CAR-T Therapy for Relapsed/Refractory B-Cell ALL: Sequencing Decisions

Elias J. Jabbour, MD: Something else about this trial that was just published in Blood Cancer Discovery is the impact of response at day 28 and at 3 months. What’s interesting about this paper is that patients who responded well and had MRD [minimal residual disease] negativity on day 28 and 3 months have a long-term survival of 70% compared with the patients who didn’t respond. It seems to me that patients who are responding may not need allotransplant.

Jae Park, MD: Correct.

Elias J. Jabbour, MD: That brings me to something very important to discuss with you today. Can we say that if somebody had a response to CAR [chimeric antigen receptor] T-cell therapy and had MRD negativity by NGS [next-generation sequencing] that we can spare the need for allotransplant?

Jae Park, MD: I think so. There’s a bit of a caveat there, where we’re beginning to know these data. Though we certainly have the most robust data with tisagenlecleucel, not every CAR is the same. We don’t have as much data analyzed for brexucabtagene. Hopefully those data will come later. But with tisagenlecleucel, given the large number of patients with longer-term follow-up, with a recent subgroup analysis in the paper, it does typically suggest deeper remission, NGS-negative, 10-6 sensitivity.

In the paper, patients who had B-cell aplasia that was maintained long term had a very long-term outcome without subsequent transplant. Very few of those patients—less than 20%—went to transplant. Based on those data, after tisagenlecleucel, for patients who have ongoing B-cell aplasia beyond 6 months and have MRD negativity by NGS at months 1 and 3, I’d feel more comfortable not recommending transplant with that particular product. Whether that can be generalized to other CD19 CARs, we have to be a little careful until we get such data.

Elias J. Jabbour, MD: You mentioned something important: that in ALL [acute lymphoblastic leukemia], you can’t wait until you get the CAR T cells. Often, patients receive bridging chemotherapy or immune therapy. It can be inotuzumab or blinatumomab. I have 2 questions for you. Do you have any objection to using blinatumomab before CAR T? And we know that blinatumomab is better in MRD negativity, or minimal disease. Do you think that even with a CAR T, it will be best used in consolidation with CAR T in patients with minimal disease or no disease?

Jae Park, MD: Those are good questions. These are real-life questions that a lot of us are dealing with. I wouldn’t choose based on a potential future need of CD19 CAR T-cell therapy, meaning deciding between blinatumomab or inotuzumab. If a patient is ready for CAR T and it’s the best therapy at that moment, we’ll choose that. But if not, with previous inotuzumab or blinatumomab, there have been many subgroup analyses that show that it may not matter as much, and certainly for adult patients.

But with the bridging therapy immediately prior to CAR T-cell therapy after CAR T-cell collection, I’d avoid blinatumomab because I’m going after CD19 and I likely need some of the antigen present to enhance the T-cell expansion and antitumor efficacy. I might not use blinatumomab immediately prior to CAR T. But the month before that, I would have no problem using it.

You asked a very important and relevant question about the inotuzumab, which has a high efficacy and high MRD negativity rate as well. Can we use that as a bridge? The dose is usually a bridge to transplant. But instead of a transplant, to avoid the VOD [veno-occlusive disease] risk, can we substitute it with the CD19 CAR T cells? That’s going to probably be the next big question that we should be asking ourselves, hopefully in a clinical trial setting. We have to see. It sounds great. Hopefully it will translate to be that way. When you ablate the B cells too much with inotuzumab when there’s no disease, we don’t know how the CAR T cells will fare in this setting.

Elias J. Jabbour, MD: There were also data presented at the 2021 ASH [American Society of Hematology] annual meeting from CHOP [Children’s Hospital of Philadelphia] about infusions of CAR T post the first CAR T infusion. Essentially, if somebody is progressing, losing the T cells, or emerging with MRD, is there room for more CAR T infusions as a maintenance or salvage?

Jae Park, MD: Relapse after CAR T-cell therapy is always challenging. As we discussed, even though the initial response rate to CAR T-cell therapy is 80%, anywhere from 40% to 60% of these patients can relapse, at least in the adult ALL setting. We’re trying to come up with ways to prevent such relapse. The reinfusion of CAR T has been looked at both ways. One is trying to act early for loss of B-cell aplasia or the recovery of normal hematogones. The CHOP group has looked at tisagenlecleucel reinfusion in children and young adults. These data presented at ASH look interesting as an oral abstract. When the reinfusion of the CAR Ts are used in that setting even before MRD relapse and morphologic relapse, but with B-cell recovery, it was effective compared with the historical control of prolonging the response duration for those patients.

If you receive reinfusion at the time of morphologic relapse, they don’t respond very well. We get about half the response rate, meaning typically about 40% response rates. Even when you do gain the response, the duration of remission is fairly short. While we can certainly try it for a subset of the patients, my general sense is that it isn’t a good long-term solution. After I give reinfusion, even if they respond, I’ll be concerned about potential relapse, and I’ll be monitoring those patients closely and trying to come up with what else we can do, whether blinatumomab or inotuzumab could be useful in this setting, if it was CAR relapse.

Elias J. Jabbour, MD: That brings me to other abstracts from ASH, including tandem CAR T—CD19 and CD22 given concomitantly or sequentially—and the Sleeping Beauty CAR T, the NK [natural killer] cell CD19. What do you think of the novel CAR Ts?

Jae Park, MD: Along the same lines, it’s one of the other ways the field is moving. Because initial response is great, but what are the ways to prevent such relapse? One such way is targeting both antigens at the same time or sequentially. We’re somewhat doing that with blinatumomab and inotuzumab, but can we do that simultaneously with this CAR T since these T cells are genetically engineered and we have a capability to do so?

There have been several batches of bispecific CAR T-cell data presented from different groups, and there are many different ways to target them. One such way is that you can generate CD19 CAR T and CD22 CAR T and infuse them together as a T-cell product, or you can engineer the CAR itself so that it has 2 arms—one to CD19 and one to CD22—and then infuse 1 CAR product that targets 2 antigens. There are many ways of doing so. The data look promising. At the minimum, they’re generating the same response rates that we’re seeing with the single-antigen–targeted CAR T, about 80% initial response.

What’s less clear at this time is whether this dual-targeted approach will prevent relapse. We need longer-term follow-up in more patients to see whether we’re achieving a deeper response either with NGS, as we talked about before, and when they relapse, whether they’re relapsing with a CD19-positive and CD22-negative, or the negative with either of these antigens. What we have seen so far in some of the data is that even when you do a bispecific, one CAR may perform better. Either will predominate, and one CAR will persist longer, either the CD19 or CD22. The antigen specificity and their sensitivity to antigen density also differs. CD22 is very sensitive to the antigen density, so you need a more antigen-dense environment for them to work well. Even though you’re infusing them, they may not work as well, so it’s similar to infusion of single-antigen–targeted CAR T. The data generated so far are very interesting and intriguing. We need a little more time to see whether that approach will be what’s going to lead to the success of less relapse in the future.

Transcript edited for clarity.