iwCAR-T 2026 | T-cell engagers & CARs for myeloma: trispecifics, anito-cel updates, & long-term CARTITUDE-4 data

Yi Lin:

Hello, everyone. I am Yi Lin from Mayo Clinic, and I’m here with my esteemed colleagues, Rakesh Popat and Ciara Freeman. And we are here at iwCAR-T 2026 and just had a wonderful session talking about all the new updates with CAR-T and T-cell engager therapies in multiple myeloma. So I’m going to start with Rakesh. You have shared some updates with data from CARTITUDE-4 with cilta-cel. Please tell us what’s the new updates from that particular CAR-T product. 

Rakesh Popat:

Yeah, absolutely. So I’ve been showing some data from the 33-month follow-up with the CARTITUDE-4 study. Just to remind everyone what the CARTITUDE-4 study was. This was a phase 3 randomized clinical trial for patients who had one to three prior lines of therapy that were lenalidomide-refractory. And they were randomized to get the standard of care, which was essentially daratumumab-pom-dex, or PVD, but most patients actually got dara-pom-dex, or a single infusion of cilta-cel preceded by lymphodepletion chemotherapy. Now, we’d already demonstrated that there was an improvement in progression-free survival, and that was shown again with the extended follow-up. And really, what we’re seeing now is that it’s also an extension in overall survival for the cilta-cel product compared to the standard of care. And then if you drill down into the data a little bit you start to get some interesting takes from the subgroup analysis. So we looked at high-risk versus standard risk and in both cases cilta-cel outperformed the standard of care. And of course, what you’re really seeing, and this was presented at ASH, is that the standard risk patients are doing extremely well. Of course, the follow-up still needs to be a little bit longer, but we’re very intrigued by the long-term data from CARTITUDE-1, where you’ve got 33% of patients who are alive and disease-free. And when you look at the standard risk curve with CARTITUDE-4, we start to think about maybe a plateau. So I thought that was super interesting. You then look at the MRD negativity rates, and that really corresponds. And I think what the take-home message from MRD is that the similar amounts of MRD negativity is seen at 10 to the minus 6 compared to 10 to the minus 5 with cilta-cel. And you don’t see that with standard of care. You see this really marked reduction down at 10 to the minus 6 compared to 10 to the minus 5 with cilta-cel and you don’t see that with standard of care. You see this really marked reduction down at 10 to the minus 6 and that’s coming through. So I think this all kind of tells you about this long-term efficacy signal that we’re seeing with these patients. Then if we flip to toxicity profile, of course we’ve reported the CRS and the ICANS rates, and infection is an issue. And, you know, there’s about a 10% NRM with infections, and all patients need to be having IVIG or hemoglobin replacement therapy. In terms of the other neurological events, I think what was reassuring was that we didn’t seem to see so much in terms of the non-ICANS motor neurotoxicity with this, but we are seeing the cranial nerve palsies. But with the extended follow-up, we’re not seeing much more coming through from that regard. We do need to take into consideration SPMs, of course, and that’s being seen, and that will become more relevant as time goes on. And then we have this emerging story, and you’ve been presenting some data, Yi, about the ALC, and if that hits over 3,000, that may be a predictor for these non-ICANS neurotoxicities, and there’s been some other real-world data looking at the efficacy of dexamethasone and how impactful that may be. But that, in a nutshell, is the summary of the CARTITUDE-4. 

Yi Lin:

Yeah, no, that’s wonderful. You know, it’s really interesting to see the therapy that has such a long, you know, progression-free survival that we actually need longer follow-up to see if it’ll truly plateau. And, you know, the MRD, of course, being a very helpful surrogate, also in the real world, to kind of understand how to potentially monitor these patients, right? But the overall survival is always the ultimate hard endpoint, and it’s really great to see that benefit. 

Rakesh Popat:

Yeah, absolutely. And I think with MRD, it’s getting more challenging with such high MRD response rates. And so actually, what we did also show in CARTITUDE-4 was the sustained MRD negativity rates, and that probably is more indicative of overall survival than the single or best measurements of MRD. 

Yi Lin:

Yes. And it’s also very reassuring, as you pointed out, that we can see the, you know, I almost call it intermediate toxicity with the delayed neurotoxicity. So not necessarily in the early part of the first months, but somewhere in that three-month window post-infusion where you may see these delayed neurotoxicity. but even in CARTITUDE-4 compared to CARTITUDE-1 where we saw less of that, and it’s reassuring to not see that beyond that window. So, you know, in earlier line, you know, having more bridging therapy options may help with that. We didn’t see that the risk completely goes away, but it’s significantly reduced. So perhaps speaking to are the T-cells more fit and, you know, other flip of the coin. So still more to learn from that perspective. But yeah, very exciting to see that data. And Ciara, you presented on anito-cel, which is another CAR-T where we’re now seeing data and we’re very excited to hopefully see this being reviewed by FDA as potentially another party coming to practice. 

Ciara Freeman:

Yes, I got the opportunity to present on the new kid on the block, as it were, and a bit more of an overview than any particular trial in and of itself. So I talked a little bit about the binding domains, which are quite different for the three products, the two that are FDA approved and anito-cel itself, which is quite different and kind of the only one in its class. It’s a D domain, which is basically a completely synthetic binder, so unlike other binders, which are derived from antibodies made by animals like mice or sometimes by llamas or sharks or camels, this is completely synthetic. So the reason that it is sort of, it was designed for simplicity and efficacy and stability. So it’s got this like three little clicking hydrophobic domains with a hydrophobic core. And basically the idea is that every single time it gets incorporated onto the cell surface, it always folds in a particular way, and that just makes it easier to transduce the cells and have those receptors on the surface. So the relative advantages that are perceived is that you don’t have to, you know, work so hard to get the CARs to express a certain number of the constructs on the surface, that they’re quite stable, that they’re maintained. And then there’s other potential benefits that are perceived in that maybe these particular binders don’t have the same degree of what’s called tonic signaling. So that’s that signaling that is occurring in the absence of the cancer cells being bound, and that that might have its own benefits, both in terms of the cells not getting exhausted too early and they also have this perceived benefit of having this on-off sort of when they bind to the target, they release it, they don’t bind quite so avidly. So there’s several sort of in-the-lab features that might convey these beneficial outcomes to using this particular domain. And so this was studied in a phase one, which was presented two years ago by Dr Bishop, who’s also at the meeting. And I’ll just share that data. That’s the long-term follow-up that we’re looking to see. So that’s median follow-up of the phase one study, heavily enriched for like high-risk patients, about 68% of patients had bad features, lots of extramedullary disease, all the things that we worry about. Overall response rate was 100%. And so that was sort of encouraging with you know lots of deep responses, complete responses seen but what’s nice is we’ve got that long follow-up so we have now out over 38 months of follow-up and the median was reached around 30 months. So now we’re seeing that signal of okay maybe this is a product that could you know deliver on that durability of follow-up that we’ve seen with cilta-cel in that same vein. So that led to the development of phase two trial. I presented the preliminary data of that at ASH two years ago, and I’ve got the update that I presented today, basically showing the full cohort of patients who were all enrolled, apheresed, and treated. The efficacy data that we’ve presented today was from the actual independent review committee. So not just investigators and what we think, but from the independent community. And again, very impressive results that we’re seeing deepening over time, 96% overall response rate, you know, CR rates now over 70%. So the real question is going to be, are we going to see that durability? Are we going to see those responses maintained? Again, high MRD levels. We’ve got spoils of myeloma now with these highly effective therapies. Now we see, you know, if we don’t see MRD negativity at a high rate, now we’re basically saying, oh, I’m not going to be interested in that. And so it’s an exciting therapy. And again, now the question is, is, you know, will this lead to a new approval, new CAR-T in our hands? And again, the questions that we’re going to, you know, ask ourselves as treating physicians is where does this sit? So, you know, the phase two trial was in patients who had had at least three or more prior lines of therapy. That is likely the approval and the label that they will get. So where does that sit in a space where we now have cilta-cel second line? We’ve got teclistamab and daratumumab second line. These are all controversies in the field. Where are we going to use anito-cel? The phase three study is imminently going to close to accrual. And so that probably will be a very similar CARTITUDE-4 story, randomizing patients anito-cel in the one to three prior lines against the standard of care. So that could, if it reads out in a short-ish time frame, sort of bring anito-cel forward. But I think it’s going to be an interesting time. The nice thing about anito-cel is that, thus far, we haven’t seen any of those concerning delayed neurotoxicities, Guillain-Barre, cranial nerve palsies, colitis, all of those things have not been reported in all the patients treated to date. So it gives us that sense that maybe this is a CAR that we could use with the same kind of high bar of efficacy that cilta-cel has set, but maybe without those concerning off-target toxicities. 

Yi Lin:

Yeah, that’s very exciting to hear. You know, I think there’s so many BCMA targeting therapies, and it’s always nice to have more tools, but they’re not all exactly the same. The engineering is different, and maybe we’re seeing some of that translate into a little bit different clinic profile. And of course, outside of trials, real-world patients, you know, we’ll need to see how that behaves there, but certainly very, very encouraging clinical trial data. They’re exciting. I think we’re all waiting to see if we can have experience for that potentially in the real world. And so speaking of more innovations that are coming, we are now seeing clinical trial data coming out with trispecific T-cell engagers. And Rakesh, can you share with us the data that you presented at the meeting? 

Rakesh Popat:

Yeah, absolutely. So we’ve seen some nice data with two trispecifics. There’s actually quite a few under development at the moment, but we’ve got probably better data with two. Ramantamig is a trispecific that targets both BCMA, CD3 and GPRC5D. And we have the early results of the first in human study. Well, we spent quite a long time trying to optimize the antibody, looking at the step-up dosing, what the optimal target dose will be, and trying to make it a more community-friendly bispecific antibody. We know how challenging giving bispecifics can be outside of tertiary centers, so we’re very keen that we could deliver this in community centers. So you have a single step-up dose and then a full dose about four days later, which was optimized at 100 milligrams sub-Q. And then the drug is given every four weeks, and roughly after about six to nine months, we had the opportunity to step down to Q8 dosing as well. So quite a patient-friendly, clinician-friendly sort of drug. We presented some of the early data at the RP2D, which was predominantly in the BCMA-GPRC5D naive population, and this is really just to try and get a feel for what this efficacy might be. And we were very much pleasantly surprised that all 27 patients responded. So we’ve got a 100% response rate there. Vast majority of patients are achieving VGPR and there are high levels of CR rates. And we also see that of all the evaluable patients, they were all MRD net neg at both 10 to the minus 5 and 10 to the minus 6. So very nice data in terms of efficacy signal. We are always considered about toxicity with a dual-targeting agent and particularly with the GPRC5D targeting because we know the effects that talquetamab have. What was interesting is, and this may relate to the specificity of the bispecific and potentially the Q4 dosing, is that we do see less GPRC5D adverse events. So particularly the oral toxicity seems to be improved. And when you look at the weight loss, you see about an average of 2.5% to 5% weight loss, whereas with talquetamab, you’re seeing 7.5% to 10%. So you’re seeing a bit of a reduction in terms of the weight loss. But otherwise, you know, I thought it was quite a well managed drug to give. It’s quite an easy one to give. Then we have the ISB trispecific. And this is interesting because it targets BCMA as well, but also CD38 as well as CD3. So it’s going a different sort of direction. And it’s been adapted in a different way so that the CD3 molecule is close to the BCMA rather than CD38 to try and remove some of the toxicities associated with CD38-targeting. Again, really nice data. We’re seeing high response rates, you know, in the 70, 80%. But what’s intriguing is that you get responses in CD38-refractory patients because that’s always a concern with the use of daratumumab and isatuximab frontline is how this would work. So we are seeing responses across the board, nice MRD negativity rates and the toxicity profile looks comparable to the other BCMA bispecifics that we see. So I think that’s quite interesting. We need to understand what the durability of the response is a little bit better. So we’re waiting on data. And particularly, I’m curious about the CD38 refractory population. Will you get that durable responses in that? So we’ll wait and see for that. I do think there’s a lot of other questions with trispecifics. And the question I always get asked is, is a trispecific better than sequential bispecifics? 

Yi Lin:

I was just going to ask you that question, yes. 

Rakesh Popat:

Or giving them both together. This is what we have to work out. 

Yi Lin:

Yeah. How are you thinking about that right now? I mean, I know the trispecific data is still evolving, but based on your experience so far on the trial, is there something you might consider from a patient selection standpoint? Or let’s see how well does trispecific continue to perform in a trial?

Rakesh Popat:

I think the bigger picture is that we’ve been trying to wrestle with the whole sequencing issue. And what I was concerned about is when you look at some of the data coming out of MonumenTAL-1 for talquetamab immediately after a BCMA bispecific, the median progression free survival was 4.9 months. Now there is some real world data now giving slightly longer durability of response, particularly if you can add some time off in between the bispecifics, but that does give you a concern about giving two back-to-back bispecifics. And so in that regard, dual targeting may have advantages in terms of the durability of response. And certainly what we’re seeing, not only is the durability of response is higher, but the overall response rate is higher. So more patients are able to respond, and then you get that more durable response. Whereas if you go sequentially, then you’ve got a 60% to 70% response rate with tec, and then, you know, you have a poorer durability later. So clearly the dual targeting does have some advantages there. 

Yi Lin:

That’s very interesting. And Ciara, I’ll let you take the last question, which is today in clinic, as you’re seeing patients in the relapsed myeloma setting, how you’re thinking about CAR-T versus T-cell engagers, the ones that are available at our disposal in standard of care practice. 

Ciara Freeman:

I love that you hit me with a challenging question. So how we and lots of other academic centers like the Mayo Clinic are considering this is, as of right now, in the second line setting, your options are, as you know, the bispecific, cilta-cel, or, you know, it’s hard to make justification given all the randomized evidence that a standard triplet, you know, is something that you would reach for. So really, it’s looking at the patient in front of you and establishing their risk profile, you know, where they live, how they are. There are certain patients for whom CAR T-cell therapy is either, you know, unfeasible or you’re concerned about, you know, side effect risk, or they have the kind of disease where you just simply can’t wait. And so in those situations, you know, the option of having a second line bispecific could be quite attractive. However, because we don’t know that patients will respond, we know that we have some nice evidence to show that if you get CAR T-cell therapy first, there is still an option for you to respond to bispecific later. And we do know that if you use bispecific first, your chances of making a very functional CAR that works well and gets the same depth of response is much less. Therefore, in general, the first option that we always have is, is this patient eligible for a clinical trial? Number two, are they fit and able to have a CAR T-cell therapy? Is that aligned with their disease, their personal and their fitness characteristics? If so, that’s our preferential choice. And then in those scenarios where CAR-T is either unfeasible or for whatever other patient or disease characteristics, we just can’t do it, then we’ll reach for the bispecific. That’s our current paradigm where we’re working at Moffitt. And I think similar to the M-Smart guidelines is aligned with that. 

Yi Lin:

Absolutely. And I think in the real world, you know, many centers exploring knowing the infection signals that for the bispecific can be higher in the even later setting because of the continuous dosing, you know, what’s the right maintenance dosing strategy? Can it be MRD directive? Can it be more of a fixed-duration? Can it be spaced out more? And that may have impact with sequencing as well. Yeah, as well. So you’ve heard the latest from the experts in the room. And stay tuned for iwCAR-T 2027, where we may have more data to guide us on how to really use these wonderful tools we have at our disposal for our patients.

This transcript is AI-generated. While we strive for accuracy, please verify this copy with the video.