Verve Therapeutics, Inc. (VERV) Earnings Call Transcript & Summary

Welcome, and thank you for joining us today for our discussion of the interim results from the heart-1 trial of VERVE-101. I'm Jen Robinson, Vice President, Investor Relations at Verve Therapeutics. Before we begin, I would like to remind you that any statements we make at this meeting that are not historical facts are forward-looking statements reflecting the current beliefs and expectations of management made pursuant to the safe harbor provisions of the Private Securities Litigation Reform Act of 1995. Examples of forward-looking statements made during this meeting include statements regarding the safety, tolerability and potential benefits of VERVE-101, the company's clinical development plans and the company's strategic plans and prospects. Forward-looking statements during this meeting involve substantial risks and uncertainties that could cause actual results to differ materially from those expressed or implied by the forward-looking statements, and we cannot assure you that our expectations will be achieved. Such risks and uncertainties include those set forth in our most recent quarterly report on Form 10-Q as well as our other current SEC filings. We assume no obligation to update or revise any forward-looking statements. Following this meeting, a replay will be made available on the company's website, www.vervetx.com. Today, we'll be joined in by Dr. Sek Kathiresan, our Co-Founder and Chief Executive Officer at Verve Therapeutics; Dr. Deepak Bhatt, Director of Mount Sinai Fuster Heart Hospital; Dr. Andrew Bellinger, our Chief Scientific Officer; and joining for Q&A will be Fred Fiedorek, our Chief Medical Officer; and Allison Dorval, our Chief Financial Officer. With that, I'll hand the mic over to Sek.

Okay. Thank you, everybody, for joining us tonight what is really we think a milestone event for, hopefully, for cardiovascular medicine. It's early, but I think the data that we just presented earlier today potentially points the way in terms of a whole new way to treat heart disease, and we're super excited about that. Today, we'll be covering the disease process, our mission and vision and then moving on to VERVE-101. We'll go through the results again, and then give you some updates in terms of our pipeline, the recent progress, and then we'll close with some Q&A. Thank you, again. Okay. Verve is focused and well positioned to realize its vision of developing single-course gene editing medicines to treat atherosclerotic cardiovascular disease or ASCVD. What you'll hear tonight is that we have human proof-of-concept for in vivo base editing technology. We have 3 product candidates against highly validated targets. We have now an emerging path, regulatory path in the U.S. with our recent FDA IND clearance for VERVE-101. We also have a new partner, a terrific partner in Eli Lilly with a shared vision, cardiovascular development expertise and strength in commercialization of products in this space. Finally, we're well capitalized with about $485 million in cash and runway into 2026. So really an exciting path forward. I'm going to turn it over to Dr. Deepak Bhatt, who's going to walk through the disease indication of heterozygous familial hypercholesterolemia and the unmet need in this indication. Dr. Bhatt?

Well, thank you. It's really wonderful to be with you all at this event and really like to congratulate Dr. Kathiresan. We've been colleagues at Harvard Medical School for many years. It's terrific to see you in this role and the success of Verve, the scientific breakthrough today, I think was tremendous and something that one day might win a Nobel prize. So really great to in some small way be part of it here. And Dr. Bellinger, who I worked with closely for many years at Brigham and Women's Hospital, wonderful seeing you now on the big stage shining just really makes me happy. Before I get started on my topic for today, disease indication, unmet need just in terms of disclosures, I don't have any financial relationship with Verve. Although after today's date, I wish I did, but I don't have any financial relationship with them. I do receive research funding from Amgen and from Novartis and Sanofi for clinical trial work on those PCSK9 inhibitors that they make as well as from Lilly having to do with tirzepatide and clinical trial work I do on that diabetes and obesity drug. So there was a huge unmet need here. By here, I mean in the U.S., I mean in the world, ASCVD, atherosclerotic cardiovascular disease remains the leading killer. That's true in the U.S. It's true worldwide. There are hundreds of millions of patients worldwide with atherosclerotic cardiovascular disease. Still one person dies every 34 seconds from cardiovascular disease in the United States, and there are about 800,000 heart attacks per year in the United States. So still a big deal. You hear a lot about advances in cardiovascular disease. It's true, but there's still an unmet need that's enormous. I don't know if you've all heard of these concepts of secondary prevention that's taking care of people that have had, for example, heart attacks, primary prevention that's treating risk factors like hyperlipidemia that can lead to heart attacks and primordial prevention preventing the development of the risk factors in the first place. And as a cardiologist, I spend a lot of my time in the secondary prevention universe treating existent identified disease and primary prevention treating risk factors. But imagine if we could really frame-shift things to primordial prevention to actually get to patients before they become patients to treat them before they actually develop the risk factors that lead to cardiovascular disease, I think the technology we're going to hear about has that potential. Maybe some of you with a medical background have heard of pack years of smoking, that's the number of packs per day a person smokes times the number of years if somebody smokes a pack a day for 20 years, that's 20 pack years of smoking, the higher the exposure the worse. This is a concept that's similar, but for cholesterol, cholesterol [ years ] for cardiovascular risk prediction and treatment. And the point that it's trying to make here in this graphic is that if a person has really high cholesterol such as from familial hypercholesterolemia, where the cholesterol is elevated right out of the gate from birth, well, that's a problem because they're exposed to those very high cholesterol levels for a long time. But actually, even if someone has more moderate degrees of elevated cholesterol but has additional cardiovascular risk factors or has actually had a heart attack that they survive, well, there as well, there's a lot of risk. And the amount of time, the number of years that you're exposed to the cholesterol elevation, just how high the cholesterol are, that combination essentially gives you an area under the curve and the higher, the worse it is. And again, potentially, if we can intervene early enough in this cycle, we might be able to put a real dent in the cardiovascular epidemic. Now does LDL cholesterol matter. Hopefully, everyone here is convinced that it does. But just in case there are any skeptics, these are some data, it's from the ODYSSEY OUTCOME trial, but there are other sources of data that show similar findings. We see that lower is indeed better. The lower the cholesterol is, the lower the risk of dying. And where things seem to sort of plateau is maybe around the 20-ish to 30-ish range sort of the cholesterols, the LDL cholesterols of a newborn baby. That's a normal cholesterol. That's a natural cholesterol. The cholesterol that I have, that everyone on this panel has, I bet most of the people here, unless you're on medicine is an abnormal cholesterol. We're all walking around with abnormal cholesterol. And some folks are walking around with really abnormal cholesterol. What's going on in the U.S. in people with coronary artery disease. I'm talking about U.S. adults, they have coronary disease. There's no debate about it. Well, in this recent publication from JAMA, we're looking at the NHANES registry. And it turns out that only 1 in 4 folks, if you use the American guidelines, where the target for cholesterol should be less than or equal to 70 milligrams per deciliter are actually achieving it. And if you use the European guidelines, which I like a little bit better, let's say, the target should be less than or equal to 55 milligrams per deciliter, only 1 in 10 green people. Everybody else there is red. So that shows that right now in the United States in people that have identified coronary disease, we're doing a lousy job of treating cholesterol. The combination of the health care system doctor and patient isn't working. And there are a lot of reasons for this. Some of this has to do with poor adherence to therapy. It's tough to take a pill every day. There are a lot of reasons that this is the case, but it is the reality. And this is why people, even with identified coronary artery disease, go on to have recurrent heart attacks, sometimes recurrent fatal heart attacks. This is a Mendelian randomization study. Maybe it's a little too late at night for this, but this is basically nature's way of randomizing patients to different genes at birth and then seeing what happens. And what this study shows is that both genes that control LDL cholesterol and genes that control triglyceride-rich lipoprotein are associated with cardiovascular risk, such that lower is better in both cases and independently so. And this really gets to the point that LDL cholesterol is really important, often the focus of atherosclerosis treatment and prevention, but triglycerides matter and, in fact, other things matter. And this is the concept of residual cardiovascular risk. So we really want to target cholesterol, get that as low as possible, if it can be safely achieved probably also inflammatory risk, thrombotic risk or important triglyceride-associated risk, Lp(a) associated risk and diabetes associated risk. So targeting all these different pathways can, in fact, have an enormous impact on that residual risk and hopefully prevent recurrent cardiovascular events. And in doing so, we can potentially put a dent in this pyramid of risk where we can hopefully move patients from, say, the primary prevention universe from going into the secondary or tertiary prevention universe where they have a heart attack or a fatal cardiovascular event. And maybe in the distant future, even intervene early enough in the primordial prevention stage to keep patients from developing that risk factor in the first place. So lots of potential when we're talking about therapies like gene editing. Finally, I'm just going to say a word about heterozygous familial hypercholesterolemia. It's a serious inherited subtype of atherosclerotic cardiovascular disease with very high LDL cholesterol levels from birth and heart attack at early ages. Just working from left to right on this slide, you can see what characterizes familial hypercholesterolemia. It's a high LDL cholesterol, a family history of high LDL cholesterol or premature atherosclerotic cardiovascular disease, people having a heart attack in an early age. It can be monogenic or polygenic. The LDL cholesterol is often greater than or equal to 190 milligrams per deciliter. That's in a lot of definitions of it. An ischemic event, like a heart attack or a stroke often manifests in these folks at age 30 to 60 in that range, sometimes earlier, sometimes even later. And there are a lot of people that are affected, at least 3 million adults in the U.S. and Europe. And I think this is a bit of an underestimate, but at least 20 million people globally. So this is a real problem. And heart attacks, obviously, are a bad thing. That's an obvious statement to make. But these are premature heart attacks that we're talking about. So we're talking about people that in the prime of their lives are being taken. So anything that can address this problem effectively could be a really huge advance. And with that, let me turn things back over to Dr. Kathiresan.

Thank you very much, Deepak. Okay. Let me walk you through a little bit of the background in terms of Verve's mission and vision. So we started by asking this question, what causes ASCVD. And as Dr. Bhatt highlighted, the key cause is high cumulative lifelong exposure to blood cholesterol, clogging the heart arteries, and cholesterol in the blood is carried in any of 3 different lipoproteins, LDL, low-density lipoprotein, triglyceride-rich lipoproteins, that's TRL or Lp(a). So these are the lipoproteins that carry cholesterol in the blood and ultimately, cumulative exposure to this cholesterol clogs the heart arteries. Now what's the solution? The solution has been highlighted by human genetics. And the solution is to keep blood cholesterol as low as possible for as long as possible. How do we know that? Because there are people who are walking around who have certain genes switched off, in one case, PCSK9 naturally switched off, leading to lifelong low levels of LDL cholesterol and resistance to ASCVD. In the case of PCSK9 deficiency, natural deficiency, it's about 50-milligram deciliter lower LDL in the blood in those individuals, about 50% lower risk of ASCVD. And importantly, absence of this gene or deficiency of this gene either in one copy or both copies, the individuals are healthy otherwise. So this gene is more or less a spare part. You can get rid of it and all you get is health. So these observations led us to wonder what if we developed a medicine that would mimic those natural resistance mutations. Now there are a number of pills and injections available right now to lower LDL, right? So we often get asked, what's the unmet need? Well, if you look at patients with heterozygous FH, only about 3% of patients worldwide with HeFH are at LDL goal. So we know that we need to get the LDL as low as possible for as long as possible. We have several medicines available right now to do that. So why is it that the majority of patients actually are not at goal? And we think a lot of that has to do with the chronic care model to treat HeFH and ASCVD. This model involves daily pills or intermittent injections administered often over decades, and that puts a heavy treatment burden on patients, providers and the health care system. And that model really does seem broken. If you think about that model and apply it to a typical patient with HeFH, this is what it might look like. On the x-axis is age, on the y-axis is the level of LDL cholesterol, you can see here, the patient's LDL is about 200 starting at birth and they -- the patient has a heart attack at age 44, so premature heart disease. Typically, the patient gets a procedure, maybe a stent to clear the clogged artery and then will get put on therapy to lower LDL. In this case, you can see they get put on statins, ultimately, a PCSK9 antibody, LDL does come down, but doesn't usually stay down. There's oscillation in the LDL over time because of issues related to the chronic care model. The patient adherence, access, health care infrastructure issues, and that lack of control, the lack of control of the cumulative exposure leads to recurrent heart disease events. And those are shown at the top, another stent, bypass surgery or sometimes even a fatal event. So what we'd like to do is replace this picture with this picture, a onetime therapy, dramatic and durable lowering of LDL to really control LDL over a lifetime. Now as I said earlier, this is a significant milestone today for Verve. With the interim data that Andrew presented earlier today really provides proof-of-concept in humans for in vivo liver base editing. And this is the time line kind of how we got to today. Verve was founded in 2018. In 2020, we developed proof-of-concept for in vivo liver base editing in nonhuman primates. Last year, we dosed our first patient with VERVE-101. And today, we presented the interim data that we'll discuss a bit more in a few minutes. Let me walk you through the preclinical data that allowed us to start our first human tests. This is VERVE-101. It's a novel base editing medicine designed to inactivate the PCSK9 gene in the liver and lower LDL cholesterol with a single DNA base pair change without cutting the DNA. The drug substances are shown on the left. It's 2 molecules, mRNA for adenine base editor and a guide RNA that localizes the editor to the PCSK9 gene. Both of these drug substances are packaged in a lipid nanoparticle, which is shown in the center, the components and then the drug is schematized all the way on the right. VERVE-101 is delivered as an intravenous infusion. I'm going to walk you through the mechanism of action. So number one, delivery by single IV infusion into the blood, the lipid nanoparticles are taken up from the blood by hepatocytes through the LDL receptor, and that's what's shown in the middle. Once inside the cell, the adenine base editor mRNA is translated to protein, the protein binds to the guide, the complex of the guide plus the protein make their way to the nucleus, ultimately scan the genome chromosome by chromosome to find -- to land at the directed spot. For us, that's in the PCSK9 gene on chromosome 1. So there's localization of the base editor to the PCSK9 gene sequence on chromosome 1. That -- at that spot, there's a single spelling change made and A-to-G to turn off the gene again without DNA cutting. That A-to-G change is in the canonical splice site in between exon 1 and intron 1. Turning off the gene, of course, will lead to decreased production of PCSK9 protein, which ultimately leads -- should lead to reduction in the blood LDL cholesterol. We've done extensive preclinical evaluation of VERVE-101 in nonhuman primates and human cells to document the efficacy in terms of protein reduction and LDL reduction, the durability, a liver-specific biodistribution as well as absence of detectable off-target editing. Here's the durability data, durability out to now 2.5 years in nonhuman primates after a single dose of VERVE-101. The x-axis here is time, on the y-axis is the LDL reduction in 2 groups of animals control as well as VERVE-101 treatment, and you can see there's about a 60% to 65% reduction in LDL that is durable that starts at 2 weeks, durable out to 2.5 years after a single treatment. This is the liver-specific biodistribution in nonhuman primates treated with VERVE-101, on the x-axis are tissues at necropsy in several groups of animals, on the y-axis is the level of PCSK9 editing. And what you can see is there is a very high level of editing in the liver and really very little in any other organ. There's a little bit in adrenal and spleen, but that's about it. So this is a liver-directed therapy, liver-specific biodistribution. Lastly, we've done extensive work on off-target. And here's just one data slide on that. On the x-axis here are a range of chromosomal locations, so about 6,000 different potential off-target sites, on the Y-axis is the level of editing in human liver cells treated with VERVE-101. So these are liver cells in a [ dish ] treated with VERVE-101. And there's no evidence of off-target editing. You can see on the x-axis, all of the potential off-target sites in black are basically centered around 0. And you have editing at the PCSK9 spot on chromosome 1, the large purple dot. These are data, again, for single base pair changes potentially at on-target and off-target sites and you're seeing exquisite on-target editing only. There's also another type of off-target event that people look at. And those are larger changes beyond single base pair changes. So these are chromosomal translocations or structural variations. So these are large pieces of DNA that we arranged, could potentially rearrange with treatment. Shown here our data, looking at chromosomal translocations or structural variants after treatment of primary human liver cells with VERVE-101. The first circle is untreated. And what you can see the red dots and the various dots also, the green lines represent different types of structural variations in this donor. So these variations exist in all of our cells. So that's there. Then you get VERVE-101 treatment in the middle. And then finally, all the way on the right is treated minus untreated. And what's important to note is this treatment does not generate any new structural variation, all the way on the right, you're not seeing chromosomal translocations or other types of events like insertions or duplications. This is something you see with base editing, the absence of chromosomal translocations or structural variants, something you typically don't see with standard CRISPR-Cas9. Standard CRISPR-Cas9 typically leads to a low rate of chromosomal issues of the sort that are shown on the left, okay? So these are -- these data are part of the reason why we chose base editing for our first few programs, we directly compared standard Cas9 with base editing in cells, mice and nonhuman primates before selecting base editing for our programs. Now with that as background, let me turn it over to Andrew to walk you through the study design as well as the interim results.

Okay. Thank you, Sek. So the heart-1 trial is a Phase Ib trial of VERVE-101 in patients with heterozygous FH. It's open-label, single ascending dose design with flexible and adaptive dose levels with about 3 to 6 patients in each cohort. Today, we are able to share the interim update with 10 patients dosed across 4 dose cohorts. We have 9 patients treated at 0.1 mgs per kg, 0.3 mgs per kg and 0.45 mgs per kg and 1 patient dosed in the 0.6 mg per kg cohort. The data cutoff date for this presentation was October 16, and the study is continuing to enroll and accrue more data. I think it's really important to pause on the decision of what kind of patients to enroll in a Phase I study. So you may be familiar with a lot of the Phase I studies that were run for siRNAs or monoclonal antibodies where typically you run a Phase I study in healthy volunteers. The FDA issued this guidance in 2022 that for gene -- genome editing products, they recommended that subjects with severe or advanced disease would be most appropriate for first-in-human gene editing trials. And they actually also called out that these patients, however, may be more predisposed to experiencing adverse events or to be receiving concomitant therapies that make it difficult to interpret safety. Now we were guided by this guidance in choosing to enroll patients with severe advanced heterozygous FH. And that was a very conscious decision. What you see here is the study design. The study population is males and females, 18 to 75 with heterozygous FH established atherosclerotic cardiovascular disease, that means they have atherosclerosis, they've had myocardial infections or strokes. They have uncontrolled LDL cholesterol that are not at goal, and that's despite maximally tolerated oral lipid-lowering therapy. The drug is administered with a premedication regimen of steroids and antihistamines that has been established in the field from other lipid nanoparticles. And it's administered as a single infusion via peripheral IV. The endpoints were safety and tolerability, pharmacokinetics of VERVE-101 and blood PCSK9 and LDL-C values, which we measured at -- as a time average from day 28 onward. The study duration is 1 year with an additional long-term follow-up study that patients roll over into per guidance from FDA and European regulators. So the participants enrolled to date, the 10 participants have had severe advanced ASCVD and a high risk for cardiovascular events. Their mean age was 54. Their mean screening LDL-cholesterol was 193. That's on treatment. There are 9 out of the 10 had a mutation in their LDL receptor. The majority of them, nearly all of them have had prior coronary revascularization. These include both coronary artery bypass grafting and prior PCIs, in some cases, multiple PCIs. And about half of them had, had a prior heart attack as well. One of them had, had a prior cardiac arrest as well. Most of them, nearly all of them were on a statin therapy. None were currently on a PCSK9 inhibitor, which was an exclusion for the study, but 2 had been previously on a PCSK9 inhibitor. So the results are on the next few slides. We saw dose-dependent reductions in PCSK9 following treatment across the 4 dose cohorts. There was minimal effect at the first 2 dose levels, but dramatic reductions in PCSK9 at the higher 2 dose levels, 47%, 59% and 84%. I think it's important to say that when you look at the PCSK9 in the blood, about 85% of it comes from the liver. The other 10% to 15% comes from other organs. We're only editing the liver. So we expected an upper bound on the PCSK9 reduction of about 85% to 90%. So in that patient where we saw 85% reduction, that suggests that we were able to edit nearly every hepatocyte and make a single base pair change in both copies of the PCSK9 gene in that individual. Looking at the LDL reductions, we saw clear dose-dependent reductions in LDL cholesterol following treatment. The reductions in the higher 2 dose cohorts were 39%, 48% and 55%. These are, again, presented as time-weighted averages after from day 28 onward through the follow-up. Each -- the bars here are by cohort and cohort average and each dot represents each patient, suggesting that the result here looks very consistently dose responsive. I just want to pause on this to show -- to say that this is the first evidence of in vivo gene editing, having a clinical benefit for patients in LDL lowering, which is, as you know, a surrogate endpoint accepted by the FDA as a clinical end point. This is the data looking at durability. On the x-axis here is time and days, on the y-axis again as percent change from baseline in LDL cholesterol, the lines are for each of the dose cohorts, the number of participants in each is at the bottom. And you see that the dose reduction, the LDL reductions that we're seeing are durable. There's a 55% reduction at day 28 in the patient -- participant -- patient at 0.6 mgs per kg, and that has been rock-steady out to day 180, 55% reduction in LDL cholesterol. Now this is very consistent with the preclinical data that we've seen, where we saw reductions of LDL cholesterol being durable out to 2.5 years. Turning to safety. The observed adverse events were largely consistent with the severe advanced ASCVD patient population that we enrolled in the study. So I'm going to focus on a few different aspects, looking at the adverse events that occurred in more than one participant, the infusion reactions were common at the higher doses, not at the lower doses. These were typically mild headaches, body aches, fever and resolved within a few hours. And these were largely expected based on prior lipid nanoparticle experience. COVID was the -- COVID and upper respiratory infection was the only other adverse event to occur in more than one participant. On the serious adverse events, we saw the 2 cardiovascular serious adverse events that we'll talk about in another slide, we saw no noncardiovascular serious adverse events. And this was really, I think a key point for us as we were thinking about this study where we did expect cardiovascular adverse events, but we're looking at the noncardiovascular events. Any treatment-related AEs Grade 3 or higher increased liver transaminases, we saw in 1 participant at 0.6 mgs per kg. We'll show you that in a second. Again, transient reversible and then the cardiovascular event that we'll talk about. So these are the ALT elevations on the left. They are dose responsive. They are transient. They are asymptomatic, and they recover within about 2 weeks to baseline. These are essentially the same as what we had seen nonclinically and nonhuman primates, and based on our discussions with KOLs and other experts on the FDA's advisory committees for a onetime therapy, we thought this was going to be not an issue. There is no bilirubin changes on the right reassuringly. So about those serious adverse events, we enrolled patients who had advanced coronary disease. The first participant who had the fatal cardiac arrest had an ischemic cardiomyopathy and have had a prior cardiac arrest. The autopsy showed severe underlying coronary artery disease as the cause of death with no evidence of pulmonary embolism, myocardial inflammation or coronary thrombus. And the investigators in DSMB reviewed this case carefully and deemed it unrelated to study treatment. In the second patient, there was an MI and nonsustained ventricular tachycardia. So this patient had a fairly complex cardiac history of having multiple prior interventions and multiple prior heart attacks. This patient, in particular, had had an unusual form of coronary disease called in-stent restenosis. So this is where you put a stent into a blockage and then smooth muscle proliferation happens inside the stent. This is different than the typical plaque that builds up in coronary arteries. It's smooth muscle constriction of the vessel. This patient had severe in-stent restenosis that had recurred 3x. Recurrent treatments have been unable to resolve it. Patients with in-stent restenosis have very high risks of adverse events, MACE events over 1 year, in some prior studies as high as 30% or more. On coronary angiography, this patient had critical left main equivalent coronary disease with in-stent restenosis lesions in both the LAD and circumflex which were felt to be preexisting prior to treatment, of course. And in particular, this patient, it turns out -- and this is unfortunate in a clinical trial, of course, had been having unstable symptoms that had not been disclosed to the investigators prior to treatment. The investigators in DSMB reviewed the case with us and determined that the MI was potentially related, largely related to the proximity of the dosing, the NSVT was unrelated and recommended that we continue dosing with no required protocol changes. So what are the next steps? We are actively enrolling in the 0.45 mg per kg and 0.6 mg per kg cohorts to complete the dose escalation phase. We have active sites in New Zealand and the U.K. With the IND clearance in the U.S. just a few weeks ago, we are working to activate U.S. sites to participate in the study as well. Our plan is to complete this ascending dose phase and then enroll an expansion cohort in 2024. And then based on a comparison of the composite data between this program and the VERVE-102 program that we'll talk about in a minute, initiate a randomized and placebo-controlled Phase II study in 2025. So in conclusion, I think what we saw today is that we demonstrated the first proof-of-concept for in vivo DNA base editing in humans. We saw dose-dependent reductions in blood PCSK9 and LDL-C following VERVE-101 infusion. The LDL-C reductions were clinically meaningful, 39%, 48% and 55% among the patients in the 2 higher dose cohorts. And we have durability data out to 6 months in the patient in the highest dose cohort. Obviously, we'll be continuing to follow those patients for years to extend that durability. And based on the preclinical data, we have every reason to believe that, that durability will extend potentially for a lifetime. And the safety profile was -- supports continued development of VERVE-101 and was largely consistent with what we expected in a Phase I study in these high-risk patient population. So we are very excited to expand the inclusion criteria as we move forward to enroll more patients and build the data set behind VERVE-101. I think Sek -- and of course, I should say thank you to, most importantly, the patients and their families who volunteered to participate in a very novel gene editing trial for their heterozygous FH. And also to the investigators and staff who have been working with us on the study and calling out, I think in particular, the NZCR Group in New Zealand, and the Richmond Pharmacology Group in London, who've been terrific investigators to work with and also to a really strong DSMB, who's been very actively involved in the study with us. So thank you.

Okay. So taking a step back, where could VERVE-101 fit into the LDL cholesterol treatment landscape. We think ultimately, a good landing point would be in patients who require LDL cholesterol lowering, deep LDL cholesterol lowering over decades. So if you think about the patients we enrolled, the average age was 54. Those patients are facing 20, 30 years of care. So single-course gene editing medicines may emerge as an option for those kind of patients to overcome the limitations of the chronic care model. Now let me move to a couple of updates beyond the data. The U.S. FDA cleared our IND application for VERVE-101 in patients with HeFH. This allows us to initiate a trial in the U.S. The FDA reviewed our complete response, which included comprehensive experiments to address preclinical requests, but even more importantly, it included the heart-1 clinical trial data set that you just saw. We have plan to activate U.S. sites for VERVE-101 development. And this is the first IND for in vivo base editing. And our experience interacting with the agency over the last year generated a lot of learnings that are going to be valuable as we move forward with all of our future programs, VERVE-102 to VERVE-101. In addition, we are thrilled to gain Eli Lilly as a collaborator for our PCSK9 and ANGPTL3 programs. As you likely know, Lilly purchased the rights to opt in for the PCSK9 and ANGPTL3 programs from Beam Therapeutics in a transaction with a total potential deal value of $600 million. Lilly's opt-in rights are characterized as follows: in exchange for paying for 33% of worldwide development costs and 50% of U.S. commercialization expenses, Lilly receives the right to 50% of U.S. profits. Ex U.S. is retained by Verve. Verve retains control of development and Verve books revenues. Lilly brings tremendous know-how in the cardiometabolic space. They bring resources. They bring expertise in late-stage drug development and commercialization for cardiovascular indications. And we have a shared vision around application of gene editing to treat HeFH as well as ASCVD. Again, we couldn't be more excited about this development. As part of diligence, Verve provided to Lilly the heart-1 clinical trial data as well as well as preclinical data for related CV programs. Now let me walk you through the rest of the pipeline. We've talked a little bit about VERVE-101. I'm going to cover VERVE-102 in a minute, both VERVE-101 and VERVE-102 target PCSK9, and they use the base editing technology. Our next target is ANGPTL3, VERVE-201. That will -- that is attempting to go after 2 indications, homozygous FH and refractory hypercholesterolemia, again, base editing technology. And we have a program with targeting LPA. And here, we'll not be using base editing technology, but rather a custom editor we're developing for this target, custom editing technology we're developing for this target. And then we have an undisclosed target liver target that's also atherosclerosis target, that currently is we're planning to use a base editor. And lastly is an undisclosed target that's partnered with Vertex. And this is the only one of our programs that is not ASCVD based. This is really a straight-up liver disease that we're partnering with Vertex to work on. So VERVE-102, so VERVE-102 is an adenine base editor, mRNA and a guide RNA, exactly the same construct as VERVE-101, but packaged in a different delivery vehicle, a GalNAc-LNP that's shown in the middle and designed, again, the edit is designed to turn off the PCSK9 gene. So VERVE-101 is differentiated from VERVE-102 in a couple of ways. VERVE-102 has 2 components of the LNP that are different. One is the ionizable lipid. For VERVE-101, the ionizable lipid is licensed from Acuitas, for VERVE-102, the ionizable lipid is licensed from Novartis. In addition, VERVE-102 has a targeting ligand, GalNAc that binds to a receptor on the surface of liver cells, ASGPR. So VERVE-102 can get into liver cells by any of 2 receptors, either the LDL receptor or ASGPR. Now these differences may lead to improved potency for VERVE-102 compared to VERVE-101 and/or tissue specificity. Here's some data for VERVE-102 in nonhuman primates. On the x-axis is time, on the y-axis is LDL reduction in controlled animals as well as VERVE-102 -- as well as VERVE-102 treated. And you can see there's a nice reduction in LDL that's durable here in this experiment out to 6 months. So we have very good data in nonhuman primates for both VERVE-101 and VERVE-102. Our plan is to simultaneously develop both, followed by the selection of one product candidate to take to Phase II. You can see on the time line here that we are currently in the midst of Part A for VERVE-101, that's the dose escalation phase. We're going to transition to the Part B, the dose expansion phase in '24. In parallel, we'll be conducting the Phase I for VERVE-102 in '24. And then based on the human data in both -- from both products, we'll select 1 of the 2 to take forward into a Phase II in 2025. Let me move to VERVE-201, VERVE-201 targets ANGPTL3, a compelling target with human genetics and pharmacology validation to lower LDL cholesterol via a mechanism that is additive to PCSK9. That's very important to emphasize. You have statins, and then you have PCSK9 inhibitors lower LDL by about 50% to 60% on top of statins. And then ANGPTL3 mechanism will lower LDL by an additional 50% on top of PCSK9 and statin. You see the product on the left, VERVE-201, which is an mRNA for the base editor, a guide now that localizes to the ANGPTL3 gene. And here, we're using our proprietary GalNAc-LNP delivery technology. In the middle, you see some preclinical data that we've shared publicly before, I'm summarizing here. We created a monkey model of a disease, homozygous familial hypercholesterolemia. We treated those monkeys with VERVE-201 and saw about a 45% to 50% reduction in LDL cholesterol, and that's the data that's shown in the middle. We're anticipating clinical trial initiation for VERVE-201 in the second half of 2024. So what does 2024 and 2025 hold for Verve? Here are the anticipated milestones across the 2 programs for 2024 and 2025. Let's start with PCSK9. For VERVE-101 in '24, we'll complete enrollment in the 2 high-dose cohorts. So that's Part A. That's the dose escalation phase, that's the 0.45 mg per kg and 0.6 mg per kg. We expect to go from the 10 patients we have right now to about 15 patients in Part A. And then in Part B, we'll be expanding one dose into a larger number of individuals. That's a dose expansion phase. And then we'll pick a dose or 2 to take to Phase II based on that. For VERVE-102, we'll initiate a Phase I in the first half of 2024 and enroll that study. For VERVE-201, '24, as I said, we'll hold a -- we'll initiate the clinical trial in the second half of '24. The anticipated milestone for 2025 are shown on the right. We'll have clinical data for both programs, VERVE-101 and VERVE-102. We will select one of the 2 to take to a randomized, placebo-controlled Phase II for the PCSK9 program. Okay. Let me close with the next couple of thoughts. Come back to the beginning. We're focused. We're well positioned to realize our vision of developing one-and-done treatments for ASCVD. We have human proof-of-concept. We have several product candidates against highly validated ASCVD targets. We have an emerging regulatory path in the U.S. with the IND clearance. We have a new partner in Lilly with a shared vision to CV development expertise as well as commercialization strength, and we're well capitalized with about $485 million in cash and a runway in 2026. So if you think about Verve, started in 2018 with the vision to develop single-course gene editing medicines. 5 years later, we have -- we've gone from concept to proof-of-concept in humans for a first-in-class base editing treatment. And in the next 5 years, we are thrilled and excited to execute on our vision to get to single-course treatments for a range of indications. Thank you. Okay. We'll now move to the Q&A portion. I think we have a couple of folks that are going to walk around with mics. And we're here to stay all night.

Whitney from Canaccord. First question, can you walk us through the dose extrapolation, I guess, from nonhuman primates to humans with VERVE-101, and just kind of how you arrived at the doses that you used? Part 2 of that question, I guess, is do you have room to dose up if you decided you wanted to with VERVE-101? And then sorry, third part is how would you think about dosing VERVE-102 based on the nonhuman primate data you showed, I think which looked about to be similar LDL lowering at 3 mgs per kg as we saw 1.5 mg per kg in the nonhuman primates with VERVE-101?

Yes. That's a -- those are great questions. On the first question about the translation from monkey to human, we were thrilled with the translation here. The monkey data predicted a safe starting dose and has predicted an efficacious dose and a range that really covered that full gamut. And so in this study, we do have the ability to go even higher above 0.6 mg per kg. And that decision will be made based on what we see with the rest of the patients at 0.45 mg per kg and the additional patients will dose at 0.6 mg per kg. Okay. And then I think, Whitney, what was your second question? Dosing of VERVE-102.

VERVE-102, I think you answered the first 2 together. But yes, VERVE-102.

VERVE-102. Yes. So the VERVE-102, I think the starting dose from VERVE-102, we've not guided to that. But I think one thing to note is this LNP is different from the first LNP, the VERVE-101 LNP. This LNP uses an ionizable lipid that's licensed from Novartis. That's VERVE-102 uses the ionizable lipid license from Novartis. That is the same ionizable lipid that is being used in Intellia's products. And if you look at the Intellia's preclinical data in nonhuman primates, the 2 doses they study in nonhuman primates are that they've reported publicly are 3 mgs per kg and 6 mgs per kg. So Whitney, I think that will give you a sense of how we might end up going in terms of its target dose.

Kostas here from BMO Capital Markets. Congrats on the data. A couple of questions from me. The first one is based on the data we saw, do you think there is any impact on the therapy from the partially impaired LDL receptors that these patients have or do you think this has no impact at all on the efficacy? And I have a couple of follow-ups.

Well, you can see that we've gotten the efficacy band that we've been talking about for now a few years, this fits squarely into the target product profile that we set out to accomplish about a 40% to 60% LDL lowering. And as you know, the current medicines that are out there, and this may be -- there may be a bit of misunderstanding about what one should expect in terms of LDL lowering with maximal PCSK9 inhibition. So the antibodies pretty much reach maximal PCSK9 inhibition. And the antibodies in HeFH patients, lower LDL in HeFH patients by about 50% to 55%. It's a little higher in non-HeFH patients. But so we're right there. We have to dose more patients at 0.45 mg per kg and 0.6 mg per kg. But we're seeing very good efficacy despite this 50% LDLR deficiency. So I think this LNP, as we hypothesized, there's enough LDL receptor in these patients to be able to take -- to be taken up and get the appropriate level of editing. Again, we have to dose more patients at these 2 dose levels, but what we're seeing so far fits within what our expectations were.

Perfect. And maybe one on the ALT elevation. Is there any way to characterize what creates those ALT elevations are? And do you think FDA will push back or other agencies will push back on these ALT elevations, cognizant that this is a onetime therapy, the ALT elevation is transient and may not be important. But any color around this ALT elevation and the extent of that would be helpful.

Yes. We've characterized it here. It's a grade 3 ALT elevation into [ 0.6 ]. Andrew, do you want to talk a little bit about...

There's pretty standard adverse event scoring criteria for laboratory abnormalities like this, and we use those standard grading systems. So this adverse event was Grade 3. It was transient. We don't think that this is going to be dose limiting for a onetime therapy. Obviously, it'd be different if it were a chronic therapy, but these levels of ALT elevation, it's transient, it's asymptomatic, there's no long-term injury to the liver. And there's some standard kind of ways of thinking about those liver risks like looking at for elevations in bilirubin, such as [indiscernible] type findings that we did not see. So we think this is a safety profile that would not preclude a broad administration profile.

And the FDA reviewed these data before allowing the U.S. study to go forward with the dose range that we proposed here.

Okay. And maybe one follow-up here. Is there any way with modeling, given the dose response to predict what dose would drive a higher grade of ALT elevation, for example, Grade 4 or is not possible?

We've seen fairly consistent dose response for ALT elevation, and so far in 10 patients, it's a limited end. But I don't think that there's a -- I guess you probably could model, but I think we don't feel like we will be dose-limited by ALT elevations is probably the best way of putting it.

Seamus Fernandez from Guggenheim. Just a few quick questions. One, a clarification on all the ALT questions. Just to keep it simple. Did you provide all of this data to FDA prior to them signing off on initiating the clinical program?

Yes.

Great. And did FDA require or ask for any protocol changes at all versus the data that you use now?

No.

Okay, that I think simplifies a lot of that. Second question is really just from an LDL and kind of efficiency of editing perspective, can you just help us understand how -- what you would view kind of -- it looks like with the 0.45 mg per kg, we got sort of the evidence of maximal editing. With GalNAc, I think we saw an extraordinary kind of PK profile from actually Lilly's LPA RNAi. And that PK would imply very, very low risk of off-target editing really just going right after the liver. So in terms of GalNAc, what would kind of keep you on this VERVE-101 path instead of kind of moving to the full GalNAc program with this and with your ANGPTL3?

Yes. I think for VERVE-101 versus VERVE-102, as I say, we're excited by both. We have good data so far in human for VERVE-101. We'll complete that. And then we'll have to see what VERVE-102 shows. We're super excited about VERVE-102 for the reasons you mentioned. The addition of GalNAc might give it more potency, might give it more specificity. I mean one of the possibilities, Seamus, is that we get similar levels of efficacy between VERVE-101 and VERVE-102, but what if VERVE-102 is able to do it at 0.3 mgs per kg. That would be a huge -- have huge impact for the product going forward. So that's the kind of possibility that we might see, but we have to show in patients coming up.

Great. Luca Issi, RBC Capital Markets here. Maybe 2 quick ones. Maybe, Andrew, if I may. Can you just talk about the MI and why was it deemed potentially drug-related? Was it just the chronological proximity to the infusion or were there any other factors that led to that conclusion? For example, did that patient also experienced material ALT elevation? Just trying to understand if this MI was actually inflammatory in nature or not is a question I think came up during the Q&A today. And then second, maybe going back to Kostas' questions. I know you're not flagging the data by genotype, but it looks like you had 1 patient that had LDL receptor that was intact. What was the knockdown in that patient versus these -- the other patients at the same dose? Again, just trying to understand whether the LDL receptor here plays a role in the update.

Yes. So the determination of relatedness on the MI was based essentially entirely on the proximity to dosing. There was no substantial difference in any of the acute markers in that patient compared to any of the other patients in the higher dose cohorts. So there was nothing unique about that patient from their response. I think it truly does signify the more unstable cardiac physiology in that patient. With respect to your second question, I think it's obviously an understandable question. It's certainly heterozygous FH patients in general, as Sek alluded to, have had slightly less LDL lowering historically in lots of other trials. So it's reasonable to kind of ask that question. It's a low end at this point, and I think -- we think it's probably not a good idea to try to over interpret [ Nf1 ] type data in that context. So...

It's just too early to do subgroup analysis when you have 10 patients and 1 versus -- but I think this is something that we can test out as we enroll more patients. But I think there was a hypothesis going in by many that the LDLR deficiency would really impact any efficacy. And clearly, that's not the case. We're getting exactly the TPP that we're looking for despite the fact that these patients have, most of them, almost all of them had an LDLR mutation.

Myles Minter from William Blair. I think people are trying to reconcile the 0.6 mg per kg, patient, obviously, the PCSK9 came knockdown and then the greatest LDL-C lowering that we saw there. My question is on -- there's a lot of patients in this trial on concomitant statin therapies. Did you monitor compliance in that particular patient that come more compliant on statin therapy throughout -- obviously, if you take statins, it's going to elevate PCSK9 but also lower...

Yes. That's a great question, and that's not an issue here. We've been very, very careful about communicating both to the physician and also the patient that the background therapy stay the same. And that's been monitored for, and there's been no change in that patient in the background therapy. I think I want to just address this kind of just with data. I think that people have asked, okay, you're seeing this 47%, but the LDL effect is maybe a little more than -- it's 55%. So what's going on there? And Andrew address this question, but we'll just bring this back during the Q&A earlier today. But the relationship between the degree of PCSK9 reduction and the degree of LDL reduction is a population relationship across hundreds of people or if not thousands of people. In any given person, that relationship actually can be quite a bit of a scatter variable. And this is data -- our data from nonhuman primates. So each dot here represents a non-human primate, on the x-axis is the percent change in PCSK9, on the Y-axis is a percent change in LDL. And you can see aggregated, there is a strong pattern of correlation, lower PCSK9, lower LDL. But there are a lot of also individual data points that could actually be representative of this patient that we just -- in [ 0.6 mg per kg ]. So look at that data point at minus [ 40 ], that animal had a minus [ 40 ] in terms of PCSK9 reduction, but an 80% reduction in LDL. So this is just to make the point that in any given individual, there could be some variability. But when we get to larger numbers, we're going to see this relationship, predictable relationship.

Right. And this has been seen before with other PCSK9 therapies as well that SKU and the individual patient response levels.

Just a quick follow-up. Did any patient on the 0.45 mg per kg or the 0.6 mg per kg achieved normal levels of LDL-C according to either ASC or...

Yes, we haven't started to look at that, and this is really too early in terms of clinical development to have that as the goal because you saw the patients are 190 when they started. So I think as we get to later stages, Phase II, Phase III, I think the proportion of patients who reach on LDL goal would be a meaningful endpoint, but it's too early to kind of go there. We're really just trying to understand PD.

Dae Gon from Stifel. And congrats on the data today. A couple -- maybe I'll start with a clarification, just going off of Seamus' earlier question on clarification. The MI as well as the other CV SAEs, were they also presented to the FDA as part of their clinical hold resolution? And how do they feel about going into sort of the earlier/less severe patients having seen that data? Maybe I'll follow up after your answer.

Yes. I guess I'll just repeat myself. Every single piece of information that you saw today has been shown to the FDA and much more. So yes, the answer is yes. And then let me just show you in terms of the -- so this guidance from February -- so March last year, I think is very relevant kind of where we have been, where we're going. So we should all just read this. So subjects with severe or advanced disease may be more willing to accept the risks of an investigational human genome editing product. However, these subjects may be predisposed to experiencing more AEs or be receiving concomitant treatments, which could make the safety or effectiveness data difficult to interpret. So then this last sentence, I think is your answer, which is, therefore, in some instances, subjects with less advanced or more moderate disease may be appropriate for inclusion in first-in-human clinical studies. And so this is the transition we're making, where we now have proof-of-concept, and we're going to be able to move to patients with less advanced to more moderate disease.

So perhaps this next question is irrelevant in that respect, but that second patient who did not provide that medical history, if you will, how do you going forward allow or I guess in-force collection of the thorough data such that these types of patients might be somewhat excluded from inclusion going forward? And then one more follow-up after that.

Fred, do you want to address that in terms of the...

Yes, it's a very good question we thought about it. And I think, again, I want to reiterate that it's the CV events, MI and other serious CV events are really not unexpected in this kind of population. And so -- but we are going to be adding as we expand to the moderate population, we'll be getting more -- a broader population to avoid that, but we're also going to be adding procedures to CT angiogram, especially in Phase I before we have the chance to do a placebo comparison to really look and understand the patients and really focus in on the ones that have a high lifetime risk, including the moderate risk and trying to avoid any with acute -- any acute risks as well.

Great. Last question. The -- there seems to be some confusion around the 0.6 mg per kg durability curve being farther out versus 0.45 mg per kg. If you could remind us sort of the trigger point, which one started first? And what was the ultimate decision point, which data kind of led you to go down to 0.45 mg per kg? Was it PCSK9, LDL, safety, all of the above?

Let me show you that in a second. May I broke it. Andrew, do you want to cover this?

Yes. So you're exactly right. The progression through the cohorts was 0.1 mgs per kg, 3 patients, 0.3 mgs per kg, 3 patients. And then we dosed a patient at 0.6 mgs per kg. As the study design was flexible adaptive dose levels, and so when we saw this patient at 0.6 mgs per kg, see a 55% reduction, that indicated that we had skipped over the dose response portion of the dose response curve. And so we wanted to electively enroll patients at 0.45 mgs per kg, so we could see the dose response curve. That is at the end of the day, the purpose of a Phase I study is not to hit 55% LDL lowering. It is to understand the dose response curve. And so that's why we did that. It is true that we saw the ALT elevation, and that indicated that we had -- we were crossing over into a different regime to a degree. So we elected to enroll some patients at 0.45 mgs per kg. We have every intention of going back to 0.6 mgs per kg as we've articulated, and we'll be doing that over the next few months.

All right. Great. Rick Bienkowski from Cantor Fitzgerald. So since we have individual patient-level data here, I was wondering, do we know which 2 patients were on statins were treated with VERVE-101 and if you saw any differences in response to PCSK9 knockdown in those patients? And my second question, I don't think we saw a time course of the PCSK9 knockdown only the time leverage there? So I was wondering if you could qualitatively talk to the durability of PCSK9 knockdown over time? And if it kind of matched what we saw with the LDL-C response?

Yes. So again, the same caveat about subgroup analysis applies here for -- with respect to 2 patients that were not on statin. It is also the case that those were 2 of the patients in earlier dose cohorts. So it's essentially impossible to kind of answer that question. With respect to PCSK9 durability, it largely was the same. So we've seen durability in the PCSK9 reductions. PCSK9, as you probably know, is a little bit more variable. So the error bars are a little bit bigger, but the exact same durability has been seen, particularly at 0.45 mg per kg and 0.6 mg per kg.

Cory from LifeSci Capital. Touching upon the really unfortunate incident of a fatal cardiac arrest, a lot of times in gene therapy studies, we sometimes can take these unfortunate events and gain some future benefit from it, whether it's through looking at biodistribution or in this case, looking at potential for off-target edits or editing efficiency. Were you able to collect any samples from that patient for data down the line? And if so, are you planning on presenting that?

No.

And second one from us, can you speak a little bit more to the potential mechanism of ALT elevations with lipid nanoparticles and thinking about the switch of ionizable lipids, is there a potential for an improved safety benefit moving from Acuitas to Novartis?

Andrew, do you want...

Yes. The -- I think the mechanism is -- of the ALT elevation is largely driven by the ionizable lipid itself, and that is largely essentially the amount of lipid that gets accumulated in the liver. So it is dose-dependent and is ionizable lipid-dependent. I can say that in our preclinical data, the Novartis lipid has less liver, less ALT elevations and presumably less liver inflammation than the VERVE-101 Acuitas lipid. And I think there, we do think that the Intellia experience with that lipid is likely relevant, where they have not seen significant ALT elevations in humans up to 1.25 mgs per kg roughly. So I think that is another potential benefit of VERVE-102. Yes.

Andrea Tan, Goldman Sachs. Maybe a follow-up question there on the lipid that's being used in VERVE-102, which you've referenced Intellia using. Just given where they are right now in terms of performing studies to assess fertility risk coming from the chemicals from that lipid, just curious if you're planning on any of these types of studies ahead of your own IND submission here to maybe preempt the FDA's request?

We have performed those studies ourselves. Yes.

I have a few questions from the webcast. The first few come from Eun Yang from Jefferies. 2 patients in the study had prior PCSK9 therapy, why did they discontinue the prior treatment and at what doses were they treated in the Verve trial?

I don't -- we don't have that off the top of our head. But I think the reasons people discontinued PCSK9 therapy, a range of different reasons in terms of costs or just a bunch of other.

I think as you know, about 50% of patients on the PCSK9 inhibitor will not be on it about 1 to 2 years later. So these patients would fall pretty typically into the pattern of past PCSK9 usage.

The second question I have from Eun is Verve is continuing to enroll patients in the 0.45 mg per kg and 0.6 mg per kg cohorts to complete the dose escalation phase. How many more patients are you planning to enroll in those cohorts?

We're planning to go to a total of 15. So we're right now at 10. So the 0.45 mg per kg will go to 6 total and the 0.6 mg per kg will go to [ 3 ].

And then I'm going to switch to Eric Joseph from JPMorgan. His question is, how, if at all, are the patient entry criteria in the heart-1 study being modified with cohort expansion?

Fred, do you want to...

Sek, was the question for the cohort expansion in the second quarter? Okay.

How the criteria being modified?

Yes. The modification is to actually go to the more moderate patients. We focused in on the heterozygous FH and severe advanced cardio -- atherosclerotic disease. But now that we've got the information we have, we are able to open it up to more moderate patients. And we're also, as I mentioned earlier, during the CT angiogram to continue to screen in Phase I.

Dae Gon from Stifel, again. Two questions on 102. Does this mean that once you do clear the CTA or IND in first half '24, you would need to do the same thing, go for the severe as the first initial set of cohort of patients? That's a no. So presumably similar as the expanded entry criteria that you're going to use, when would you decide for your Part C, the 0.1 mg per kg and 0.3 mg per kg dose patients, when would you decide on the final go-forward dose? And when would they be redosed with the new now go-forward dose?

The idea for Part C is to pick the dose that we take forward to dose expansion. So once we have that defined, that would potentially be the dose they get. In terms of the timing of the Part C, Andrew, do you want to...

Yes, those [ are pretty staggered ] and then we initiate Part C after Part B is initiated.

Luca Issi, RBC Capital Markets. One big picture question here, maybe for our KOL. I think when we speak to some of your colleagues, there are clearly 2 camps. There's Camp A, clearly enthusiastic about the promise of one-and-done approach that can solve compliance. However, there is Camp B that notes there's lots of other drugs out there to lower cholesterol and then maybe Verve is over-playing a little bit the compliance issue. You're clearly Camp A, but what do you think -- what kind of data, if any, do you think can be -- need to see in order to become Camp A?

Yes, I think that's a great question. And I guess you're right. I am in Camp A, but that's how physicians often are -- some are really enthusiastic about new approaches. Others are skeptical. That's just human nature and physicians are no different. I think the key though when we're talking about LDL control is what real life shows and it shows that a good chunk of patients are off their generic statins. Why is that? You can't blame cost there. The drug is cheap. It's just for a variety of reasons that patients even ones that have had heart attacks just aren't 50% are off their statin in a year as the registries repeatedly show. Dr. Bellinger already mentioned with the PCSK9 inhibitors, a similar sort of issue of patients coming off, their costs might have a little bit to do with it too. But certainly, it's just the idea of acknowledging that one is sick and getting repeated pills, repeated injections. So I do think the idea of a one-and-done really will be appealing to patients and to physicians. As far as the ones that are skeptical, I think durability data is going to be what convinces them. So the data out to 6 months looks pretty durable to me. And if that continues with longer follow-up, that's the sort of data, I think will make skeptics into believers.

This is Kostas from BMO Capital Markets. Maybe a couple of questions here, more like a general level base editing. I think we are forgetting a little bit about, this is the first time that we see in vivo data from our base editing. Any high level or even detailed lessons learned around base editing that we didn't know before? And maybe how would you compare it now to other approaches, CRISPR-Cas9 or potentially other approaches that are emerging? And then a follow-up irrelevant to the first one is based on the time line you presented, when exactly will Lilly be deciding whether or not to opt in for the programs that you can potentially partner?

In terms of the Lilly question, so they have the ability to opt in to VERVE-101 at the end of Phase I. They also have the ability to opt in at the end -- for VERVE-102 at the end of Phase I as well. So they're -- basically, each product, they have the ability. Our sense is that they'll likely make that decision after seeing both VERVE-101 and VERVE-102 data. So that's likely when they will decide. Yes, that would be about right. Your first question is?

On base editing.

It works. We didn't know that before. And so I think that's a huge step for that technology. We're going after this disease. I think we're finding efficacy. I think there's a range of other diseases that could be approached now with base editing for liver at least even. So I think it puts that technology in a good place.

And maybe one more. We saw that there are different approaches emerging in cardiovascular disease. One is epigenetic editing, and we recently saw some data here at the conference. Any comments around epigenetic editing and maybe pros and cons around that versus base editing?

Yes, no, I don't want to go too far into these other technologies. I think everybody is trying to develop specific areas of benefit that they might have over others. I think we feel very comfortable with the unmet need here why we're doing this. There is a huge unmet need, despite all that's available, this number here, right, 3% goal, I mean, that's the fact. So I think -- and we have an approach of one-and-done that I think could solve this or help solve -- help -- at least provide another option for this. And so -- but this is just one technology base editing. Others may come in and say, all right, I have this other technology for one-and-done as well, and I think that's great. Ultimately, if they can get it to work, it would be great for patients to have multiple options. Okay. All right. Wonderful. Thank you so much for joining us on this very special day, and look forward to staying connected. Thank you.