CRISPR Therapeutics AG (CRSP) Earnings Call Transcript & Summary

Good afternoon, everyone, and thank you for joining me at the third day of the Needham Healthcare Conference. My name is Gil Blum, and I am a senior biotech analyst with a focus on immuno-oncology. Today, I have the pleasure of having with us Sam Kulkarni, the CEO of CRISPR Therapeutics. [Operator Instructions]

So let's start with transfusion-dependent beta thalassemia and sickle cell. I think a lot of investors are pretty focused on this. And again, congratulations of the filing being accepted. That's I think it's a big move for CRISPR and to Vertex.

Yes. Thank you, Gil, for having us at this conference. We're quite excited about where we stand today. It's hard to believe that we're nearly 8 years into the life of the company, but only 10 years since the technology, CRISPR technology was elucidated and developed for gene editing and to stand here where we filed the first BLA of its kind in the entire world is quite a proud moment for CRISPR, but also for the -- an important moment for the entire field. It does feel like 1985 when the first proteins or biologics were brought to bear as medicines, which sparked off a whole category of medicines, which now comprise over half of the entire pharma market. So we're quite excited, both Vertex and us not just in terms of having filed the BLA, but the prospect of bringing these medicines to patients in need. And we do think that there'll be a lot more patients that could benefit from this medicine than how people are thinking about or handicapping it. And -- but there's a lot of work to do ahead of us as well. So anyway, all in all, very excited about where exa-cel stands but also very excited about everything else that's coming behind that in terms of our pipeline.

Okay. A couple of technical questions regarding filing. So it looks like a regulatory process for exa-cel. So a little ahead in Europe compared to the U.S. Are there any different regulatory requirement in different jurisdictions to ex U.S?

By and large, they are very similar, I would say, between the U.S. and the European authorities. I do think that there are some specific -- there could be some specific differences here and there in terms of CMC matters, but generally very similar. I think the fact that we began -- we completed our European filings in December last year versus completing our filings in March for the U.S. FDA, is just a matter of providing all the data that was necessary for each of the agencies.

Okay. Just to remind everyone now that the filing is complete, what are the time lines for the review?

Typically, what is the U.S. FDA, what you have is a window of time, about 60 days where they can review the filing and either accept it or not accept it. I think to do when they accept it, you also find out whether you have a priority review or not and what the PDUFA date might be. We also find out whether there's potentially an Ad Comm required that doesn't have to be at the 60-day mark you can find out later. But I think having the BLA accepted is a pretty important milestone. And then I think in the Europe -- for Europe, I think there's ongoing dialogue after you file and questions that you answer. But overall, I think that, that time line for approval typically takes 9 to 12 months in Europe. But given that this is a disease with a lot of unmet need, we're hoping that people act to a sense of urgency.

Okay. And maybe because we discussed it in brief here, any handicapping on a potential Advisory Committee is a bit of a first of its kind.

It is the first of its kind, and an Advisory Committee is it could be a good thing for us because it would demonstrate or would highlight all the risk -- the way physicians think about risk-benefits and could provide tailwinds in terms of a potential launch. So I think at this point, it's very hard for us to say whether an outcome is necessary or not or the FDA will ask for it. Given it's the first of its kind, there's a -- I would have said that there's a good chance there will be in an Ad Comm, but we don't know at this point.

Regardless, you will be fully prepared.

Absolutely, absolutely. And we do view the Ad Comm as a positive thing.

Okay. And maybe a bit of a kind of an outside looking in question. Do you see changes in attitudes, time lines with recent structural changes in the FDA to OTAT? And a lot of people have spoken about Peter Mark's involvement. I'd love to hear your opinion.

It's hard to say for one case, case-by-case basis. I think we generally -- I've always maintained that we've seen a very supportive agency and the people of the FDA are trying to do the right thing for these medicines to bring medicines to patients safely. I think there are always going to be some disagreement between companies and the agency in terms of CMC requirements or other preclinical assays and preclinical studies. But overall, it's hard for me to say if there's been any recent changes, but I continue to see a supportive environment.

Okay. And maybe from a commercial preparedness side, what can you tell us about your inventory set up ahead of the launch? And do you expect there to be an initial bolus of patients? How do you see some of these dynamics?

Yes. I think the key question is capacity rather than inventory. I think since it's a patient-by-patient basis of drug manufacturing, the only inventory we're talking about is Cas9 or guides or the necessary materials that go into manufacturing. But I think from a capacity standpoint, we do have the capacity to support whatever demand we may see in the launch, even if it's above what we expect. And over time, we're going to continue expanding that footprint because of the capacity. And we don't want to build too much capacity off the bat, but at the same time, we never want to be capacity limited.

That all makes a lot of sense. What learnings you say the company has garnered from others experience like bluebird's experience in the space, things like pricing in Europe?

Yes. I think lessons mainly, I think, on the clinical development, regulatory path. I think on the commercial side, I wouldn't say that there's much to work off of. I think I'm not sure the bluebird experience in Germany is a big read-through on willingness to pay for a therapy like this. I think we feel very confident, so does Vertex around the value that this medicine brings to patients the pharmacoeconomic benefit, and I think there's going to be support across the spectrum to bring this therapy to patients. In terms of the clinical path, the regulatory path, I think why I would take is gene editing is a much more reliable, consistent way of manufacturing where the CMC requirements and the CMC needs are more facile than anything that involves a virus. I think we've seen this for lentiviral-based products. We've seen this for AAV-based products. And outside of the path that AveXis had, which was a pretty linear path towards approval, any other company going towards a BLA using viruses have had a bit of a tortuous path. And that's not because any of these companies are good or not good. It's because with viruses, there's inherent variability from a CMC standpoint. And that's something that we're getting -- everyone is getting a handle on in the industry, but it's not straightforward. Gene editing, on the contrary is actually much more consistent. If you get the Cas9 and guide into the cell and to the nucleus, you get the edit. I think that gives you a level of reliability and a very limited variability that will give the agency's comfort that everybody is getting the right product. And ultimately, that you're not going to see the same manufacturing-driven variability to drive clinical variability in patients.

Maybe spending another moment here on pricing and benefit that I'm sure you've seen the recent published ISA report in this space that does suggest that a high premium is warranted. What are your thoughts there? You feel it's pretty supportive, I assume?

Yes, it's very encouraging that even the draft report has a relatively high price point. I think they obviously will do a final report. At some point, this summer, and they're soliciting opinions and input, but at the same time, I think it's the starting point is a favorable starting point. It just illustrates what we've been saying all this time alone, which is the pharmacoeconomic benefit alone can support a very high price. Forget the entire value-based argument. And you layer that on with the value you're bringing to these patients and to society. And I think you're going to create net economic benefit for the system from this in a relatively unambiguous way, and I think that's what's going to garner a lot of support from the payers as well as we launch this therapy.

And one last question that we do get often from investors, which goes back to capacity and how many patients can receive treatment per year on launch?

Well, I think what we're gearing up from a capacity standpoint. And capacity, when I say capacity is twofold, capacity of the qualified treatment centers to handle number of transplant procedures per year and the capacity from our manufacturing standpoint to produce or manufacture the drug for all these patients. And I think from both those standpoints, we want to get to a point where we can comfortably have capacity for hundreds of patients per year and eventually pave the way for even more beyond that, but that's what we're building. I think from a demand equation, and the payer side, I think we'll have to see where all comes out. But I do expect the demand equation is going to be stronger than people anticipate as well. The one thing to note for everyone is that I think the revenue recognition comes at the point of infusion. So there's going to be a little bit of a time lag between time we launch, bring the patients into the funnel and by the time we actually infuse them, but that -- regardless of that, the key marker to look for is how many patients are coming into the funnel where we harvest their cells because once their cells are harvested, they're manufactured, we suspect there's very little drop-off from that point to patients actually getting infused.

That also makes sense. So another strategy to expand the market would be to use milder overall conditioning regimens. There's been some notable failures in this recently. What are your thoughts? And how is CRISPR working in this direction?

Yes. Both Vertex and us have made this a top priority given the value we see with exa-cel to bring the gentle conditioning agent to the market. The interesting observation I have is that over the last few years, we've seen some failures. And the reason for that is these agents were being developed for indications other than sickle cell and thalassemia. The PK/PD profiles were all designed to work in AML or some rare disease like SCID. But in sickle cell and thalassemia, what we need our antibodies or ADCs that have a fast clearance, so they don't impact the drug product but sufficient hit that you create enough room in a bone marrow to enable enough sufficient chimerism. And so that's sort of the requirement, which may be different than say, what you need for allotransplant with SCID or for AML. And I think that's why we're doing a very much of a demand spec-based build of the assets we're going to develop here. And that gives us greater confidence that it's going to work where others have not worked.

Okay. So maybe a broader question. There's significant investment involved in rolling out any gene and cell therapy, particularly on the CMC and manufacturing side. How much of a moat do you think from a competitive standpoint is having in-house stabilities compared to companies that use contract manufacturing?

No, it's a delicate balance. I think if you ask people 3 years ago, everyone said, you have to build in-house manufacturing, and what we've seen is capital destruction. Everyone's built expensive in-house manufacturing, which now is selling for pennies on the dollar. But I think what we did is, I think -- so that we want to be bespoke. I think the key know-how is the process development and analytical characterization for the products. The actual manufacturing of the product is, once you figure out the process, you can be done externally without much issue. So we're going to have a hybrid model where all the early products are developed, for instance, in our own manufacturing facility like CTX112 or CTX131 for the early clinical trials, and then they will be -- once they're fully established, we could do that in a CDMO because it's more capital efficient versus us doing it ourselves. And then same for mRNA or same for AAVs. Our manufacturing facility can produce all these. But we don't want to turn be the ones operating at scale for manufacturing eventually because it's a different competency and CDMO is pretty good at it. So we're going to have a cyber model. I don't know that it per se creates a big moat. I think the biggest value-add we see in having our own GMP facility 30 minutes away from our R&D headquarters is that the bench to bedside translation happens much faster. As soon as we start making an asset at the bench side, we start thinking about how to manufacture it at GMP scale or in a GMP environment. And so our typical time line from concept to IND in some cases, has been less than 18 months, which is unimaginable if you're going to do this at a CDMO.

Okay. I'd like to switch gears and talk more about immuno-oncology. So let's start with the B-cell lymphoma to CTX110 and 112. The most recent data updates have indicated about a 40% CR rate with about 20% of patients maintaining kind of long-term activity. And investors have noted that these results come in below what you would expect for autologous CAR-Ts. Now I'd you like to spend some time differentiating the kind of patients that would go in to allogeneic therapy versus an autologous one. I know this has been a bit a point of contention for investors, and I think it's worth clarifying.

Yes. And also apples-and-oranges when you compare the clinical trial data. I think when the autologous therapies came out and then the early trials, you've got patients that didn't have huge tumor burden nor were they fast progressing tumors. If you look at the intent-to-treat population for Kymriah, for instance, of 170 patients that they publish, only 110 got the CAR-T. Off the ones that didn't get the CAR-T, somewhere manufacturing failure, other were just fast progressors. And in our instance, we're getting much sicker patients in our trials. And so it's not really an apples-to-apples comparison to say that the 1 and 5 durable CR that we have, 20% is far lower than autologous CAR-Ts because in the real world, my sense is it's somewhere between 1 and 4 and 1 and 3 for autologous CAR-T. So somewhere in the 27%, 28% range for durable CRs, okay? That's my own. It's not like there's a systematic study done. That's my own take on what I've seen across different economic micro centers. So that means that we're not that far away. And we have all the convenience of allogeneic therapy. Mind you, and that data was with 1 dose, right? So now if you have 2 doses, and we've shown that safety profile remains the same, odds are that we end up with better efficacy and hopefully going to have a better efficacy. And that gets us even closer to this -- to what we're seeing in the real world with autologous therapies. This is all the CD19 context. With our CTX112, I do think that those CAR-Ts going to be much more potent and you could end up even enhancing upon the durable CR rates that you've seen with 110 or with the first-generation programmers, which could be equal or better than auto CAR-Ts. So the limit of improvement in allo CAR-Ts is a lot higher than the limited improvement in autologous therapies.

And that kind of brings me to the next question. If you had to set up bar for your ongoing pivotal study, but where do you think that would be? Is it -- should we compare it all to autologous CAR-Ts? Should we compare more internally? Should we -- what is the best way to think about this?

Well, the bar to approvability in my mind, I think where we are right now, the 1% and 5% or 20% durable CR rate, is actually an approvable profile for CD19 allo CAR-T because there are going to be a significant set of types of patients that are going to benefit from that therapy being available. But there's a market sensitivity. If we're at 20% versus 25%, that translates to me to a difference of being a $0.5 billion drug to a potentially a $1.5 billion to $2 billion drug. And that's, I think, the commercial sensitivity in that efficacy. But I mean, if our worst cases were $0.5 billion drug, that's not a bad place to be because our profitability on allo CAR-T is a lot higher than what you've seen with autologous CAR-Ts.

Okay. And kind of last focus on 110 before we switch to 112. Let's say everything goes well. Where do you think 110 would slot ends up in the therapeutic algorithm.

Well, I think we're developing for relapse refractory settings right now. But in fact, with my future hat on when say, 8 years from now, what does that sequence of therapy is going to look like for CD19 positive malignancies? My suspicion is that allo CAR-Ts and bispecifics are going to move up in lines of therapy at the expense of auto CAR-Ts. You could very well have a patient get an allo CAR-T for tumor debulking followed by a few cycles of bispecifics as maintenance as the first line of treatment, whether it's bispecifics or Rituxan as maintenance because if it's early-stage tumor, it doesn't justify doing auto CAR-T and the expenses associated with it. You also want to treat it right away, so you don't want to wait too long. And I think bispecifics are going to be moving up in line of therapy as well. So I don't know what the path is going to be in terms of development. If you go into high-risk front line, et cetera. But eventually, I think allo CAR-T and bispecifics are going to jockey out auto CAR-Ts in the grand scheme of things.

Okay. So let's spend some time on CTX112, recently saw the paper from [ Carl June ] Lab on Regnase-1, it seems that both of the edits that are in 112 are designed to enhance the health of donated T cells and improve their expansion in tumor microenvironment. Just to help me understand how important is suppression in the TME for a liquid tumor.

It is important. I think there are 2 aspects to these edits. One is overcoming exhaustion because of factors in the TME that are suppressing the activity of CAR-Ts. The second is just the CAR-Ts own health and robustness in multiplying and replicating, right? I think -- so what we're seeing and we just presented these data at AACR is that with the Regnase-1 edit and TGFBR2, much higher fraction of the CAR-Ts were the central memory phenotype. And it remained naive several days into an animal experiment with exposure to the tumor, right? And what that tells me is we have a window of about a month before these CAR-Ts are rejected by the patient's immune system. In that time, if you can maximize the activity of the CAR-Ts, that's what's going to get us the most deep responses that could lead to durable responses. That's sort of the thesis behind it. And that's why we're going after the edit. And in fact, Regnase-1 is a lot of notoriety with the recent publication from Carl June, but TGFBR2 is also going to get a lot more notoriety in the next few months because at AACR, you saw GPC3 auto CAR-T with the TGFBR2 disruption, which, by the way, we also have a program there. And that look pretty interesting. And there's other programs which are also employing TGFBR2 knockouts. I would say, TGF-beta suppression broadly because there's different edits you can make, whether it's TGFBR2 knockout or some others to achieve the same desired effect, but all in all, I think the cytokine space in as effectors of T-cell activity or influencers of T-cell activity is going to become a much more steady field and much more in the news.

Okay. Now assuming that CTX112 is successful, so do you see it replacing CTX110 over time? Or do you think maybe both of them are going to be coexisting, but in different patient populations?

It could. We don't know yet at this point. I think our thesis is we -- it makes sense for us to move forward full speed ahead on CTX110. Just going through that BLA is going to give us a lot of experience and actually launching in a market is going to teach us a lot even whether CTX112 works or not. And if CTX112 is much better, obviously, we try to bring that to the market as soon as possible as well. So this is a new paradigm in the industry because typically, we'll pick one or the other. But I think the synergies of working on both are quite important here for us and far greater than people imagine. So we'll continue to push for both programs.

So moving on to CTX130 and 131, just starting with the T-cell lymphoma space. So we definitely saw some interesting activity there. And is there a likelihood of an expansion study becoming registration or this is very high unmet need in T-cell lymphoma?

Absolutely. That's what we're trying to clarify with the regulators. I think in T-cell lymphoma, there literally are no options for patients. There's an agent that was approved 10 years ago on an accelerated basis that never had a confirmatory study, for instance, or others that only treat CD30 postive T-cell lymphomas. And so in that environment, a single-arm study could certainly get us to registration approval. And I think the numbers of patients require not dramatically high. I think the treatment paradigm in T-cell lymphomas, we need to establish essentially the multi-dose regimen at this point. We need to -- we're seeing very encouraging activity, much better than what the benchmark is. But I think our goal is to get to that point where we can say we're in a registrational trial sooner or later but it does require a little more FDA discussion.

And one of the challenges of targeting CD70 is it does, it can result in T-cell aplasia. How is this being managed in your studies?

Interestingly, we haven't -- while there is some T-cell aplasia, we haven't seen an increase of risk of infections or any sort of effects you see when people are immunosuppressed. I think -- so we haven't had to deal with any of the safety issues essentially because of targeting CD70.

And again, switching gears to CTX130 in RCC, again, a really interesting study. Can you remind us what the T-cell expansion looked like in that one responding patient?

Yes. So we haven't, first of all, we're super encouraged that we have the first demonstrated CR, complete response, in a solid tumor with an allo CAR-T. In fact, the first demonstrated complete response with any CAR-T in the world. It was awareness in case. I think -- so we're encouraged by the activity against the target in RCC. We haven't disclosed the T-cell expansion for that 1 patient. I think we do generally with CR, we do see a reasonable T-cell expansion, but it's not always predictive. It doesn't always mean that if you have hired tumor expansion of CAR-Ts that you'll get a CR or even a PR. The reason for that is this all was down to a effector to tumor ratio. When the tumor is very bulky, sometimes we see very high expansion, but we don't see responses because the tumor is so bulky, and you don't get complete -- or you don't get deep responses. But when you do see some response. You don't see complete response sometimes. But sometimes there's a sweet spot where the tumor is bulky enough where you see quite a bit of expansion, but you also see complete killing of the tumor where you see CRs. So it's a little bit hard to say. And in fact, this data, this has been borne out with all the autologous CAR-Ts as well. Initially, there was a thesis that expansion is the single biggest predictor of complete response and has not been the case.

It's more area under the curve. So the dynamics are slightly different, right? If you get a very high peak Cmax usually associated with tox.

Well, associated with tox, but also in the case that with case we're seeing sometimes associated with quicker exhaustion. So it could all the -- you're right about the year-end of the curve. But we have now data for CAR-T expansion in about 150 patients across all our trials. The bigger data set then most other companies in all of the T-cell space. And we're learning quite a bit about what it means for from a PK/PD standpoint. Well, how predictive things can be. And we haven't disclosed all these data, but it gives us a big leg up in terms of how we develop the next set of products.

Excellent. So going back to the 2 edits we discussed. Do you expect them and CTX131 to support MF expansion, support activity within the tumor microenvironment, maybe both?

Well, I think the -- typically, when you infuse the cells, they have to go find the tumor to start expanding. And I think then they expand and they come back into the bloodstream where they're measured. And what we wanted to see is greater expansion because the cells are somewhat more central memory phenotype. And then for them to be able to retain their effector activity for longer even as those cells differentiate from early naive cells to the more effector like state that they can keep going 15, 20 days post treatment. And that's what's going to get us a durable response.

The longer term surveillance.

Yes. I don't know if you want to call surveillance or to your point, just enter into the curve, longer-term killing.

Pressure. Longer-term pressure.

Exactly. We may never get surveillance for 3 months unless we have other persistence edits. But I think we would get enhanced killing for longer than what we're seeing with our first-gen CAR-T, which is about 14, 15 days.

Okay. I'd like to spend a little time also on the regenerative medicine angle. You've had some interesting reports recently on a new deal with Vertex, new partner -- new partner programs with [ VXA180 ] and VX264. What can you tell you about us about this?

Well, we're very excited about our regen franchise. Again, speaking of world first, we were the first, to my knowledge, to have treated patients with edited cells derived from IPCs. I think that's a pretty big feat. And that brings together the 2012 Nobel prize with iPS derived cells or creating iPS cells with the 2020 Noble Prize on using CRISPR Cas9. I think that's 2 major technology waves that have created -- can actually create significant value. Now what we have with Vertex is a 2-part partnership. On one hand, we're working in a 50-50 co-development fashion with the CTX211. And the thesis here is that we take stem cells that are not fully differentiated, partially differentiated, but their fate are determined and implant that into a device and implant the device into the patient, subcu. And the goal is to get the cells to fully differentiated inside the patient's body and then produce insulin, endogenous insulin that's going to reduce their insulin requirements and hopefully make insulin impendent. All this while without the need for immunosuppression, right? So that's one part of the program. The other part of the program, a partnership we have is that we've licensed our IP around CRISPR/Cas9 editing to Vertex to pursue -- further pursue their own efforts with a different cell line where they differentiate the cells to fully -- to the -- to what's called S6 stage or fully differentiated state or islet cells where you can inject those cells, either as naked cells or in the device to achieve the same end goal in the end result. And I think having multiple shots and goals is a very good thing for us. And we look forward to advancing our programs as well as the Vertex programs advancing at the same time.

So there should not be any read-through between these 2 verticals.

Well, at this point, no read through between what the Vertex programs are doing versus ours. I think we just want -- from both our standpoint and work at standpoint, I think the more shots we have in this very attractive market, where we have the first-mover advantage, the better off it's going to be.

Okay. And kind of on the data end, what should we expect this year from the earlier programs so VCTX210 and 211.

Yes. 210 was a safety run-in. I think this was, as I said earlier, the first time that we put edited iPS to ourselves or edited stem cells or [indiscernible] cells into patients. And that's an important technological advancement, but we wanted to demonstrate the safety of it. And so we'll work with the agency to say that how do we do a safety run with the lower dose of cells in the device to prove the safety of this approach, which allowed us to move into 211. With 211, we have a higher dose of cells, and we have a more optimized clone to produce greater level of insulin in these patients. And what we want to see is data for a handful of patients with at least 6 months of follow-up. And the reason I say 6 months is it takes these cells about 3 to 4 months to differentiate fully inside the patient's body, and only then do they start producing insulin as measured by C-peptide production, which is generated when proinsulin is broken down into insulin. And so we do want to wait 6 months for these patients to show whether steady-state insulin production might be. And that's the data that we're looking for. Whether we get there by the end of this year or whether it's early next year, I don't know.

Okay. I'd like to spend the last few minutes that we have before we take a couple of questions from the audience on your in vivo programs. So maybe you could shed a little light on the reasoning behind going forward with ANGPTL3 as your initial target?

Yes. ANGPTL3 is going to be a very important target in cardiovascular medicine. You have several pharma companies working on it now and so is LTLLA or LTA. These are both the risk factors that could be very important in cardiovascular disease and especially mitigating cardiovascular disease. So ANGPTL3 is a protein that actually works by inhibiting lipoprotein lipase, which acts on triglyceride-rich lipoproteins. And so inhibiting activity, I think actually helps reduce triglycerides by almost 60% in some clinical trials or by assessing natural history data and also reduce your cholesterol levels, LDL cholesterol levels. And this is all with a single dose of CRISPR Cas9-based medicine that could lead to this level of reduction. So we're moving forward very quickly on as fast as we can on these programs. What I said is we hope to be in the clinic with our new programs this year for at least one program and then quickly follow that with the second program in LTA.

Okay. And is the strategy to focus initially on smallish kind of genetically well-defined indications and maybe expand later? Or maybe on the [ agnostic ] approach?

Absolutely. I think what we want to do is start with indications where but there's homozygous familial hypercholesterolemia or patients that have uncontrolled triglyceridemias with certain manifestations, where we can have for relatively easy biomarkers. I think people with very high LPA levels, that's what we're going after first so that we can hopefully have single-arm trials that get us to approval and then follow that up with confirmatory trials.

All right. We are coming up on time. I do have one question from the audience, and this is specific to recent ICER report. The question is where is ICER getting their information about the safety of exa-cel?

Yes, it's unclear. ICER doesn't have access to any of the nonpublic clinical trial data. They do have -- they have an advisory committee, where they have experts who may or may not be exposed or be part of the trials that are ongoing. But unclear to us what their sources of data are, but I would say that they generally don't have access to nonpublic data.

Excellent. Sam, thank you very much for taking the time today. Great discussion.

Yes, Thank you very much for having us.