To have my first fireside chat with Autolus. We have the CEO, Christian, here with me today. Welcome, welcome to Miami.
Thanks for having us.
Maybe before I jump into a whole set of questions we have, you could just give us a quick introduction to yourself and the company.
Yeah, sure. You know, we started Autolus 10 years ago, just got through our first approval end of last year. It's a company focused on the CAR-T space, where the first product that we developed for acute leukemia. It's a spin-off from University College London. The company is located in the U.K., but with quite a significant presence in the U.S. as we're launching in the U.S. since the latter part of last year. My own background, I've been doing this for quite a long time. I was before in the T-cell engager space, kind of with Micromet, pioneered that space and got the first product onto the market, transitioned the company and sold it to Amgen. Obviously, that program has taken off very nicely.
It's got sort of the additional second layer with tarlatamab that followed, which was very nice to see the team continue to be very successful. There has been probably 20 some years looking to find ways to talk T-cells into attacking tumors.
In IO.
Sort of what I've been sort of focusing on.
Your first approved product, Obe-cel, so I thought we'd talk about it commercially and then move into the pipeline and other products. Can you just help us understand how it's how is the launch tracking towards your ingoing expectations?
Yeah, we got the approval on November 8 last year. It was the first approval for a CAR-T therapy without a REMS. There was no obligation to collect data and report the data on CRS and on ICANS, which is obviously a reflection of the quite unusual safety profile that we have with the product. We're very pleased with that because that's an important step also in the field to get to products that have a lot better safety profile. As we're sort of getting to the actual approval time point, we had 30 centers that were ready to get activated. As you know, the centers have to be accredited that deliver CAR-T therapies.
There is quite a range of agreements that you have to put in place, trainings that have to go, have to be set up for a center to be able to deliver CAR-T therapies. That was ready to the point where with the label received from the agency, the final step of the activation could be initiated. That time is really driven mostly by the centers themselves because it is mostly their internal processes of getting a new therapy into their internal processes and get it taken up. That can vary a bit in terms of the time. We are tracking really well. We were already at the beginning of the year, first week of January, at around 20, early, low 20s in terms of number of centers that were activated. We are now above 30 centers that are activated.
We continue to actually bring centers on board with the goal to have approximately 60 centers by the end of the year. To put that in perspective, 30 centers give you approximately 60% of the patient access in the U.S. 60 centers will get you beyond 90% in this indication. We're rolling very effectively. I think we've been very pleased with the speed at which the centers actually activated, which obviously is very much driven by their own incentive to actually get the product in their systems. What was very helpful in the process was that we did not only have obviously the approval at the beginning of November, but also at the beginning of December, we had the clinical study that was basically the foundation for the approval. The FELIX study got published in the New England Journal.
Two weeks after the publication, the product was actually included in the NCCN guidelines. The importance of that is that at that point, it becomes a recommended therapy. As you actually start with a new product and you launch a new product, obviously the payers need to evaluate the product. They do not start with the product being on policy. It is an individual patient assessment. Actually having the ability to refer back to the guidelines as a recommended therapy is incredibly helpful in that process. Those are all elements that really helped us get off the ground very quickly. We are very pleased with the dynamic that we are seeing in the market with a lot of centers actually starting to be already on several patients included in therapy, which is very promising.
Because of the ALL indication, are these patients more naturally seen in these academic settings? Are all these centers already giving other CAR-T, expanding anything here?
The disease setting, just to maybe spend a moment on that. Acute lymphoblastic leukemia is a very, very aggressive form of leukemia. It originates in the bone marrow and it basically crowds out the marrow and renders the patients immune suppressed as a consequence of the disease itself. The way we treat it is actually in the front line, it's with high-dose chemotherapy. That high-dose chemotherapy actually lasts up to about 18 months to 24 months for adult patients. Kids can go up to three years. It is an enormously challenging disease because you have a cell that's driving the disease that is very, very close to a stem cell. It has an enormous proliferative potential. Managing these patients is actually very difficult because you already start with a patient that's highly immune compromised. You make that worse using high-dose chemotherapy.
A lot of these patients need a lot of support. They may end up in ICU to be managed on the standard of care. As a consequence, these patients tend to get aggregated into the academic transplant centers as the places that can manage this level of complexity of care. Those are also the centers that actually are very experienced with CAR-T therapy in general.
Obviously, having a product like ours fits very well because it gives them a new opportunity and a new modality to work with, but also gives them a product that is very well manageable, which is important because obviously the time it takes to manage a patient and the involvement that you have to manage a patient has a huge impact on the throughput the center can manage, the number of patients they can manage, and frankly, the amount of benefit they can generate overall. It is a very important factor for the centers to actually have products that they can manage well that does not require them to go to ICU and actually have basically patients that they need to manage for a few weeks until they get through their crisis. That is why the product resonates extremely well with the physicians.
Got it. Once you're onboarding these ATCs, how is it going? Once it's approved and they have their internal processes, how are you seeing the ramp-up of Obe-cel use in each ATC?
The disease that we're looking at is obviously one of the forms of leukemia that is on the rare side. We have a total of about 1,500 patients in the relapsed refractory setting in the US. This is not comparable to multiple myeloma. It's a substantially smaller indication, but it's an incredibly high-need medication indication and one where once the patient actually presents with a relapse, you don't have much time. You need to act. You can't actually sort of control these patients and then try to manage them for a while. You really have to address the disease head-on. What that means is that there is not a bolus of patients that are sort of built up at the time of launch. It's a much more gradual, incidence-driven kind of dynamic that you see with the patients coming in.
We've been very pleased with the number of patients actually coming in through those centers. One of the key things that we're looking for is not only actually a center to have their first patient, but actually to have repeat use at a center. It is when you see repeat use at the center that you start to understand that that center actually is starting to really deploy this therapy because there is obviously experience, positive experience that gets basically reinforced through the subsequent patients that are being treated. That is the dynamic that's really the key dynamic in terms of the take-up from a demand perspective. Very pleased with the start, incredibly motivated centers. It is a community of physicians that deals with a group of patients that for most of them, they're going to lose them. It's a very, very tough environment.
Physicians are quick to adopt because they need therapies that work. That is actually very helpful from a launch perspective because it's not a situation where you have a lot of other options that work well in the setting. You might actually just gradually try something new, but you're kind of okay with what you got. This is not where this is. This is really a setting where you need alternatives that really give you a different outlook for these patients.
Can you talk a moment about the outcomes and maybe how they compare to the other options, including the other CAR-T?
One of the key premises behind Obe-cel or AUCATZYL is that what we were observing when we started out in the field was that we obviously did see remarkable levels of activity from CAR-T programs, but we also did see a substantial amount of toxicity. It was quite clear that that shouldn't be the case because the killing, actually, when you think about the cell's modality, the cells engage, they kill the target cell. That's a very quick process. What we're seeing is that when that happens with a normal T-cell, there's actually very limited cytokine release that happens and very limited amount of activation or over-activation that happens of the T-cell. It's a very economic process. What we did see with the original CAR-T programs is that that was actually not the case with the original designs.
They did actually manage to get a kill, but then were basically getting stuck on the target cell. As a consequence of these cell-cell interactions, the CAR-T cell, after delivering the kill, got highly over-activated, produces a lot of cytokines that have no clinical benefit, and were driving a lot of toxicity. What we were looking at was to find a way to dissociate the desired activity from the toxicity. What we realized, what we needed to create was a product that could engage very selectively, deliver the kill, but then also have an ability to disengage rapidly. Technically, what that required was a product that actually had a fast off-rate. It was a very unusual phenotype and feature that we actually had to design into the product that gives us now in the clinic a very different profile.
Our adverse event profile looks very different. When we look at, in this very, very challenging population, we have high-grade CRS levels in about 2% of the patients, which compares to other programs. As an example, Kymriah in pediatric patients in the ELIANA study had 47% of the kids in high-grade CRS. We have 2% on the adults, which are much more susceptible to those adverse events. We have high-grade neurological toxicities, which are transient and fully reversible, in the range of about 7%. That is, these are numbers that are actually below the numbers for Blincyto, which is the standard of care now in this setting. They're definitely way below any other CAR-T program that's been explored in this indication.
When we then go into our experience in non-Hodgkin's lymphoma as an example, we see actually that in those patients, we have no neurological toxicity and we have no high-grade CRS. It gives you a very clear understanding for how this profile really translates. Now, it doesn't only translate actually from a safety perspective. It also actually made the product more active because the cells were not getting stuck after the initial kill. What that allows the CAR-T cells to do is to re-engage. You get actually more rapid serial killing. Actually your kill rate goes up, not down. The growth of the CAR-T cells does not get inhibited by the cells being stuck on the target cell. You get also more CAR-T cells formed in vivo. Our peak, as well as our area under the curve, is much different.
Combined with the fact that we're not actually exhausting the cells, we see persistence in these patients up to three, four years, and we have seen continued persistence even beyond that. It gives us an ability to have extremely high level of activity, sustained pressure on the leukemia, but arrive at that level of activity without actually inducing the level of toxicity that was seen previously.
You're now taking this to other indications. I wonder if you could just give us an overview of the plans in further indications in oncology, but also the potential why you're excited about it for autoimmune disease.
Right. So we're interested in sort of, we see two opportunities or sets of opportunities for the product. On the oncology side, one of the things that we did observe in the acute leukemia setting was that we actually got patients in long-term survival. We see that when you look at the survival curve, the curve actually stabilizes, goes horizontal. We see it on EFS. We see it on the survival curve. We see that actually with the patients not receiving any additional therapy after the CAR-T. That's very unusual. It's what we were all hoping to see at some point in this disease setting. We see it clearly with Obe-cel. We have an ability to really have a very profound impact.
What we also did see is that the level of tumor burden at the time of dosing was highly predictive on the outcome for the patients. If you actually manage to get the tumor burden relatively low before you dose, you have a very high probability that your event-free survival will stabilize at a very high level. As an example, if you're below 5% tumor burden, your event-free survival stabilizes between around 65%. That's obviously a very remarkable level of outcome. The sweet spot, when you think about it from that perspective, to maximize outcome is to go in patients that are relapsing or have initial disease and actually have a low level of disease burden before you dose, with the goal to get long-term outcomes in these patients as sort of a definitive treatment. That's the sweet spot of the product.
There are places in non-Hodgkin's lymphoma where you have similar situations where indeed you get to low disease burden, but you have still a high probability of relapse, which are based on certain remaining levels of cells, which are typically referred to as minimal residual disease. Those are areas that we think are particularly suitable for the product and would fit the profile very well. The second area we're looking at is in autoimmune disease. In autoimmune disease, we've seen quite a remarkable set of data initially out of compassionate use data from a team in Germany that was able to show that you could basically get a reset of the autoimmunity by removing the CD19 positive cells.
The surprising part of the data was not that you could remove the memory B cells and impact longer term, but that you can actually get a very fast effect by removing the plasmablasts, which are early forms of plasma cells that are still expressing CD19. That was part of the biology that was unexpected. It was quite remarkable in terms of its consequence because what they were able to show is that you could manage these patients from a very refractory state to a point where they actually had a significant improvement in their clinical scores within a matter of about a month to three months' time.
What you want to have when you then think about this, what you want to have is a product that actually can do this, that is very safe and can be administered ideally in at least a hospital outpatient setting as a way to sort of actually provide broader access to the type of therapy. We think our product has an exceptional profile in terms of the safety as well as the efficacy and the ability to remove CD19 positive cells. We've shown that we got the vast majority of our responders, as an example, in ALL are at below 10 to the minus 6 level of MRD. It is basically not measurable levels of MRD. You get very, very deep responses, which tells you a lot about the activity of the product. Obviously, we get there with an excellent safety profile.
We think the product is in an excellent position there. The other aspect, which is, I think, attractive, is that it's the only product actually that's being explored in autoimmune disease today from the CAR-T segment that actually has an approval anywhere else, that has an actual commercial manufacturing base, that has a presence in clinics commercially. That is a key, is what we believe of key importance because obviously it reduces the burden in terms of what you have to show for the product. You have already a significant safety database.
Also, when you think about it from an investment perspective, you do not actually have the burden of actually having to invest in actual manufacturing capacity and capability, et cetera, and also do not have to take the risk associated with the manufacturing and CMC packages, which have been notoriously challenging in the field to get right. We think we are in a very unique position with a remarkable product, and we are going to have early data in April that we will plan to present at the R&D event that we are planning, and then obviously more substantial follow-up also from a time perspective, second half of the year.
In the April event, what should we expect in terms of patient sizes and follow-up?
What we've done is that we initially wanted to actually establish a fixed dose to dose the patients at. The original work that was done at the University of Erlangen was based on a pediatric CAR-T program. In children, you basically dose at the level of a million cells per kilogram, which is what basically the Kymriah dosing scheme that was developed for that product. We had actually used the same dosing scheme in our pediatric work that we've done with the program. We had an opportunity also to compare, frankly, the data that we have versus the data that the team in Erlangen had. It was very clear we had, if anything, obviously the better safety data. We had a bit longer persistence, but overall quite a similar overall profile for the product that we were seeing. Very encouraging from that perspective.
Now, the patients that we actually want to treat are obviously not nine-month-old babies that we need to sort of make sure we don't overdose, but we're dealing with adults, mostly teenagers and adults. You can actually go to fixed dose. We translated the pediatric dose, which we had established at a million cells per kilogram for pediatric ALL, moved that to a fixed dose of 50 million cells. The initial cohort is six patients that we're going to be reporting on. We believe in those patients already, I think pretty well kind of what the behavior of the product is, the behavior with regards to the impact on the cell types that you need to remove and the impact on the clinical outcomes for these patients as well.
I think it will give us a first look and a first feel for the product and its behavior in severe autoimmune patients.
Of numerous pharmacodynamic markers that you and others are measuring, which do you think are actually most correlated with clinical activity? For instance, we saw with Roche's REGENCY trial with obinutuzumab, it can get pretty deep depletion of B cells, but I think the activity that was in lupus nephritis was still pretty modest. Reflecting on that and all the other markers, which ones do you think is most important?
Right. Obviously when it comes to very deep depletion, the most sensitive way of measuring that is what we do in oncology. To get where we are with the oncology trials where we showed we're below 10 to the minus 6 for almost all the responders, 96% of the responders were at that level. Obviously it gives us probably the most stringent test for removal of CD19 positive cells that you could run. There's no corollary to that in autoimmune disease. Obviously what we know from oncology, a monoclonal antibody versus a bispecific versus a CAR-T, there is a very significant difference in outcome because they do not achieve the same level of completeness of removal. To get to cure in ALL, you have to completely remove those cells. There's no in-between there. We get to that level with the CAR-T.
We didn't get to that level with T cell engagers. We certainly didn't get there with ADCs, and we didn't get there with monoclonals. That is the element of the perspective where you can measure that in a very, I think, rigid and solid and robust way. On the markers that I think were particularly interesting in the autoimmune space is, first of all, you want to have an understanding whether your product actually behaves the way it should. We've seen quite a bit of variability in the data that was published last year for the various programs, quite surprising at some level, the level of discrepancies that we saw.
What you want to see is very consistent expansion, and you want to see a very consistent level of at least initial persistence of the product because you need a certain minimum amount of time for those CAR-T cells to do the job. It's a cell therapy. It takes a bit of time. Cell migration, cell killing takes time. You need a minimal amount of presence in these cells. That's the first thing you're going to observe. How consistent is that across the patients and what you can demonstrate? The second thing, you obviously want to see a correlation on B cell depletion. It will not be at the stringency measured that in autoimmune as you would do in oncology, but it gives you, I think, a sense of the dynamic and the kinetics of that.
The third area you're going to be looking at is the impact on autoreactive antibodies. You want to see double-strand DNA antibodies as an example that those actually start dropping. It takes obviously a bit of time, but you want to start to see those drop. Those antibodies have a half-life of give or take a month. You want to see the impact on the disease scores. Now, the disease scores in a lot of the autoimmune indications are somewhat complex because they're composite scores. You have elements that actually are coming from organ function. The impact on organ function is one set of components. You have an impact that looks or an element that looks on the level of autoreactive antibodies that are in circulation. There are patient and physician reported outcomes that you see.
What's important is when you look at these is you need to look at actually deconvolute these absolute scores and actually look at the individual impact that you have on the components that make up the score. That's important because what we'll have as we go into sort of a real-world setting is you will have a variability with regards to the level of tissue damage you will have in these patients. If an organ is damaged, for example, a kidney to a certain level and you have a certain level of fibrosis reached, you will not be able to revert that by stopping the inflammatory component. There's going to be an element that might be residual.
We've seen that in some of the data that was published last year where some of the patients didn't go to a SLEDAI score of zero because, frankly, they had a level of damage in their kidney you couldn't get there. The reason why there was an expectation around that came from the original work, which was very young patients with very limited impact on their organs where you could still get the full recovery of the organ. That's not to be expected if you have patients that are further down the line in terms of age or in terms of duration of disease where you wouldn't actually get to that kind of a situation. I think you need to look at that very carefully and understand the dynamic.
Those are things we're going to be looking at because I think that sort of goes to the point of what's clinically meaningful and what are really the areas you're going to impact with this therapy for these patients.
We shouldn't just say, oh, the DORIS score doesn't match, it's not as good. You really have to get under there.
You literally have to look at the.
You really have to look at the inflammation and maybe that damaged fibrosis over time might resolve a little bit or without the continual insult, but it will take a long time.
It will depend on the nature of the patients. You can have patients that are very close to dialysis. A patient who's close to dialysis, you're probably not going to get reverted. You're going to get approved over time, but you will not get a full reversal. A patient that's early on, a young patient early on, you're probably going to get a full reversal. Those are things that I think the market was sort of probably overestimating from these compassionate use data points, which are obviously not a clinical trial, which I think got lost in the conversation quite a bit. That I think is an area to look at and to understand. What we did see certainly across all other programs is a consistency in the impact in these patients, which is really encouraging.
Got it. A lot of people believe, and I think you also alluded to this, that for cell therapy to be successful in autoimmune disease, you really have to get to the referring centers. You have to be able to be used by the rheumatologists or neurologists or nephrologists, depending on the disease. I think listening to some of those doctors talk, they'd like to take chemo completely out of the equation. Do you think that that's going to be necessary to have to broadly reach patients, or do you see this evolving over time in terms of the collaboration, and where can Obe-cel fit?
I think with any new modality that gets into rheumatology or neurology, there is a transition that takes place. If you go back to rituximab entering into RA, rituximab was initially delivered by the hematologist-oncologist. As the rheumatologist got more familiar with the therapy and did realize actually these are adverse events they can manage, there was obviously a strong economic incentive for the rheumatologist to actually take on and adopt the therapy and deliver the therapy themselves. That was a transition that we did see taking place when rituximab actually got to market. It was very clear that that took place. In the CAR-T field, there is probably an element of that. What we're seeing is that a lot of the centers that are actually delivering CAR-T have basically a unit within the center that delivers CAR-T therapy.
That unit actually collaborates with the hematologists, with the oncologists, but also with the folks on the autoimmune side on rheumatology, nephrology, neurology. There are some of those models that are already being established. As an example, if you would go to Stanford, Stanford actually has capability built up to support outside of oncology. It is still the CAR-T center. There is a shared level of infrastructure that is being used and knowledge that is being used and also exchanged. There is a clear dedication to sort of actually serve this additional group of patients. I think that is where I would expect the starting point would be. Depending on the profile of the products, I think there is an opportunity to go further out and actually increase the presence in centers further out.
That depends a lot on the profile of the products, how you need to manage the patients, and whether that's economical, frankly, to do. I think that's the transition we'll be seeing is a hospital outpatient setting, which I think will be very important. In addition, using the initial, leveraging the initial infrastructure, and then eventually there's a possibility to move further out. We also have to be clear that the patients we're talking about are patients that have severe forms of their disease. We're not talking about moderate forms of the disease going on a CAR-T therapy. That would be not health economically sensible.
In the severe forms, which are patients that tend to be much closer monitored also at the larger centers, there is also actually a pretty good overlap between the presence of the patients at the centers and the centers overlap with CAR-T experience in other fields.
Got it. In the last minute, I just want to talk one question. We talked a lot about this in our pre-call, and then I put it at the end probably because we're both really interested in it, is your manufacturing. I wonder if you could talk about how your Nucleus facility is differentiated from other manufacturers and what that means in terms of your COGS and tracking towards profitability for the program.
Now, manufacturing a product for individual patients is the hardest part in this field, industrializing that process. The challenge of the industrialization is frankly that you start with highly variable starting materials. You might have patients that have come out of a leukemia crisis. You might have patients that have had relatively recent exposure to high-dose chemotherapy and so on and so forth. They may have leukemic effusions where most of the cells you could collect were leukemic cells. Your starting material is very, very challenging with a huge amount of variability around it. What your manufacturing process needs to deliver is a product with a very, very tight set of specifications. Managing that and doing that consistently requires quite a remarkable process, manufacturing process. You need to actually invest a lot in process development and characterization.
You also need to create an environment where you can automate a large part of this manufacturing process because you do it literally for each patient. The only way you get economies of scale is if you have a level of automation, and then you can actually scale that accordingly. When you think about that type of facility, that facility today, with the exception of a few companies that are active in the field, does not exist. There is no CDMO infrastructure that can really do that at scale and has the capability of doing that at a level that is actually economical. The decision we took is actually to set up our own manufacturing facility, which we did in the U.K. That facility we built from groundbreaking to actually full BLA filing in less than 24 months.
Build it and fully validate it. Then with exactly three years from groundbreaking, we got the approval for the product. We did a lot of things differently to get that timeline, which is about half of the time that pretty much anyone else needed. The importance, however, of the facility is that it gives us an ability and the level of control over the process, the operating model, and so on that allows us to get to an attractive level of cost of goods. That's really critical in this space so that you actually have a reasonable level of profitability and, in fact, do not only create fantastic clinical outcomes, but also actually have, frankly, a business that you can actually build on.
Yeah. Can you tell us that the sort of track to profitability, have you guided that? The scale you need?
We haven't guided on that. As you can imagine, when you build a facility, that facility is built for a certain scale. In our case, it's about 2,000 products that we have earmarked for to support the ALL market, which is about two-thirds of the patients between the US and Europe with acute lymphoblastic leukemia. That's what the facility is designed to do. You have to reach a certain level of use of the facility to get to a place where your overhead costs are no longer key elements or key factors driving your overall cost of goods. That's what you have to get to. That's several hundred patients you have to start getting to where that actually starts to become asymptotic as a component of cost of goods. That's really what the dynamic is.
The drive for us is, on the one hand, obviously make sure we drive our launch as efficiently as we can, but also as we go forward, add additional indications to actually max out the capacity and then, if need be, obviously think about expanding capacity beyond that.
Got it. Great. Thank you very much. I appreciate it.
Yeah. Real pleasure. Thank you.