The Midwest vs City of Hope: A Face-Off on CAR T in DLBCL

Where does chimeric antigen receptor (CAR) T-cell therapy belong in diffuse large B-cell lymphoma (DLBCL)? Team City of Hope and Team Midwest battled it out, using the latest data and research to defend their stance on CAR T-cell therapy, in a CancerNetwork-hosted Face-Off. At the 2026 Tandem Meetings of the American Society for Transplantation and Cellular Therapy and the Center for International Blood and Marrow Transplant Research, experts from the US formed into 2 teams, one consisting of City of Hope professionals and another of experts at centers across the Midwestern US states. They discussed the latest data, pressing questions, and clinical scenarios.

Round 1: Data Presentation

Team City of Hope on Real-World Second-Line CAR T-Cell Therapy vs SOC

Presented by Geoffrey Shouse, DO, PhD

Real-world results from a study using patients with large B-cell lymphoma from the DESCAR T and REALYSA registries who were mirrored to patients in the phase 3 ZUMA-7 trial (NCT03391466) were Team City of Hope’s data of choice.¹ The study identified 659 patients with early relapsed/refractory LBCL, which was stratified into 356 patients from DESCAR T who received second-line axicabtagene ciloleucel (axi-cel; Yescarta) and 132 patients from REALYSA who received second-line standard of care (SOC).

The mean age of patients receiving axi-cel was 58.6 years vs 65.9 years for those who received SOC; 91.3% vs 75.8% had LBCL, 82.0% vs 80.3% had stage III or IV disease, and 12.1% and 30.3% had an ECOG performance status of 2 or higher.

The study showed that there was a benefit in event-free survival (EFS), the primary end point, at 6, 12, and 18 months with axicabtagene ciloleucel compared with SOC. Respectively, the values were 55% vs 25%, 46% vs 16%, and 42% vs 14%. Overall survival (OS), however, did not show significance in favor of axi-cel, only a slight nonspecific trend toward improvement; at 6, 12, and 18 months, the respective values were 77% vs 74%, 64% vs 57%, and 57% vs 50%.

Following second-line treatment, 194 patients received further treatment, and data were available for 121. Notably, only 10.8% of patients who received salvage treatment underwent autologous stem cell transplantation.

Team Midwest’s Rebuttal

Graff / Do you think this is indicative of what we do see in the real world? We saw the clear EFS benefit of doing CAR T over SOC. Is this something that reiterates what we’re doing now in the real world?

Shah / Often, real-world data are reassurance that what we see in clinical trials is what we see when we put that into real practice. We have great clinical trial data for second-line CAR T vs SOC. We have 2 very large clinical trials that showed the benefit of CAR T, with an OS benefit reported for one of these trials. This data set is just more evidence that we need to get people to CAR T when they have relapse DLBCL, and we need to get them there as quickly as possible.

Graff / We see that with the third line, when those patients who did get SOC relapsed, a number of them did wind up receiving axi-cel, which was what it was compared against in the initial analysis. Do you have anything else to add from what you just presented?

Herrera / No. The lack of a survival benefit here probably reflects that in ZUMA-7, patients weren’t getting axi-cel or CAR T cells in the third line, whereas here, certainly in the real world, they were allowed to. I don’t think that necessarily takes away from what we saw in the clinical trials. In general, it is strong evidence that getting patients to CAR T cells is providing a huge advantage in terms of disease control and remission status.

Team Midwest on Second-Line Axi-Cel vs Liso-Cel in LBCL

Presented by Olalekan O. Oluwole, MD, MPH, MBBS

Team Midwest shared a retrospective study that evaluated axi-cel vs lisocabtagene maraleucel (liso-cel; Breyanzi) in second-line LBCL.² Of 555 patients who received leukapheresis, 447 received axi-cel and 108 received liso-cel; of them, 404 and87 patients were evaluable.

The median age at leukapheresis was 63.0 years in the axi-cel group vs 71.0 years in the liso-cel group; 76.1% vs 66.7% had primary refractory disease, 84.5% vs 75.5% had Ann Arbor stage III or IV disease, and 11.4% vs 22.4% had an ECOG performance status of at least 2. The median time from last treatment to infusion was 87.5 days (IQR, 62-188) with axi-cel vs 113 days (IQR, 78-260) with liso-cel. In the axi-cel group, 44.8% experienced complete or partial responses after bridging therapy compared with 50.6% in the liso-cel group.

Regarding efficacy from leukapheresis, the median progression- free survival (PFS) was 8.1 months (95% CI, 2.0-14.2) with axi-cel compared with 9.8 months (95% CI, 7.3-12.2) with liso-cel (P = .94). Oluwole attributed this difference to the selection bias that is implicit with both products. Regarding OS, the 6- and 12-month rates with axi-cel were 80.37% (95% CI, 76.22%-83.87%) and 65.01% (95% CI, 59.91%-69.63%), respectively; for liso-cel, the respective rates were 80.55% (95% CI, 69.52%-87.92%) and 71.57% (95% CI, 59.36%-80.69%).

The incidence of main post–CAR T toxicities revealed that cytokine release syndrome occurred in 89% of the axi-cel group and 49% of the liso-cel group (P <.01), with grade 3 or higher events occurring in 5% and 1%, respectively (P <.148). Immune effector cell–associated neurotoxicity syndrome occurred in 53% vs 9% (P <.001), with grade 3 or higher instances being noted in 19% and 1% (P <.001). It was also noted that intensive care unit admissions were reported by 21.6% of the axi-cel group and 2.50% of the liso-cel group (P <.001).

Team City of Hope’s Rebuttal

Graff / I think it’s really interesting that if you look at this abstract at face value and you don’t think about where the patients were infused, where the manufacturing sites are, or the processing time, you would look at that and say, “Oh my God, these products are not the same.” There’s clearly a benefit with one over the other, but you can’t look at it in a vacuum. You have to take all of those things into consideration. Does this abstract change your decision-making?

Budde / I don’t think so. The real-world data really are limited by patient selection. Intrinsically, it’s not a randomized trial. I think it supports that there’s a role for both products.

Round 2: Hot Topics

When disease biology directs speed, should high-risk patients with DLBCL be referred for CAR T earlier?

Deol / What most of the centers do…people who are treating these patients in the community have started reaching out to the centers early on for patients who have high-risk disease, high-grade B-cell lymphoma. We’re lucky in Michigan to have strong collaboration with our community partners. We do a lot of educational sessions, and we’ve had a lot of communication in terms of identifying, early on, patients who might not be doing well or have issues in terms of their disease response. [It’s] to try to get those patients to see us so we can all have a discussion on the appropriate therapy, whether you want to include [polatuzumab vedotin (Polivy)] in the up-front treatment or not, whether you want to think about what needs to be done if the patients don’t respond well. The patients, at that time, if they have that information about sequencing therapy and know what to expect, it can give them some time to prepare in terms of what to do in case their disease behaves poorly.

Herrera / It seems like Team Midwest was arguing the point for me, so I’ll just take those points on my tab. But it’s really clear. There’s 1 therapy that’s associated with a survival benefit in the second line in patients with primary refractory disease or early relapse, and that’s CAR T cells. We have a lot of real-world data showing some replicating in a way in what we saw on clinical trials. Patients want to live. Doctors want to cure aggressive lymphoma. So early CAR T referral, regardless of patient characteristics, is really critical. There are plenty of data now to suggest that T-cell fitness is better the earlier we’re leukapheresing patients and using those leukapheresis products. Patients are going to do better with a lower tumor burden, so really getting those patients in early before they’re so heavily treated. Honestly, having that CAR T-cell referral be early so that we can really get started on the logistics of planning [is important]. They may need holding therapy or bridging therapy. They may need various interventions where you can collaborate with your referring physician in the community. But getting that referral done quickly so you can get a jump on getting those CAR T cells collected and infused is going to hopefully get you to where a patient wants to be, which is cured of their lymphoma.

Graff / Well, and speaking of bridging therapy, I think that’s what you just said about the earlier the better, that there needs to be a true multidisciplinary discussion when that patient relapses, you’re speaking to the treatment center right away. Because what you give today can affect tomorrow. And I think that is so critical. When you talk about early referral, it’s not just, “Oh, I’m going to give 2 cycles of blah, blah, blah, and then I’m going to call my CAR T partner.” No. You need to do it now and do it together. Would you disagree?

Oluwole / We should back up a little bit because the frontline therapies right now for large cell lymphoma are pretty good, especially for the low [international prognostic index (IPI)], whereas it is true to do early CAR T referral, should we do that for everybody? Maybe there’s a subset that is low tumor burden, with very low IPI and early-stage disease—and we know that R-CHOP [rituximab, cyclophosphamide, doxorubicin, vincristine, and prednisone] may be a 70% to 80% chance of cure. Perhaps some thought process in that would also help with the patient-specific referral. [But] the high IPI, the high tumor burden, the non–[germinal center B-like] type, as you’re starting them on R-CHOP, getting them to have some form of consult with the CAR T center would be a good complement.

Shah / I think the City of Hope is, just by the nature of their name, overly optimistic, just full of a lot of hope. We want early CAR T referrals, but we know it’s not happening. We know that, for years, we keep talking about this, that x number of patients should be getting CAR T, and a fraction of them get it. What we need are better, specific patient-guided treatment options so that the physicians in the community can identify the right patients who need that early referral. To simply say that all patients should get an early CAR T referral is just going to set us up for additional failure when we aren’t even getting to the patients who need it. Having patient-specific treatment sequencing is a realistic goal. I’m full of hope, too, but I’m also a realist.

Should efficacy, vein-to-vein, and safety profile impact CAR T selection in LBCL?

Shouse / When we think about factors that help us decide which CAR T product to give a patient, we do think about efficacy, vein-to-vein time, as well as the safety profiles. Those are probably the 3 most important parts. But as we move forward in time, some of those aspects have become maybe a little bit less important. With efficacy, for example, I don’t think there’s a clear winner when we look at the data—the real-world data or the trial data. Although there’s never been a head-to-head study, I don’t think that anyone can make a clear argument that one product is better than the other in terms of response and in terms of survival. It just comes down to a matter of preference or other factors. When it comes to vein-to-vein time, however, I think we all know that axi-cel does come out of the box faster. It’s got a shorter vein-to-vein time, and…manufacturing success is also previously better for axi-cel. As many of us know, however, the prior unfair cutoff by the FDA for liso-cel product viability has changed, allowing for the liso-cel manufacturing quality or approval to be much higher now than it was before. All the real-world data we’re talking about [has been] prior to that change. Now, the manufacturing success rate, in my experience, is much better with liso-cel, and probably comparable to axi-cel. When we talk about vein-to-vein time, it’s become less of an important issue with all the novel bridging therapies we have. We’re able to control disease a lot better. We also know that bridging therapy, when it’s successful, improves outcomes in CAR T.

Out of the 3 topics we have, safety is probably the most important. It doesn’t seem like it matters which study you look at; liso-cel is statistically superior in terms of safety, especially with neurologic toxicity. When I think of a young, fit patient, they can tolerate axi-cel, but if liso-cel works just as well, why would you make a patient take it? Why would you put a patient through an increased risk of toxicity? A lot of us give the patients back after 100 days. We really don’t know the long-term effects of high-grade neurotoxicity on long-term cognitive function. You’re talking about 20% to 30%-plus high-grade neurotoxicity rates with axi-cel and much, much lower [rates] with liso-cel. Are you really doing your patient a service by giving them something that’s more toxic without a clear efficacy benefit?

Shah / Let’s start with what the actual question is first, right? It wasn’t which CAR T is best in DLBCL. It is: What is the CAR T decision in rapidly progressing disease? This patient needs CAR T fast. You have 1 shot to get him a CAR T. Are you really going to put all your money on liso-cel and wait 4 weeks on a product that maybe comes in specification, or are you going to get the patient the product that you can get them the quickest that shows great efficacy? Now, you brought up that we have great drugs to bridge. I prefer not to bridge because, as you said, when bridging works, they do better, but that is the bias in itself. Obviously, diseases that respond tend to continue to respond to other therapies. My concern in this patient with rapidly progressing disease is that I have 1 shot to give him a curative-intent therapy. Dr Herrera mentioned the OS benefit, which I believe was also only shown with axi-cel in a publication. We need to give a therapy that’s going to save their life. In this situation with rapidly progressive DLBCL, I find it hard to convince myself to wait around and hope that the liso-cel product comes in specification in an [adequate] time period because, yes, there might be a difference in safety, and I have to admit that, but there’s no safety question at all if they die waiting for the CAR T product.

Shouse / Vein-to-vein time is only a portion of the waiting time, and not even the largest portion for most patients. I still bridge with axi-cel if I have to. In a patient with rapidly progressive disease, you can’t wait. If it works, then the likelihood that they’re going to make it to axi-cel and the likelihood they’re going to make it 1 more week to wait for liso-cel, in my experience, is not that big of a difference.

Round 3: Clinical Scenarios

Scenario 1:

Graff / [This] patient is 2.5 hours from the nearest CAR T center in a rural area and has an adult daughter and spouse available for caregiving. He’s willing and able to relocate temporarily for treatment but has limited access to local subspecialty oncology services. What would you do with this patient?

Team City of Hope

Herrera / That’s an interesting question. In terms of whether they should have gotten something different as part of their frontline therapy, we don’t necessarily have a cell of origin here, but they were a high-intermediate-risk IPI. In the [phase 3 POLARIX trial (NCT03274492)], there did appear to be some benefit to patients who got Pola-R-CHP [polatuzumab vedotin, rituximab, cyclophosphamide, doxorubicin, and prednisone] in the frontline setting. To be honest with you, I’m not too fussed about the frontline therapy they got. This sounds like a bad-acting, aggressive large B-cell lymphoma. One thing that’s become clear as we’ve both discussed these points is that in a patient with symptomatic disease and rising lactate dehydrogenase [levels], time is critical.

First of all, we would all agree that, in a primary refractory patient, we were aiming to infuse CAR T cells into this patient. In terms of a choice of product, axi-cel has the fastest vein-to-vein time. We’ve discussed this point before. In a 58-year-old patient, I don’t have as many of these concerns as an older, frailer patient. We’re going to assume that they’re in pretty good shape…. I would think this is maybe the one scenario where there’s an advantage to using axi-cel. [There’s] similar efficacy, and you really want to get the patient infused quickly. To the social and access considerations, this isn’t a patient who you have that 14-day, get-them-back-to-the-community timeline in mind because there are pretty limited services locally. There’s less of an advantage for liso-cel there. It sounds like there’s a pretty good support system. The patient can come to your treatment area from their rural area. You want to make sure they’re getting those CAR T cells. I’d certainly be trying to get them to CAR T cells, probably with some holding therapy or bridging therapy, depending on when I can get a leukapheresis slot. I’d probably be thinking about axi-cel in this situation.

Graff / [Team Midwest], is there anything different that you would have done?

Deol / I agree that we would want to get this patient to CAR T-cell therapy as quickly as possible; [we seem] to have the resources. There are programs that can help patients come from far away and stay close to the treatment center. That’s been a big help for some of our patients coming from farther out. They and their caregiver can stay close to the cancer center and get the benefit of that program, too.

Scenario 2:

Graff / Patient lives alone in a major metropolitan area. Uncertain availability of a dedicated caregiver, but curious about outpatient CAR T programs. Employed full time and is concerned about prolonged treatment disruption. Taking all of that—characteristics, age, and the social and access considerations—into account, Dr Deol, what would you do?

Deol / This is another patient who ideally should get CAR T-cell therapy because we’ve seen OS benefit in this situation for patients who relapse early after initial treatment, as we discussed with the ZUMA-7 data. This is a patient who should hear about that benefit. He wants to get the best possible outcome and wants to get back to doing work without disruptions. Getting him to CAR T-cell therapy would be the right thing to do. We occasionally come across patients like that who might not have a dedicated caregiver right away, but after they hear about what is needed and what the requirements are, they can talk to friends and family members. We’ve had some members from church who have volunteered to help with patients and get them through that first few weeks to a month or so of the requirement for having caregiver support. For this patient, I would consider getting them to CAR T-cell therapy. Depending upon what is happening, he doesn’t seem to be very symptomatic at this time, so there are no issues in terms of requiring any bridging therapy right away. We can get him to CAR T-cell therapy, and I would prefer to get him to axi-cel, if I can.

Graff / Would you do something differently, [Team City of Hope]?

Budde / No, it’s a really good discussion. One of the things that’s missing for this and the case before is really what the patient’s philosophy is. What is the patient’s tolerance of a life-threatening event? Those are pretty important to discuss with the patient.

Herrera / In this case, the patient is asking about outpatient CAR T programs. If there are caregiver concerns, perhaps a thought about liso-cel, since the safety profile is more favorable. It’s worth thinking about, especially when you hear about what they want their life to be.

References

1. Brisou G, Belot A, Bachy E, et al. Comparison of axicabtagene ciloleucel to standard regimens as second line treatment for large B cell lymphoma in real life: a LYSA study from DESCAR-T and REALYSA registries. Blood. 2025;146(suppl 1):2764. doi:10.1182/blood-2025-2764

2. Brisou G, Bachy E, Gat E, et al. Comparison of axi-cel versus liso-cel as 2nd line therapy for relapsed/refractory large B-cell lymphoma in real-life: a LYSA study from the DESCAR-T registry. Blood. 2025;146(suppl 1):127. doi:10.1182/blood-2025-127