iwCAR-T 2026 | Dual-targeting CARs in LBCL: trial updates and discussing the role of bridging therapy

Gloria Iacoboni:

I’m Gloria Iacoboni. I’m a hematologist at Vall d’Hebron Hospital in Barcelona, Spain, and I’m very happy to be sharing this roundtable discussion for VJHemOnc, joined by many colleagues. You’d like to introduce yourselves? 

Saurabh Dahiya:

Absolutely. My name is Saurabh Dahiya. I’m an associate professor of medicine and the director of the cell therapy program at Stanford. Pleasure to be here. 

Nirav Shah:

Nirav Shah, Medical College of Wisconsin. 

Krish Patel:

Krish Patel from Sarah Cannon Research Institute in Nashville. 

Akil Merchant:

Akil Merchant at Cedars-Sinai Medical Center in Los Angeles. 

Gloria Iacoboni:

So we just had a very engaging discussion talking about dual targeting CAR T-cell constructs in lymphoma and how each of the constructs and all of the trials you presented kind of add a specific tweak aside from the CD19/CD20 dual targeting. So maybe you’d like to touch upon that briefly, like in each of your trials, you know, what does your construct add to the novel, to the standard, you know, CAR T-cell products we have available? 

Saurabh Dahiya:

Yeah, absolutely. So I presented on the data on KITE-753, which is a dual-targeting CD19/CD20 CAR T-cell product. Additionally, besides dual-targeting, what this product does is it’s also a rapidly manufactured product. It’s early harvest that is done to preserve memory T-cell in the product that is infused. What we showed was a fairly high efficacy rate with objective response rate of close to 82% with a CR rate nearly at 80% in these relapsed/refractory large B-cell lymphoma patients. Additionally, the safety of this product was preserved, was quite good, with low CRS and low ICANS incidence and no grade 3 or grade 4 CRS events that were seen at the highest dose level. 

Nirav Shah:

Yeah, so I think this was a really interesting session. And rather than starting about all the things that are different and what we disagree about, I think the four of us are all believers in a philosophy that targeting more than one antigen might be a better way to improve outcomes in lymphoma. I think that’s something we can all agree about. What we all have differently is how we’re doing that. And so, you know, obviously great data, Saurabh, about the product from Kite, rapidly manufactured. What we are doing with the product called zamtocabtagene, it’s a tandem CAR, so sort of built differently than the one we just talked about. But it’s also a different manufacturing platform, a different manufacturing concept of not only a fresh apheresis, but a fresh delivery of CAR T-cells with manufacturing starting during the lymphodepletion. And so, you know, we hear a lot about rapid manufacturing, but here there’s true vein-to-vein time of 12 to 14 days because you actually start lymphodepletion at risk in patients during that manufacturing process. And, you know, I presented updated data from two phase two studies. One was randomized in transplant ineligible patients, you know, that were frailer and sort of got GEMOX as a standard of care versus zamtocabtagene, and there was a clear, you know, progression-free survival and EFS benefit. And we also showed data from the U.S. third-line study that looked at this as a single-arm trial in patients with relapsed/refractory diffuse large B-cell lymphoma. There, the overall response rate was in the 70s. CR rate was in the 50s. But again, you know, we also demonstrated a platform that hasn’t been done before. We were able to deliver fresh CAR T-cells throughout the United States and throughout Europe, which I think is just showing that you don’t have to cryopreserve, that there’s more than one modality to be able to deliver CAR-T. And so, you know, we’re excited by this data, but I was also excited by what all of you guys had with your slightly different platforms. And so let me hand it off to you, Krish. 

Krish Patel:

Yeah, thanks, Nirav. So I presented data from the Phase 1B of JNJ4496, known as prizla-cel. And this is another tandem dual-target CAR T-cell. And what’s, I think, a little bit differentiated about that particular CAR-T is the CD20 binder is distinct from other CD20 binders that we use, both in the other dual-target CARs, but also in bispecific antibodies and rituximab. So, you know, whether that’s going to make a long-term difference remains to be seen. That’ll ultimately be borne out in clinical trials. But what is exciting about that product is in essentially second-line and later large B-cell lymphoma patients, we saw a complete response rate across the board, high-risk patients very much enriched in the trial, of 80-plus percent. And so I think that depth of response ultimately is a good signal that dual targeting is important. The durability is very good as well. And so, you know, whether this will take us over the finish line above CD19 CARs is very much the subject of those randomized studies that I think each of us alluded to. 

Akil Merchant:

Yeah, I presented on ronde-cel, which was formerly LYL314, and also a dual-targeted tandem CAR. So the unique thing about ronde-cel is the selection for CD62L, which selects for naive and memory cells. My takeaway was that the CR rates and toxicity are pretty comparable. I think the thing that we’re waiting to see is the durability. And we have this hypothesis that the selection of naive cells is going to, and I showed some correlative data, translational data, that we’re getting very good expansion and preservation of function several months out after treating patients. And we were hoping that that might add to the durability question, which I think we’re all eagerly awaiting the follow-up data. 

Gloria Iacoboni:

So that’s great. Thank you for presenting the data. So I want my thoughts upon safety first. In terms of CRS and ICANS management, did any of these trials, these phase one, two trials leading into the phase three, did any of them use like prophylactic dexamethasone or any prophylactic strategies or were there just like an, you know, early intervention strategy at CRS grade one? 

Saurabh Dahiya:

So for the KITE-753product and KITE-363 product, the previous iteration of the product, which was longer manufacturing, there were no prophylactic steroids or prophylactic other immunosuppressive drugs that were used. It was largely based on the ASCO Bianca Santomasso 2021 guideline paper. And it was a guideline, and the investigators used their institutional protocols to manage toxicity. But just broadly speaking, I think the toxicity profile was quite good with the KITE-753 product. And the overall usage of tocilizumab and corticosteroids for the few patients that develop grade 1 or grade 2 toxicities were quite low. Nearly half of the patients did not even develop cytokine release syndrome. And close to 70% of the patients did not develop any neurotox or ICANS. So nothing unique about toxicity management for the KITE-753 product. 

Nirav Shah:

Yeah with zamtocabtagene, I mean, I think what’s interesting, I think the next generation CARs, we sort of have handled this a lot better, right? Like we’re not, I don’t think we’re going to see that level of toxicity for two reasons. One, we know how to better manage it. And two, I think we’re just designing these CARs a little bit better. So with zamtocabtagene, the grade three CR rates across two large studies were less than 5%. Same with grade three plus ICANS. So I think that the acute toxicity phase is much better. There was no prophylactic steroids or anakinra or tocilizumab allowed. It was all treatment per institutional guidelines, but we all treat earlier, right? I mean, we treat often the first fever and so they don’t accelerate to higher grade toxicities. So I think overall, it looks really favorable. And I think that’s just sort of the way the field is moving. 

Krish Patel:

Yeah, I would agree. And I think to your point, Nirav, all of them are 4-1BB costim to some degree, and so that safety profile seems very comparable, consistent. And similarly, in the JNJ study, that maybe early intervention approach is what we used. Didn’t include prophylaxis, but, you know, for some patients with early grade one CRS, could include intervention, but the general use of tocilizumab and dexamethasone was low. 

Akil Merchant:

Yeah. So with ronde-cel, we actually have sort of an interesting experiment because they didn’t start with prophylaxis, and about halfway through, they instituted prophylaxis, so we can actually compare the before and after. And so the prophylaxis is like the ZUMA – It’s 10 of dex and 0-1-2. CRS was always good. There was no grade 3 CRS in either. ICANS in the first half, this looks pretty much like liso-cel, so it’s pretty reasonable, and it’s about half, roughly, with the prophylactic dex. 

Nirav Shah:

That’s interesting, because the prophylactic dex actually didn’t change the rate of ICANS, is my understanding, right? 

Akil Merchant:

Well yeah, that’s what we’re seeing.

Nirav Shah:

But in the KITE studies, you would know probably better than anybody, but you’re seeing that it does help in your situation?

Akil Merchant:

At least that’s what the data are. I mean, it’s still smallish numbers, right? But that’s what it looks like, yeah. 

Gloria Iacoboni:

So is that what you’re doing for phase three? Are you moving forward? 

Akil Merchant:

Yes. So now the way ronde-cel is going to be developed is with the three days of steroids. So in the pivotal study, that’s built in. And in the randomized study, it’s yes for ronde-cel, and then it doesn’t speak to, because you give standard of care. So if you use prophylactic dex, if it’s your axi-cel or whatever, then you can do it. 

Gloria Iacoboni:

So you can do whatever with standard of care? 

Akil Merchant:

You can do whatever for the standard of care. 

Gloria Iacoboni:

But if a patient is randomized to ronde-cel, they have to get the prophylactic dex. 

Akil Merchant:

They do get the three days of dex. Exactly. 

Gloria Iacoboni:

And I think that’s interesting because moving forward from just CRS grade three or higher, I think just CRS any grade is also important when we’re talking about potentially treating those patients outpatient, which is where the field is probably moving towards. And to that regard, Krish, I saw that 80% of patients with the prizlo-cel product, you know, developed CRS. So that may be a bit high. We want to consider these patients for the outpatient management. That’s some things we’ve talked about. 

Krish Patel:

Yeah. So the study did allow… the phase 1B study is ongoing, it does allow outpatient delivery. And so largely what we saw was about 54% grade 1 CRS, so a single fever that resolved usually, I think, if I recall, the median time to resolution was about 24 hours. So I think depending on what your usual care infrastructure is, if you’re today delivering CD19 CAR T-cells in the outpatient setting, I think the safety profile of prizlo-cel is conducive to that and allows that. But I think your point, Gloria, is important is that the field as a whole is moving towards that. So that’s something we’re looking for, that consistent ability to deliver outpatient CAR T-cell amongst all of these products. And so I think grade one is manageable. And we hope that that grade two CRS and above risk for all of these products will be low. 

Gloria Iacoboni:

And then I think it’s unique to zamto-cel starting LDC at risk. That’s not in any of the other trials, if I understand correctly. It’s, you know, unique to that product. So I think, is that something that we could potentially consider in the future with other trials or other products? It’s something that’s being discussed, you know, given the rate of manufacturing success currently. I mean, why not try that with other products and shorten the vein-to-vein time? I’m not, you know, maybe you can speak to that if that’s something you have discussed within your trials. 

Nirav Shah:

Well, I’m curious because, you know, I know you talked about shorter manufacturing. Does that impact your vein-to-vein time, Saurabh? 

Saurabh Dahiya:

Yeah, the vein-to-vein time was shorter. It was about 13 days as compared to what was done for this similar vector but longer manufacturing. So it was shorter as compared to the 363 product. But it’s somewhat not fully elucidated just yet and a bit controversial. In my opinion, I think what matters is optimizing the patient fully before they’re infused. So collecting the patient, bridging if needed for a bulky patient to get them to a debulked state, and then coming in with the CAR is a good strategy as opposed to whenever the product is ready and just go in, which is a reasonable strategy for a lot of chemo refractory patients. But optimizing the patient, I think, is critical. Second point, I think the intent with the 753 product is primarily to preserve the T-cell fitness. Less time on culture, less time ex vivo expansion, preserving the memory phenotype that there is, and retaining the effector function that there is. I think that’s the main intent with the rapid manufacturing platform. I know the word rapid has a connotation, just delivered rapidly. But I think the intent there is to have the better fit product available. 

Akil Merchant:

I think the challenge in designing these trials is you want them to reflect what will be the ultimate use practice. Like in my practice, I often take patients to CAR with no measurable disease because if I can get them into an optimal state, I know they’re going to relapse, then I know that their CAR, the CAR works best with lower disease burden. That’s not good for a trial, I need something to measure. And so, but, you know, so that’s why this whole bridging, no bridging thing, I think it’s a challenge because at least in our practice, if we can deliver bridging, we try to deliver effective bridging. I wouldn’t hold bridging back because it was in my practice, even if the trial was done without bridging. So I don’t know how people feel about that. We’re very pro-bridging in our institution. So I think the biology suggests you want to optimize the target effect duration. You want to get that disease as low as possible. 

Gloria Iacoboni:

And just out of curiosity, like where are your preferred bridging regimens in the current landscape where we have so many options? 

Akil Merchant:

Yeah, I have not yet ventured into the T-cell engager, although my myeloma colleagues love that. It makes a lot of sense. But since we usually have overlapping antigens, at least with the CD19 and 20, but I usually use a lot of pola and radiation. 

Gloria Iacoboni:

If they didn’t get pola in the front line? If they didn’t get pola-R-CHOP?

Akil Merchant:

Yeah, I mean, if they didn’t get pola in the front line or if there’s a gap, I think. And we stay in for 79B, which seems to be pretty robust. It doesn’t get lost. 

Nirav Shah:

I struggle with the whole bridging question. You know, I think there’s, I get that school of thought, but the people that respond to bridging respond to CAR. Like, yeah, you know, and my concern is there’s a group of patients when you try to bridge, they actually fall off and they never get the CAR, right? Because they blow through bridging and then they might become ineligible or too sick to be able to get the CAR-T. And so you never are able to capture that patient population because they sort of fall off. But, you know, the people that respond to bridging, like I agree with you, like I feel great. Right? And, you know, it really comes down to… so I guess like when, you know, we do a lot of IIT work with, you know, zamto-cel and, you know, we actually manufacture and deliver the CAR in eight days. And so my message, you know, or philosophy is just, all right, they’ve demonstrated that they need a CAR. Get them to CAR as quickly as possible. I haven’t delayed… I guess my question to you guys is, are you delaying your CAR in order to give bridging? Because that’s what you were sort of implying, right? 

Akil Merchant:

I mean, that’s why I’ve never been hot on that rapid turnaround, because I would want to optimize the patient. Like, you know, I come from an allotransplant experience, right, where we know you want to get the patient in their optimal disease state, not to tackle the toughest disease. And so the idea that they’re going to CAR, so I’ve got to give the CAR now, that’s not how we do our other practice. That’s your final consolidative, curative regimen when the patient is in their best days. 

Saurabh Dahiya:

I think there’s enough heterogeneity in our large B-cell lymphoma patients where we kind of have to match the solution with the right problem. If there are patients who do not have a kinetically fast-moving disease, they have a few sites that are chemorefractory and they’re not symptomatic, if there’s no need for bridging, I kind of feel that I think just when your CAR is available, infuse it as early as possible. However, if there are patients where you need to stabilize them and there’s need, I think you do need bridging in those particular situations. I think, again, just to double down on the fact that the comment that I made earlier is the intent with rapid manufacturing is not rapid turnaround. It’s to preserve the T-cell fitness that there is. And that really does translate into a very different profile of these products. The YTB product from Novartis, a very distinct PK that there is. The 753 product, the expansion is just fantastically high, and it’s delayed. Its peak expansion is day 15. So your cytopenias from lymphoid depletion are resolved. The peak expansion is happening late. It’s kind of pushed out. And that really does make a difference in toxicity in the patient experience ultimately. So again, the intent is, with rapid manufacturing, I think, again, for most products is to preserve fitness. 

Krish Patel:

I’ll just make a short comment. I think that the challenge is exactly like you said, Saurabh. You know, when patients are symptomatic and have progression, you need to bridge. I think bridging in the… So bulk disease is not all the same either. You can have patients with bulk disease who are not kinetically progressive. I think this is where the field has to try to identify and use tools to say, okay, beyond debulking disease, are we also making impactful changes like we heard about inflammatory milieu, right? So if we debulk the disease rather than just lower disease state, are we doing things that are actually impactful to the tumor microenvironment that help the CAR function? And we just don’t have those tools today. 

Akil Merchant:

And safety, too, right? 

Krish Patel:

And safety, right. So those are all things that I think we have to continue to work through. 

Saurabh Dahiya:

It is amazing that 10 years into it, we’re still debating about bridging therapy. And five years ago, this lymphoma consortium work showed that bridging is actually associated with poorer outcomes. Ultimately, it was basically confounding by indication. Patients who need it [overlapping conversation]. 

Nirav Shah:

That’s exactly what I’m saying. 

Akil Merchant:

So that’s what I like… Bridging should be mandated or not allowed? 

Nirav Shah:

Excluded, right. 

Akil Merchant:

By choice. Every time you give a choice, you add that. 

Nirav Shah:

I think that makes a lot of sense because, again, if you respond to bridging, guess what? You’re going to respond better to CAR. And when you don’t, what is the chances you’re going to… You know, if you’re blowing through whatever you get and then you give the CAR, I mean, it’s probably less likely to work. And so that’s the, I mean, that’s why we don’t, we can’t agree on this topic. 

Akil Merchant:

But I have never, I mean, you made a hypothetical that a patient got bridging and then progressed and couldn’t get CAR. I’ve never had that experience. 

Krish Patel:

I have had that experience. 

Akil Merchant:

That you delayed though, but you didn’t have a choice. They progressed, they were going to progress anyway. 

Krish Patel:

No, I think it’s not even just progression. I’ve had patients who got bridging therapy and then had febrile neutropenia, got septic, and got sick. So it’s a rare event. But I think there is that element of bridging where you need to stay on the path. And so to your point, Gloria, I think that’s a struggle because we prefer less intensive bridging. But what’s the sweet spot? What’s enough to help? 

Nirav Shah:

I think it’s just two different philosophies. Like, do you just get, do you want that CAR that could be made, right? Or an in vivo one day, right? Where you’re not going to bridge them, you’re just going to inject them, right? Theoretically, is that the best methodology? Or do you actually, if you want to give effective bridging, I think you have to use your best class of drugs, which would be bispecifics. But they still require require a ramp up. And so are you delaying then your CAR to ramp up a bispecific? And I don’t know the answer to that. 

Gloria Iacoboni:

I mean, and there are some interesting trials looking at a rapid ramp up of bispecifics. So there was the French data where they reached like the full dose of glofitamab in like seven days. So I think- Yeah, we rapidly ramp up. Definitely that’s in the future, trying to ramp up those bispecifics in a safe but more rapid way. And kind of in this line of trying to understand further, I think it’s interesting how the different randomized phase three trials moving forward are going to require CD19 and or CD20 testing and biopsies at time of potential relapse. So you mentioned, Akil, in your data, that the phase three randomized trial did not require CD19 or CD20 testing. 

Akil Merchant:

Yeah. Well, I think, so we should get into that. I think that’s an important thing because you have experts here who’ve done the testing. But when you look at IHC, which is essentially very poorly quantitative with a limit of detection probably that doesn’t match the molecules per cell that CAR can be effective at versus flow or quantified flow. I think that really makes that challenge. 

Gloria Iacoboni:

What is the optimum methodology to do that? So in terms of the other randomized trials, are they looking at CD19 and/or CD20? Is that, you know, an inclusion criteria? 

Saurabh Dahiya:

It’s not a mandated requirement by either IHC or flow to, with the KITE trial, to have expression as entry criteria into the study. However, I think it is an important correlative question that we need to, as a field, answer both at the entry into the studies as well as time of progression. Most of the trials are not mandating biopsies at the progression time, which I think is a missed opportunity, especially in our randomized controlled trial when we are really trying to tease out is one better than the other or not. And without doing the biopsy, I think it would be, yes, I think clinical hard points I think will declare themselves, but I think we are at probably like 65% CR rate currently with our current chemo plus current CARs that are there. It’s going to be better than single antigen targeting. But I think we really do need to understand about mechanism of relapse with these therapies. 

Gloria Iacoboni:

We could learn so much, I think, from looking at those biopsies in both arms, like comparing antigen loss. And those who get in the CD19, axi-cel, and those who get in the CD19/C20, KITE-753 know, comparing those two would be interesting. And just perhaps as a final thought, we touched upon this in the discussion. I think right now we have phase one, phase two data. I think it’s important to, you know, be cautious when we’re comparing, you know, safety and efficacy because the patient populations are so heterogeneous across, you know, the different studies. So it’s important, you know, when we’re looking at those, you know, summary tables to perhaps try to put in some of those baseline features, you know, how many patients were primary refractory or the time from last treatment to, you know, entering the clinical trial. Other features that can capture, you know, the aggressiveness of the underlying disease to better understand what patient population were included into each of these trials. And the phase three trials moving forward, as you mentioned, also looking at duration of response, not only CR rates, but, you know, what does the relapse rate, you know, look like? And the late onset toxicity is, I think, all of this is going to be very important. 

Nirav Shah:

What’d be really interesting in our presentation is if we started including like inclusion criteria, right? And washouts, right? They’re so different between studies. And those things, I think, often determine what your patient population looks like just as much as cutoffs of age and creatinine clearance and things like that. Because it’s just each one of these strategies is so different, right? And so it would help us better understand that patient population. 

Gloria Iacoboni:

Yeah, and sometimes I think we don’t go into the weeds enough to look at those baseline characteristics. 

Nirav Shah:

We like the tables. [laughs]

Gloria Iacoboni:

We like the tables with the CR rates. And it’s like, this seems more efficacious, but we’re not really understanding the patient populations. 

Akil Merchant:

We may have an opportunity, because the randomized trials will have, at least hopefully, the comparator arms in common, it might be an opportunity to sort of, the natural history experiment to just study the comparator arms from you said try to answer how some of these these eligibility criteria shape the population.

Nirav Shah:

I think this is a… either I mean you know zamto-cel sort of, you know, we did our sort of pivotal trials before this all came out. But for the three of these studies, I think the amount of data you’re going to generate. And it’s actually the right trial. Right? It’s a standard of care CAR versus what could be better. I think it’s going to be incredibly exciting. 

Gloria Iacoboni:

Great. Thank you. So with this, we’ll finalize our roundtable and hope you enjoy the rest of the meeting. Thank you very much.

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