i exon in a given target. One of which can be optimized for a particular molecule, the other of which cannot, and we actually see this through our optimization. So early on, this is the key question. Can we get that selectivity? And if we do, that leads to an advanced program. If we don't, we move on to either another i exon, another target, another sequence.

And then Anu, do you want to talk a little bit about, again, the confidence in the threshold for lowering or MSH3?

Yes. I think up to 50%, as I said, there is enough human data, which was then followed up and validated using animals. And I think in terms of safety, one of the things that we're not going to do is we're not going to knock out MSH3. We're going to go up to 50%, where we believe we will get the most efficacy and we will not have any risk of cancer. And in terms of like looking at MSH3 in that pathway, there is enough redundancy. So MSH3 does not really play a role in the single mismatch repair that's like more critical, that's MSH2 and MSH6. That's why, for example, MSH2 is also a genetic modifier identified in HD but people are not pursuing MSH2 to because of the fact that it is indeed an essential gene, and it will cause cancer. Same thing with MLH1. So one thing that has really emerged from again, in terms of efficacy strong efficacy data and strong safety data that in that order, like what are the top 3 mismatch repair proteins that would be ideal for somatic -- targeting somatic expansion, #1, MSH3. Number two, PMS1, #3 MLH3. So in terms of, again, safety we're not going to go up and then knock down -- knock out MSH3. We will knock down up to 50%. And the beauty of small molecules is we can do that. We can titrate down that reduction to 50%, and we can do that informally throughout the brain. So we're very confident that this 50% reduction should not have it in cancer risks.

Terrific. I think we have time for 1 more quick question or so as a quick question, and then we'll move on to the next. Paul?

Paul Choi with Goldman Sachs. My first question, maybe for Matt is on the hemoglobin programs. And just given that there's so many late-stage and or commercial programs, can you maybe talk about where even if fetal hemoglobin is a validated target, where you see your program potentially fitting in, in the various spaces, whether it's access to gene therapy or other modalities and just sort of where you ultimately envision the program? And then on the brain tumors program, historically, PD-1s haven't shown much activity there. So can you maybe just talk about what is the rationale behind combining your asset there, even if it is brain penetrant with sort of the PD-1 program?

I'm going to scan for both briefly because we're going to move on to the next part. The first question is I would hardly say the commercial space right now is crowded for sickle cell disease. I know there's a number of therapies in development, and I think being able to offer a highly selective oral therapy I think it will fit very well and we have a very important option in the therapeutic landscape. And as we mentioned earlier, we will continue to monitor that. This is an early stage program as we bring it forward. We'll continue to see the how and if the commercial landscape evolves and then do a reassessment at that point. I think one of the compelling value propositions of the splicing platform is that it is oral, and we believe very strongly in the relative benefits of having highly selective and specific titratable oral therapies. Your second question, in terms of PD-1 and brain tumors. I think one of the questions that we'll look at as we move forward is could the synergy of our target along with PD-1s, actually bring them to a level that together we can target brain [ mestastases ] or brain tumors where clearly, there's a high unmet need in being able to have at least an oral therapy that can add to the momentarium and provide better efficacy, I think would be certainly a very interesting and important opportunity. So with that, thank you, Chris, Anu, and thank you, everyone, for the questions. There'll be time after words as well for those in the room, we can do informal Q&A.

We're now going to shift gears and move to presentation of our inflammation and ferroptosis platforms. These are different set of programs. We have clinical -- early clinical stage as well as preclinical stage programs that we're excited to share with you. So let me ask Jeff Trimmer to come to the stage and kick us off.

All right. Thank you for the introduction, Matt. I'm Jeff Trimmer, I'm the site head for our research facility in Mountainview, California, where we're actively working on early and late-stage preclinical programs. Today, I'll present an overview of the inflammation and ferroptosis platform and then provide more details about our 2 most advanced preclinical programs. I'll then hand it over to Mayzie to present our clinical programs. So the inflammation and ferroptosis platform was designed specifically to deliver compounds that target specialized enzyme hubs that are essential regulators of biological functions, such as inflammation, energy production and oxidative stress. At the core of the platform is our unique small molecule library that contains compound specifically selected to target these enzyme hubs and deliver differentiated or first-in-class compounds. It's worth noting that this chemical library is distinct from the PTSeek library that Chris introduced earlier today. Today, we are advancing multiple preclinical and clinical programs for both CNS and non-CNS indications. And Mayzie and I will be highlighting these 4 today. Starting with our most advanced preclinical program targeting 15-LO ferroptosis to deliver a disease-modifying treatment for Parkinson's disease. Next is our NRF2 activation program to mitigate inflammation, oxidative stress and prevent cell death for both CNS and non-CNS indications. Then we have our 2 clinical programs in NLRP3 inhibition and DHODH inhibition, which are advancing 2 differentiated selective and highly potent compounds for the indications listed here. I'll begin with our Parkinson's disease program, which is targeting 15-lipoxygenase and ferroptosis. Ferroptosis, which is a key pathway in Parkinson's disease pathology is a newly identified form of program cell death. It's characterized by oxidative stress, iron accumulation and the formation of highly reactive lipid peroxides. Together, these processes elicit oxidative damage, cell membrane disruption and activate inflammatory pathways culminating in neuronal cell death. Our ferroptosis has been implicated in CNS and numerous other chronic disease indications, making it at a particularly attractive therapeutic strategy for PTC. 15-lipoxygenase is a key regulator of ferroptosis, and we have previously demonstrated that targeting this enzyme by inhibiting it, prevents cell injury and activation of neuroinflammatory pathways. Within our compound library, we have unique chemical scaffolds that inhibit 15-LO and have been shown to simultaneously block ferroptotic cell death and activation of neuroinflammatory pathways. The important role of ferroptosis in 15-LO and Parkinson's disease is becoming quite well known. As a scientific community is increasingly recognizing the association between ferroptosis and Parkinson's disease. And over the last 10 years, interest in the specific disease association has grown from a single publication in 2015 to over 150 this year alone. Shown here is a figure from a paper published last year demonstrating the strong link between ferroptosis and several key aspects of Parkinson's disease, including oxidative damage, activation of neuroinflammatory pathways and alpha-synuclein aggregation.. So through our strategy of targeting 15-lipoxygenase in ferroptosis, we are able to simultaneously address fundamental disease processes, including [ astrocyte and microglia ] cell activation, leading to neuroinflammation; aggregation of alpha-synuclein, which is the primary component of [ LuAbodies ]; oxidative stress that results in depletion of the endogenous glutathione pool; and finally, neuronal cell death. Now when taken together, this provides a compelling mechanistic rationale for targeting 115-LO and ferroptosis as a therapeutic strategy for Parkinson's disease. In the following slides, I'll share data demonstrating the effects of 15-LO ferroptosis compounds on several of these disease processes, both in in vitro and in vivo Parkinson's disease models. Beginning with neuroinflammation. Neuroinflammation is a hallmark of Parkinson's disease occurring from the activation of pro-inflammatory astrocytes and glial cells. Now shown here are images of astrocytes that in the bottom 2 panels have been exposed to a [ propheraptotic ] challenge. Note the astrocyte in the middle panel displays the classic fried egg morphology of [ astrogliosis ], which is characterized by retraction of the neuritic processes and expansion of the cell body. Compare that to the bottom panel. In the bottom panel, the cells look unchanged compared to the control, retaining our normal morphology and not converting to a pro-inflammatory state. We've quantified the changes in cell volume in the figure here on the right where you can see that the 50% increase in volume associated with astrogliosis is completely blocked by the addition of our compound. Following this, we set out to demonstrate the targeting 15-LO in ferroptosis results in an improvement in a second fundamental disease process, protein aggregation. Phosphorylation and aggregation of alpha-synuclein leads to the formation of Lewy bodies. Again, a defining feature of Parkinson's disease. Shown here is a neuronal coal culture using cells expressing human alpha-synuclein, where the bottom 2 panels again, have been exposed to a pro-therapeutic challenge. We've added a red staining antibody that binds the phosphorylated alpha-synuclein, and you can clearly see in the vehicle panel a significant increase in phosphorylation. Again, compare that to the bottom panel that like in the previous astrocyte data looks nearly identical to the control, suggesting that this approach blocks phosphorylation of alpha-synuclein. Shown here is looking at alpha-synuclein aggregation. And we can clearly see that with the vehicle-treated cells, there was a significant threefold increase in aggregation that again, was completely blocked by the addition of our compound. Now we were the first to report last year that targeting 15-LO in ferroptosis prevents phosphorylation of synuclein, aggregation of synuclein in activation of pro-inflammatory pathways. We've also, at this point, looked at similar models of glutathione depletion and neuronal cell survival and seen a similar effect of protection. So from here, we went on to look at gold standard animal models commonly used in Parkinson's disease research. Shown here are the results of a study using the Line 61 mouse, a human alpha-synuclein expressing transgenic mouse model of Parkinson's disease. In this study, we performed a proteomic analysis to characterize the effect of alpha-synuclein phosphorylation in aggregation on structural and functional proteins across 4 critical neuronal pathways including neuronal structure, plasticity, transmission and synapse formation. Shown in the gray bars, moving from left to right, we saw dramatic reductions in concentrations of 60 proteins in these 4 critical neuronal pathways in the control of transgenic animals. The loss of protein in all 4 pathways was dose dependently prevented with the addition of our compound over the course of 4 weeks. These data clearly demonstrate that preventing neuroinflammation and alpha-synuclein aggregation confirms or confers neuronal protection in an animal model of Parkinson's disease. The final question now is does this preservation of structural and functional protein translate to an improvement in locomotive function or behavior? And the short answer is yes. This is the Parkinson's disease model where the dopaminergic neurons of the substantia nigra, the area of the brain affected in Parkinson's disease have been exposed to a propheraptotic agent, cydoxydopamine. The typical behavior of an animal when it's placed an enclosure a cage is to try to escape. As shown in the video on the left, the vehicle-treated animal makes no effort or attempt to get out of its confinement. Compare that to the treated animal on the right, which is displaying the typical behavior of trying to fit, find a way out, rearing on its time lags in displaying typical surveillance behavior. Preservation of this behavior demonstrates that the structural protection that we saw on the previous slide, confirms locomotor behavior in this particular disease model. The loss of which locomotive behavior is a defining symptom in Parkinson's disease. So to summarize the program, we've established a proof of concept across multiple disease models. We are currently scheduled to select a development candidate in the first quarter of next year with a Phase I healthy volunteer study planned for the second half of next year. Concurrently, we've initiated a translational biomarker program to facilitate and accelerate the clinical program. With that, I'll review our NRF2 activation program. NRF2 is an intrinsic transcription factor that regulates cellular response, such as stress response and inflammation and we're developing NRF2 compounds that feature a differentiated mechanism of action, which confirms, improved selectivity and modulation of both cellular stress response and inflammation response compared to other non-selective NRF2 activators. In the following slides, I'll share data demonstrating the effects of NRF2 activation in in vitro and in vivo models selected to validate and differentiate our mechanism of activation. In our initial experiments, they were designed to evaluate our NRF2 activators in a variety of cell types and then benchmark them against an existing approved NRF2 activator omaveloxolone. In these experiments, we specifically selected 4 different human cell types, representing both peripheral and neural inflammation and subjected them to a pro-inflammatory challenge. As you can see, the PTC compound modulated cytokine release, while no effect was observed with addition of omaveloxolone. Mechanistically, this is explained by RNA seat data confirming that our mode of NRF2 activation modulates the expression of genes that produce cytokines, chemokines and other molecules that regulate inflammation response while omaveloxolone does not appear to do so. We then followed up this study with a more comprehensive proteomic analysis of NRF2 activation and cytokine production in human microglia cells. Using a similar inflammation challenge is on the previous slide, we again compared our compound to omaveloxolone. Starting with the figure on the left, the PTC compound and omaveloxolone both increased the canonical NRF2 target proteins or the proteins that are directly downstream or regulated by NRF2. However, moving to the middle and right panel, we see that the PTC compound significantly increase the anti-inflammatory proteins while decreasing pro inflammatory proteins to a much greater extent than was observed with omaveloxolone. So after demonstrating these effects in vitro, we selected animal models with a well-characterized phenotype to establish a PK/PD relationship and gather POC data to further validate and differentiate our mechanism of action. Shown here is the relapsing remitting MS mouse model, in which a toxin is introduced that causes activation of T cells, a pro-inflammatory response and demyelination of the motor neurons. The animals are observed for 21 days following the introduction of the toxin and a clinical score is recorded. A clinical score of 0 indicates there's absolutely no presentation of disease. So you can see in the vehicle and the omaveloxolone treated arms that around day 12, there started to be frank presentation of disease they continue to worsen over the duration of the study. Compare that to the compound PTC compound treated arm, where there was basically no increase in the clinical score throughout the 21-day duration of the study. We then analyze data from multiple studies to determine the correlation between the clinical score and NFL or neurofilament light. Neurofilament light is a biomarker of neurodegeneration that is commonly used in clinical studies as a surrogate for clinical protection. Not surprising, the lower the NfL, the lower the clinical score and the lowest levels were observed in the PTC compound treated arms. From here, we look at a therapeutic treatment strategy, where administration of compound did not commence until we had frank presentation of disease or an increase in the clinical score on day 11 or 12. At this point, we started administering compound daily, and we saw that the worsening of disease plateaued compared to the vehicle-treated animals. Suggesting that this mode of NRF2 activation can be protective even in the face of existing disease. We next evaluated our NRF2 mode of activation in a renal injury model characterized by inflammation, oxidative stress and fibrosis. In this study, we compared the protective effects of our compound to bardoxolone, a nonselective NRF2 activator. In this model, it recapitulates numerous features of renal injury associated with chronic kidney disease, such as fibrosis, [indiscernible] fibroblast activation, formation of significant crystals leading to impaired renal function. Starting on the left, we measure blood urea nitrogen to creatinine ratio, which is a biomarker of renal injury. And you can clearly see that after 7 days of dosing, there was a significant reduction in the PTC treated animals compared to vehicle and bardoxolone, which demonstrated severe kidney injury. Moving to the right, we look at body weight or changes in body weight over the duration of the study, where a reduction in body weight is indicative of poor general health. And we can see, again, the administration of our compound was identical to the change in body weight and control and much and significantly less than what was observed in the vehicle in bardoxolone-treated animals. Clearly suggesting that our differentiated mode of activation is superior than the nonselective NRF2 activator. In the final study that I'll present, we assess the protective effects of NRF2 activation in a mouse model of primary -- or pulmonary fibrosis. This is characterized by structural damage, changes in lung morphology and activation of fibroblasts leading to fibrosis. Starting here on the left, after 21 days of dosing, the PTC compound mitigated the increase in lung weight observed in the vehicle group. This is indicative of a reduction in inflammation in the lung and edema. On the right is a figure showing the effect of compound administration on hydroxy proline, a known marker of fibrosis. And you can see that the nearly 4x increase observed in hydroxyproline in the vehicle-treated animal was completely blocked by the addition of our compound, further validating our differentiated mechanism of NRF2 activation. So to summarize, we've identified and optimized a novel class of NRF2 activators. We've established proof of concept across multiple disease models. And currently, we are scheduled to deliver a development candidate in the second half of next year. So with that, I'd like to turn it over to Mayzie Johnston to introduce our clinical programs, and thank you for your attention.

Thank you, Jeff, and good morning, everyone. My name is Mayzie Johnston, and I have been working in drug development now for the past 20 years and with several years in the inflammation space. I've had the great privilege to work at PTC for the past 4 years, and I lead our clinical stage inflammation programs. So today, I'm happy to share updates on both our NLRP3 and our DHODH programs. . So I'll begin with NLRP3. So NLRP3 as part of the innate immune system and is an intracellular sensor protein predominantly found in macrophages. The role of NLRP3 is to detect danger signals, such as infection, stress or tissue damage. So if you think about it, NLRP3 is the body's smoke detector for cellular stress. Now once these danger signals are detected, NLRP3 has been activated, triggering the assembly and activation of the NLRP3 inflamasome, leading to elevated levels of pro-inflammatory cytokines IL-1beta and IL-18. Now while this pathway is crucial to immunity, when dysregulated, it can lead to a wide range of autoimmune and inflammatory disorders, making it a very attractive therapeutic target in drug development. So NLRP3 inhibitors block the ability of the NLRP3 protein to assemble into the inflammasome. So PTC 612 is a highly potent and selective NLRP3 inhibitor with novel chemistry and favorable drug properties compared to other NLRP3 inhibitors in development. It has a unique non-cell funnel urea structure, potentially leading to greater safety and tolerability. PTC612 has demonstrated efficacy in multiple in vitro and in vivo preclinical models with evidence of NLRP3 pathway inhibition and a well-defined PK/PD relationship. So one of the first things that we did was to establish in vitro validation of activity in several different cellular models. So IL-1 beta, as I mentioned, is a potent pro-inflammatory cytokine, predominantly produced by activated macrophages, neutrophils and monocytes and PTC612 inhibits IL-1 beta in multiple cellular models after challenge with known NLRP3 agonists. So in THP-1 monocytic cells, you can see very potent inhibition of IL-1 beta with an IC50 of less than 1 nanomolar. Now typically, greater concentrations are needed to inhibit activity in whole blood due to high levels of protein and red blood cells. Yet even in whole blood, we see very potent inhibition of IL-1 beta again with an IC50 of less than 10 nanomolar. So you may ask, well, how does this compare to other NLRP3 inhibitors? So I'm glad you asked. So we benchmarked our NLRP3 activity against other in-class compounds. Here again, we looked at activity in the THP-1 monocytic cells, where you can see PTC612 is roughly 13 to 200x more potent than these other compounds. So in human whole blood, again, PTC compares favorably to the other compounds in development with potency 1.3 to 18x greater. So now that we established in vitro activity, our next step was to evaluate PTC 612 and a representative model of pulmonary inflammation and fibrosis to demonstrate proof of concept in inflammatory lung disease. Now in this model that we're going to talk about, [ intratrial bleomycin ] is administered to mice causing an acute lung injury which triggers inflammatory response marked by an increase in these pro-inflammatory cytokines, leading to a [ barent ] wound healing and fibrosis. So we ran 2 separate studies. The first is a prophylactic model where PTC612 was given 1 day prior to the administration of bleomycin. Now what we see here is a very nice dose response for the reduction in fibrosis. So then we ran a separate study to determine if PTC612 could reduce fibrosis when given after the initial insult from bleomycin has occurred. And again, we see a nice dose-dependent reduction in fibrosis with normalization to that of control. So while we're not showing it today, we also conducted several other in vivo studies, which demonstrated efficacy and proof of NLRP3 pathway inhibition and other in vivo studies similar to what we've observed here. So in summary, we have conducted studies to benchmark our NLRP3 inhibitor against other compounds in development. And we believe we have a best-in-class compound with greater potency and novel chemistry. PTC612 is a highly potent and selective inhibitor of NLRP3 demonstrating a significant reduction in proinflammatory cytokines and efficacy in, in vivo and in vitro preclinical models. So IND-enabling studies are ongoing with our first-in-human Phase I study planned for the first half of 2026. In parallel, we are finalizing indication selection in inflammatory pulmonary diseases. So now I'm going to shift our attention to our Phase II-ready DHODH program. So DHODH or [ dihydroorotate dehydrogenase ] plays a critical role in inflammation by regulating the proliferation and differentiation of activated T cells and thereby modulating pro-inflammatory cytokine production. DHODH is responsible for the production of pyrimidine nucleotides, which are essential for the survival of activated T cells. So basically, DHODH inhibitors essentially starve activated T cells by depleting the pool of available pyrimidine nucleotides. The central role of DHODH in regulating the immune response makes it an attractive target for a wide range of inflammatory and autoimmune diseases. In fact, DHODH inhibitors have demonstrated efficacy in a multitude of diseases, but they have limited widespread use due to their toxicity. And this toxicity is likely associated with the lack of specificity and associated off-target effects. Now some of these toxicities include neutropenia, GI distress, alopecia and hepatotoxicity. So PTC844 is a novel second-generation DHODH inhibitor and has greater potency and selectivity than other DHODH inhibitors, including our first-generation inhibitor. This greater potency and selectivity will potentially improve the safety and tolerability profile. So PTC844 blocks the ability of activated T cells to proliferate and differentiate, resulting in the suppression of pro-inflammatory Th1 and Th17 cells, which are implicated in many autoimmune diseases. It is important to note that there is no effect on resting differentiated cells, okay, only highly active proliferating cells. So again, similar to what we did with our NLRP3 program, the first thing we did was to benchmark PTC844 against other DHODH inhibitors for both potency and selectivity. In a cellular assay looking at proliferation, PTC844 has better potency compared to both vidofludimus and teriflunomide, which is the active metabolite of [ leflinomide ]. PTC inhibits proliferation at an IC50 of 1 nanomolar compared to roughly 5,000 and greater than 10,000 for both vidofludimus and teriflunomide. So to determine selectivity of these compounds, we did a uridine rescue. The inhibition of proliferation is truly DHODH dependent, then uridine, which is an exogenous pyrimidine should rescue the cells. And this is exactly what we see here with PTC844. At the IC50 and higher concentrations, you see that proliferation goes to 0. For vidofludimus and teriflunomide, supplementation with uridine does not completely rescue the cells. And this suggests that there's some other non DHODH mechanism contributing to the inhibition of proliferation. I just want to point out again that this potency and selectivity of PTC844 is important because we can achieve high levels of DHODH inhibition and do it in a selective manner, which ultimately should translate to better efficacy and a greater safety and tolerability profile. So we also look at the effect of PTC844 on IL-17 in PBMCs. Again, IL-17 is a pro-inflammatory cytokine produced by differentiated T cells. And as you can see across a number of concentrations, PTC844 has equivalent activity, suggesting we have achieved a critical threshold of DHODH inhibition at the lowest concentration, again, demonstrating this extremely potent inhibition of DHODH. There's a theme here, extremely potent inhibition of DHODH. So the next step was to test our potent and selective DHODH inhibitor in various reliable and reproducible inflammatory and autoimmune disease models to establish proof of concept which can then be translated into a wide range of potential clinical indications. So one example that we're sharing today is the murine MS model, which mimics the aspects of multiple sclerosis by triggering a T cell-mediated immune response, leading to inflammation, demyelination and neurological symptoms. So the higher the clinical score, the worst the symptomatology. And in 2 distinct experiments, PTC844 was comparable to [ fingolimod ], also known as [ Gilenya ], which is one of the gold standards for the treatment of MS. So we also looked at PTC844 in an antigen induced model of rheumatoid arthritis. Now this is a standard model of RA. And in this model, rabbits were sensitized to [ elbalbumib ]. [ Acersensitization ovalbumin ] was injected into the knee of these rabbits, which triggers then an inflammatory or immune response, which mimics RA. And here on the left, we are measuring the absolute change in the knee diameter. And for vehicle-treated animals, you can see a consistent increase in the size of the knee whereas with the treatment with methotrexate or PTC844, you see less swelling and stabilization over time. Now I will point out that the high dose, the 6 mg per kg dose is actually equivalent to that of methotrexate. And in fact, the lines are superimposable. Here on the right, we're looking at the percent change in diameter of the knee at the end of the study. And once again, you can see that PTC at 6 mg per kg is equivalent to methotrexate. Now this is important because the 6 mg per kg dose represents the lowest dose that we are going to move forward in our Phase II clinical trials. So this gives us great confidence in the fact that we'll be able to achieve even greater efficacy. So now in addition to the in vitro and in vivo proof-of-concept studies, we also conducted the standard GLP toxicology studies. The positive proof of concept, along with GLP tox supported advancement to the first-in-human study, which is a Phase I single and multiple ascending dose study. Single ascending doses up to 45 milligrams and multiple ascending doses up to 27 milligrams were evaluated in 64 subjects to assess safety, PK and PD. So now to determine target engagement, we measure DHO or dihydroorotate. DHO is a substrate of DHODH. So when we block DHODH, levels of DHO will increase. As you can see here, even at the lowest concentration, we are seeing increases in DHO and these increase in a dose-dependent manner, again, showing that we have demonstrated target engagement. Now based on efficacy observed in many in vivo animal studies, we are targeting DHO levels of roughly 30,000 and above for clinical efficacy. And as you can see here in our multiple ascending dose study, we're achieving these levels at our mid- and high dose. Importantly, no serious adverse events or clinically relevant findings on physical exams, ECGs or lab values were reported in the study. So in summary, we have demonstrated in vivo and in vitro POC and greater potency and selectivity with PTC844 compared to other in-class compounds. PTC844 has demonstrated safety in both GLP toxicology studies and in Phase I. Currently, we have an ongoing food effect study, and we are finalizing indication selection. We are planning to initiate a Phase IIa PK/PD study in mid-2026. So you've heard today about the strength and innovation across PTC's exciting pipeline. And Jeff and I enjoyed providing an update on our inflammation platform. But to close, we have leveraged our unique small molecule library to deliver differentiated or first-in-class therapies across 4 distinct and highly clinically relevant targets. 115-LO, NRF2 and NLRP3 and DHODH. All 4 programs are advancing with key milestones planned for 2026. So thank you for your time. And now I'll turn it back over to Matt.

Thank you very much, Jeff and Mayzie, for your presentations.

We are going to open up Q&A in the section. I know we're a little over time, but we certainly want to allow for appropriate Q&A and discussion.

This is Ben Burnett from Wells Fargo. I want to ask on the DHODH program. I thought that the data was interesting in the -- for the Phase I. What is the time course from the older assets where you saw hepatotoxicity and GI distress occur?

The older DHODH, yes. So that was, I think, ancient history for a lot of us here, but I realize that's still out there. And I think a lot of the earlier things were -- those side effects were observed at very, very high doses in early oncology studies were, again, far beyond, I think, anything over time that was looked at in subsequent clinical work. I think one of the things that Mayzie highlighted that's critically important is with this new second generation of molecules like PTC844, we have so much greater potency and specificity for the DHODH target that it's allowing us to get significant levels of DHODH inhibition doing so with on-target specificity and safety and tolerability. So I would just briefly summarize and say that this is far and away magnitudes better, as you'd expect from a 2.0 versus an early 1.0.

Kristen Kluska at Cantor. What degree of effect does 15-LO 5LO have on modifying [ LRRK2 ] relative to other targets, similar stage development in these preclinical studies? And then what other synergistic effects does this target offer that others in Parkinson's lack?

Okay. So what makes this particularly attractive compared to other approaches is that we're operating upstream of some of the fundamental disease processes that have been reported like synuclein aggregation or neuroinflammation. So we're capturing these fundamental disease processes right at the start. So I think that offers a particularly meaningful advantage of hitting the fundamental pathway that is upstream of neuroinflammation, protein aggregation, glutathione depletion, [ evocative ] stress and cell that. So that makes us a particularly attractive therapeutic strategy for us. And your first question was, how does it relate to LRKK2?

Yes. Just the degree of effect that it has on LRKK2 specifically relative to other targets that are looking at this in a different way.

I would say the magnitude of effect on the other aspects of the disease pathophysiology are much stronger compared to what it would be doing to LRKK2 for example.

I think Jeff emphasized some of the work on LRKK2 in Parkin and [ PINK 1 ], which are targeting specific typically mitochondrial mutations can be important, particularly if you have LRKK2 specific pathology. But when you consider what we're able to do with by targeting keratosis through 15-LO is the fact that you have these distinct pathways of pathophysiology that each independently have been implicated in Parkinson's disease pathogenesis and the ability through one single target access each of them, including mitochondrial dysfunction is quite frankly, exciting.

Paul Choi with Goldman Sachs. I think my first question on the NLRP3 program for Mayzie. Can you maybe comment a little bit on for the pulmonary fibrosis animal work you've done? Have you done any testing with the standard of care agents like [ EzratOfab ]? Can you maybe comment on combinability with the current standard of care? And then my second question for Matt is sort of beyond IPF as you think about developmental areas there are a couple of other candidates in the clinic targeting recurrent pericarditis, other things like that. Just sort of curious your appetite for exploring beyond fibrotic diseases, let's say, cardiovascular, other areas?

Sorry. Yes. So specifically, yes, the pulmonary fibrosis model that I shared with you, both the prophylactic and the therapeutic, we actually did benchmark that to [ natetinib ], and we had equivalent results. So it just wasn't shown on the slide. Now we have not looked at the combined ability yet, but that obviously is something that we would consider if we were to go into IPF, we would actually look at on background of standard of care therapy.

Yes. Just for the comment that one of the things we were doing in -- we were always -- when we were bringing this program forward, given the link between NLRP3 and [ flobosome ] and pulmonary pathology, we thought that, that would be a very interesting place to look, of course, with an idea to what more rare disease indications and high unmet needs. So a lot of the early in vivo work was POC work in well-characterized pulmonary fibrosis models to establish the PK/PD relationship and get the necessary proof of concept necessary to give us confidence to bring this forward. So we're still finalizing what would be the first indication, but certainly we're aware that the NLRP3 inflammasome -- first of all, we know a lot of people are looking at it and we implicating a number of different organ systems and we'll get a first indication in mind and then start to think about what could a second and third indication look like, again, using the usual filters we talked about, developability, competitive landscape and things like that. I think again, one thing we're particularly excited about with this compound is at least the benchmarking work we've done to date shows superior potency in part to the chemical structure as Mayzie alluded to, which allows us to move into a space where there may be others find a niche that we think is well suited to PTC's wheelhouse and be able to make an important contribution to the field.

Jo Thome from TD Cowen. Just a quick question on the targeting 15-LO because I know [ peticanonin etlixisad ] also targeted that same mechanism. So I guess how differentiated are these compounds from those? Maybe what are the differences? And did you see similar impacts preclinically if you looked at [ vatiquinone ] and some of those others?

I'll just make a general comment and then I'll let Jeff go a little bit deeper. I mean I think the ticket on was really 1.0 point out. And the teams have spent a number of years if you think about the time from when [ vatiquinone ] came forward, the target was elucidated to being able to, just as you'd expect over time, have more specific, more potent by distributed molecules that can give us confidence that we can have a significant effect and be the first [ antitheraptotic ] therapy for Parkinson's disease. Jeff, I don't know if you want to comment any more on some of the...

I would pile on by saying we're down to single-digit animal potency now, which is a significant advancement.

Judah Frommer from Morgan Stanley. Maybe just one high-level one on the PD program. So I think it's fair to say this feels like a more targeted R&D approach for PTC than what we've heard in years past. So can you talk a bit about what's changed within the broader organization to support this focus, how financial resources and cash flow breakeven guidance could be impacted as hopefully multiple programs move forward. And then just a follow-up on PD. Would you characterize learnings from 857 kind of similar to how do you talk about [indiscernible].

So on the first question, look, I think we -- a couple of years ago, we undertook a significant restructuring and really refocused and fundamentally change the way we run the company. One important aspect of that was focused in execution. I think what you've seen today, certainly with the splicing team and we're incredibly proud to have such dedicated and accomplished scientists who've been at PTC sometimes over 2 decades is in the case of Chris, and who's been here for 18 years and have seasoned experience, researchers and developers like Mayzie and Jeff. And a lot of it was saying, we've got this incredible talent have incredible potential. I mean I think if you think about the splicing platform, this is an incredibly innovative and valuable RNA technology platform that, quite frankly, has been under cultivated. Just on self-reflection and reflection of the organization. You have incredible talent, incredible potential, and this is something that we felt needed to be focused, dedicated and get our best minds on this, and that's what we're doing. And when we talk about bringing the same rigor and focus to the research organization that we brought across the company, we now have, just like we do in the -- for the business, we have quarterly business reviews, where we'll sit down with the commercial teams, and they'll explain to us, how did the quarter go. What are goals for next quarter? And how do we get there? We're doing the same thing in research now. We have quarterly research reviews where the teams sit down and say, okay, what were the goals for this quarter? How we met them? Well, I haven't we met that more of the goals for next quarter? And I think it's really combining what is immense talent, immense experience, right amount of discipline and focus, and I think this is what you get, and we're incredibly excited about it. In terms of financial resources, look, we've been made very clear once we started on this journey a couple of years ago that we're going to move the company towards cash flow breakeven. I think one of the most incredible things to think about is we're doing all this with this significantly. I mean a fraction of the research budget the company previously had. So again, this is, I think, what brings discipline focus can bring you and also being very thoughtful about how we allocate expenses for research and early development. We remain committed, as we've said, to continue to reduce OpEx. We expect when we give OpEx guidance at JPMorgan in January you expect to hear that it's going to be -- there'll be a reduction going from 1 year to the other. And we'll also talk about our journey towards cash flow breakeven, which we remain incredibly confident. And even with this -- even if all of these research programs were successful which I think would be tremendous, but not -- I mean not everything will succeed, but we're well positioned. And the other thing just to emphasize as well, certainly, when you hear about a program like Parkinson's disease. So we talk about a cancer target in splicing. We've said all along, we've got science that we believe can be incredibly important for the development of therapies, both that PTC can handle the development and commercialization for and others that would be appropriate for strategic partnerships. We want sort of an intel and side model where we'd like to put our science inside a larger indications through strategic partnerships where we can still contribute in advancing highly innovative and impactful therapies for diseases of high unmet need even if they're not again in our development and commercial wheelhouse. And then your other question was Gen 1, Gen 2, particularly on 857, again, I think similar to what we talked about a lot of important lessons they are in terms of being able to evolve and have much higher potent and specific therapies.

This is Brian Cheng from JPMorgan. Can you confirm whether your NLRP3 program is a covalent or a noncovalent binder? Given the spectrum of covalent binding and non-covalent binding for NLRP3 in the space, what's your take on both binding approaches? And then lastly, how does the ability to penetrate a blood bring barrier differ from others?

So let me answer your second question first. So PTC612 is peripherally acting, it is not CNS penetrate. We do have a library of compounds where we do actually have CNS penetrant and LRP 3 inhibitors, but this one specifically is being developed for peripheral indication. For your first question, I believe it's a noncovalent NLRP3 binding. Yes. yes, the chemist back there is nodding yes. I thought I had that right. Yes. Thank you, Matt.

Any other questions? Terrific. Thank you, both. And I want to thank everyone again for attending our research event today. We've been incredibly excited to share the tremendous progress we've made across both our splicing platform as well as our inflammation ferroptosis platform. We look forward to providing updates as we look to move these compounds forward and to really incredibly excited about the company we've been able to build over the past couple of years and really position ourselves not only for near-term success, but long-term success. Thank you all again for joining. Those who are in the room who have informal discussion and lunch available in those online, thank you again for calling in and joining us today.