Okay. Hello, everyone. My name is Daina Graybosch. My team and I here at Leerink Partners cover immuno-oncology, including cell therapy. And under the cell therapy part of our coverage, I'm really excited to welcome Scott, the CEO of Fate Therapeutics.
Thank you.
Thank you for your time.
Thank you. Thank you for having us. Appreciate it.
We are gonna jump right in and spend a half an hour talking about allo-iPSC-derived CAR T and CAR NK. Because it's the topic du jour, we're gonna start with autoimmune disease.
Okay.
But I do want to save some time for some oncology questions as well.
Sure. Happy to.
Okay, so first, we've seen data in autoimmune disease from a few autologous CAR T, and I wonder, what's the value proposition for an allo CAR T and your product, FT819, in particular, in autoimmune disease?
Yeah, so far, I think what we've seen in relatively small number of patients with the autologous CD19-targeted products is very remarkable. We have seen essentially a full reset of the immune system and long-term drug-free remissions, which I think has surprised most folks in terms of the profoundness of the responses. I think as terrific as that has been for a small number of patients, it is only a small number of patients, and the paradigm that has been used to treat these small numbers of patients is much more of an oncology paradigm.
Mm-hmm.
And so it involves the typical CAR T cell treatment, which we've all, you know, come to love in oncology, but you're talking about taking patients off immunosuppressive therapy. You're talking about performing apheresis, hospitalizing patients for 10-14 days, and taking them to academic medical centers. And at the end of the day, I would say that that's not the right treatment paradigm for a patient with autoimmunity, and you've talked about this a lot in your research. Even for the oncology paradigm, we faced significant logistics in trying to serve all the patients that would deserve CAR T cell therapy, and there are significantly more patients in the autoimmunity space. Folks have talked about the TAM being 10x larger than oncology.
In order to reach these patients and deliver those types of profound resets, I think we're gonna need a different model for that.
When I hear you say that, I think allo NK. Which is you-
What about allo T?
But, but-
What's wrong with allo T?
Tell me why I'm wrong. Like, how can allo T also change the paradigm across that inpatient? Obviously, apheresis-
Yeah.
but inpatient immunosuppressant. Some of it is still a T cell, after all.
It is a T, it is a T cell, after all, which I like that it's, FT819 is a T cell. So let's be fair, the medical precedent that's been achieved to date is, in fact, with a T cell. And so I think there's, including in oncology, right? I think generally speaking, in oncology, we would agree that T cells have outperformed NK cells. Doesn't mean that that will be true in autoimmunity, but yes, that I think that's been true in oncology. And so I think there's a lot to like about for us with FT819, starting with our iPS-derived CAR T cell program. I think when I look at the data sets that we've generated to date with FT819 in oncology, it wasn't that difficult of a decision to decide to start with that in autoimmunity.
If we look at, for instance, the behavior of FT819 when we deliver it to a patient in the oncology setting, we see your typical autologous CAR T cell profile. We see cells begin to expand rapidly within the first week. We see PK peak at day eight to 11. So clearly, we see CD19-mediated expansion. We see cells persist through the first two to three weeks, and we see that in a dose-dependent manner because we've done dose escalation. We see kinetics of B-cell depletion, rapid and suppression through day 30. So knowing a little bit about what we think might be the mechanisms in autoimmunity, that have been shown to date, I think, you know, we looked at it and said, "Gosh, there's a, there's a lot to like in starting with FT819 in autoimmunity.
Do you still risk having to be in an academic center because of the T cell expansion and the toxicity, or do you think that that's just gonna be different in autoimmune?
I think, rightly so, the FDA is being cautious in bringing CAR T cell therapy to autoimmune patients.
Mm-hmm.
I think, for most of us, I believe they are mandating that we treat patients as if they were an oncology patient, so taking them to an academic medical center and hospitalizing them. Although we only have three days of hospitalization under our protocol, which I think is different. So I look at that, and I think that from, from FT819's perspective, I do think we have the potential to begin to bring that product candidate outside of the academic medical center and more into the community. We have certainly been able to do that with our NK cell products, and so I think that the safety profile we've seen to date with FT819 has, in fact, also been differentiated. So we see, we see no ICANS, we see no GVHD, and we've seen, in a small number of patients, mild CRS.
And so I think we have a safety profile that's also differentiated and can support movement away from the academic medical center and into the community, even with a FT819 being a T cell.
So why take FT522 as well? It's autoimmune.
So FT522, there's a lot to like about FT522. So FT522 is a CAR NK cell CD19-targeted. And I'll start by saying there's a lot we still don't know about autoimmunity. Part of the value proposition of an off-the-shelf cell therapy is the potential to move away from cy/flu conditioning, which has been given today with CAR T cells, including our own FT819. Because we have an iPSC platform, because we can introduce multiple, multiple pieces of synthetic biology into our cells, FT522 includes a novel but we think very important and differentiating piece of functionality. We refer to it as an allodefense receptor. So we've engineered into the iPS cell line for FT522 an allodefense receptor, which is essentially a CAR that targets 4-1BB.
This piece of functionality was specifically designed to overcome the requirement to give patients cy/flu conditioning in order for the cell to thrive and be active. And so what we see is preclinically, when we are able to deliver FT522 in preclinical models, we deliver FT522 in the background of a competing allogeneic system.
Mm.
We actually see FT522 become activated off of allogeneic T cells and thrive. So this suggests or indicates to us that FT522 can be given without cy/flu and can actually be activated and potentiated by a patient's existing immune system. So we would prefer FT522 to be given in the background of a patient's immune system without cy/flu conditioning.
Can you take those same edits and apply it to a T cell? So I think you said in your last earnings call, you can take the 4-1BB ADR. Can you also take a cytokine support and-
Yes.
Move away from cy/flu of a T cell?
Yes. So, we absolutely have. It's not the same. We don't use the same. In FT522, we rely on an IL-15 receptor fusion.
Yep.
When we are moving on the T cell side, we are utilizing an IL-7 receptor fusion. So you can absolutely build in cytokine support into your T cell constructs. In addition, as you mentioned, folks have asked us, "Hey, can you embed ADR functionality into your T cell platform?" And we have done that. So we have multiple master cell lines that incorporate ADR technology on the T cell side as well.
Where is that into when a construct like that would come into the clinic?
I would say, I would say in the, well, be conservative, first half of 2025.
Got it.
I think you could expect to see our first ADR-armed T cell product, IND submission.
Why in autoimmune, are you focusing on SLE first, versus some people are going more narrow for nephritis-
Yep.
And some people are going broader for multiple autoimmune diseases?
Learnings. So again, I think it's early days in autoimmunity. Certainly, want to engender learning as much as we can, as quickly as we can. And so treating different types of patients with SLE, I think, is of value, number one. Number two, I also agree with that, the B-cell-mediated mechanisms of depletion and reset may be applicable to other autoimmune diseases. So expansion into other autoimmune disease, at least B-cell-mediated autoimmune diseases, is of value to us with both 819 as well as 522.
Can you talk about... We've talked a lot theory here and a little bit from oncology, like the safety profile. What's your strongest preclinical support for both or either 819 and 522 in autoimmune disease?
I think, I mean, the cells do not necessarily know what type of disease they're going into. Let's keep that in mind, right? So the target is a CD19 bad-acting cell, whether that's an aggressive DLBCL cell, whether that's an indolent lymphoma cell, or whether that's a normal B cell that has been programmed to secrete aberrant antibodies. So I look at all the preclinical data at some basic level as being relevant to: Can your products essentially clear and control CD19-positive bad-acting T cells? I think when I look at the preclinical data, and I relate this a little bit to the clinical data that's been seen out of the folks in Germany, certainly, depletion of B cells in circulation is important. I'm not diminishing that.
But I think what's true in solid tumors, and also likely true in autoimmunity, is there are bad-acting CD19-expressing cells that are residing in secondary and tertiary compartments. And so, when I look at, for instance, preclinical models, some of the most interesting preclinical models with respect to proof of concept in autoimmunity and the ability to reach into those tissue is actually on the solid tumor side.
Mm.
So, you know, you can create model systems of solid tumors where, you know, cells are required to home... cells are required to infiltrate. They're required to leave the blood, essentially, find their target, attack, and attack their target. And so some of the data that we've seen on the solid tumor side, I think, is actually most interesting as it speaks to autoimmunity and your ability to reach those secondary and tertiary compartments. And again, I would say, as much as we like NK cells, preclinically, we have seen, at least in the solid tumor setting, T cells can-
Mm.
Outcompete NK cells with respect to homing and infiltration and activity in more difficult-to-reach compartments. Now, to be fair, if you add a monoclonal antibody and provide additional targeting for the NK cell to synergize with, it's a fair fight.
Does that mean you'll quickly move to adding a monoclonal antibody with FT522?
I think it's one of the differentiation features of an NK cell, is that. And again, a lot we don't know with respect to autoimmunity, but certainly there is an element of disease that is associated with the plasma cell compartment, as an example, whether that be sort of the plasma cells or long-lived sort of plasmablasts, plasmacytes. FT819, FT522 certainly does have the ability to combine with monoclonal antibody and reach all the way from an early precursor B cell, down, all the way down the line into a plasma cell. And I do think, as you know, there are folks developing CAR-19 bi-CAR-BCMA dual-targeted cells, and discuss a value proposition about the breadth of their reach-
Mm.
into sort of the different aberrant autoantibody compartments. And I think FT522 does bring that to the table, and it's something we certainly look at when we submit the IND, the potential for monotherapy, but also the potential to combine with monoclonal antibody and reach deeper into the plasma cell compartment.
You sort of have alluded to on your earnings calls that you might take different development strategy. If I take what you're saying now, with those past comments, I'd say more likely 522, you'll pick a different disease that maybe would benefit from a broader targeting?
I'd like to be able to run more of a basket study with FT522. Yes.
Mm.
Yes, and identify diseases that may have more of a plasma cell component to them as an element of differentiation.
Yeah. You have data coming up with FT819 at ASGCT in May, and you said that some of that will be pharmacodynamic data of relevance to autoimmune. I'd wonder if you could talk about what type of data, and then why do you believe, you know, some PD data in oncology is a surrogate for how well your cell will perform in autoimmune?
Yeah, I mean, touched on this a little bit, right? So the product going into the patient, I don't necessarily think the product knows what disease the patient has. It's all about-
Yeah
... can this product engage with the target? And in this case, in autoimmunity, the bad acting target is a CD19 expressing B cell. And so I do think the read-through, if you will, in oncology, where you have CD19 targeting CD19 expressing CD19 CAR, targeting CD19-expressing malignancies, is a really good read-through with respect to your ability to deplete CD19 positive B cells, suppress them for a period of time, and reach into tissue where these bad-acting B cells may be harbored in autoimmunity. So I think all of that is relevant. I mean, I think the folks with the oncology data on here, to the extent there's open questions, have you developed a cell?
Do you have a CAR construct where that cell can home, where that cell can infiltrate, where that cell can have activity in tissue? I mean, gosh, you can look at PET scan, or sorry, CT scans in the oncology setting and have confidence that your cell can reach and deplete tumors in those types of tissues.
Do we expect CT scans?
Yeah.
Of cells?
I mean, I think that's part of it. I mean, I think, well, you know, certainly, like, look, B-cell depletion in the blood is an important proxy.
Yes.
Is that, the whole story, is that sort of sufficient? It's probably not sufficient. There are these bad acting compartments that reside in other tissues of the body, whether they're in the bone marrow or the liver, what have you.
Yeah.
And so being able to demonstrate that your product can reach those areas and be active, I think is really important.
One question we get is, if you look at some of the data of CAR NK, we've got it in, on the updated MD Anderson data-
Mm-hmm.
with their CAR NK.
Mm.
They have in the supplement, you know, B-cell depletion by patient.
Mm-hmm.
I think the really notable thing about that chart, in addition to it being nonlinear on both axes-
Mm
... is that it, the patients start with pretty low baseline.
Yes.
Is that a problem? You know, higher normal baseline could act differently.
Absolutely. So I do think that the data that has been shown to date, in oncology, has been, at some level, colored by the fact that almost all of us treating patients with B-cell malignancies, the B-cell counts in the blood at time zero are very low.
Yeah. Yeah.
Right? And, and that's not necessarily the case in autoimmunity. Not, not that they're high in autoimmunity, because you can look at the Schett data, right?
Yeah.
Most patients have relatively low to normal B-cell counts. In oncology, though, I think another thing that I didn't touch on, but provides us confidence is, you know, we do have a lot of data with, for instance, FT819.
Mm-hmm.
And certainly in oncology, you might expect that at baseline there are patients with like as a relative to autoimmunity, super physiological levels of B cells in the blood. And you can start getting a sense of what capability and capacity your product has-
Mm.
With respect to depletion. So it's one thing to say, "Hey, at time zero, we have 20 B cells in the blood, and we can deplete 20 B cells." It's another thing to say, "Gosh, at time zero in the blood, there's 20,000 B cells.
Wow! Yeah.
We can deplete 20,000 B cells. It's another thing, as an example, to say, and, you know, we'll look to speak more of this at ASGCT, there are patients that B cell counts do not go down with cy/flu. So some of the discussion that's been talked about in autoimmunity, what's the contribution of cy/flu-
Mm.
to B cell depletion?
Interesting.
Right? There are certain patients we have where cy/flu has not depleted B cells, and so you can identify specifically in those instances what's the activity of-
Interesting.
For instance, FT819 or one of our CAR NK cell products with respect to B cell depletion.
That's sort of something you could only see that contribution in oncology?
Yes.
Where cy/flu are probably always gonna be active-
Yes.
-in, uh-
Yes. And so, we were surprised at some level in beginning to prepare for ASGCT, where we saw certain patients that were essentially resistant to B cell depletion with cy/flu.
Yeah. I think, the big question that often comes up is the iPSC source.
Yeah.
So what gives you confidence that iT and iNK are behaving similarly to just regular allo T, donor-derived, not regular, sorry, donor-derived allo?
Sure. I think, preclinically, I mean, as you know, I mean, we do exhaustive comparisons against the primary counterparts, whether it's a primary NK or a primary T. And it's not just in terms of in vitro or in vivo experiments. We've done gene expression mapping, and so we feel very comfortable that... I'm not saying we're perfectly recapitulating biology. It was never the goal-
Yeah
... to perfectly recapitulate biology. But when I look generally across the field with respect to, gosh, how are our NK cells comparing to others in the NK cell field? Certainly, I'll put our product candidates up against any other NK cell program. And then on the T cell side, again, I think at least when we deliver our product to a patient, we're seeing FT819 looks very different than, for instance-
Yeah
... our NK cells. We see very different patterns of behavior, and we see FT819 performing like you would expect an autologous CAR T cell to perform, with respect to homing out of the blood, expansion, and peak PK. And so when I look at it, I think, look, we were never into an academic exercise about recapitulating sort of NK cell-
Yeah
... biology in its truest form. Let's be serious. All of us are engineering in-
Yeah
... synthetic biology into these cells. At some level, it's the synthetic biology that's performing. I am certain, though, our NK cells perform differently than our T cells, so, you know, at some basic level, we are making very different types of effector cells, one on an NK cell side and one on a T cell side, and they perform and look differently.
Can we talk about enrolling these studies?
Yeah.
I think it's been for me a bit of an obsession, like what's really driving the challenges, and is this a signal of commercial challenges to come? Because it's about the rooms, you know, collaborating with these, you know, ivory tower transplant docs. That's always gonna be an issue, and how are you mitigating that risk?
I'll go back to a comment I said. If we're going to treat autoimmune patients like oncology patients, which has been done largely to date-
Yeah
... that is a limiter on autoimmunity and the potential for autoimmunity. So I believe autoimmunity will require a completely different therapeutic paradigm to reach and penetrate that market.
Yeah.
This is where I think the reality of it is, that you will need a mass-produced product. You will need a product that can be produced in a cost-effective manner. You will need a product that is capable of reaching patients, outpatient, in the community. You will need a product that, for all intents and purposes, is a cell therapy that looks like a monoclonal antibody, which is exactly what our platform is.
Even with cy/flu for FT819?
I do think cy/flu is also a limiter.
Mm.
I do think that the right solution for these autoimmune patients is to move away from cy/flu.
Yeah.
And I think we will test this with FT819. So we will test our ability to take FT819 off of cy/flu and see its performance. But we also have the platform, like we've first-
Yeah
... taken the step with FT522, whether it be with NK cells or T cells, where we think we have solutions where we, our cell therapies can thrive without cy/flu conditioning.
Actually, it brings me to probably my most scientifically-
Uh-oh.
intense. I think you've said that with standard-dose cy/flu, FT819 looks like an auto CAR T. That's really different than the other first-generation allo CAR T, where all of those have had to require enhanced lymphodepletion to get similar PK, or some sort of passive allo evasion, and you have neither of those. So what's up?
Well, I don't think we have the same response rates that, for instance, Yescarta .
Oh, I think you said PK.
Yeah. Yes, yes. I just want to be clear. I don't think FT819 is outcompeting-
Got it
an autologous CAR T cell in aggressive lymphoma, for instance. It's, it's not. But is it behaving like-
Got it
You would expect a CAR T cell? Yes, it is behaving in the manner that I would expect a CAR T cell would behave. I'm not an expert on others' programs in terms of how their allogeneic cells have performed with standard conditioning. Most, as you know, have not even tried standard conditioning.
Mm.
Most of them immediately went to enhanced conditioning to drive therapeutic efficacy. We started with standard conditioning. We thought that was the right thing to do, and while I will say I don't think we're outcompeting their autologous counterparts with respect to response, we are certainly seeing a profile that does look with respect to sort of how they behave with respect to expansion, PK, dose-dependent activity, homing to tissue, clearance of tissue, or clearance of tumors, in secondary and tertiary tissues, much like you would an autologous counterpart. Now, there's one caveat to what I just said. The reality is, we are not treating the same patient that the autologous CAR T cell programs are treating.
Yeah.
Half the patients, more than half the patients in our study with FT819 previously, previously received CAR T cell therapy.
Yeah.
So the patients we're treating are not CD19 naive with respect to targeted treatment. So I wouldn't necessarily expect, I think based on all the data that's out there, I wouldn't necessarily expect FT819 to be similar to a CD19-naive targeted therapy.
How do you give FT819 without lymphodepletion? Isn't it immediately just gonna get removed from people's NK cells?
No. So, I think the interesting thing about the autoimmunity space as is a standard of treatment is immunosuppressive therapy.
Mm.
Patients in autoimmunity today are treated with immunosuppressive therapies. I believe that an allogeneic cell therapy, like FT819, could be added to an immunosuppressive therapy, and allow the cell to thrive and clear B cells.
So, so far in the academic-
As an example, Cytoxan-
Yeah
is given in multiple cycles.
I get it.
... to patients, right? So for instance, Cytoxan is given once a month outpatient for six months to control B cell activity.
So you-
Cytoxan's immunosuppressive. You could imagine adding a cell therapy to an immunosuppressive regimen to essentially drive more rapid kinetics and deplete deeper depletion of B cells to allow for immune reset.
So instead of having to have patients stop their standard of care because they're getting cy/flu, you keep them on their standard of care, and you see if you can give this?
You add cells. It's not dissimilar to an oncology approach-
Yeah
in solid tumors, where, for instance, you're typically adding on to standard of care to significantly improve efficacy. And so a next step away from cy/flu would be, okay, your standard of care in a community setting involves an immunosuppressive therapy. We're gonna bring cells to that and now achieve, potentially achieve a profoundly different outcome, i.e., immune reset.
I'm gonna ask one oncology question. It's gonna be on BCMA.
Okay.
'Cause you and I have talked about this-
Yes
... so I'm interested in what you're seeing.
Yes.
I think maybe surprising in MM with the BCMA CAR T, I think, for me, it's pretty clear you need good bridging therapy, and the auto CAR Ts are doing much better when you have appropriate bridging, and then you maintain that, which isn't exactly the same thing we've seen with CD19. You know, with FT576, have you learned from that? Are you gonna have bridging? Can we interpret these data, or is there gonna be-
So we don't have-
More?
We don't have bridging therapy. So I mean, I've obviously, you and I have talked about this. I find this a really interesting observation that potentially the CAR T cell therapies, the autologous CAR T cell therapies, have been aided-
Yeah
... in their ability to drive responses with bridging therapy. That's in conflict at some basic level with one of the benefits of an off-the-shelf cell therapy, where you can intervene immediately with your cell, and you don't require the patient to go on to experience bridging therapy prior to administration. That could be a headwind, as an example-
Yeah
... for something like FT576, because you're asking FT576 to do all the heavy lifting right out of the gate, as opposed to in the autologous paradigm, what it really is, is bridging therapy-
Yep
... to help deplete and control tumor, followed by cy/flu, followed by cells.
Yes.
It's really different than, "Hey, we're just bringing FT576 to the table." And so, yeah, that's a potential headwind for what we've done with FT576 in the study.
Why not change what you do? See how-
I mean, right now, we are-
More auto-
So we are in a dose escalation, and we are trying to understand activity of FT576-
Right
... including the BCMA binder, right? We have a novel BCMA binder, which, another half-hour conversation, that could be brought into autoimmunity.
Yeah.
But as you know, there are significant differences with respect to how BCMA binders have behaved-
Yep
... in oncology. And so one of the things we're just trying to validate is we have a novel BCMA binder. As we've increased dose, do we see increasing activity with the BCMA-directed therapy?
Awesome. Well, thank you very much.
Thank you.
Thank you, everybody, for your attention.
Thank you so much. Thank you.
Thank you.