One, my name is Maury Raycroft, and I'm one of the Biotech Analyst at Jefferies. It's with great pleasure that I'd like to welcome Sam Kulkarni, the CEO of CRISPR Therapeutics. Thanks so much for joining us today, Sam.
Thank you for having us.
We're going to do fireside chat format, so maybe for those who are new to the story, if you can give a one-minute intro to CRISPR?
Yes. I'll try in one minute. CRISPR's been around for over a decade now as a company, and we've gone through different phases. The phase I of the company was to get CASGEVY on the market, and we successfully did that with our partner, Vertex. That's being commercialized by Vertex right now, and the trajectory looks really strong for CASGEVY as it gains scale. That's an ex vivo CRISPR-engineered cell product that essentially replaces the faulty hematopoietic stem cells.
Beyond that, the next generation of products, there's three assets that have shown encouraging data that hopefully should be phase III-ready by the end of the year or early next year. This is namely CTX310. This is for liver editing of ANGPTL3 that reduces your LDL cholesterol and triglycerides by 50% almost with a single injection or single infusion.
We showed some promising data at AHA last year. The second asset is an ex vivo allogeneic CAR T named zugo-cel, and we're developing this across a number of autoimmune indications. Again, we'll have data by the end of this year, but I think it's a very promising asset that's completely undervalued in general in the sector of autoimmune.
The third asset is an siRNA, a long-acting siRNA, once every six-month injection of, it's called CTX611. It's against Factor XI. It's basically, think of it as a blood thinner, but with no bleeding risk, or less bleeding risk. That is applicable in many, many different situations, and that could apply to hundreds of thousands of patients or millions.
A potential blockbuster in the making, and that had encouraging phase I data and by the end of this year, we'll have phase II data in a TKA setting. That's the second layer of the business. The third layer of the business, obviously, is everything else beyond that. We have two assets that are going into the clinic or have gone into the clinic. One is for refractory hypertension. You have patients who take four or five medications a day to manage their blood pressure.
You could do a single infusion and reduce their blood pressure dramatically, and have a major difference in how their lives look. The second is for a rare disease named Alpha-1 Antitrypsin. That one, I think we have a best-in-class product, whether it's gene editing or RNA editing, and I'll talk more about that.
Lastly, I think we have an Lp(a)-reducing asset, CTX321, where we've gated the further development of it based on a large trial from Novartis named Lp(a)HORIZON, which we'll read out this year, hopefully. That's sort of the third layer, then obviously, we have a lot more exciting things like in vivo CAR Ts and in vivo HSC beyond that. All in all, I think we have a very rich and broad pipeline. I think all those layers that I talked about, I think it looks very encouraging as we get through the rest of this year.
Got it. What'd you say right after in vivo CAR Ts?
In vivo HSC editing.
Got it. Okay.
This is a direct hematopoietic editing with a single injection. We showed monkey data where it was durable. We could get almost 50% editing of hematopoietic stem cells. That would mean that you can treat all the million sickle cell patients around the world versus right now at the price point we have, you're looking at a much smaller subset in the U.S. and Western Europe and Middle East perhaps that are getting treated.
Right. Yeah. Makes sense. That was a great overview. I want to dive into the different programs that you mentioned there, maybe starting with CTX310 for ANGPTL3. You mentioned the data at AHA last year where you showed close to 50% LDL reduction and then triglyceride reduction of 55% after a single shot. What's the current status of the program and how many sHTG and refractory hypercholesterolemia patients and duration of follow-up can we expect in your second-half update?
Yeah. We haven't said exactly how many, but when the report of the data, we had 15 patients worth of data, and only three were at the highest dose level where we showed efficacy across the dose levels, but the best efficacy at the highest dose.
We went into an expansion phase where we're looking at different types of patients, whether it's severe hypertriglyceridemia, where there's very high triglycerides, but they may be okay on LDL and other parameters. We have refractory hypercholesterolemia where in fact, we had two patients who were already on PCSK9 therapies that had a further 40% drop in LDL cholesterol after ANGPTL3.
Refractory hypercholesterolemia is all the patients, whether it's they're on PCSK9 or other therapies that are refractory to other therapies. There's mixed dyslipidemias, people who have maybe not severe high triglycerides but reasonably high triglycerides but also have reasonably high LDL. There's different populations, and last obviously is the homozygous familial hypercholesterolemia.
We're trying to get more patients in each of these subsets so we can actually get a little better handle on what the right dose is for each subset. If you are in Kalshi or Polymarket, you'd bet that the most obvious phase III to do would be severe hypertriglyceridemia because there's not many options for very high triglycerides, and you want to do a one and done. I think we're evaluating the data in this phase I-B, and we're also seeing what the best dose level is, and that will dictate what the phase III design looks like. It's not super complex. It's just we need to do more patients.
Got it. Is there anything specific in patient baseline profiles that you're focused on at this point, and any anecdotal observations from the study that you can share?
Yeah. Every patient's journey has been very interesting to observe. We got to talk to one of the patients who got dosed. The person had nine heart attacks. Okay. In one of the latest heart attacks, luckily, he had his dog with him, and the dog went and called the neighbors to rescue this guy. He's tried every other therapy, including PCSK9s, and these are people who really need benefit, and that's something that works durably. If they're not compliant, even for one cycle of siRNA or something like that, they end up with these spikes, which is not good.
A one-and-done therapy for these kinds of patients is really interesting, and the fact that ANGPTL3 works on top of PCSK9 certainly is a very synergistic combination in the long run. I think ANGPTL3 will emerge as the best target.
Within severe hypertriglyceridemia, I do think it's a superior target to APOC3, and the reason for that is there's been observations of liver fat increases with APOC3 that you don't see with ANGPTL3. ANGPTL3, you actually see the converse. You see a reduction of liver fat. Again, it's not big numbers, but you see a reduction. I do think ANGPTL3 might end up being a very, very powerful target in medicine.
Got it. You mentioned the PCSK9s, and at AHA, you had some interesting observations in those patients who were on PCSK9s. Anything new on that front? Could you potentially enrich for those patients, and could that even be a regulatory path forward there?
Yeah, absolutely. That's why the post-refractory, I guess you can call it hypercholesterolemia uncontrolled on PCSK9. They're not really refractory. There's effect of PCSK9, but they're not controlled, is an interesting population where you could get a faster approval, potentially. We're looking at all these populations. We have a steering committee composed of the top cardiologists in the country, including Dr. Steven Nissen. I think we're getting the right advice to then be able to do the phase IIIs.
Got it. What percentage of hepatocytes are successfully edited at dose level four? I don't know what you can say on that front. Is the magnitude of lipid reduction consistent with that level of editing, or is there functional amplification?
I think, at the high dose levels, what we've shown with our LNP platform is the highest dose level we're getting nearly 100% editing of the liver. It's hard to confirm that because we can't biopsy someone. If you do all the modeling, you would expect that ANGPTL3 reduction would be about 80% if you do 100% editing, and that's because some ANGPTL3 is produced in kidneys and other organs. It's non-zero. I think at the maximum doses, you've reached 80% ANGPTL3 reduction, which kind of confirms our hypothesis that all the liver cells are edited.
Got it. Makes sense. Maybe talk about next steps for the program, and is there a clear path to an accelerated approval strategy, or how should we think about what this registrational study could look like?
Yeah, I think there's a lot of benchmarks to look at, but I do think with gene editing, you probably don't need as many patients. For instance, for the siRNA trials or the ASO trials for severe hypertriglyceridemia, you had about 1,000-patient trial. You may not need 1,000-patient trial for gene editing because the effect is so profound.
You can still rely on acute pancreatitis events, et cetera, as secondary markers for these trials. I think for populations that you just alluded to, which is potentially refractory hypercholesterolemia or even some mixed dyslipidemias, we could just use LDL as a biomarker for approval and get an accelerated approval.
I think we don't want to talk too much about our strategy right now, but by the end of this year, once we meet with regulators and get the go-ahead, that's when we'll talk about what the impending phase IIIs are.
Got it. Makes sense. Let's shift gears to Lp(a). A lot of interest and excitement around what Novartis is going to show in the second half of this year when it comes to MACE benefits. What are your expectations on the outcomes, and how will those data inform your strategy?
Yeah, I think the entire industry's watching the HORIZON trial with a lot of anticipation. I think this is the first trial. Everyone believes, and it's known through epidemiology studies, that if you have high Lp(a), it's almost 6x more atherogenic than LDL cholesterol. If you're born with very high Lp(a), you have a high risk of MACE events, and we continue to see that.
What has not been ever shown in a randomized control study is whether the reduction of Lp(a) pharmacologically will reduce that risk. For what subsets and is it for every person or not? We don't know that, but I think HORIZON trial will show that. Unfortunately, the asset, the HORIZON trial is based off an ASO that's not the best reducer of Lp(a), right? You're only getting about 70%-ish reduction.
It may not be the perfect asset or the drug to demonstrate this effect, but I do believe we're going to see reasonable reduction of risk, whether that ends up being 15%, 20%, we'll see. We'll find out. I'm actually particularly more curious about the subset analysis because my personal hypothesis is that high Lp(a), high LDL will certainly have a greater benefit versus high Lp(a), low LDL. The rationale for that is I think LDL's a bad actor that starts some of the inflammation and nucleates the plaques, and then Lp(a) logs onto it.
If you have low LDL, you may not get that, but if you have high LDL, you certainly have more of those nucleating events. Let's see what happens with the HORIZON trial, but we're ready to go with our-- We have CTX320, which had about 73% reduction of Lp(a) at the highest dose, but we have CTX321, which is targeted to get 90% reduction at the highest dose. We'll see which one we advance forward based on HORIZON.
Got it. Makes sense. You've got some optionality there based on the HORIZON data. For Alpha-1 Antitrypsin, you mentioned it could be best in class. You've shown preclinical data with your SyNTase editor, and you're planning to enter the clinic middle of this year. Can you talk about the patient population you aim to enroll and the bar for success relative to the base and prime editors?
Yeah. I think, the reason we're very bullish about our A1AT asset is. This disease, for those who are new to it, which is basically you have one mutation in your alpha-1 gene that creates what's called a Z allele. It creates a misfolded protein that doesn't do its job, and that has a protective function in the lungs, essentially, that is missing against elastase, and that's what causes the disease.
In all our preclinical studies, the patients that have this Z allele have about, let's say, five to six micromolar of alpha-1 enzyme produced. The question is, how much do you need to increase it? If you look at this model that's very predictive, it's called a rat PiZ model. It basically recapitulates the Z alleles in a rat, not a mouse, in a rat.
In that model, I think our competitors showed about 2.5x improvement in the baseline level of AAT production. If the baseline were five, you'd expect that you'd get about 13 or 14 micromolar of AAT production. If a baseline's six, it's a little higher than that. That's what's recapitulating in the clinical trials in humans. Our preclinical models show that there was nearly a 5x improvement over baseline.
Would that recapitulate in humans? Well, let's see. We're going to put that in clinical trials soon. If it is restoring AAT to about 20 or 25 micromolar, which is what normal people have, that's effectively a cure. That's why we're pretty excited about that editing. That also is our first foray into SyNTase editing.
This is an extremely precise editor that's, right now, shown to be much more potent than the prime editing that was described in literature. We've made a number of improvements to make it that much more efficient. I think we're going to carry that forward across a number of our other programs, too, that we have not disclosed yet. A very powerful editing platform that applies to both base editing as well as RT-based editing.
Got it. That makes sense. From a regulatory standpoint, Beam has helped de-risk the space by getting a clear path for alpha-1. How does that influence your clinical trial design and potential for an accelerated approval path? When you think about timelines versus Beam and other competitors, where do you see CRISPR?
I think Beam's done a great job de-risking the regulatory path, and I think we'll follow in their footsteps in a way. It's a pretty big population, so I don't think being two years behind is going to be a significant deterrent to a commercial launch. There's lots of patients that can benefit from gene editing, the more options there are for patients, the better off we are as a society.
Got it. Okay. From a COGS perspective, how do you think about COGS per dose for this program, and how do you contextualize this versus chronic therapies like RNAi or chronic modality?
I think the COGS question's going to go away for gene editing.
Yeah
It's getting lower and lower over time. It may end up being, if it's only a one-time dose, you're going to end up with a higher gross margin overall, because you don't have to manufacture for every dose eventually. Let's say an siRNA is priced at $20,000 per year, and you price gene editing at $80,000-$100,000 for once in a lifetime, your COGS are for one dose, whereas siRNA COGS are going to be for several doses over time.
One, there's a great pharmacoeconomic benefit for society. If you think about a 20-year-old with high Lp(a), you're going to have this person take 100 doses of siRNA. I don't think it makes sense. You should do a one-time gene edit. The margins support normal drug-like margins.
Yeah. It makes sense. Let's shift gears to CTX611. You've got the long-acting, every six-month dosing for Factor XI siRNA asset, and this is for thromboembolic disorders. We'll get a first look at phase II data in patients undergoing total knee arthroplasty in the second half of this year. How much data is going to be in the update, and how are you setting expectations around this readout?
For those who are new to this asset, CTX611, our bodies have two pathways of clot formation. One's called intrinsic, and one is called extrinsic. If you hit your nose on something and you're bleeding, or if you have an injury, scrape your knees, there is a tissue factor pathway that ultimately forms the fiber and the clots. There's the intrinsic pathway that's more dangerous that typically results in thrombus formation. It's a Factor XI pathway that goes from Factor XII, Factor XI, then down to Factor Xa.
Factor Xa is the node. Factor X is the node where both the pathways meet. DOACs, which these were amazing medicines like Xarelto that were created against Factor Xa, multibillion-dollar drugs, act at that node. They both shut down both pathways of clotting.
That's what leads to increased bleed risk if you take a blood thinner. You're on a blood thinner for AFib and you hit your nose on something, you're going to end up in the ER. What we're doing is by going with Factor XI, you don't shut down the extrinsic pathway. You still will get a clot if you have the tissue factor pathway active, but you basically shut down any of the risks of thrombus formation.
The TKA study or total knee arthroplasty study is the perfect way to test this hypothesis because you get a knee replacement. You're immobile, you have very high chances of clot formation or thrombus formation. There's chances of pulmonary embolism and all that, and people generally take enoxaparin for that. Because you can't take a DOAC, because you've had surgery, you're going to bleed if you take a DOAC.
If you take a Factor XI, you would have a much lower chance of these embolisms or thrombus formation or VTE formation, but still prevent that bleeding. That's why this has been a good baselining to see how good a therapy is in the space, which milvexian did, and abelacimab and Regeneron all did TKA studies. That's not the end market. That's the most promising.
It's just the stepping stone towards very attractive markets, and there are two big studies reading out from other therapies. One is called Librexia for milvexian. That's going to read out hopefully this year. There's a study called LILAC from Novartis for abelacimab. One for a small molecule, one for an antibody. If those two studies are positive, we're looking at a $20 billion category here.
We may end up having the best drug in the category because it's once every six months, it's an siRNA. You actually have reversibility because if you have an siRNA, you're on an antibody, you can't just give somebody plasma and hope that they start clotting again because the antibody is going to work on that too. With the siRNA, you can do that.
You can just give factor and it works. There's reversibility, and because you're cutting off Factor XI at the root, you never have a chance of Factor XIa formation. All the drugs, antibodies, and small molecules are acting on F actor XIa. If you just take that root out, you never have that branch issue, essentially.
Several reasons to believe we have the best-in-class asset in what could be a very attractive category, and we're two readouts away from establishing this as a very attractive category. Bayer's success with asundexian and secondary stroke prevention is a great bellwether for this whole category.
Got it. In your readout, is there a specific bleeding event rate and VTE prevention benchmark that you want to achieve for it?
I think there's no specific. You're just comparing versus enoxaparin to see how much reduction in VTE rate you have. If you look at some of the antibodies, you had 15% for enoxaparin arm, and it was 7% or something for the drug arm. Something in that range. We just want to show the drug is active in this segment. It's sort of a gating thing to then start a phase III, essentially.
Okay. Where are you at with the phase II? What kind of visibility do you have into the study, and how many patients could be in that?
This is a 400+ patient trial, global trial. It's a big trial. Enrollment's going well by what we've said so far. We're on track to put top-line data out end of this year.
By end of this year. Okay. For the two competitor readouts, what are your expectations for those, and how does that read through to CRISPR?
Yeah, I think the two readouts, one is Librexia, which is for AFib patients. What happens for these AFib patients is because of the anomalies with how their hearts beat, the blood flow is such that you increase the chances of clot formation, et cetera, in the compartments of the heart and elsewhere. They generally go on these blood thinners if you're on AFib. The Librexia trial is for all AFib patients, whereas LILAC is for the DOAC-ineligible AFib patients, and even that is a big market, right? Very high chance that LILAC is successful. Librexia, hopefully successful as well, and I think that's category-defining, essentially.
Right. Makes sense. If they miss in the studies, I guess, is there still a chance that CTX611 could do better?
Absolutely, because it's a better agent. I think it's much higher inhibition of Factor XI than some of the small molecules. There's different assays to measure it, so it's hard to say apples to apples at this point, but we think we have best-in-class asset.
Got it. Let's shift gears to zugo-cel, your next-gen CD19 program. We're expecting a comprehensive update second half of this year for both oncology and autoimmune. How are we setting expectations for that update?
Yeah. We're proud of what we've done with zugo-cel. Pound for pound, we're getting the same activity as an autologous CAR T with an allogeneic CAR T, and the cost of goods with the allogeneic CAR T is sub-$10,000 per patient. It's immediately available, and on the autoimmune indications, we're only a year behind the autologous players, right?
A year, maybe a year and a half behind, but we're going to catch up very quickly. My suspicion is if you looked at the end of last year, you had maybe 60, 70 patients dosed with autologous CAR T and autoimmune disease and, like, five patients dosed with allogeneic CAR T. This year, by the end of this year, you're going to see that there's almost 2x more patients dosed with allogeneic CAR T as opposed to autologous CAR T.
I think it's going to flip very quickly because it's going to move that much faster, you're getting deep B-cell resets. We showed data for two lupus patients in our update. These patients are several months out, completely drug-free. They can't believe it. These patients, they've all lived their lives. They started getting lupus in their early 30s, and they've always taken multiple drugs all the time, and they still feel really crappy because these diseases have all sorts of manifestations, and now they're living normally.
Patients who couldn't walk 10 ft with some trouble now can run several miles. Profound changes in how they live their life, and I think the effects cannot be ignored. I think more data will come out, and it'll support this, but we think zugo-cel will also go into phase IIIs in the not-too-distant future.
Got it. That's going to be for autoimmune or oncology?
For autoimmune.
Yeah.
I think oncology's a tougher putt because there are so many indications, so many agents already approved. Autologous is already very freely available for most patients in the U.S. in oncology. We're making our bet in oncology with LNP-based integrating CAR Ts.
You saw recent data from a company called Kelonia, which Lilly acquired, and there was data from Legend yesterday, and these were viral-based permanent CAR Ts. What we're going to do is an LNP-based integrating CAR T that's a simple injection, and it should create essentially the same as autologous CAR Ts and get rid of your cancers. In the not-too-distant future, we can get that to clinic as well.
Got it. Okay. For the autoimmune data update, what are you going to be able to show there for that data update?
We haven't said how many number of patients and things like that, but we'll have a reasonable number of patients across the first trial we're doing, which is lupus, scleroderma, and myositis. There'll be open trials in ITP and wAIHA as well, and we now have a trial open for neuroimmune indications.
You look at things like MS, right, and people think, Oh, here's a disease that's solved, or NMO, and you take these B-cell depleters like Ocrevus or Kesimpta , and you deplete your B-cells. For those patients who allow measurement of their CSF fluid, you look at the oligoclonal bands, and the disease is still there. Something's still slowly chewing up their brain cells while they think they're cured. It's not a cure.
I think with CAR Ts, I think you're going to go get to the root of this, into the brain, into the CSF, into the spine, and get rid of those pathologic B-cells once and for all. That, I think, is truly a cure for these patients. I think we're expanding zugo-cel across all these indications.
Targeting CD19, is that sufficient, or do you think you could need a dual targeting just to prevent relapse?
Seems like it's sufficient because, yes, there's some activity with BCMA, but at some point, the mature B-cells die. You want to get rid of the clones at the baseline with CD19, and that seems to do the trick.
Got it. Wanted to ask on CASGEVY, what's the latest that you're seeing there? It seems like momentum's definitely picking up. There's been about 500 cell collections to date. Talk about how many of those are going to convert to getting treated.
They should all convert. I think patients, when they start a journey, they should all end up getting dosed eventually. There's maybe one or isolated cases here or there, but if you look at today, Vertex announced that they had initiated more than 100 patients in the last quarter, right? That's more than a patient a day that's getting initiated, and that's all forward revenue.
If you just did that math, that's $800 million of revenue, right? That's in the future. I'm convinced CASGEVY's going to be a multi-billion-dollar opportunity, and with scale will come profitability, and with scale will come efficiencies and everything else, too. I think we're the only ones approved outside the U.S. In the U.S., we're going to get major market share as well. I think that's on a very good trajectory as well.
Okay. I think we're out of time. Maybe in closing, if you want to highlight key catalysts ahead investors should be focused on.
Yeah. I think I would just say for investors, most investors are focused on CASGEVY, but we have six other assets reading out in the next 12 months. Data will tell the tale, but in many ways, I think people should also start looking beyond CASGEVY as we expand our portfolio into several exciting assets.
Got it. Thanks so much for joining us today, Sam.
Thank you.