Welcome, everyone, to the day two 2023 Jefferies Healthcare Conference. My name is Roger Song, one of the Senior Biotech Analysts in the US. Our next presenting company is Omega Therapeutics, CEO, Mahesh Karande. He will do a few slides, and then he will join me for a fireside chat. Welcome. Morning. I'm Mahesh Karande, CEO and President of Omega Therapeutics. Roger and Jefferies, thank you very much for having us here. Would love to walk you through a few slides at the beginning to sort of tell you the Omega story, and then I think we are going to get into a Q&A. Omega Therapeutics. This is my mandatory disclaimer for forward-looking statement. Omega Therapeutics is really pioneering the next generation of medicine. If you think about drug development over the last 100 years, right, it has occurred in the reverse order of the central dogma of biology. As we know, small molecules and large molecules pretty much dominated most of our lives, right? In the past decade and a half, people have moved up the central dogma and took aim at mRNA, when non-coding RNA companies came about, and several of them had, you know, good success in specific therapeutic areas and indications. If you think about the Modernas of the world or the Alnylam of the world, tremendous scientific progress, right? More recently, people took aim directly at DNA with gene therapy and editing, and I think that also has worked well. Science has advanced tremendously, but again, in narrow ways, you know, in areas where potentially there is no option, gene therapy works. Gene editing, you know, has a long-term liability along with gene therapy. If you really think about, you know, the central dogma, step back. What Omega is doing is looking at what actually controls the central dogma of biology, what controls transcription, what controls expression of genes and the protein that is formed, and that is really epigenomics or epigenetics. What we are doing is we have taken aim at the overall epigenomic control system of the central dogma and have been working on programmable epigenomic mRNA medicines. I'll attempt to walk you through some of the science there, right? One of the things that we asked ourselves when the company was founded is, you know, what if epigenetics worked through a central control system? Because if you think of mother nature, nature never does anything in ones or twos. There's always a systematic way of approaching things, right? If there was such a system, what if we could drug that system, control it, and thereby control gene expression? That would be extremely powerful, and that could lead to, you know, transformative medicines. As it turns out, nature has organized genes and their regulatory elements in these conserved three-dimensional loops of chromatin, called insulated genomic domains. Insulated genomic domains by themselves act as fundamental regulators for the genes that sit in it, right? There can be single or multiple genes in an IGD. There are about 15,000 IGDs that are distributed across the 23 chromosomes and are ubiquitous in every cell of the body. All the cells, all of us have the same IGDs, right? These are conserved pretty much across species, right? And within species, you know, they are obviously homogenous. In lower order species, you might have a little bit of different sequences to these, you know, DNA elements that sit within these IGDs. Between non-human primates and humans, you know, mother nature has created a system that is pretty much intact, and that's tremendous, right? That's evolution. The beauty of these IGDs is, at the base of these IGDs, of this three-dimensional structure, and think about it as a, you know, as a hot air balloon structure, sort of a simple illustration, as we have shown here, but it can be florets, it can be infinity loops. What they are characterized by is at the base of this structure, you have these CTCF proteins that come together, seen there in red, at the base of that structure. They bind, and they create an insulation such that transcriptional activity of genes or transcriptional control of genes that sit inside the IGD is fully contained within the IGD. What that means is only the regulators that sit inside the IGD control expression of those genes. The second, and another very important fact, is that all of these regulators have sequences. Think of them as genomic sequences that are unique. Now you have 15,000 IGDs, right? Each having tens or sometimes even hundreds of genomic sequences that are unique. Mother nature uses that to lay epigenetic marks and thereby control gene expression. What we have done is really taken that into account and created, you know, our Omega platform to drug these IGDs, you know, with our genomic medicine, and I, you know, I'll explain that a little bit. First and foremost, right, these epigenomic regulatory locations that I talked about, think of them as epigenomic zip codes. We call them EpiZips, and these are proprietary targets that Omega has created. You know, just think of them as hundreds of thousands of targets that nobody has, that we can precisely go to and control gene expression in that IGD, right? How do we do this? How do we actually then, you know, create that, the medicine? Our medicine is composed of two things. First of all, this is the first systematic use of mRNA therapeutic. You know, we call them Omega Epigenomic Controllers, yeah, because they control the epigenome, right? mRNA expresses two proteins inside the nucleus, right? One is the DNA-binding domain. That is our own proprietary DNA-binding domain, that, you know, focuses on 21 nucleotides and goes and attaches to a very specific regulatory element of our choosing, right? That we'll decide genomic, you know, through our computational genomic work up front. At that point, it expresses a second protein, which creates an epigenetic change, and that's an epigenomic effector, right? The effector is borrowed from a vast repertoire, you know, our library of epigenomic controllers that we have created. Think of them as up or down, right, control for different duration. The most important thing here is that, you know, fundamentally, if you think about disease occurs because genes are either overexpressed or underexpressed. At the IGD level, pre-transcriptionally, if you are able to control gene expression and bring it back to a normal level, you will resolve disease. The way we do this is through these Omega Epigenomic Controllers, whether, you know, where the controller controls gene expression, brings it back to a normal level. That's the modulation, up or down. It's not an on or off switch, but it's a modulation, bringing it back to the right normal level of expression for the duration that we choose. The duration is really decided based on disease. For example, in oncology, you know, for our first program in the clinic, we are using, you know, a two week duration because that's what oncologists are used to, right? In terms of actual practice. We could use a six month duration, but potentially that would disrupt, you know, clinical practice, and that's not what we are trying to do. For a chronic disease, we can easily go for six months or nine months, and we have data that we have demonstrated that these work for six months or nine months, depending on the epigenomic, you know, actual epigenomic effector that we use, the epigenomic change that we are making. Whether it's, you know, DNA methylation or demethylation, or acetylation, or histone modification, we get to choose that, and we get to multiplex to different locations within the IGD. As opposed to going only to a promoter and making a change there, we have the entire regulatory machinery that controls the single or multiple genes that sit in the IGD at our disposal, and that's what we do with these epiZeps, right? When I put this together, our platform really is, you know, biology. We figured out the biology, and then we use the best technology to tackle the biology. That's what Omega platform is. At the base of this is world-class computational genomics and data sciences, which is extremely important. We have a tremendous team, you know, well-versed in AI, machine learning, you know, large data science. The reason this is important is, you know, on this slide, right? If you think of drug development, particularly the discovery process, right? You know, obviously, in the age of new medicines and genomic medicines, et cetera, you know, we have turned sort of the high-throughput, small molecule process, which takes years and, you know, hundreds of thousands of constructs before something hits, completely on its head. These are designed, engineered, programmable, and we incorporate properties in them that we want for that disease. Think about it for a second, right? What we do is we start off with a biologically known target. Take our c-MYC program, right, which we all know has been undruggable for 40 years, right? The biology of c-MYC is very well known. Think of c-MYC. When we started this program, we looked at the IGD where the c-MYC gene sits. We understood the IGD. We figured out all of the regulatory elements that control c-MYC expression, you know, in cancer cells, in hepatocellular carcinoma, for example, which is our first program, right? We figured out which of those regulators have what level of control on c-MYC. We decided that these are the two or three regulators we want to act at, and we created constructs, again, computationally, of our final design for our controllers, right? That took us about three to four weeks. We started designing these and actually building these, right? To get five to 10 constructs, took us another three to six weeks. At the end of two months, two and a half months, we have five to 10 putative controllers that could be your final drug, already ready to be tested. We started in vitro testing and in vivo testing, right? At the end of about sic to nine months, we were able to figure out which of those is the best and the most optimized to control c-MYC to the level we want, right, from a homogeneity standpoint across hepatocellular carcinoma. We declared that as a development candidate, right? Think about it. This is an already optimized development candidate that we declared, right? That was, I would say, about a year after we started working, nine months after we started working on it. Then we began IND-enabling studies, right? We got our IND cleared. Let me step back. For our c-MYC program, which is in the clinic, from scratch to getting the IND cleared, was 27 months. Right? Now, you know, you can ask me, "Well, what about the other programs in your pipeline? Why aren't they at that same stage?" Well, first of all, you know, I think they probably are. We are a small company. You know, if we want, we can probably go after 50 targets and 50 programs today. We just don't have the resources and the money for that. That's a whole different discussion. At the same time, the other important thing here is delivery as well. In this case, we had the, you know, the liver delivery available that allowed us. Our second program that is leading is in lung. We are at the forefront of lung delivery, and that's c-MYC in non-small cell lung cancer. I'll show you the pipeline in a minute. Before we get to the pipeline, right? Look, our approach has very, very differentiated and distinct advantages. If you think from the platform standpoint, we are leveraging all epigenetic mechanisms, right? Which allow us to control gene expression, and tune the gene up or down to the level that we want to bring it back to a normal range of expression, right? Broad applicability to nearly all diseases, because if you think about disease occurs because genes are either overexpressed or underexpressed. That means we should be able to go after all genes. They all sit in IGDs. It's pretty, you know, inductive thinking. We have actually demonstrated that, right? Then I just talked about rapid prosecution and, you know, scalability of the platform, right? Now, from a controller standpoint, we don't edit. We don't use tools that could nick DNA. We don't change nucleic acid sequences. You know, our IND was cleared in one month for a completely new modality. The FDA was very comfortable that they cleared our IND in one month. By the way, that's not just the safety piece, that's also CMC, right? We have a tremendous team that actually figured the CMC out, because that's the bane of biotech existence, typically, right? One of the key things of our drugs is also a separation of PK and PD. We drug this once. For c-MYC, we are drugging it once, right? Every two weeks. If it's a chronic therapy, we'll drug it once. We have data that shows that our controller lasts for six months, the effect of the controller. This is important, you know. The LNP that is delivered in the mRNA and the two proteins that it expresses, all degrade within a matter of couple of days, maximum four days. What stays behind is the epigenetic change we made and the effect. Think about combination from an oncology standpoint. The only toxicity is LNP toxicity, which is pretty well known. The drug is gone. The effect stays. We can freely combine, you know, with a bunch of different therapeutics in oncology or otherwise, right? Huge, huge advantage of how we are doing this. Pre-transcriptional control allows us, particularly in oncology, to overcome things like autoregulation, which is, by the way, why MYC drugs have failed. Well, small molecules have failed because there's no binding protein. You know, non-coding RNAs have probably failed because they cannot control c-MYC because it autoregulates and you run out of a therapeutic index, a therapeutic window. You can't put enough drug in the system. We don't have that problem, right? Oh, sorry, I went back. You know, I won't belabor this slide, but I made the point, we have looked at every possible disease process, Omega can work in every possible disease process. That's actually important, because what that allows us to do is super scalability for this company, which is outlined in our pipeline. If you think about our pipeline, we set this up early, four years ago, right? It's in oncology. You know, our leading program is in the clinic. We're super excited. We'll be getting data later this year. We'll be talking about data. You know, our second program, which is contemporaneous with 3 other programs, is in c-MYC for non-small cell lung cancer. That's a lung-targeting LNP that is different from the liver-targeting LNP. That's lung tropic. You know, our team has been working on lipid nanoparticles internally. You know, it's a phenomenal scientific team that is at the cutting edge of LNP, but we also partner outside. We have other programs in regenerative medicine, in, you know, monogenic disease and multigenic disease. 2 things. Think about regenerative medicine as almost as opposite of oncology. In oncology, you have uncontrolled cell growth, which we are controlling epigenetically. In regenerative medicine, think of master regulators that you can actually control cellular programming with, that you can control cell growth. Our HNF4α program does exactly that. HNF4α is a master regulator of hepatocyte function. We are able to regenerate hepatocytes, and we are able to regenerate hepatocyte function, and our data shows that we have been able to do that in animal models, right? That's incredible. Multigenic disease, I'll make one point. You know, nature is very, very smart. The way it has filed genes, if there are multiple genes in an IGD, typically there are anywhere from one to 10 genes in an IGD. Some of them are monogenic, like the MYC one. Our program in CXCL1, CXCL2, CXCL3, and IL-8, which are chemokines, they all sit in one loop, and they all work in tandem. We can control all four with a single therapeutic, as opposed to a monoclonal antibody, which can control a protein only of one at a time. Where you would need four mAbs, we can do it with one therapeutic. That's a huge advantage, right? That's really our pipeline, and that's it. Thank you very much. I'm sure, you know, Roger has a bunch of questions for me. Thank you. You want to just. Join me there. Yeah. Awesome. Yeah, thanks for the presentations. Absolutely, very exciting technology, broad application. Maybe we can maybe drill down a little bit of your pipeline. First of all, You have a so much, kind of a option, kind out there for your platform. Why you choose MYC, liver cancer as your lead program, and what are the proof concept data already generated pre-clinically? Of course, you are not just a pre-clinical company, you are a clinical company, so you are doing the phase I, and what should we expect from the phase I? Maybe start from the. Yeah. Yeah. That's a great question, Roger. Look, I mean, I think, you know, as I said, c-MYC has remained undruggable for years, right. Everybody knows, everybody who studies oncology knows that c-MYC is considered to be the holy grail in gene in oncology. It's implicated in over, I would say, over 50% of solid tumors and pretty much 100% of metastatic cancer, right. It has remained undruggable for two major reasons. Three reasons. One is that its protein lacks a binding pocket, so it's very difficult to directly drug it with a small molecule or a large molecule, right. Secondly, you know, the half-life of its mRNA and protein is so short that, you know, it's again, at the mRNA level to control is, it is difficult. The other piece of it is that, you know, it is one of the classic oncogenes which autoregulates. It's a perfect example. If you tamp down its protein or its mRNA 100%. Think of it, right? In today's world, we think of on or off, right, because of editing. Small molecules and, you know, non-coding RNAs also typically work in with that principle, that you have a completely aggregate expression. You completely aggregate expression of c-MYC, what happens? Is it autoregulates and cranks out more, and more, and more. At one point, you can't put enough drug in the system to control it, right? That's why it has remained undruggable. We thought about it. Look, you know, when we set this up, we obviously had a bunch of different, you know, targets we could go after. We were studying 50 targets. We whittled it down to 20. Then we decided on this so as to study the breadth and depth of our platform. Coming to c-MYC, we realized that we could control c-MYC pre-transcriptionally, bring it back to a level of expression. You know, to segue into the data, right? Look, what happens with c-MYC in cancer is cancer cells get overly addicted to c-MYC. You know, the expression is turned on high, right? Normal cells require a certain basal level of c-MYC for normal operation and c -MYC is one of the more fundamental genes which controls many pathways in its, you know, in downstream, right? You don't want to fully turn it off, because what that will do is normal cells will starve and then the gene will autoregulate. You know, all our work, as well as work that Gerard Evan and others have done, have shown that all you need to do is bring the overexpression down by upwards of 50%, not shut it down to 100% or zero, right? All you need to do is bring it down to 50%. That's what we are doing. That induces apoptosis, right? We have demonstrated that in our preclinical data over and over again in different models. The other thing which actually, you know, this does is, t his is a hypothesis that we have proven out preclinically. If you look at our ASCO poster, you know, our CSO, Tom, presented a really nice poster, which essentially did another thing beyond just intrinsic cell death by apoptosis, right? One of the theories really is a hypothesis that we've had, is that if you down-regulate c-MYC for a certain period of time, you reset cellular programs. What that does is, you know, potentially those cancer cells forget that they're cancer cells and they don't regenerate. Actually, look at our poster, we have demonstrated that. We actually demonstrate, we, you know, these were, you know, mice with intact immune systems, which represents, you know, sort of what we would see in the clinic with humans, right? Immunocompetent mice. We treated them, and then for a certain period, I think it was two weeks, we didn't do anything with them. After that period, we re-implanted hepatocellular carcinoma tumors in them. Those tumors didn't take. Would you call it curative? I don't know. We'll prove it, but the data points towards something pretty cool, right? We also implanted them with non-HCC MYC-driven tumors, NSCLC. Those group, right. What we have done is because we down-regulated MYC in HCC cells, right, those cells created a memory that they were tumor-free. That's incredible. This is the kind of data that we have generated preclinically. Segueing into the clinical trial, right? Our clinical trial is running. Everything that we have demonstrated preclinically from, you know, target engagement to the making the epigenetic change, to measuring the mRNA change, to measuring the protein change, and eventually looking at the correlation between that and tumor killing, and this newer stuff that I just talked about, all of that is what we are planning to demonstrate in our clinical trial. We are super excited about this clinical trial. That's fascinating. This anti-tumor memory, that's very cool. Maybe that's back to this, the whole notion, like, you know, epigenetic, it can be inherited kind of moving forward once you correct this in the upstream, this kind of way they translate the protein. Maybe downstream, they can actually maintain this anti-tumor. Absolutely. Yeah. Look, I think it comes down to choosing the right mechanism. Think about it. Mother Nature, right, turns genes on or off for periods of time, right? That's how we are all functioning. At the same time, genes are turned on or off constitutively for long periods of time, right? What we have done is completely co-opted that system in a very intelligent way. We actually understand the system. This science was only delineated in 2016 by Rick Young and Noubar and David Berry, who are founders of this company, and this company was created. We have taken that, I think, you know, we've supercharged that science to the point that, you know, we understand this really, really well, but at the same time, I believe we are still scratching the surface. Think about it. Nature can turn genes on or off, or express them at different level for long periods of time. We have demonstrated control through, you know, six months, more than that, with a single therapeutic, right? If you ask me, "Can you show two years?" Well, I say, "Well, let's wait for two years." Right? We have data for six months. We'll have data for more time, right? Can we do that even longer? Absolutely." We're just scratching the surface. Awesome. Okay, maybe just last minute. What should we expect in the second half, this data readouts, you know, as investors, and, you know, what's the expectation we should set for the initial data from that? Look, I think from an initial, this is obviously a phase I study. The most important thing that we are studying is safety, right? Because this is the first time anybody is studying this. That's one. Secondly, we have a really solid translational plan for mechanistic data, right? Target engagement, epigenetic change, mRNA change. Think about PK and PD. Those are definitely things that we would demonstrate. You know, obviously, the trial is running as a monotherapy with a dose escalation monotherapy. At the same time, you know, we have preclinical data we've combined with checkpoint inhibitors. We have combined with, you know, TKIs. At some point, we'll do combination work. In 2023, right, the data will span across sort of monotherapy, that I just described, and, you know, we will be presenting at, you know, some conference later in the year. Awesome. Great. Thank you. Thank you, Mahesh. Yeah. Thank you very much. Yeah. Thank you, everyone.