Alnylam Pharmaceuticals, Inc. (ALNY) Earnings Call Transcript & Summary

Welcome back to our afternoon session. First we're going to start with a panel discussion with Vir Biotechnology and Alnylam Pharmaceuticals. From Alnylam we have Vasant Jadhav, VP of Research and from Vir Biotechnology we have Phil Pang, Chief Medical Officer. Thank you both for joining us today.

And I think the best place to begin is to just give us a background on Alnylam's sRNA platform, as well as, tell us about the Vir and Alnylam collaboration in HBV. How far back does this go? How closely did the R&D teams work together? And then we'll have some follow-ups from there.

Sounds good, Patrick. Hey, it's a pleasure to be here. My -- again, my name is Vasant Jadhav from Alnylam. So when we think about Alnylam and the technology to make RNAi therapeutics goes way back. We founded in 2002, and the key challenge obviously was the delivery of sRNAs. And there are 2 approaches that we have come up with. One is lipid nanoparticle. And the second one is the conjugates, the gamma conjugates for liver. Now within the gamma conjugates, we have made kind of tremendous progress going from the initial Standard Template Chemistry to all the way what we now call as Enhanced Stabilization Chemistry Plus. So this has really happened in last, I would say, from 2010 onwards when we first really had the breakthrough in the conjugate technology. And with the initial focus on genetically validated diseases and those kind of targets with biomarkers, when the data came from ALN PCS. So this was our compound for PCSK9 before it became known as inclisiran and now with Novartis via Medco. So for that one, we had our conjugate in the Enhanced Stabilization Chemistry, the ESC Chemistry as ALN PCS and the Phase I data from that study came out somewhere around 2016. And that really gave us the confidence that we can take our technology and have very robust activity and long-lasting feeding. I mean to the point now that the inclisiran is for the every 6 months dosing after the first 2 doses given every 3 months. So with that data coming out in 2016 and its safety profile, along with many other programs that we have on ESC conjugate, gave us the confidence that we can go into the broader indications. And this is the time when we started thinking about then another broader indication, and that's where HBV came in. So as a company, we started working on Hepatitis B Virus program way back then. But I just wanted to highlight that ALN PCS was really an example for us, really feel confident that we can move into these kind of programs. So with HBV program, and in relation with the Vir, it started in, I believe, October 2017 -- I have to be -- keep honest with self -- with the date, sometimes you don't remember, but I believe that is correct. October 2017, when the collaboration started between Vir and Alnylam in the antiviral space. And HBV was one of the cornerstones of that collaboration. By that time, we had worked on HBV with our program known back then, ALN-HBV or HBV01. Now as we were advancing our technology, the inclisiran data was looking very good, but we had a couple of programs where we saw transient LFT elevations. And it happened to be in the HBV01, one of them. And these were, again, asymptomatic in healthy volunteers. Small elevations are seen in few of the healthy volunteers. We have done extensive preclinical work in this space. And under hypothesis that this is most likely happening because of seer-mediated off-target effects. So we thought if you're going in the broader indication like HBV, and we have a good understanding of what might be the potential reason of this small LFT changes happening with ALN-HBV, why not fix it? Because we have a ways of mitigating that and that's what really led to HBV02, now known as VIR-2218. So VIR-2218 is exact same sequence, but it has a key change. The key change means, it has a seed destabilization strategy in the middle of the seed, more specifically in position 7, has this glycol nucleic acid. So what this does is by having the seed modification, it minimizes the unintended off-target effects of this compound. And we have shown that this strategy works across the board for many sequences. And by doing so, by reducing or by literally taking -- I mean the RNA-seq picture looks very, very clean. And by retaining on target, we are really improving the specificity. We were so glad to see the data coming out of VIR-2218 studies in healthy volunteers as well as later in the patients that the ALT response, the LFT profile is completely different than what we are seeing with HBV01. It's an amazing story of kind of going from bench to bedside, seeing some observations and then fixing it and going back from bedside to bench. So we feel that with our ESC+ technology with improved specificity and having -- retaining all the benefits of potency, durability we have created a very compelling molecule, and thus, we believe, is panning out in the current ongoing studies.

That's really helpful, and answers a few of my follow-ups. So I guess what -- the first one I have then on the follow-ups is why was one siRNA trigger appropriate for HBV? And how concerned should we be about viral state?

Yes. No, that is a very important question. And also how -- maybe Phil can chime in a little later on, on this one. So the key thing with the HBV and the patients that will be treated with VIR-2218 would be under the NUC background therapy. We know with this nucleoside inhibitors or reverse transcriptase inhibitors, the replication of the virus is really kept under check. It's really the production of the S antigen and other things. So we think with this technology, especially for the HBV and the background therapy of NUCs, we believe that the single siRNA approach is, in fact, more appropriate, from CMC point of view, from a convenience point of view. And that makes the profile of this program more compelling in our minds. So why go to 2, if 1 is more than sufficient? Phil, do you want to add in here?

No, I think that's exactly appropriate. I mean in the end, you want to use the fewest number of drugs to achieve the maximal impact. Nucleoside reverse transcriptase inhibitors have decades-long history of being excellent at viral suppression. And the goal here is not viral suppression, the goal here is immunologic cure, known as a functional cure. And getting there with one siRNA that is targeted to the correct part of the genome that is able to result in significant hepatitis B surface antigen declines, we believe will sort of remove the brake on the immune system and allow for that immunologic control.

So does HBV solely infect hepatocytes or other liver cells also involved?

So the current literature -- although, of course, there's always ongoing science, the current literature best supports that it is exclusively replicating in hepatocytes. Now of course, you can find some residual RNA DNA in different type of cell types, but whether or not it's actively able to replicate in those cell types is not entirely clear. So the field currently accepts it as hepatocyte exclusive replication.

Yes. Just to add on that, I -- yes, the NTCP receptor, right, the HBV receptor is pretty much expressed in the hepatocytes. So that's kind of goes hand-in-hand that the viral entry that is happening through the receptor is also hepatocyte express. So that's also kind of -- is a supporting evidence for it.

Got it. Great. Okay. So can HBV replicate by either integrated DNA or cccDNA? And how do we know VIR-2218 is suppressing Variance virions and/or subviral particles generated by both INT DNA and cccDNA?

It's a great question, Patrick. So let me begin by taking a slight step back and saying, the answer to your first question is that productive virions, meaning a virion that you could extract from someone's blood, inject into someone else and actually cause disease can only be created from cccDNA. Integrated DNA does not create full-length RNA transcripts that are able to result in productive infection. What integrated DNA does is it's a truncated form of the full genome when it's transcribed and largely produces hepatitis B surface antigen, although there is some belief that it can also create some of the other viral proteins, but not all that are necessary. But with that having been said, since you can get surface antigen created both from Integrated DNA and cccDNA, the question is, how do you know you're knocking down both since it has 2 different sources? And the answer to that actually comes from pioneering research done by Arrowhead as well as by Gilead, which shows that in E -- Hepatitis E antigen negative patients, the majority of surface antigen is produced from Integrated DNA. Conversely in E antigen positive patients, the majority of service antigen comes from cccDNA. So therefore, if you were able to knock down surface antigen in both E antigen negative and E antigen positive patients, you have, by definition, knocked it down from both Integrated DNA and cccDNA. And I believe we are the only siRNA that as a single siRNA is able to -- is currently able to knock down both sources while also targeting the X region of the genome. And the reason why I bring that up is because the virus has evolved over millions of years, only a few genes. And from an antiviral perspective, knocking down as many of them as possible seems to make sense.

Yes. No, that does make sense. And it's helpful. So just a few follow-ups on the data presented at ESO. So just -- first, just regarding inclusion criteria, patients were on NRTI treatment for at least 6 months. Can you tell us how long on average these patients have been on NRTI? And also why you have not enrolled NUC-naive patients in the trial?

So we are capturing the NRTI length but only in -- not in the most granular fashion, so I can say, almost obviously, by definition, everyone is greater than 6 months. Most people have had been on it for at least a few years. And the reason for that is that, as I said earlier, we believe HBV is an immunologic disease and patients are in different immunologic states. And so first off, when you're doing an experiment, you want to first control as many variables as possible. And so having patients on an NRTI for 6 months means that they've probably achieved stable DNA, HBV DNA suppression, and their disease is well controlled, and they are probably in a more uniform immunologic state. We are definitely interested in studying NUC-naive patients and that's a relatively straightforward study. Instead of waiting for them to be on NUCs for a long time, you simply start the siRNA and the NUC at the same time, and that is definitely something we plan to explore.

Got it. And so in the prior KOL session, we'd heard a bit about some of the baseline characteristics and how ethnicity is important, genotype is important. So can you talk -- most of the patients were male and of Asian descent and had a surface antigen of 3 to 4 log, whether E antigen positive or negative. And so I think you've touched on why we would want to see E antigen positive or negative and why that's important. But just generally, why were these baseline characteristics chosen? Generally, we had heard that it's more difficult to achieve functional [indiscernible] patients of Asian descent and in the relevant genotypes there. So I'm curious, presumably, this is why and presumably, we would see good effect in other patients, but I'm curious as to what, kind of, the reasons here. And I guess also, is this kind of -- is this going to represent the baseline characteristics in future studies?

So I would say that maybe as a philosophy, we should take a step back and say, genotypes B and C, which are more represented in the Asia Pac region, are more -- thought to be more challenging to treat with interferon. It is not clear actually what will be most difficult-to-treat -- to achieve functional cure with these new modalities. So I would say that we began with no [indiscernible] bias, except for the bias that we want to get, frankly, quality data as soon as possible. And the reality is that the magnitude of the epidemic in the Asia Pac region is largest and it's easiest to find well treated patients who have a good -- investigators with a good track record and patients who have a known history in that region. So that's why it resulted in many patients of Asian descent. The male-female split was, honestly, just random chance. It's not something we selected for. But in terms of the surface antigen, of course, if you have people with very low surface antigen, at least with regard to interferon therapy that is associated with a higher rate of spontaneous clearance as well as -- well, in spontaneous clearance as well as clearance with interferon, but we don't know what's going to happen in -- with these new modalities. And so all we did was to make sure that they had a minimum surface antigen of greater than 150 IUs per ml. And the rationale there was this, if you don't have a signal, you can't show knockdown. So the desire here wasn't to cherry-pick people with very low surface antigen and try and see if we could cure somebody, it was let's really understand this molecule in terms of a dose response.

Maybe Phil, if I can add to that?

Yes, sure.

I think one of the things to also highlight, kind of came up earlier as well, VIR-2218 is targeting the X region away from the integration hot spot. I mean, that's one of the main things. But also, it's a highly conserved sequence, is part of the highly conserved reason, one of the most conserved parts of the virus. So it would be hard for the mutations to happen there. And secondly, this sequence is highly conserved across the genotypes and so we have no reason to believe why RNAi won't work against any of those genotypes. In fact, just the in vitro reporter studies that we have done, we do see activity across all of those ones. But to Phil's point, this needs to be seen in actual patients. These differences that you kind of mentioned were more for the other modalities, but for RNAi kind of modalities, needs to be seen, but we have no reason to believe why the siRNA can't go after those genotype viruses.

Got it. And so I guess, could you talk about the safety profile, it appeared to be pretty a benign safety profile. But there was some headaches in the study so is that concerning to you? Is there anything else about safety profile that investors should be aware of?

Honestly, I -- of course, when it comes to safety, there's no substitute for lots of patients, right? And I think maybe the first thing I would say is that in a disease that has 292 million people that you are trying to eventually help with their disease, the safety profile will be key, and that goes back to what Vasant was saying earlier, which is that this siRNA with its ESC+ technology has the potential to have one of the best therapeutic indexes out there. And obviously, when you have 1 versus 2 siRNAs, you have half the chance of having an off-target effect. Now with that as a context, of course, until we've dosed hundreds of thousands of patients, we won't be able to see if any difference between the different siRNAs. With regard to your question about headache, and specifically, as I said, it's too soon to say for sure, but I personally am not concerned. These are very small numbers. And all the headaches were Grade 1 or Grade 2. And if you look at -- and then if you try to ask yourself what the mechanism could potentially be, even though it seems implausible, is -- the drug is going largely to the liver, if not exclusively to the liver. It's rapidly cleared out of the bloodstream. And so you would have to postulate some type of class effect. And the reality is, if you look at the data from other siRNAs, like patisiran, givosiran, inclisiran, there really is no suggestion of some type of a neurologic or headache associated effect on a class level. I don't know, Vasant, do you want to add to that?

Yes. I think Phil you said it. I mean, that's where in the earlier part, we're talking more about the inclisiran data, the safety profile of that molecule. I mean, to just put it in a few words when the data came from ORION with more than 3,500 or so patients, the safety profile of that molecule was similar to placebo. I mean, there wasn't anything different with the placebo group. So we don't believe that as this technology would have something like headaches or anything, we haven't really seen that.

So in the first session today, it was mentioned that the guidelines may be, kind of, moving or maybe could be a bit too restrictive. And so with these questions, that's, kind of, the context in which I'm asking them. So is it the expectation that patients who demonstrate surface antigen less than 100 IU per ml at week 24 could eventually clear surface antigen entirely? And then can you reconcile that level, which is -- it's consistent with what we see with siRNA with what we understand regulators define as functional cure being sustained, undetectable, HBV surface antigen and HBV DNA and serum with or without [indiscernible] conversion into the anti-hepatitis B antibody -- S antibody after completion of finite course of treatment. So can you just, kind of, reconcile, first, I mean, is it -- how -- are these guidelines still sort of being adjusted? And is this the level that we think if it's below 100, yes, that could actually clear surface antigen entirely once treatment has stopped. And then how does that reconcile to, I guess, how the regulators are thinking about this?

So Patrick, I'm going to be maybe a little controversial here and say that I think the regulators are completely correct. The data suggests that if you clear surface antigen that is associated with a strong clinical benefit, the reduction in liver cirrhosis and the risk of Hepatocellular carcinoma. So that has to be the target that we're shooting for, right? I mean the regulators didn't just pull that out of the sky, they basically looked at the clinical data and said, this is what makes sense. So the real question is, is a higher target of any value and I want to be very careful here, which is we need to be making sure that we connect what the data is historically to how it was achieved, and therefore, what -- whether or not there's any read-through to the current modalities. And so people who have a lower hepatitis B surface antigen naturally have a higher rate of clearance and patients who have a lower surface antigen to begin with, do better on interferon therapy. It is not entirely clear to me that knocking it down simply by knocking down the levels in your hepatocytes and blood will directly correlate to that data because, of course, you're breaking the link between the biomarker and the cccDNA. What I mean by that is people who naturally have lower surface antigen probably have less amount of cccDNA. When you knock down the surface antigen by knocking down the mRNA, you're not changing the amount of cccDNA in that patient. So when cutoffs like 100 are used, I believe they are merely exemplary. And what I mean by that is that they are a sign that we are achieving a biological effect, and we are making a significant impact but there's no threshold that predicts functional cure right now because no one's cured a lot of patients. We're talking onesies and twosies in every single clinical trial. So you can't have a predictor of something if you haven't even achieved that something to begin with. So I think the regulators are on the right path. I think that is the standard we should hold ourselves to. I think that this is also why I think that combination therapy is what's going to matter the most. I think that siRNA therapy by itself is maybe curative in a small minority patients with certain immunologic characteristics, but I think we're really going to need to move to combination therapy. And that's why, as you and I have discussed, we think that VIR-3434, which is, to our knowledge, the only vaccinal antibody in development for HBV, has the potential to be a transformative asset when combined with VIR-2218.

Yes. It's a great segue. And so VIR-3434. So first, just can you tell us about the antibody platform, how it was formed and what differentiates yours -- antibody platform from others out there?

It's a great question again, Patrick. And so maybe one of the -- Vir was founded in a very unique fashion in that we began with a large asset -- just a large amount of cash, a field we wanted to go into infectious disease. And then we went out there and we scoured the world for what we thought were the most impressive collaborations, licenses or companies we could work with. And one of those companies was Humabs. It's a small company in Southern Switzerland, founded by a giant in the B-cell field known as Antonio Lanzavecchia, who, in addition to being a member of the National Academy of Sciences, really got there because he transformed our understanding of how to isolate monoclonal antibodies from humans. And I would say that the best example of this platform and how it's proven itself is a lot of people have seen the news recently for Regeneron getting a cocktail of antibodies approved for Ebola. Well, Humabs created mAb114, which is the antibody licensed by the NIH and Ridgeback and it was a single antibody that did as well as the 3 antibody cocktail from Regeneron in curing people of mid- to late-stage Ebola. So that is a clinical example of the kind of unique antibodies that this platform can find and we've done the same thing for our flu antibody, 2482, which we were able to isolate an antibody that covers all major strains of influenza over the last 100 years. And therefore, we believe will protect against even future pandemics. It's also what allowed us to identify 3434, a monoclonal antibody that covers all 10 genotypes of HBV, as well as all the major variants. And so those are sort of examples of it. And really, what differentiates it is it's -- they are fully human antibodies with a large-scale screening process based on functional characteristics, which we think are important.

So I think that an antibody connect is an entry inhibitor is intuitive. We've seen some evidence in HTV infection of the benefit of the entry inhibitor approach. So I'm curious in HBV, how helpful could entry inhibition be by the time a patient is chronically infected?

So when we think about chronic infection, we need to think of it as a dynamic state, right? It's not -- all the cells are infected. They stay infected, and that's it, right? So -- and in fact, no one has accurately characterized probably because it is dynamic, how many, in fact, the hepatocytes are in any given individual, it varies so much. So in this dynamic process, what we believe is important is to break the cycle. So if you think of it this way, given that HBV is a hematologic disease, if you really awaken the immune system and kill off already infected cells, the worst-case would be is if there's ongoing infection because when you kill off the already infected cells, the new hepatocytes that appear get infected, you kill those off and you start killing off the entire liver. And that's called Fulminant hepatitis. So breaking the cycle is really key, both from a safety perspective as well as from an antiviral perspective. And then finally, I'd like to point out that entry and inhibition is probably an untapped and underexplored area of HBV biology. And what I mean by that is that, as you pointed out, in hepatitis Delta, they combined an entry inhibitor with interferon. And surprisingly, that combination cured people of their hepatitis B. So I think that there is something unique. And I think that for all these reasons, we're very excited about 3434, which is not just an entry inhibitor, also knocks down hepatitis B surface antigen and as we'll talk about, hopefully, has an FC modification, which allows it to act potentially as a therapeutic vaccine.

Yes. That was actually my next question is, does the antibody have a factor function? And if so, discuss the significance of this engineering on the FC region? And is it impacting both NK cells and T-cells.

So when one typically thinks about a factor function, one automatically starts to think about the oncology field, right, because this is a well-trodden area, whether or not it's afucosylation and these sort of things. And that's the ability -- so when we typically think of a factor function, we think of the ability to recruit either NK cells or macrophages to mediate what we call ADCC or ADCP. That is definitely intact in 3434. But what we think is even more important is it's engineering to not only increase ADCC and ADCP but to increase its ability to activate macrophages. And so therefore, its ability to act as AT cell vaccine. And so there was a great paper that just came out by Dr. Jeff Ravetch at Rockefeller in Nature on October 8, entitled FC Optimized Antibodies Elicit CD8 Immunity to Viral Infection. And that was actually some of the original data we saw 2.5 years ago, which led Vir to license the rights to this particular mutation for all of infectious disease. Because you can imagine, it's game-changing if you can turn every antibody into a T-cell vaccine. We're using this in HBV, and we're also interested in using this in COVID-19.

Got it. And so can you tell us, are HBV virions and/or the sub viral particles, are those expressed on hepatocytes? Or are they present in the tissue and/or the serum?

So the Virion -- so let me take a step back and say, how well expressed surface antigen is on the cell surface of hepatocytes is controversial. So if you're asking me, is it absolutely known that ADCC will occur in the setting of an hepatocyte? The answer is we don't know yet. That is something we need to explore. Some investigators like Ulla Protzer, who is well-regarded in the field is absolutely certain from her data that it is expressed on the surface of hepatocytes. Other investigators are less certain. But I would say that the importance of the T cell impact is different, of course, because you're not waiting for expression on the cell surface, what you're waiting for is the antigen to presented in an MHC Class I complex. And that is definitely certain. It is definitely presented in MHC complexes. That's actually how you get T-cell activation and/or senescence. With regard to where else it is found? I think that this is the key difference, which is that the virions and the sub viral particles go everywhere. And -- but I think they are -- the actual antigens from an intracellular perspective and therefore, to be presented, are going to be found largely in hepatocytes. Now of course, there is cost presentation and so there's excellent data coming out of Antonio Bertoletti's Lab, showing that cross presentation of antigens found in the blood actually result in B-cell dysfunction. And that's the great thing about this vaccinal mutation, which is called GAALIE in Dr. Ravetch's paper, we often refer [ colloquial ] as XX2 is that it should stimulate CD4 and CD8 T-cells and therefore, should have an impact both on B-cell dysfunction and T-cell dysfunction.

Got it. That's helpful. And then so with what assays or methods has Vir evaluated potential emergence of escape mutants? And what did that work demonstrate?

So the reality is we haven't seen any escape mutants yet, but we have a small data set clinically. But that honestly is not that surprising because as Vasant suggested earlier, everyone's on a NUC. And frankly, most of them are on second generation, third-generation NUCs like tenofovir or entecavir -- sorry, Baraclude. And so the likelihood that you're going to get an escape mutant is vanishingly small. And then you've added on an siRNA to that. So we haven't seen any clinically. It is, of course, possible to generate them in a tissue culture when you give very low doses of the siRNA and very high doses of the virus, but the fitness of those mutations and the clinical relevance is going to be hard to imagine in the setting of an NRTI.

Got it. So in vitro, Vir-3434 demonstrates approximately 5,000 folds greater potency than HbI gene neutralization assays. 3434 is better, able to prevent the spread of HBV to uninfected cells in vivo compared to HbI gene. So tell us about this preclinical study that's conducted. How well does this model translate to humans? And how often are compounds evaluated with it? And how did VIR-3434 compare?

Great question. So Patrick, so let me separate it. It's actually -- those statements are linked to 2 different experiments. One is an in vitro experiment where you can do a dose response and compare it to hepatitis B immunoglobulin and show that it is 5,000 fold more potent in its ability to protect hepatocytes from being infected. The in vivo data was actually using in mouse with a humanized liver. And so basically, you just destroy the mouse liver, repopulate it with human hepatocytes and see replication in those hepatocytes. It is a very good model virologically in the sense that you replicate the entire viral life cycle. It is not a good model immunologically because, of course, in order for them to survive with humanized -- with human liver cells, they need to be immunosuppressed. So I would say that in terms of the antiviral potency, it's excellent. It is additive with NRTIs. And I would say that there's no reason to believe that as a novel mechanism of action, it won't have an important role to play. I mean if you think about NUCs in general, NUCs knock down viral replication, and therefore, indirectly block viral entry. And one of the observations that is made in the coinfected Delta B field is that you will basically be able to suppress ALT signals by giving an entry inhibitor, which suggests that part of the ALT process that we see is actually due to the infection of new hepatocytes and their destruction. And if you can prevent that, we think that, that's also beneficial.

Got it. Yes, that makes sense. And then -- so let's talk about the Phase II trial. So this has 3 parts. Part A is a single standing dose study. Part B is assessing 3434 plus NRTI, though in chronic hep B patients with S antigen less than 1,000 IU per ml. Optional part C it evaluates patients with higher levels of surface antigen. So first, why is you're focusing on patients stable on NRTIs? Would we not expect to see a profound decline in surface antigen in these patients? And why have the limit of less than 1,000?

Yes. So a lot of questions there, so I'll take it in pieces. First off, the NRTIs, as I mentioned earlier, are creating a higher barrier of resistance. But the other thing they're doing is -- one of the fascinating observations, this stems from a conversation I had with Anna Lok, one of the past presidents of AASLD, what she said to me is, "What happens when you have a patient in, let's say, fulminant hepatitis?" You give them a NUC and you can often take them off the transplant list, which means somehow, the presence of an antiviral is able to dampen down an old progressive immune response. And the belief there is, again, when I talked about chasing your tail in this cycle, which is that it's all great if the immune system is killing off the infected hepatocytes, but if they're killing off the newly infected hepatocytes that just got there, you're going to end up in a problem. So there's a safety benefit to having the patients on NRTIs in addition to the prevention of viral breakthrough. With regard to the 1,000 IUs per ml, remember, this is a drug that is going to be used as part of a cocktail with the siRNAs. And trying to achieve a hepatitis B surface antigen threshold that's in 1,000, as you know, from the data, is quite straightforward for an siRNA. So once we achieve that surface antigen knockdown with the siRNA, we would then follow with 3434. So this felt like an appropriate way to gauge the kind of impact you would have. Now for patients who have 5 or 6 orders of magnitude of surface antigen, this is again why we want to pair it with the siRNA. The amount of antibody you would have to give just from a stoichiometry perspective would be just enormously high. So this is sort of a clinically relevant situation, which we believe is a proxy for the situation which you would pre dose with an siRNA. And so that was one of the reasons. Of course, for safety, we also want to test what happens at higher surface antigen levels.

Yes, that's interesting. And so how -- what will it determine if the optional Part C is initiated? And what level of surface antigen do you anticipate to be enrolled in that part?

So we're going to take all-comers in Part C in terms of their surface antigen. And effectively, what we're going to be doing is, I don't see any reason why we would not enroll that cohort. I think it's just not going to be gating to starting of combination therapy. The more information we can get, the better. And I think that's important to know, but it's certainly not a critical path to further development.

So in -- assuming we move forward to Phase III combination trials, would Vir anticipate NUC naive patients then would have to first be treated with NRTI for a period of time before being treated with 3434 and/or siRNA?

Definitely not, period. What I mean by that is that we'll certainly be testing patients who are already on NUCs and adding 2218 plus 3434. But we're also have to test patients who are not on NUCs and starting them on the siRNA plus the NUC at the same time and then adding on 3434. Now people don't know yet, which is the easier population to treat, which is the easier population to cure. Obviously, there are many more people not on NUCs than on NUCs, so you need to be able to address both populations. Whether or not we would have to do a Phase II before the Phase III, or we can do it simply in Phase III would be really a question of what kind of efficacy and what kind of safety do we demonstrate in our earlier trials.

Got it. So just kind of moving on to the clinical development strategy. Regarding the combination strategy, we have several studies that are planned. We have VIR-2218 plus NRTI plus PEG interferon, and presumably also 2218 plus NRTI plus 3434. So PEG interferon when administered it's usually administered for out a year in some patients, they can have tolerability issues. So I guess, the first question is, do you think combination with RNA interference is that the ideas that this treatment duration can be reduced or maybe in the dosing?

I completely think it can be reduced. And so I think that this is -- one of the things that's underappreciated, people here pegylated interferon and they think, "Ah". But why don't people want pegylated interferon? Well, it's 2 reasons. One, it's a year-long of therapy and the side effects are duration dependent. That was clearly shown actually in hepatitis C, where you compare 12, 24 and 48 weeks. And the people who took it for 12 weeks had virtually hardly any problems compared to the people who took it for 48 weeks. But then you have to take something for 48 weeks, which gives you maybe a 3% chance of cure, who's going to want that? But that doesn't mean that if, for example, you would be able to reduce that interferon duration to 12 or 24 weeks and achieve cure rates of 30% to 50% that people would not be interested. I think that they are 2 different paradigms entirely. But I would say that our strategy is interferon is available off the shelf, we can immediately see if knocking down surface antigen translates into higher functional rates of cure with that particular immunomodulator, because 3434 is a year behind. And then we will follow that with 2218 plus 3434. Now of course, 2218 plus interferon will be the fastest path to registration, and we'll have to see. How does that data look compared to the emerging Phase II data from 3434 and 2218, to decide whether or not we're going to do this in 2 successive waves that are close in time to each other or simply wait for what can be a potentially interferon free regimen, which, of course, is still preferred to your short course of interferon.

Right. Yes. No, that makes sense. Okay. So then just with the Phase II combo trial with 2218 and interferon, this is also being conducted in patients stable in NRTI, which we've already covered. So -- but I'm just curious, could you discuss the design for the study? Any notable inclusions or exclusions beyond this that we should be aware of? How the dosing of 50 mg or 200 mg type via 6 doses was decided on. And any -- what are the expectations for adverse events with the primary endpoint?

So let me start with the end. I think that with regard to interferon adverse -- so with regard to adverse events -- with respect to interferon adverse events, we would expect them to be less simply because the duration is 12 or 24 weeks, not 48 weeks. For the siRNA, we haven't seen an AE of special interest as they call. So we would have to wait for more data, but there's nothing we're expecting from that perspective. But from an efficacy perspective, obviously, what we're hoping for is functional cure. But in the meantime, before we even get to functional cure, which is off all medications, what we want to see is that people get to undetectable on treatment and that is what we're hoping to see that that is our hope for next year. So that is in terms of where we want to go from a safety and efficacy perspective. In terms of the trial design, the inclusion/exclusion criteria are very standard for this type of trial. There's nothing notable. We are excluding patients who have advanced cirrhosis, these sort of things is often the case in these early-stage trials. With regard to the dosing, many people have asked us, well, you saw a dose response between 20 and 200, why not go higher? And I think what this ignores is the frequency of dosing, which is that we believe that the half-life of this siRNA in the liver, not the plasma, but in the liver, is well in excess of 30 days. So if you give it every month, what's going to happen is the amount in the liver is just going to continue to stack up. So the question is, is 200 milligrams given 6 times going to result in any different exposure than, let's say, 400 milligrams given twice or 600 milligrams given once? And our modeling suggests that 200 times 6 is going to result in an excellent level of exposure. Now the other thing is that when you give a bigger dose less frequently, often more of it just gets cleared out of the blood and through the kidney. So this is sort of a happy place. Now that doesn't exclude us looking at higher doses for a shorter duration of time. But first things first, let's cure a bunch of people and not to optimize how to deliver it in a more simplistic way. So that was how the 200 times 6 was chosen. And the 50-milligram dose was to bracket it. I mean we need to really understand what the dose response is, both for regulatory purposes as well as for biological purposes. And being able to understand if 50 milligrams given 6x results in an exposure that in the presence of an immunomodulator might have an impact, right? We always think you got to maximize everything. That's true in an antiviral setting, whether or not that's true in an immunomodulatory setting is not clear. So -- and I'd like to actually take that to highlight a collaboration we are planning with the NIDDK or the NIH where we're actually planning on getting serial liver biopsies as well as FNAs to really understand what is happening immunologically in the liver rather than in the peripheral blood to be able to best optimize future therapy.

Right. That's interesting. And that's helpful. So would Vir also have to conduct studies or show data in NUC naive patients before moving to a pivotal program?

I don't think so. I think that it really will depend on the safety and the risk we would have to take. We could do a lead into a Phase III. We could go directly into Phase III and start with a smaller number of patients before we expand it to a larger number of patients. Really, it's going to be a discussion with regulators as to what we think is necessary, both from a safety and efficacy perspective. But obviously, those are things we're constantly thinking about. And obviously, now would be a reasonable time to start really considering whether or not those things need to happen in parallel or could happen in a staggered fashion. But we are very mindful that the NUC naive population is not something we plan to leave on the table.

Right. That makes sense. Okay. So can you discuss the AAV HBV mouse model data that evaluated 2218 in combination with 3434? How synergistic was this approach? And what are the implications for clearance of surface antigen in humans?

So it's funny. We have the privilege of having our Chief Business Officer also be a chemist. And so he likes to pound me over the head and say the word synergistic has a very specific biologic and chemical term and be careful. So I'm going to shy away from the word synergistic and simply say, when you give the siRNA alone, you get about a log decline. When you give the monoclonal loan, you get about half a log. And when you combine the 2, you get 3. But we haven't formally shown synergies. So I'll shy away from that word. But clearly, 1 plus 0.5 is more than -- 3 is more than 1 plus 0.5. Now the question is not can we knock it down, which was very impressive, I agree, but can it stay down? Will it keep it from coming back? And this is the key part, which is after you knock down the surface antigen, if you don't reawaken the immune system, it will just come back. And of course, you can't test that in an AAV mouse since it doesn't have a human immune system. So really, the question for us is, can you achieve the impressive knockdowns, which we did. And now in a human can you reawaken the immune system so that it goes down to 0 and stays down. And that's exactly what we're hoping for. And this goes back to the reasons why we keep focusing on the immunologic aspects of HBV, whether it's 2218 removing the stop sign on the immune system or 3434 or interferon stepping on the gas of the immune system so that we can stimulate the immune system that will control hepatitis B.

Yes. So just thinking about the combination trials and how you're moving these forward, it sounded as if -- well, I think it's known that we would prefer 3434 plus 2218 over interferon. But at the same time, it's possible or probable that 2218 plus PEG interferon is the quickest path forward to approval. So if that's the understanding, some KOLs have stated, we need to see something like 25% to 30% functional cure rate and the novel treatments for approval. I guess, first, I would want to know, if you agree with that. Then I'm curious if you think if 2218 plus PEG interferon plus NRTI, can achieve this rate at least?

So I'm going to agree with the KOLs, which happens only about 50% of the time for me. So I would say that definitely a 20% to 30% bar, I think is a very reasonable bar. And I think the question has to be nuanced, though, which is not just can you achieve that bar, but in rich patient populations, right? Because if -- not if, because hepatitis B is an immunologically diverse disease, you may be able to achieve 40% or 50% in one small subpopulation and 10% in another. So defining that subpopulation will be huge. But that's also why we think 3434 is potentially even better than interferon in the sense that it is a much more specific immunologic response. And you're actually giving them the immune response that you hope that they need and therefore, will be less sensitive to these different immunologic populations. So 2218 plus interferon, we hope it can achieve greater than 30%. We think it will. But of course, we don't know until we have data. It will be about which population we show it in, but then 3434 not only has the possibility of achieving higher rates of cure, but also in a broader patient population.

And so then what would the rate be when 3434 is added? What would the rate of functional cure be?

I have absolutely no idea, Pat. I wish I could predict that. If I knew that, we'd all be in a happier place. I mean I -- it's impossible to speculate. There's no reason to believe that it should be -- I like to say that the biology of HBV still remains complex. I think that right after the HCV success, everyone turned to HBV and thought it was going to be a knock off in 1 or 2 years. I think that there was a lot of over exuberance and maybe a misunderstanding of how hepatitis B is so different than hepatitis C. It is not an RNA virus replicating in a cytoplasm for which a pure antiviral approach would work. It is an immunological disease caused by the virus, and therefore, we needed fundamentally different approaches. And so I think the field took a left turn, focusing mostly on antivirals to begin with. And now you can see most of the field is shifting to immunologic mechanisms. And I think that, that is the right place to go. But how much push from the immune system we can expect in which populations, I just think it's too early to say.

Got it. That's really helpful. And I think with the last few minutes that we have here, I could turn to Vasant for a few. So I guess Vir has been running the program and the data is very interesting. We're looking forward to the combination studies and the data generated there. I'm curious, what does Alnylam need to see in the Phase II combination studies up into the program? And how will the program -- how the clinical development program -- and I guess this question really is to both of you, how will it be managed for when Alnylam ops in?

Yes. Thanks, Patrick. I mean, obviously, a very, very important question, but we can begin with saying the data that we have seen so far has been impressive with VIR-2218 and the profile that we have seen with this molecule in terms of the safety. So we remain very, very, very enthusiastic. And for our technology making a move from genetically validated diseases or, kind of, orphan disease space that we started our technology first and moving into the broader indications, this would seem like a place to go, right? So a lot more discussions would need to happen more on the corporate level. But I mean we are very, very enthusiastic with the way the program has advanced so far and also more importantly, the relationship as well with Vir. I mean when we think about working together, it has been seamless, absolutely seamless. Sometimes I say that is it doesn't feel like 2 different companies. And I'm -- I mean it, not just saying it for the sake of saying it here on this forum, but that has been the experience. I mean with the productivity of our RNAi technology, there are so many opportunities we can pursue. And we have been very judicious in finding the partners, collaborators. Regeneron is an example of that one in the CNS and the ocular space. And Vir -- similarly, Vir in that place in the antiviral field where they would take -- both players are bringing the best they can from their strengths. Alnylam from the technology side, and Vir from the antiviral field, the knowledge about it and how to push these things into the clinic. So we remain very excited and not only for the program, but working together with Vir.

I just want to second that and say it has truly been a -- Vir is born out of collaborations, whether or not it's with GSK now for COVID or Alnylam for hepatitis B or many other things. And I have to say that working with Alnylam has been incredibly seamless. I mean there's no -- oh, this is -- don't ask Alnylam this before you ask this person that or all these steering committees. We just call each other up, and it's like calling up a Vir colleague who's living on a different floor. So it's really a seamless interaction. There's just never anything that drives our decisions besides what's best for the molecule and what we think is scientifically most appropriate. And I think that, that's a common value we share, which is something that makes really working together such a pleasure.

Indeed, and I mean, just to add, I mean, we are thrilled that this would be an opportunity for us -- for Alnylam to have opportunity like VIR-2218, HBV02, so really excited. And another is just an example, and we just had this Oligonucleotide Therapeutics Society Meeting. It's one of the biggest meeting in the space of Oligo Therapeutics. And there as well, we're going back to back. So I presented on the technology side and how the data came out from VIR-2218 and comparing it with earlier HBV01 version, and then Phil talked about the clinical data. So it was very similar to the discussion that we are having now in the very much scientific meetings as well.

Right. Okay. Well, good luck to getting the studies enrolled. We're looking forward to the data. And thank you very much for participating in our Virtual HBV Conference. And to everyone else, the next session starts in a few minutes. Thanks so much.

Thank you so much, Patrick.

Thank you.