Good morning, everyone. I'm Farzin Haque, one of the biotech analysts at Jefferies. It's my pleasure to introduce Geoff McDonough, CEO of Generation Bio. He's going to talk about an exciting way to deliver siRNA to T cells for the development of that approach for the treatment of autoimmune diseases. Welcome, Jeff. Farzin, thank you. Good morning, everybody. It's really a pleasure to be here. Really appreciate the invitation to speak about what Generation Bio is up to, really looking to change what's possible for patients living with T cell driven autoimmune diseases. As a reminder, Generation Bio is a public company, so I'll be making some forward-looking statements today. Not everything that we plan to do will come to pass. We are creating, as Farzin said, a first-in-class opportunity to bring targeted siRNA therapies to address T cell driven autoimmune disease. This combines a uniquely selective reducible delivery system for T cells with the ability of siRNA to knock down undruggable T cell targets. We're building this approach to the clinic in order to bring our first IND forward in the back half of 2026. This idea of bringing a validated payload to new tissues is not new. Alnylam began siRNA development as a therapeutic platform and has done so brilliantly, moving from the liver to the CNS and to the eye. The ability to change the targeting modality in conjugates and bring siRNA to new tissues was expanded by Arrowhead, adding antibodies against TfR1 to access skeletal muscle. What we're bringing forward is the ability to selectively get siRNA into T cells. I'll come back to why that's been such a limitation for the field. Suffice to say, it's never been possible before now to get siRNA to work in T cells with conjugate delivery. It's a new field for siRNA as it relates to T cells and autoimmunity. Why are we interested in T cells? They sit at the nexus between innate and adaptive immunity, and they are the driver in both the initiation and maintenance phase of several autoimmune inflammatory diseases. They represent a really powerful target across a spectrum of indications where there remains significant unmet need. We are focused on high-value indications where our approach could uniquely improve on the standard of care. Obviously, we are interested in indications where the predominant pathology is driven by T cell activity. There are about a dozen of these indications, 6 of which I have used here as illustrative examples. Interestingly, what they have in common is a therapeutic ceiling where usually about a third of patients gain benefit from current standard of care, or they have safety limitations related to the off-target activity of non-selective modalities. There are several gaps also in patient experience. If you look across the range of these example diseases, alopecia and vitiligo in the skin, they're both significantly debilitating diseases for which there are a very limited set of therapeutics today that have very significant off-target side effects. They have a very low degree of penetration, so represent a significant opportunity with almost 3 million patients between them. Ulcerative colitis and Crohn's disease have a very similar story, obviously a broader array of existing therapies, but where also a therapeutic ceiling exists, about a third of patients achieve an initial response with existing therapy. Lots of room to do a better job in that disease. For primary biliary cholangitis, there is no disease-modifying therapy. It's obviously a disease where interface inflammation driven by T cells is a key driver. We really like this array of disease opportunities for a selective approach to T cells. Finally, for type 1 diabetes, the opportunity to arrest the T cell driven destruction of islet cells early in disease and preserve islet function and insulin independence potentially for a significant period of time is obviously a huge unmet need and I think a really interesting opportunity. How do we think about the ability to intervene on T cells? T cells are an incredibly powerful driver of our natural immune responses. That power comes from their ability to be activated and to proliferate substantially. As T cells proliferate, they differentiate into effector subtypes, CTLs or CD8 cell types. Each of these subtypes, TH1, TH2, TH17, are CD4 effector cells that are all very potent drivers and maintainers of our immune response and also of our autoimmune and autoinflammatory responses. Tregs is the subset that's there to bring natural activation of T cells under control and to draw those processes to a close. We have very good medicines that get at the activation and proliferation of T cells. They work to knock down or to deplete the activity of a broad set of T cells. These are drugs like tacrolimus or high-dose glucocorticoids or the JAK inhibitors. The downside of these classes of medicines is not only that they're non-selective, but they drive very significant off-target and long-lasting off-target side effects. We have more specific downstream drugs that work within each of these effector pathways. They work well, and they tend to drive this kind of therapeutic ceiling because there is significant networking and crosstalk across these different pathways. Knocking one of them down in disease often leads to the ongoing activity and potency of remaining effector pathways. What we're interested in doing here is reaching T cells selectively without touching other cells of the immune system or other parenchymal cell types, really getting rid of that off-target side effect profile and delivering siRNA to reach intracellular targets in T cells. Obviously, antibodies can be selective for T cells, but they can't hit intracellular targets. Small molecules and degraders can hit T cells, but not selectively. They hit several other cell types. This is a unique niche where we can have T cell selectivity and access to intracellular targets with the precision and pharmacology offered by siRNA. How do we do this? The core innovation of Generation Bio has to do with our cell-targeted lipid nanoparticle delivery system, which uniquely dials out non-specific delivery to the liver and spleen and dials in cell-selective delivery with the use of a biological ligand, which we conjugate to the surface of the LNP. We recently gave a scientific presentation, which is available on our website, that explains how we've achieved this. It is related to unique components in the LNP that drive these core characteristics. Namely, the selectivity is achieved with the use of a novel anchored polymer, which prevents the non-selective delivery of our LNP or uptake of our LNP by the liver and the spleen. That same anchored polymer also allows for our LNP system to be reducible without the accelerated antibody-driven clearance that's typical for the use of anchored polymers like PEG. This is a major breakthrough in the ability to both achieve selective avoidance of the liver and spleen and maintain reducibility. Finally, we're working with a proprietary ionizable lipid that, in combination with our polymer, maintains the potency of the endosomal escape of this system, which is a core driver of therapeutic translation. Just to exemplify some of these characteristics in data, this is all primate data, really showing the efficiency, selectivity, and specificity of our system. In the left panel, you can see that in primates at six hours, we're delivering less than 1% in total to the liver and the spleen combined. Traditional LNPs have about 99% of every dose taken up non-specifically by the liver and the spleen. This offers an enormous tailwind in terms of potency, but also in terms of reducing the variability and delivery that's characteristic of non-selective LNPs. The middle panel shows how selective the system is. When our LNP does not have a ligand, we get zero uptake in T cells. All of our uptake is ligand-dependent. Finally, in terms of specificity, you can see here we are using monocytes as an example cell type, but this is true across all immune cell types. We get no ligand-driven uptake. We are only reaching T cells, which, of course, is our target. We also have highly selective and efficient delivery. You can see here that we are reaching between 95%-100% of target T cells with a single dose in humanized mice across both CD8 and CD4 subtypes, as I was mentioning before. We are very efficient at getting to T cells even when they are the tiny minority of cell types in a tissue. What you're seeing in the panels on the right, in the top row, this is a scrambled siRNA. You're seeing a mixed background of cells in the mouse liver, most of which are hepatocytes. In the middle panel, you see that some of those cell types are T cells. In the right panel, you can see that in the top row, all of those human T cells in the background of the liver are expressing our tool protein, beta 2 microglobulin. In the lower panel, you can see even with this very small and sparse presence of T cells, we can identify those T cells and knock down the human B2M only in those T cells. This is a very efficient system and a very selective system for getting at T cells even when they're tissue resident, which is really important because many of these effector T cell subtypes are most powerful in driving and maintaining disease in the tissues. I mentioned earlier that conjugates have not been effective in reaching T cells historically. There's a variety of reasons for that, but it probably is an integration of receptor efficiency and the leakiness of the endosome for conjugated siRNA. It's just not sufficient to get high enough levels to load the risk complex in T cells. Our system, because it's LNP in origin, is designed to open the endosome and to discourage the endosomal contents. The idea of a selective system just seemed very well suited to this problem for siRNA delivery to T cells. Indeed, that's what our early work has shown. Again, using an siRNA against beta 2 microglobulin, a highly expressed housekeeping protein in T cells, you can see in the left panel that we get very potent nanomolar level knockdown in vitro of beta 2 microglobulin. In vivo, in the middle panel, at very low doses, 0.1 mg per kilo, we're getting essentially 100% knockdown across a range of relatively traditional chemistries for siRNA. It's durable. In a model system where you're getting activation and proliferation of T cells, we're getting up to two to three weeks of knockdown, which in comparable preclinical models translates to every month or every other month dosing in humans. We're very excited to start to translate some of this data to actual autoimmune targets. We're showing here in the left panel that we're able to achieve knockdown in both naive, resting, and activated T cells. It's really important because traditionally it's been much easier to reach activated T cells and not so in resting or naive T cells. For this particular target, where we would expect to prevent cytokine release, we do so. There's a functional consequence of knockdown in these cells. In a two-week view of T cell expansion, we're able to fully abrogate the ability of T cells to expand without depleting them. That's really important because the ability to therapeutically reverse your intervention over time is useful. If the T cells are present and functional, they can recover function quickly. Some nice early progress here in translating this system to therapeutically relevant targets. This ability to selectively reach T cells is just the beginning of what our cell-targeted lipid nanoparticle system can do. We have shown that we can deliver selectively to liver cells and T cells, as I've just shown you, but also to HSCs. I think this ability to translate the core characteristics of the system across multiple cell types is something that over time can become quite interesting for additional tissues. This combination of selective delivery and SI, I think, opens the ability to go very precisely after the role that T cells play in autoimmune disease. We are on track to announce our lead targets and a pool of indications in August this year, which will then allow us to unfurl a series of milestones on the back half of that on the journey towards our first IND in the back half of 2027. Farzin, that's all I have prepared. I know that you have at least a couple of questions. Obviously, welcome those online and in the room as well. Thank you. Thank you, Jeff. There are no other examples of targets being pursued by other companies, as you mentioned, in the T cell autoimmune space. What has been the bottleneck, and are there any technical challenges in you mentioned some of the delivery aspects, but are there any other reasons that can potentially not be pursued? I think delivery has been the overwhelming one as it relates to the ability to target T cells selectively. One of the reasons that selective delivery with antibodies, for example, has been a limitation in T cells is that all of the targets that are available for antibodies obviously are extracellular. It has been difficult to target discrete surface proteins on T cells without causing activation of the targeted T cells. Often the cell surface proteins are designed to cross-link and co-activate with other associated cell surface proteins. Antibodies targeted against the T cell surface have this liability of driving unwanted activation of T cells. From an extracellular perspective, I think that has really been the biggest limitation. For small molecules and including degraders, I think for targets that are limited to T cells, T cell selective therapy has been possible and I think works very well. The challenge is that the most relevant intracellular targets in T cells are broadly expressed in other cell types. This is what you see expressed in classes like JAK inhibitors, where you want to knock down JAK activity in T cells, but you get, as a consequence of JAK inhibition, significant off-target knockdown or inhibition of JAK activity in other important cell types. I think if you look at the use of siRNA in T cells, that's been an idea that's been around for a long time. You can find tool siRNA molecules in the scientific literature as far back as the early 2000s, but they're all being introduced into T cell systems with electroporation for the simple reason that they just don't get in in high enough quantities to the endosome. Once they're into the endosome, they don't get out efficiently enough to really inhibit the risk complex. I think that is what has left this as an untapped white space for a period of time. Makes sense. There has been a lot of BD activities around RNAi recently, notably from the Biogen, CT, CRISPR series, and then AVE and ADRIX stills. How do you view the competitive landscape evolving and Generation Bio's positioning in the long run from a BD standpoint, especially with the CDL platform? Yeah, I mean, I think all the progress and confidence in siRNA is a good thing both for us and for the field. This is an approach that has been broadly validated in a series of cell types and tissues. The journey is not over. It's not over for the liver or for the CNS or for the eye. I think what we're excited about is to have a unique entry point to the beginning of that journey in T cells and other autoimmune or other immune cell types in autoimmune diseases. We don't see technology on the horizon that can allow for siRNA with a conjugate mechanism to work in immune cells. As long as that barrier exists for the activity of our LNP system, I think we're going to have a strong and durable competitive advantage there. We're happy for all the progress with siRNA, and I think we're excited about what we can unlock in T cells with our approach. Can you give us a sense of the—you gave us a sense of some of the range of autoimmune indications you aim to pursue, but is there something that would give you more differentiated than the antibody approaches out there, like in terms of the particular patient populations you aim to pursue or risk populations? Sure. I mean, Sure. I mean, yeah, that's right. I think maybe I'll answer your question a couple of different ways. I think for indications like PBC and type 1 diabetes, in both of those indications, there are no disease-modifying therapies available. There's nothing that really can go at the core activity of T cell-driven self-cell destruction. There, I think we'd be looking for the differentiation of a first disease-modifying approach to those indications. In a disease like ulcerative colitis, the bar has been set at about a 30% remission or response rate. I think maybe partly what's in your question, Farzin, is it's just not interesting to enter ulcerative colitis if you're going to have a 30% response rate. What's interesting is the ability to have a much greater response rate and over time to see that response rate be durable. One of the challenges in ulcerative colitis is tachyphylaxis. You get even among those patients who are in that 30% who respond, they do not tend to maintain their response over time. The last thing I would say is we have access to targets with this approach that have differentiated expression between effector T cell subsets and regulatory T cell subsets. What that may unlock is the opportunity to broadly reduce effector T cell function while preserving Treg function. I think where that leads to differentiation in a variety of fields is if the pharmacodynamic effect of reducing effector T cells and allowing Tregs to do their job can be durable beyond the therapeutic effect. For example, if you could give three or four months of therapy where you're changing the balance of T cell activity, could you then move to a much less frequent maintenance regimen in diseases like PBC or type 1 diabetes if you've been able to get the balance of T cells reset and have it be durable? I think we're seeing those kinetics in CAR Ts that are designed to deplete B cells and reset them. We're curious to know if some of the targets that we can uniquely access can recapitulate those dynamics in T cells. I think these are the three kind of areas that we're focused on in terms of differentiation. Makes sense. For the siRNA targets you're pursuing, are you saying more if they're specifically expressed in certain T cell or NK cell subsets at this point? No, it's a good question because our delivery system obviously drives targeting through the expression of the target receptor in T cells. The particular receptor that we're targeting is broadly expressed in CD8s, CD4s, and NK cells. To begin with, the targets that we've selected are targets that are expressed in all of those three cell types. Obviously, in autoimmune disease, it's the CD4s and CD8s that are the most relevant for the impacts. Got it. In terms of catalysts, could you have NHP data this year? Is there potential for accelerating the IND filing from second half 2026 to potentially earlier? Yeah, great question. I don't think there's a potential to accelerate the IND filing timeline. I'll start with your second question first. As of new platform, we just have a core of work that needs to be done in a certain timeframe and sequence for the first filing. Obviously, we can imagine that that would lead to a very accelerated ability to move into additional indications over time after we get that first one in hand. Yes, as we build a base of preclinical data for our lead targets and indications, there'll be further both mouse data and NHP data over the course of the next 12 months or so. We want to really build up the I showed some really interesting primate data here that shows the differentiated selectivity and efficiency of the system. We'd like to build that same database with knockdown targets in primates as well. One of the nice things about T cells is you can bring them from an animal or a human for that matter and show that with ex vivo stimulation, you've changed the behavior both in terms of proliferation and differentiation in T cells. We will be developing that data in primates and in human samples over time. You mentioned you'll have the update in August. That will be the company update in August, yeah. Got it. Any status update for the Moderna collaboration for the in vivo CAR Ts? The collaboration is still proceeding very nicely. We are in the process of enabling Moderna to take the delivery system forward under their own power for their own development of their own programs. We are just crossing the two-year anniversary in the collaboration. I think both of us are really delighted with the progress that we've made together. To close off, basically, cash position and upcoming catalyst, I guess, the IND filing? Yeah. We ended last quarter with about $150 million in cash. It's sufficient to drive the company into the second half of 2027. Great. Thank you so much. Thank you for the questions, Farzin. Thank you for your attention. Good day.