CAR-T Topic Panel: EULAR 2026 Save

Hello everyone, welcome to RheumNow's panel on CAR T-cell and cellular depletion therapies. This is a special session post EULAR 2025 in London where we were at last week looking at all the latest and greatest in rheumatology research and drug development. It was a busy week. This panel of experts were out there looking at this category to see what was new and this is a hot new area. So let's get into it. We'll do two rounds of our favorite presentations. I'm Jack Cush from Dallas, Texas. Janet — Janet Pope, London, Ontario, Canada. Yusuf — hello, Dr. Yusuf Yusuf from Leeds, United Kingdom. And Al — Al Kim, St. Louis, Missouri. Okay, let's get into it. Let's begin with Yusuf. What was your favorite from the meeting?

Yeah, so one of my favorites is abstract POS0079. So this is about allogeneic CAR T-cell therapy. The product is from Fate Therapeutics, FT819. So the background of this is that we can improve the scalability of the CAR T-cell therapy, also whether it can reduce the manufacturing time from the autologous approach and also reduce cost.

So in the development of this FT819, they do have two regimens — one is Regimen A and also Regimen B. The difference between the two regimens: Regimen B doesn't have any conditioning. In Regimen A, which is the one presented in the poster, there are also two conditioning regimens that they compare — one is using low-dose cyclophosphamide and another one is using bendamustine.

So in terms of the development they were looking into various connective tissue diseases including lupus, scleroderma, and vasculitis, but in the poster that they presented there are only 16 patients with SLE. These patients are severe refractory and had lupus nephritis, and some of them didn't. The longest follow-up that they did have is up to 24 weeks.

So in this study, in terms of the conditioning, what they could see was if you use the bendamustine they tend to have a more rapid B-cell depletion and tend to have more rapid clinical efficacy compared to the cyclophosphamide. In terms of the efficacy, in almost all patients they responded, and the safety profile looks good as well. There were only four patients out of 16 who had CRS grade 2 or less — so nothing above that, no ICANS — and importantly with the allogeneic approach there is no evidence of graft-versus-host disease in up to two years of follow-up, and as well there is no hypogammaglobulinemia.

So I think this is really interesting to see that by using low conditioning with bendamustine we do see favorable efficacy and safety, and it'll be interesting at the next conference if they could present larger numbers using the no-conditioning Regimen B.

Yeah, there are so many aspects to the initiation and the conduct of these trials that we forget that these early phase trials are more about safety than they are about efficacy. And you mentioned the acronyms ICANS and CRS. Al, do you want to address those as major safety issues in these trials?

Yeah, sure. So I think that the experience from oncology has clearly shown us that cytokine release syndrome, or CRS, and immune effector cell-associated neurotoxicity syndrome, or ICANS, respectively, are the main reasons why people have to be hospitalized for these therapies, which dramatically increases costs. These are very easily treatable conditions if they're recognized early, but they pose major toxicity if not recognized early — in fact they can lead to mortality, particularly ICANS. I think that these are really important proof-of-principle findings, generally, that they are occurring in less frequency and less severity in autoimmune patients compared to oncology patients, which is encouraging.

Yeah. And I like that the allogeneic approach seems to be very popular here. It's got advantages, and the use of conditioning is sort of the "do we or don't we" question — you know, it's sort of like me standing in front of the mirror before I go out on Friday night about what I'm going to wear — but there needs to be some clear choices about that, and I think this particular data was good. Anybody else have a take on this study? Al, I think you were also interested in this.

Yeah. So I give Fate Therapeutics a lot of credit for doing several different things with this trial that are unusual — this modified reduced lymphoid depletion, no one else is doing that, including the no lymphodepletion arm. Of note, they did present that data a week ago at ASGCT in Boston, and they showed three out of three got SRI-4 and only two out of three got LLDAS. And I think what you were seeing with this pattern, both Regimen A and Regimen B, is not as good as full lymphodepletion,
right? and for lupus. And I think this is an interesting precedent that they're starting to set up. Um, but the other companies can't show that because they haven't, you know, taken the chance to do something like a reduced lymphodepletion therapy. Yeah. And Janet, are you — what are you looking for as far as outcomes in these studies? Are you looking just to meet an endpoint or are you looking for something superlative? Well, I mean, we're always looking, as you said earlier, for safety, but I think we're really looking for um not just short-term benefit, but I think we're looking for long-term uh true immune reset. I don't think people are going to be in remission forever, but drug-free remission or um getting to DORIS remission, I think should be an outcome, at least in lupus trials. So I'm looking for something beyond what maybe the T-cell engagers might promise us and use last word on this where the authors think they're going to go with this. Yeah. So I think uh they're going to expand the numbers for that regimen B with no conditioning. Uh and after that I think um they are going to go for um phase two B you know with comparator groups. Yeah. Interesting. Al what's your favorite? Yeah. So um uh it's very closely related to the previous poster. It's poster 0694. It is describing an autologous CAR T that's a CD19 BCMA uh dual targeting in lupus and lupus arthritis called ICG318. Um the targeting aspect is not that interesting actually and the autologous is obviously not that interesting but they also chose a limited lymphodepletion therapy just using cyclophosphamide, no fludarabine as you just talked about. The efficacy is not as good as if you do a full lymphodepletion but what's unique about this particular CAR T product is that it's also armored, i.e. it carries additional payload and it encodes actually an IL-15 with an IL-15 receptor alpha chain that allows it to make its own IL-15 and use it on the same cell. IL-15 is important because it's one of the two cytokines that allows for CAR T cells to expand and survive once they are delivered into a body. And that's the reason why we have to do lymphodepletion in the first place because if there are too many T cells there's too much competition for IL-15. So this mitigates this in some ways by being able to supply its own IL-15. So here they took 10 patients um with lupus nephritis and I think what's fascinating about this is the fact that they had nine out of 10 that had um DORIS remission which is quite remarkable. Um there was CRS — nine out of 10 of them had CRS but all of them were grade one, no ICANs, and due to the fact that they're targeting BCMA they did get a lot of infections — they had four out of 10, three COs, one UTI, and everyone was hypogammaglobulinemic which is again going to be a feature of BCMA. So again I think what the other problem we're seeing with these abstracts is that multiple variables are being moved at the same time so there's lymphodepletion reduction here with the IL-15 that's armored but they're also targeting BCMA and you have to wonder which one's controlling what in terms of the efficacy. But I think if we can extrapolate from our full lymphodepletion in lupus targeting CD19, it seems that CD19 is sufficient for lupus. I'm not 100% sure what we add with BCMA. But nevertheless, I think again there are treatment strategies that are out there being used in oncology now being copied and pasted into rheumatology to be able to try to reduce lymphodepletion because the number one reason why we can't enroll a patient for CAR T is the lymphodepletion. They don't want that. Yeah, it's well it's a scary, really scary aspect. I think it scares the doctors maybe more than it does even the patients and and that's a gigantic problem. Um the problem sometimes is you mentioning that um they got CRS — cytokine release syndrome. Does that mean they also needed to be treated with tocilizumab? Yeah. So the authors didn't report that at least in the public facing data. Um and so we were unable to obtain whether or not they needed therapy. But um you know I I think that uh this is an interesting question moving forward is whether or not we can find ways to be able to either risk stratify or even identify in real time through biomarkers those that may be requiring therapy and as a result need to be hospitalized versus those that can stay purely as an outpatient. Again, we're moving many steps ahead of where we're at right now. But I think that is kind of what the goal we're thinking about. When you cause hypogammaglobulinemia in these people, a number of them are going to need IVIG. Oh yeah. Got to carry it in your pocket. I mean, literally these pe — I mean and and it's a problem with BCMA, right? I mean, we know this is going to be an issue. And so it's really interesting like I I I don't know how that conversation you know occurs — that you need to be revaccinated against everything including your childhood
vaccines. Right. Right. All right. That's uh again it's um these are big steps um these trials. So, I think that's important and and I think it's important in getting um uh um a workforce of people that are comfortable using these therapies on these patients that really is going to advance the science.

Um, and we we need to have this workforce because when this all started, you know, with Schett's cohort and whatnot, you know, I think one of the craziest things was the scramble for new investigators for new drugs and who had a lot of us had treat, you know, a lot of experience with biologics and and inpatient infusions, whatnot, but this was a whole new game. But we have now a lot of competent people doing this research. I think it's encouraging.

Janet, what what do you have? So, I have something that actually goes along with what we've just discussed. So, this was uh poster 0678, and I picked it because I want to know which CAR T in lupus to drive, so to speak. So, this was a a smart little comparison of two CD19 CAR T therapies in lupus. And what they were trying to do, they had five patients in one group and seven from the other. So they got all the raw data to really compare allogeneic versus autologous. So there was five in one group, seven in the other. So this isn't randomized. These are people coming out of the trials. So it's a total of uh 12 patients, you know, almost half in one group, half in the other.

So they wanted to compare the molecular responses from the two groups and looking at the improved cytokine signatures. Obviously trying to find how many um like are you on single cell seq etc. having mostly only naive B cells of the ones that would be considered pathogenic. So they also wanted to look at interferon and neutrophils.

So the answer is I don't really know but they're they're similar but they're not identical. So as a for instance they have all these graphs on myeloid cell compartment B cell compartment um the NK cells and what's happening there and comparing autologous with allogeneic they there are definitely different little clusters they are not fully overlapping but in my mind not knowing totally how to read this it looks like they would be in the same sort of logarithmic change so in my mind it looked kind of the same even though they weren't identical.

They also looked at um how you would change um other other you know T cell talk and uh cells. So they looked at um ones less likely to be involved but they looked at monocytes at various um time points as well and they were looking at um some other things that might be kind of more interesting to us.

So they looked at then the bottom line is the effect size of the lupus um I'll say immune reset which is really going to be a composite endpoint of a lot um pre versus post CAR T and frankly as far as I could tell was of the 12 patients with lupus it looks like in general whether you got allogeneic or autologous that the immune reset looked pretty darn robust in all those groups. It wasn't 100% overlap again and some confidence intervals around it. They went on to look at interferon and plasma cells and neutrophils and again there were some changes um between the groups that got the one kind of CAR T and the others but they were mostly quite in line.

So I think the bottom line is here it might not matter at least from this uh study whether you get allogeneic or autologous you're probably going to be uh appropriately immune reset but there's another thing of what we just talked about is if all these had um lymphodepletion protocols that were the same that would make them more the same and I'm pretty sure on these two studies they they lymphodepleted so we just talked about less lymphodepletion or none you're going to get less um sort of your naive um CD uh cells being present at the end. So I think it's it's a good question but and they tried to answer it but for me as not really knowing all the all the pharmacokinetics and also all the cell biology at the cellular level I think it doesn't really tell us the answer.

So my question is how are we going to know that we've accomplished this concept of immune reset. Is it a phenotypic characterization or is it can it be genotypic or based on transcriptomes? I mean uh Janet do you have a feel for this?

Right. So because these are both lupus trials I mean the other thing that in all the lupus trials they're looking at which is neither phenotypic but it's downstream they're looking at full neutralization of double-stranded DNA normalization of complements um looking at some of the ENA autoantibodies and saying they should disappear in a certain time frame and I it's not always by 100 days but they should go down rapidly and then they kind of lower down more slowly after that.

So, I I'm not exactly sure. I think you'd probably also want to say um how many of what used to be the pathogenic branch of the variety of our B cells are now basically um naive. Um Al though, would you say something
different? Sort of speaking from the bench side of things. Yeah, you know, I think we still are struggling to actually define this. I think we've loosely used all these changes in the immune system that were detectable to define immune reset. You know, in fact, Paul Bernetto, who's chief medical officer at Hinge Bio, and myself, we're co-heading an effort through the Lupus Research Alliance to actually try to figure out how to define immune reset. I mean, we have to start from the beginning. But I agree with you, Janet. I think from a molecular perspective, in terms of subsets, yeah, more naive than memory. In terms of other molecular aspects, we're looking at B-cell repertoire reset. In other words, the B-cell repertoire becomes much more heterogenic, less oligoclonic, which makes sense because the autoimmune response is very selective and then once you get rid of it and now you homogenize things, you know, it looks reset. But does that actually mean it's going to drive efficacy? I don't think we have that answer yet either. Probably yes, probably. But that's the holy grail. Yeah. In all of this, we better start defining it.

So yeah, let me move on to my chosen presentation. This was an oral presentation OP00008, the COMPARE trial — CAR T-cell therapy in rheumatoid arthritis patients. They introduced the data on six patients who had refractory RA that was ACPA positive. The patients entering the trial had failed generally around four or five biologics and targeted synthetics, ranging from three to eight. They had very active disease, generally in the high disease activity category, and as high as a DAS28 of 6.2. All of them were pre-treated with fludarabine and cyclophosphamide and then received their one infusion of something called — I don't remember — midocatab, what was it called? Autoleucel? My mio — we call it mivoa auto — yeah, mvo auto, thank you so much. They need to shorten that up for me especially. They did, you know, as expected, show full B-cell depletion, significant neutropenia along the way, and from the podium spoke as if this was the second coming. But you know, they showed a steady decline in disease activity. There was some CRS; there wasn't much in the way of ICANS. It seemed to be fairly well tolerated. But when they talked about that, you know, the first thing he said was everybody achieved remission — I thought, oh, that's not quite what I read, but okay. And then you see the steady decline in DAS28 scores, and I just went straight to the graphic that showed the ACR 20/50/70 scores at week 12, 24, and 52. At week 12, only 50% of people achieved an ACR20, and the best responses were out at one year where, you know, now these are significant numbers — and again this is an uncontrolled observational study.

Now they do have other patients enrolled, like another dozen patients that are enrolled. They did not present any data on those. They did show that rheumatoid factors and ACPA levels dropped, but they didn't go to zero in everyone. On the other hand, you know, some of my colleagues were upset that ACPA didn't drop very much at all, but you know, those people that dropped — the ones that started out an ACPA of 3,400 dropped to 200, so they had significant drops in autoantibodies, and this was associated with efficacy. But this is, for me, why I went here: CAR T-cell and cellular depletion therapy seems like it's the great hope for autoimmune therapy and it's being applied across the board, and we've seen great results with lupus and really encouraging results with scleroderma. I'm not so sure about myositis, and I'm not at all sure about RA, and this says it might could work in RA, but I don't think it's quite the magical bullet it's looked like in lupus.

So I'd be really interested in the three of you and your comments on this report because you certainly are following this data more than I. Al, what do you think? Yeah. So I'm so happy you presented this abstract because I think it highlights a whole bunch of different things that you nicely covered. And the other one too that we just transitioned off of is that, you know, immune reset appears to be happening in these people, but we're getting very variable responses — some are at best getting MDA, some of them do hit remission, but they're all hitting immune reset. So I think the disease state itself appears to play a critical role — i.e., are you even B-cell responsive? Right? Multiple sclerosis being another example, anti-synthetase syndrome being another example, where we're seeing that there are certain diseases that you can reset their immune status but it doesn't seem to matter. Yeah. What did you think? Yeah. So I think, like what Alan and yourself mentioned as well, I think there's quite a heterogeneity in terms of
the immune pathogenesis. So it may be RA is more varied. So that's why it's harder, you know, to target — you know, how much depletion you can achieve with all this, you know, CAR T or BCMA and all sorts — but I think there might be other immune things that need to be targeted. So maybe a little bit lagging behind compared to, for example, lupus and scleroderma.

Yeah, so Janet, you're a clinical trials researcher. Someone comes to you with an allo CD19 prep or a BCMA, and you can enroll anybody you want, you know, autoimmune disease, and they'll give you four categories. Are you enrolling your RA patients in this?

Well, so far with the data we have, I was underwhelmed. By the way, the guy presented it really well, but it was a little bit discouraging. And by the way, these patients were not all the most difficult to treat. They had in fact failed things. So they were difficult to treat, but they hadn't been recalcitrant to everything.

So here's what I think. If I were going to enroll an RA patient, I'd say let's turn the clock back. Let's look at an earlier RA with not much damage but has not responded to a lot of stuff, because the chance of responding well to your fourth advanced therapy if you haven't been on everything — csDMARDs in the last three — is pretty sparse. So it's a law of diminishing returns. But it brings up a concept of maybe immune dimming, not reset, which I will talk about after.

Okay, let's move on to our last selection. You want to give us yours?

Yeah. Yep. So my quick one is presented at the late-breaking abstracts. So LB00004. Two is a company and three is a crowd. So this is a progression from BiTE — which is the bispecific T-cell engager — to now development of TriTE, so T-R-I-T-E, trispecific antibodies. So usually the antigen would be CD19 and CD3, which is the molecule of the BiTE to bring it together and cause apoptosis from the T-cell activation. But this time they added CD28 as well. The reason for adding CD28 is to provide co-stimulation, and this will then induce more T-cell activation and also reduce T-cell exhaustion. So it's really priming for that.

And this early results data were presented in patients with lupus nephritis. There were 12 of them — severe refractory, so had multiple failures to biological therapies before. Because it's still early, they did have data for eight patients out of 12 at six months, and also have about six patients up to week 36. So at week 20, at six months, all of them had SRI-4 response, whereas at week 36 there were five out of six who responded. I think there are two with overlapping data, so they finished at 12 months and they seem to sustain SRI-4 response there.

So it's early preliminary results. There's not much significant safety signal compared to other BiTEs you might expect, but something that we are looking to in the future. And also I would expect, if we are following the oncology space, in the next year or so we might also get a TriKE — which is where they replace CD28 with an NK cell — in a TriKE. So that will be a way forward.

So these are parenterally administered, I assume. Yes. Yeah.

Okay. And other advantages to this form of therapy over CAR T?

So I think the advantage is easier to administer, and also just trying to minimize the risk of T-cell exhaustion. That's the reason for adding CD28 to make three. But we still don't know — we need longer follow-up data — whether patients would relapse or whether repeat dosing would be effective as well.

Do you see there are companies that are struggling with whether they should be developing CAR T or these bispecific T-cell engagers instead?

Yeah, I mean, you mentioned this too, but there's no lymphodepletion conditioning with engagers, and that's a very attractive reason why people are moving into engagers — on top of the fact it's much cheaper to make. But their efficacy hasn't been as robust as the CAR Ts at the first-generation level. I would consider this to be a second generation, because they are trying to mitigate some of the issues of the first generation inducing T-cell anergy that can happen with the first generation. So this is I think an important proof of principle, but it's certainly not the answer right now.

Okay, let's move on. Al, what's your last one?

Yeah, so the last one I'm doing is abstract LB000003. And this is an interesting abstract because they actually had multiple abstracts on a product called AB-1001. And what this is is
trying to address the question of how can we deliver cell therapy as an outpatient completely, right? Avoid everything inpatient. And so what this approach is is that it is an NK cell therapy. It's not CAR NK cell. There's nothing modified in the NK cell itself. Although the NK cell is not normal, it comes from donors that have a high affinity naturally occurring variant in CD16, which is the Fc receptor. So this will enhance ADCC through NK cells. Now you may have heard this story before because obinutuzumab and other type 2 monoclonal antibodies do it the exact same way but from the opposite ends of the lens. They're modifying the Fc tail to enhance their affinity to normal CD16. Here you have enhanced CD16 that is normal, is present in about 10% of the human population. So they studied it in RA which is obviously a more difficult disease to study and unfortunately you still need LDC or lymphodepleting chemotherapy. Here they use flu/cy but remarkably they had zero cases of CRS and zero cases of ICANS in all the patients they treated. I think they treated at least six, and they did see people improve varying amounts. Some people went into remission, others went into moderate disease activity, but of those that were treated, three out of the four that were on DMARDs were able to stop. Three out of three that were on prednisone at screening were able to stop and they also saw some results in N-of-ones in Sjögren's and in scleroderma. So it's kind of an interesting approach. Again, I don't know if this is the final answer, but I thought I would never see an abstract about pure cell therapy for at least three to five years, but we got it this year.

Are they taking the tack that this is going to be easier to administer? Yeah, I think so. Right. And again, this would be the right approach to do using these particular NK cells because this binding affinity business for NK cells is really really important. Of note, every patient had to get rituximab in order for the NK cell to act. I forgot to mention that. They have tested this with obinutuzumab too in lupus but did not see increased toxicity. So there seems to be a ceiling effect here. But the bottom line is that you add your favorite monoclonal antibody to your favorite cell and you can chase them with these NK cells and deplete them. Interesting. All right. Very novel. That's why it was a late breaker at EULAR 2025.

Janet gets to bat cleanup. Right. So I'm talking about Dr. Hector Chinoy, who gave the WIN — which is What Is New — for inflammatory myositis. It was June 5th, a packed room in that great big room. So I'm going to tell you maybe three points. So the first is — I'm quoting him — he said immune reset may be asking too much and we need to consider immune dimming, especially in our patients with SARDs and multi-drug failure. So I heard a lot about immune dimming. It seemed every time that a T-cell engager didn't do as well, they said, "Oh, we're immune dimming and then maybe you can put them back on the drugs and they'll be better because we haven't fully reset them." So that's one thing.

He also said we really probably have to rethink not only what SARD patients go in, but — because he was talking then about inflammatory myositis — which inflammatory myositis patients would you put in? So the recalcitrant ones are probably going to be bad news in that they might not be responsive. They might even have a higher burden of — I'll call it epitope spreading for lack of a better word — it just might be the totally wrong group. He was asked about what about the MDA5 who already have early ILD, and he said yeah maybe that would be a good candidate, but he didn't sound fully optimistic on any of the immune reset thinking in the early days of inflammatory myositis.

And then the final thing is these are very heterogeneous patients, so I don't think there's going to be a one-size-fits-all even if you're going to lymphodeplete and then give what you think — even if it's an autologous CAR, not an allogeneic, in these patients. So to me it was an excellent review. He also had beautiful slides. He spoke really well but it was a little bit sobering for me. I think it put me back into reality a bit.

Nomenclature clarification — SARDs — did you — oh sorry, SARDs, the systemic autoimmune rheumatic disease patients, not SARS as in COVID SARS, or as in slow-acting rheumatic — you know, they were called the slow-acting rheumatic drugs — but the systemic autoimmune rheumatic disease patients.

So why is it that — I think it's odd to me as an outsider to myositis — that the CARs have not done as well in inflammatory myositis? Could be for the heterogeneity that you point out, Janet, but on the other hand — and this is true, one related whatever — the dermatomyositis patients with ILD respond really
well to Jack's and you know any of our diseases with nothing really works the best we can do is is level them off and keep a stable FVC but you can turn things around in those people so it's clearly a different kind of disease I guess is what I'm saying. But does anybody have insight as to why they're harder to treat? I'm going to give you a guess and I don't know that it's true. This is like off-the-wall and not proven or anything, but if you're doing lupus nephritis, I mean, it is a systemic autoimmune disease and you do have antibodies and stuff, but look at the surface area within the target organ of two kidneys. I realize we want the rest of their lupus good, too. If you think of inflammatory myositis, your muscles have huge blood flow, but you've got all these — because unless we're talking distal muscle like inclusion body, which we're not, we're talking proximal muscle, but I think the surface area and the blood flow and the clearance out of these muscles — every time you move, you're probably changing clearance, changing what penetrates in and out. It's a lot of volume. Maybe you need far higher doses, but I'm making it up. I don't have a clue. Right now it's king of the hill and someone's got to knock it down. Anybody have another idea? No. Yeah, I think it's due to the heterogeneity that you were talking about just now. We're talking about inflammatory myopathies. So you know you have the dermatomyositis subtype and then you have the MDA5 antibodies, for example, as well as Jo-1 and everything. So it's quite hugely varied and I'm sure there are various pathogeneses — maybe just targeting B cells only is not going to be enough.

So Al, before we came on you had mentioned that you thought that the cellular depletion research area was really hot at ACR, but boy it got a whole lot hotter here at EULAR. What are you expecting going forward in the next six months? Um, more diversification, and I think this is going to make it extraordinarily complicated to summarize. There's so much innovation going on both in CAR constructs, CAR engineering, and also T-cell engager engineering, and you mentioned one of those tri-specifics. There's no real bounds on the innovation when you look at oncology and their experience, and so these are rapidly coming over into autoimmunity. It's going to be somewhat overwhelming because the best idea won't be fully tested and won't win — there are other complicated reasons, i.e. venture capitalist investors and angel investors that are muddying the waters. I think that's the reason why these types of platforms are important — we can really apply rigor to make sure that the wrong narrative isn't put forward.

And how far are we away from a randomized controlled trial, placebo-controlled, active-controlled? I'm surprised that some of the phase two trials that are out there remain single-arm open-label. And I think right now it seems like the FDA is making a position that we'll have to figure that out ourselves later. And hard to recruit in an RCT because people who want CAR T are going to drop out if they're randomized. Even if it's a pretty decent control arm, they're going to drop out and say, "Well, I'm going to this other center because they're going to fly me there and do it." So I think patients are going to have difficulty unless — oh, in a year you can cross over — and they might even drop out then. So I'm not saying that it's wrong. I think we need controlled studies but it's very difficult to recruit them.

Yeah, I don't think we get smart until we have controlled trials. We're still in discovery, which is a really important phase, and we've come a long way in two years. So I want to thank our panel for a really great discussion on a really complex subject. I want to encourage the audience to tune in to more of these post-EULAR panel discussions. Take care.