All right. Good afternoon, and thank you for joining us at Guggenheim Second SMID Cap Conference. I am Debjit one of Therapeutics Analyst here, and joining me from Beam is CEO John Evans. John, thank you for your time despite the miserable weather.
Great to be here.
I know there is a huge amount of focus, or let me take a step back. A year and a half ago, or two years ago, I know the company used to get a lot of flack because INDs cleared, or INDs are not getting cleared, but other patients. 2025 seems to be that year where you'll have data from not one program, but up to three. With that, maybe I'll hand it over to you for a quick rundown.
Yeah, it's a great point. You bring us back to some of our transitional moments. So yeah, Beam is developing next-generation gene editing technology in CRISPR, it's called base editing. And so this is an improvement, we think, on the first-generation technologies, which do an amazing job targeting within the genome precisely, but they can only make cuts. It's double-stranded breaks. And base editing is a more, we think, elegant, precise, and efficient way to make more controlled, single-base changes in the genome. The technology builds on CRISPR, so it wasn't even published for about four years after the original CRISPR wave. And so we've been, in some ways, playing catch-up in terms of our execution. But I think we've sort of finally arrived, and I feel really good about that on multiple levels. One of them is execution, just as you noted.
At this point, we have four different drugs with regulatory approval for treatment in the clinic. That includes both ex vivo and in vivo, seven different countries, or five different countries, seven different approvals around the world. We've built manufacturing. Our clinical engine is up and running. Just as you noted, it was a long time to get the first patients in, and sickle. It was coming out of COVID. It's a difficult process, of course, in that case. At this point, we have over 40 patients enrolled in that trial. It's really moving quickly now. I think part of that is we've been able to build the engine fully. Part of it is, of course, the enthusiasm for the drug and what's been seen with it.
And so we are very much in, I think, that important window of time for the company, where we have our first clinical data now coming out that is finally, hopefully, going to show that base editing does indeed bring new things to the table and open up new possibilities for transformative medicines relative to what has been possible in the past. I think we started that in December with BEAM-101, and that was at ASH for sickle cell disease and showed on a variety of levels. We feel that there's good evidence there, obviously early days with seven patients of data, but good evidence that there's real differentiation from base editing in sickle cell disease. That then transitions us to this year when we'll have our first in vivo data with Alpha-1 antitrypsin deficiency with BEAM-302, and that'll be in the first half of this year.
And then on the heels of that, of course, sickle continues to progress. We'll have our BEAM-301 program for glycogen storage disease, another in vivo program, which will be dosing patients soon here in the U.S. And so it's a very catalyst-rich period for us, obviously, that we're in now over the next 6 to 12 months. And we think it sets us up, hopefully, to help lead the field and create some exciting medicines for patients.
This is obviously not the first time you're going to see in vivo base editing data. We've had proof of concept from Verve. Obviously, the discussion immediately switched over to LNP and the formulation. You guys invest a lot of time and effort on formulation. Maybe just walk us through the LNP angle of it.
Yes. Yeah, so LNPs are critical. So LNPs are one of our favored ways to deliver within the body. They are, of course, now well validated with vaccines and Moderna and Pfizer and others. And when you inject them into the bloodstream, they go to the liver. And this is a great way to carry our gene editing machinery into hepatocytes, which is the cell that we want to target for a lot of liver diseases. So the pioneers in this space, it begins with Alnylam. The first LNP is using short oligos as the payload, and that's Onpattro for TTR. It was then really Moderna who pioneered how do you package large, long mRNAs into these LNPs. And it's tricky. You have to do a lot to make that work, but it is possible, and they've now shown that.
Then more recently, we then are in the editing era. And so you have Intellia having shown editing in the liver with CRISPR, the first-generation CRISPR, the nuclease, two different programs there. And then Verve, with their first program, showed successful base editing. That was sort of using our base editor that we had given to them to go after these cardiovascular disease targets in their LNP, and they did indeed show editing. So we're building on a lot of, I think, validation and de-risking and progress in the field. In terms of Beam's approach, we actually build on a lot of that heritage. In fact, one of the early leaders of the company is Pino Ciaramella. He came from Moderna, and we actually have a lot of ex-Moderna people at the company. So I think we have a very good deep heritage in how to think about building LNPs.
It is a bit of an art, not a science. You have to really know what you're doing. The process development really matters. The manufacturing really matters. We've put a lot of care and effort into our LNP. I think that clearly, if you build a good LNP, it can be successful in the clinic. Our job is to show some data that shows what our LNPs can do.
So would you agree with that, that the LNPs are just as important in, especially in Alpha-1 antitrypsin, considering that these patients could have compromised hepatocyte function? And then sort of the hurdle becomes higher?
Yeah, I think, I don't know, the hurdle is higher, although I would agree with what you're saying that it's something that we're going to look at, so in fact, we've designed the trial to try to control for that and make sure that we have a clear signal on both fronts, so just maybe a little bit of background on Alpha-1 as a disease. You have really a spectrum of two problems in Alpha-1, so every patient has this sort of single letter misspelling in the gene, and they're creating a mutant form of the protein. It's called the Z allele or Z mutation, Z protein, and so that protein is building up in the liver toxically, and it's getting stuck there, and it basically causes liver toxicity.
It's also because it's getting stuck there. It's not being secreted to the bloodstream where it's supposed to be protecting from degradation, your lungs. And it's an important protective element of your blood. And so patients basically have a progressive lung dysfunction and progressive liver failure. And different patients will appear in different places on that spectrum. Most patients are primarily lung with some amount of underlying liver involvement. Every patient will have some liver. There's a minority of patients who are very heavy liver. And there's probably environmental or other factors that contribute to where on that spectrum patients may lie. So you're absolutely right. Our LNPs are liver-targeted. And so what we've done is we've got a Part A in the trial where we're going to treat that majority of patients who are not the most severe liver patients. They're more lung. We'll do dose escalation.
We'll get the safety, tolerability, and efficacy, and then we will go back and do a Part B where we test in those patients we've excluded so far who have the really, really challenging livers. All that said, there's evidence both from other players in the field, like in the Moderna era, as well as our own preclinical studies that suggest that we actually don't expect to see much of a difference. In our preclinical studies, we dosed animals with very mature disease, cirrhotic livers, very, very stressed livers, and we didn't see any difference either in efficacy or in tolerability there, but nonetheless, good to do the clinical trial in this way so we can make sure we have a clean visibility into what contribution those livers may be giving to the tolerability profile.
Got it. Now, this could be a two-part question, so bear with me. We are expecting two to three cohorts of data. So let's say three patients per cohort, up to nine patients. Do you think low teens is the right kind of metric? Or given the Intellia experience of redosing, you don't really need to push the dose, but come back in and then correct or take the dose higher or basically redose again. Should be thought of as a course of treatment, not a one-and-done.
Yes. So I would say, so first the bar and then kind of the regimen. Let's answer those two in order. So I think the bar for us, we are aiming to get to 11. It isn't actually like the number 11 is a bit arbitrary, and it's somewhat mythical in the field. But it comes from the fact that we know the clinical genetics here is so clear. Patients who have the disease are ZZ patients. They have Alpha-1 levels in the four to six range, total Alpha-1, Z only. They're not making any M. M is the normal protein. So there are other genotypes, basically carriers, who have one copy of the Z mutation, but not the other, who have different profiles. And so SZ is a profile where S is kind of an intermediate allele. And they don't have progressive disease. They don't have classical Alpha-1.
And they live in the 8-12 and up range. And that's basically where the 11 comes from. If you can get patients to that zone or higher, they should be out of Alpha-1 as a disease. It should be a disease-modifying outcome. And so that's the goal. And so I think that we'd like to get there. I think if we can be on the way there, that would be a good outcome. Because if we have a clear path to dose escalating all the way to 11, that would be great. If we can get there in this data set, even better. We'll see. But that's sort of the line in the sand. And I think although it is arbitrary to a degree, it has some regulatory precedent. It's been used for approvals before in the past. And I think the field is very focused on it.
So I think that's a very good target. That would be the place we get to. We know we have a drug. So then your second question?
Could you come back and read this?
Redose it, yes. So I think, yeah, so over the course of the Phase 1, call it 12-18 months, we want to fully explore the dose and schedule for the product. So we're doing this initial dose escalation now. That's as a single dose. And so our first data readout, as you rightly noted, will be across, call it six to nine patients, two to three cohorts. We clearly can keep dose escalating from there, is one option. We also can come back and redose if we wanted to add efficacy. Our preclinical pharmacology is designed so that on a single dose, we have the potential to get to 11. That was always the goal of the product when we were dialing in potency. But at the same time, we're not going to leave efficacy on the table. If we can add more AAT, then all the better.
And that may be a mix of continuing to explore higher doses and/or continuing to explore a multi-dose regimen. But based on the preclinical studies, we don't think we need those things in order to get to that target zone.
So how should we think about the dose? Because across the field, mostly because of all the data that Intellia has generated so far, it's basically about 0.7 mg per kg or whatever, flat dose of 50-55 mg. Is that the dose range we should think about in the Alpha-1 antitrypsin program?
Yeah, so we're going to test four dose cohorts in the Part A, anywhere from, I've said, sort of 0 to 1 mg per kg. That's a reasonable range, and it begins on the lower end of that range, obviously. And we're selecting a dose there that we hope will have some biological activity. That's important ethically for patients who enter the trial, but then for sure, as you dose escalate to higher, you would expect to see higher levels of efficacy and climb that range. There's formally no reason you can't keep going if it's tolerated. I mean, that's something we'll explore. We tend to at Beam be believers that LNPs generally won't be used above 1 mg per kg. I think that that's sort of what you're saying. We haven't seen others do that.
You would get into a regime where you'd start to be a little more worried about the Cmax toxicity of an LNP. Whereas within that range, I think that at this point, the field has some pretty good experience with the tolerability and long-term utility of those kinds of doses. So I think that's the right way to think about it.
Got it. And obviously, there's been some developments on the RNA editing front. Great. We're getting some proof of biology. What does that mean for Beam from a hurdle perspective?
I think the Wave data, the RNA editing data that you're noting, got to, I think it was 10.8, I assume a sort of peak after a single dose. In the case of a Wave, you had mentioned multi-dose editing before. With gene editing, if we were to do a second dose, you would just stack the editing. But it's still a lifelong effect. You're still a single course. You're sort of done with the drug once you've gotten to a certain level of editing. And that's, of course, the advantage of gene editing. With RNA editing, it's chronic for life. The key question is, how frequently does that have to be given? And what is the trough level you achieve? And what is that mean value?
So at the very least, the initial data that they gave was up around 11 for at least the single dose peak. And so we'll have to see what the multi-dose data hold. But that's basically the same bar that we have always felt exists. So I think it's great for them to have gotten there. It is proof of biology. I would agree with that. Obviously, different mechanism. We remain very focused to get to that same level or above, which is the 11 micromolar. And in so doing, I think I do think we still like our TPP, if you will, our target product profile. So we would be offering a one-time potentially transformative therapy that patients could take. We're fixing the disease at its root cause at the DNA level.
Every edit we make will turn a Z mutated allele into an M normal allele, which means you're going to be producing normal protein for the first time, and you're going to be making less Z, hopefully, and Z is a bad actor in the body, and we'll be under normal regulation. So if the gene needs to turn on, which it does when you're sick or infected, then our edited gene will crank out more mRNA. You'll get much more protein, and all of that will work the way the body intended, so I think it's a simple, elegant, and ultimately, it's doing all the things that I think are on the checklist for patients with this disease, so we think that's quite competitive with what you're seeing from other approaches in the field, and as long as we can get to that efficacious threshold.
So if we are, let's say, settled somewhere in the high teens, when you go through the full exercise of dose escalation, settle on a dose, where do you think the regulatory bar is going to be? Is that going to be just based on expression? Or do you need to show functional outcomes? And when you think about eventually getting this product to market, is there a niche, low-hanging patient population that's going to go for this versus the broader market?
Yeah, it's a great question. I think for sure, we are already thinking a lot about registration pathways in the disease. And that's an area that we will work through as our data comes out and then as we work with regulators on those questions. I have often said that I do think that this disease and our mechanism lend themselves to a conversation about faster paths to market. I myself, when I was at Agios, we were part of several drugs that got approved on a relatively accelerated pathway, four years from dosing to approval, these types of things. Our Chief Medical Officer also has been part of that at Alnylam. And so we think a lot about those things. And generally, the kinds of precision medicines that open that door a bit are when you are on the fundamental driver of the disease, because the FDA knows that.
And they'll give you credit for that. And where you have plentiful biomarkers to measure to show the level of your effect. And you know there's a lot of clinical genetics or history that you know that the level of those markers should predict benefit. And that's exactly what Alpha-1 has. So now the last piece of the puzzle is you need great data. So I think that we'll have to see what we get. And then that's a conversation to have with regulators. But certainly, I think that's a possible biomarker-driven approval pathway, possibly accelerated approval, these sorts of things. We would certainly be exploring that. Either way, you're still going to go back and show functional benefit. And so that will have a couple of different flavors. The main one would be the lung. And there's a variety of different endpoints that are in view to the field.
And that's been an active area where the patient groups and the key opinion leaders have been driving that conversation. We have our own favorites. One of the ones that we like is CT densitometry, where you're literally measuring the density of lung tissue and showing that you're stopping any deterioration of that, much like bone density and osteoporosis. So I think that's an elegant one. But there are others. And again, that's a conversation that we would have with regulators in the context of the data that we will generate.
So if I was to draw a parallel in the cardiovascular side, LDL-C is a great surrogate. But for real reimbursement, you still need to have an outcomes data. So in that context, we haven't yet seen CBER being that flexible on the genome editing side. It's already been on the gene therapy side. So is this like the first program where you start seeing that? And the densitometry data, how long will that take to collect?
Yeah. So I think that I would say, I mean, gene editing, of course, has only just come on the scene. So I think the regulatory precedents there are fewer. Although I would say, I think the FDA was quite enthusiastic about CASGEVY with sickle cell disease. I mean, I know they had a panel to talk about off target. But I thought it seemed clear to me that the basis of the approval was a lock. And they were quite enthusiastic. So we actually see CBER from Peter Marks on down being quite constructive overall for these types of therapies. So I think that's a real positive. In terms of reimbursement, it's a fair point. I mean, heart disease is another size category altogether.
Totally.
And that's a place where you have very mature drugs that are pretty effective, like statins. And so when an RNAi knockdown or something comes along, payers will look to keep people on statins as long as they can. I think that Alpha-1 is different. There is not a mature disease-modifying set of options for these patients. And they are in very severe urgent need of care. And it is still an orphan genetic population as opposed to half of the country. So I think that we would be in a different kind of reimbursement conversation. Nonetheless, in this day and age, you have to show value over time. And so I think that we would still have a plan to generate that kind of information.
Your question about the duration of the endpoint, so something like CT densitometry, that could be a year, year and a half follow-up, kind of reasonable types of numbers. And there's lots of natural history data that is accumulating now and will come out from other studies in the field that we can use for that sort of thing. And again, there are other options as well that we can look at.
Got it. And from an IND perspective for this program, following the same path as, let's say, Intellia, get enough data, get the dose, and then talk to the FDA, not just on the IND, but also the registration study?
Yeah, it's actually not quite that, so there's actually nothing stopping us from having gone to the U.S. upfront with this one, and you can actually see that by the fact that BEAM-301, our second program for GSD, is already in the U.S., so it wasn't like we have a roadblock with the FDA for any reason, and that was actually a very successful application. It was approved on the first try, and now we're open and screening patients, so with Alpha-1, it was different, so it was not the regulators. It's actually the patients, so the patients in the U.S. are all on augmentation therapy, and they don't like to discontinue it, and so we want them to, because we need a clean sort of background. We need to know their levels without augmentation, and then we would treat.
And then the whole point, you want to see M get produced for the first time. And hopefully, Z goes down. And so that would be a lot easier conversation to have if we have some PD data in hand. And so the basic thought was, start ex-U.S. where generally augmentation is not available, generate that data over the course of Phase 1, then come to the U.S. And it's a much easier conversation. So I think less about the regulators, more about the physician-patient conversation in the U.S. to bring them some data that they can hopefully get enthusiastic about, stop augmentation, and then get on the trial.
Got it. Switching back to that LNP, the original question, is it the same LNP you're using across all the programs? Or every program has a different unique LNP?
So for the two programs that are in the clinic now, 302 and 301, it's the identical LNP. So they work the same. So certainly, there would be read-through from, for instance, 302 to 301 if we can show tolerability and efficacy there. I think for future liver programs, some of which haven't been disclosed yet, but we're obviously hard at work on a whole pipeline within the liver, I think most of what we are doing would be consistent. We do have the opportunity to swap out the lipid. We've licensed the lipid in one case. So we have our own lipids. At some point, we may make that change. But that's a minor element. Everything else about our LNP capabilities, the process development, the formulations would be identical.
So this is, in many ways, what I'm very excited about with this whole field, which is these are programmable medicines. And so the thing you're actually injecting is almost identical from one drug to the next. You're literally just changing the sequence of a few letters in the mRNA or the guide RNA. And yet you have a totally different medicine. And what that means is the toxicity, the manufacturing, the regulatory package, all of that is the same program to program. So if we're successful in de-risking with an initial data set, you then have a lot of confidence that you can then imagine the follow-up LNPs to follow, if it's to the same tissue, would have a much higher probability of technical success.
Well, actually, Scott Gottlieb was kind of talking about the same thing. Get the FDA to recognize that you don't need to do the same preclinical tox package all over again when all you're doing is one small change. The rest of the thing is exactly the same.
In fact, we actually already do some of that. I think that is a definite direction. Peter Marks has been a leader on this. We are already doing it. Some of our packages, we use data from 301 in the 302 package, for instance, because it's the same LNP, even though the payload is different and vice versa. Some of that already they will allow you to do. I think that's going to gain momentum. More and more, we will have the flexibility to go faster to new opportunities without having to reinvent the wheel every time to the extent that we're using similar, well-validated reagents. I think that's a vision of where the future is going to go.
In the last few seconds, in terms of when you're thinking about the next targets or hepatic targets, Street has obviously not given credit to any of the gene editing companies if there is a, say, siRNA therapy available. So how are you thinking about the next indications?
Yeah, it's a great question. Look, I think gene editing is going to be a major and disruptive force in medicine across a wide range of indications. Nonetheless, I would acknowledge that there is a challenge when you're going up against incumbents that are very well-trodden and successful at knocking things down, things like small molecules, antibodies, and RNAi now, and that's clearly been a place where there's some difficulty. I still expect those products to have utility. One time is still, there's a lot to be said for that, but I think it's fair to say we have pushed ourselves maybe more towards places where there's more differentiation relative to what is available for patients for exactly that reason. We think that Alpha-1 and 301 are good paradigms of the kinds of liver programs we're looking to bring forward where there isn't a lot for patients.
We can do something that no one else can do. And it's that precision medicine profile where early in Phase 1, we can see hopefully a signal of success and then have a relatively rapid path to getting it to.