Thank you guys for being here. Super excited to have an old friend join us for a new fireside chat.
Yes, thank you for having us.
It's great to have you. We all want to learn prime medicine and prime editing, but I'll let you kick things off.
Thanks, Emmer. Again, thank you for having us. Pleasure to be here in Miami, escape the weather in New York. Yeah, what can I start with? Just maybe telling you a little bit about prime editing.
I think maybe just at the outset, for a lot of folks, there's gene therapy, there's gene editing, there's base editing, there's prime editing. I think everything beyond gene therapy, folks are not quite sure what it is, how is it different, but also, is it even practical, and are there direct? Maybe let's start very high level, and we can get into much more specifics.
Yeah, so I think about just what is prime editing. Just think about it as the most versatile and the safest way to edit the genome. Now, let's talk about why that is. When I think about sort of the gene editing technology, and we start with CRISPR, which is an incredible breakthrough. Now, for the first time, with very, very high efficiency, you can go in and make specific changes to DNA. Like, that's just an incredible, obviously, Nobel-winning breakthrough. What is CRISPR editing good at? It makes a double-stranded break, right? It is very good at knocking something out. You can find a very specific location in the DNA. You can make that double-stranded break.
You can have that impact, and it's very good at knocking out genes or knocking out control elements if you want to maybe turn something up, like you do in an example with sickle cell disease. There are limitations to that. You can get a lot of off-target editing, where you can find similar sequences, where you can get additional double-stranded breaks outside of that sort of primary area where you want to edit. You can also get what are called indels, right? When everything kind of gets put back together after you make that double-stranded break, you'll have other base pairs that get inserted, so you can cause frame shifts and other things. You can get proteins that can get created that are not natural that the body can then react to.
Immunogenicity is.
Immunogenicity is a risk. With base editing, this was, again, an incredible discovery in David Liu's lab, that now you can add a deaminase to the mix. You've added an enzyme to the mix. You've changed it from a CRISPR enzyme to a modified Cas enzyme that now makes a single-stranded break instead of that double-stranded break. You can change one letter to another. For the first time, you can actually rewrite the genetic code. It is limited to one letter changed to another. You can change a purine to a purine or a pyrimidine to a pyrimidine, but you can't do the reverse. It is limited to kind of repairing or correcting four base pairs. You have much lower off-target editing because you're making that single-stranded break, although it still does happen with base editing.
The second thing is you do get something called bystander edits. If you're changing an A to a C, as an example, and you have multiple A's within that editing window, then you can have multiple A's that get changed to C's. The protein, in some cases, has additional changes, so it's not exactly wild type, like the case in their Alpha-1 program, as an example.
The exact site is not as specific as what you're saying. If there's more A's around it.
The site is specific, but in that editing window, see, that editing window where that deaminase can act might be 10-15 base pairs long or 7-10 base pairs long. If there's any additional kind of A's within that editing window, that has the potential to be converted, right? Because the deaminase can act on it.
I see. Got it.
With prime editing, also in David Liu's lab, said, "Okay, now I'm going to add a reverse transcriptase to the mix." I have a reverse transcriptase. Now in my guide RNA, right, the same guide that you're using to bind to a specific sequence, I'm going to put in a DNA template, right? DNA base pairs that can now get written in via that reverse transcriptase right into the genome and make permanent corrections. You're not limited by correcting one letter. You can do actually much larger insertions, but you can correct what are called transversion mutations, transition mutations. We can fix frame shift mutations. We can do what we call hotspot editing, where we can fix multiple mutations with one editor.
With our PASSIGE technology, we can do even large gene insertions or multi-kilobase insertions through putting in a very specific landing pad and through a DNA donor very specifically put in where we want that DNA to be inserted, like we do with our BMS collaboration for ex vivo CAR T cell therapies.
Got it. Got it. Okay. That is very helpful for background. Maybe just before we proceed, is there an industry standard for off-target editing?
I don't know about necessarily an industry standard. I think every company is going to come up with the assays that they're going to use to measure off-target editing. I know for Prime, we did an extremely extensive off-target package when it came to our first program that we took to IND for chronic granulomatous disease. Essentially, we didn't see any evidence of off-target editing. I think from what I've seen, we have even more assays than we've seen even some of the other companies come out with or that we've seen. We are doing as extensive as an off-target analysis as we can do. To the point of we don't see anything where the FDA is, "You don't even have to test this clinically because you don't see anything in any of your analysis.
Is it fair to say that a lot of those assays are probably more consistent between Beam and Prime, given it's the same lab where the technologies came from?
I think we've all created the assays to test off-target, not within that lab, but actually on our own. Like, that's something we've built internally. It's something when we have a, so when we nominate a lead, when we have a drug candidate that we're ultimately going to take into IND enabling studies, we've gone through that full off-target analysis to know that that editor does not have any off-target liability.
Got it. And so on this off-target, just so I understand it in a little more detail, this is done on an in vitro basis, obviously?
Yeah, many of these are in vitro assays, but you can do in vivo testing as well.
Has that ever been done?
Have we ever done that? Yes.
I see. Just so I know, how does that get done in practice in vivo? You just.
You just do the same assays once you've done those corrections, and you can do them on multiple different cell types. It is not only that you're looking at, you're obviously looking at off-target within the genome, but you're also looking at off-target within different tissues. You also want to make sure that you're also editing the right tissues.
Got it. In vitro.
In our liver program, we want to see liver editing, but we do not want to see editing, for an example, in the gallbladders and other things.
Right, right, right. Okay, got it. As it relates to sort of gauging that off-target, in vitro and in vivo both have been done, but the number of target sites that's chosen, is it limited to 200, 300, 400? Theoretically, it could be broader than that. Everybody has a certain defined set of target sites they're screening for.
Yeah, I mean, I know we go as extensive as we can. I don't know the exact number that we do, but I know it's as extensive as an off-target analysis as we can do without knowing the full details.
Okay. Have off-target effects when the off-targets were seen on some of the CRISPR stuff, was the assays materially different and/or the number of target sites chosen materially different?
Not that I know of.
Okay. They were seeing it fairly consistently in vitro in whatever sample size they went after.
I think it depends program by program. I don't think you can blanket statement and say, like, every sort of CRISPR approach is going to have off-target. We can all optimize to try and reduce or eliminate off-target. I don't think you can blanket it. Yes, I think we're all using sort of similar assays, although I'm sure there's differences to kind of evaluate this.
Okay. Maybe just zooming in now on the Prime Medicine side of the story, first, maybe higher level sort of broader corporate strategy. I know there have been changes. I know you took over on the company. I think you also made some pretty significant changes on the operating structure priority list. Maybe just catch us up on what happened, what triggered that. Were you there prior to that?
Yeah, so I joined the company in January of 2024. Many of these changes you're talking about happened sort of in two stages. One in September of 2024, and then probably again once I became CEO in May of this year. We really focused the company on what I would call areas where I really see value creation sort of the near to medium term. When I look at kind of where our pipeline was, when I joined, we had about 18 programs in the pipeline. I think a lot of them were we were making a lot of great progress across the entire pipeline. Again, it's such a versatile gene editing technology that there's so many different places we could take it to. I think in reality, there's higher probability.
I wanted to find places where there's high probability of technical success, and there's high probability of commercial success. When we distilled that all down, we've got great delivery vehicles to hepatocytes in the liver. That is kind of where we started. Where are there true unmet needs that we can go after? For Wilson disease, there really is no other gene editing technology that can go after these mutations. You can't do it with CRISPR. You can't do it with base editing, given these are not transition mutations. This is something that was sort of perfect for prime editing. It's a high unmet need. There have not been drugs for this disease for decades.
We talked to tons of KOLs and patients, and I can tell you they are really excited for a therapy that can really change the course of this disease, which today they're on zinc salts, they're on chelators lifelong, they're on low copper diets, and even with that, patients can still progress. This is something that can truly be hopefully a cure for these patients and really make a difference. For alpha-1 antitrypsin deficiency, it's a competitive space for sure. I really strongly believe that prime editing is the best approach for this disease because we can really take a patient back to true wild type protein under endogenous control, which is important for this specific protein in the body. There are many other companies doing prime editing. You saw a deal yesterday between Tessera and Regeneron as an example.
Tessera, we're fairly certain, is doing prime editing. CRISPR announced a program recently. They call it Synthase, but it's still prime editing. People can use sort of different terms for their technologies, but in these cases, they're really just sort of copying David Liu's and Andrew Anzalone's sort of seminal technology and just really doing what we're doing. We want to be the leader with our prime editing program in A1AT, but really believe we'll get paid one way or another, whether it's through our program or others.
Allan, just remind me, when the cuts happened, what was cut and what was not cut?
Yeah.
We know what was not cut. What was cut?
In terms of programs?
Yeah.
Yeah. Obviously, what we're really focusing on, again, Wilson's, alpha-1, cystic fibrosis, and our ex vivo CAR T program with BMS. The things that we deprioritize are ocular programs, our hearing loss programs, and our programs in neurological disease and other ex vivo outside of the BMS collaboration, ex vivo HSC programs. In terms of where I see a lot of value in the future, because I do think a lot of these programs should go forward, we actually had a lot of success within these programs. It wasn't like we weren't—we could have had a retinitis pigmentosa IND already had we pushed that forward. There was another liver program that was super tiny that we didn't take forward.
Are you partnering that, or are you just keeping that in place?
We'll look to partner some of those programs. I think in some of the areas where the commercial opportunity is really small, that might be a little bit more challenging. There are other areas where the commercial opportunity is higher. I think if you think about neuro disease, I mean, there are hundreds of different genetic diseases in the brain that we think would be amenable to prime editing. We've generated some really promising in vivo data when it comes to neurological disease. I think it's still a place where we've seen AAV be more successful with direct injection lately. We're seeing that sort of delivery problem get solved real time. I don't think we're necessarily all the way there yet, but I think we're getting there.
I would like to see our neuro programs partnered and move forward because I think they can make a significant difference in the lives of patients if we can get that done. There are a lot of really interesting follow-on liver programs that were not paused per se that I think can move a lot faster and cheaper as we can leverage what we have done in Wilson disease and alpha-1. There are a lot of good areas, exciting areas to continue to invest in, but we have really focused the company on where we think we can build value.
Got it. I want to sort of dial in now on the liver more specifically. Maybe let's start with Tessera because that's what Street saw, a high-profile partnership. I guess maybe before we even go there, it sounds like you definitely want to develop this standalone, and you're not looking for a partner for some of these advanced programs right now. Is that reasonable?
Yeah. I mean, look, I always say there's nothing sacred, especially when you're thinking about the number of diseases that we can ultimately go after with this technology. When I think of where prime editing can be in 10 years, we're not going to be a one product, one disease company. That being said, I see a lot of value that we can create on our own with Wilson and even with alpha-1. I am always happy to, again, nothing sacred if there's a big enough number in front of us and the better return for shareholders to partner something, then we'll always consider that partnership. I do see a lot of value for us in taking these programs forward. Partnering these programs would require pretty significant economics for us to do that today.
I remember last time we spoke, you mentioned your goal is double digits on efficacy for AAT programs. Where's your head at sort of on what you're looking to accomplish, and where do you think Tessera lands on some of these?
Yeah, I mean, it's hard to say from a competitive standpoint because everyone's using different preclinical models. So it's always, unless you're doing apples to apples, true comparisons in the same model with the exact same editor and delivery system that another company's used to know exactly where everyone is. From what I've seen, at least at face value, the programs look fairly comparable across the board when you look at CRISPR, Tessera, and Prime that are all, again, doing prime editing. Ultimately, it's going to depend not just on the prime editor being used, but it's going to depend on the delivery vehicle. At least from what we've seen preclinically, we've got an LNP that may have a wider therapeutic index from at least what we've seen initially compared to, as an example, what we've seen with maybe the Acuitas LNP that we've measured preclinically.
That could allow us to dose higher, and that could be a way to differentiate. But we'll really have to see how these things do in the clinic.
Excellent. Okay. What is the timing like for you versus Tessera?
Yeah, so I think what they announced yesterday is they plan to be in the clinic soon. I would take that to mean, and I think maybe they said before year-end. Call it, we've said an IND or CTA mid next year. So you're about, call it six months, give or take.
Excellent. Six months separate.
Fairly close in time.
Got it. Timelines-wise, it sounds like your Wilson disease is ahead, correct?
Yeah. So the Wilson's, we've commented on a first half IND. So that's slightly ahead, yes.
You could have proof of concept type data by later next year?
We're guiding to data for both programs in 2027.
Okay. Will the initial studies be on open label?
All of the studies will be open label.
Okay. You would know theoretically later next year on where the data is tracking?
The studies will be open label.
Okay. Okay. Fair enough. Fair enough. Remind me again, what's the trial design for the Wilson's?
Yeah. For Wilson disease, it'll be obviously a dose escalation study. We'll be looking to find the optimal biologic dose. You're obviously going to look at safety, but we are going to test a lot of different efficacy measures initially in that trial.
Is there a certain type of Wilson disease patients you're taking?
We're not going to, at least initially, we're not going to go after patients that have decompensated liver disease, which is still a minority of the patients. They'll be a little bit milder than we can go after compensated, but probably initially we'll go even earlier. Essentially, we're just looking at patients with specific mutations, right? The first mutation we're going after is called 1069Q, which is in about 30-50% of the Caucasian population. They'll have to be genotype for that mutation. We're looking at a number of biomarkers that can really speak to activity. One is copper PET studies, which we're not going to mandate in every patient, but you can see the data in our deck on our website. You can see almost complete reversion back to wild type after these mice have been treated with our editor PM577.
You can look at, there's an enzyme level that you can look at in the blood called ceruloplasmin, which is really low in patients even on standard of care, the ability to actually increase that once you've sort of normalized these patients' enzyme levels in the liver. There are a number of different biomarkers we can look at. Ultimately, you want to get these patients off of standard of care and show that they're really being maintained.
Got it. Allan, just so I understand from a development perspective, is it mutation-specific editors you're making, or could you have a combo of multiple editors in one shot? How's that going to work?
If I understand the question correctly, we'll have 1069Q will be its own editor, right?
That's the IND.
That's the IND. We are going to have follow-on editors that we believe will go in the same IND. The only thing changing, the LNP is the same. The only thing changing is the guide.
Wouldn't that make a new IND then?
No, it would not.
Has that been done before by Prime on other programs?
It has not been done by Prime because this would be our first in vivo program that goes into the clinic. We have gotten feedback that you can do multiple mutations within an IND. We have gotten that feedback not just on this program, but we have actually had that on other programs as well that we did not end up taking forward, but we are thinking of taking forward.
Then practically speaking, just fast forwarding from a, you have proof of concept in 2027, you do a quick registrational on this. The commercial presentation looks like what? Different RNA guides in the same injection without having a screen from a mutation?
Not the same injection. It's a.
Your mutation analysis.
Yeah. If you think about the drug product, the drug product is slightly different, right? Because you're going to have the guide, the guide will be different, right? If we use a nicking guide, that will be different. Everything else is essentially the same, right? The drug product is slightly different, but it's still the same LNP. The expectation is, and I think the FDA, even what they put out recently in the New England Journal article, if you have a read, kind of really sort of speaks to this type of sort of plausible mechanism and a way to kind of get multiple mutations forward. The idea would be, yes, we can prove this preclinically, whether it's an in vivo study, then ultimately take that into the clinic.
The hope eventually is that we just believe there's enough in vitro, the clinical translation that we don't have to even do these in vivo studies.
Excellent. Excellent. Fantastic. Unless we missed anything today, I feel like we're in good shape. Good luck. Sounds like it's a huge year coming up.
It is. No, great questions. Really appreciate it.
Let's stay in close, Josh. Thank you again, Allan.
Thank you.
Thank you.
Questions, man.
I'm going to introduce you to our CEO, actually.
Yeah.
Please do.