Caron Jacobson, MD: In our series, we didn’t do the propensity score analysis, so we weren’t able to subtract the high-risk disease characteristics. But we also saw a decrease in overall survival and an increase in nonrelapse mortality—so treatment-related mortality. And some of those are infectious. So the sort of piggybacking of bridging therapy to lymphodepleting chemotherapy may be particularly harmful for these heavily pretreated patients.

David Maloney, MD, PhD: There’s another option here. Bridging therapy may actually have an impact on lymphodepletion. If you come into lymphodepletion with different milieu or different lymphocyte count, the efficacy of the lymphodepletion may be important. We have found that the intensity of the depth of lymphodepletion can impact outcomes with general CAR [chimeric antigen receptor] T-cell therapy. We’ll get to that at a later time, but these things may all be completely connected and not easily separated.

Caron Jacobson, MD: There’s 1 other point I want to make about bridging therapy. All these analyses with bridging therapy are before some of the newer drugs that we’ve seen come to market recently and presented on at the ASH [American Society of Hematology Annual Meeting & Exposition] this year. There are drugs that are more effective now and more effective at debulking. I think we’ll get to a point later where, just like in ALL [acute lymphocytic leukemia], we’re learning that tumor volume is an important predictor of response in lymphoma. And so if we are giving someone a more effective bridging therapy and can actually debulk them, we may improve outcomes. It’s hard to know. Right now, we just don’t know. If it’s something inherent, maybe it would just be a Band-Aid for something in the disease that leads to lack of response. Or is it actually something about the tumor volume itself?

Max Topp, MD: But there’s risk with that, because all these new therapies that you mentioned are actually curative. We end up seeing patients in a later relapsed stage. From the way I’ve seen things done previously, we would have a hard time generating a product for the patient. It’s positive, but also it could be quite detrimental to the success story.

Caron Jacobson, MD: Sure, and I couldn’t agree with you more. I just want to clarify that I was talking about using that after apheresis.

Jason Westin, MD: We’ll talk about other CAR T cells in a moment, but the other CAR T-cell studies use bridging therapy. ZUMA-1 with axicabtagene ciloleucel was a little different. It did not allow for bridging therapy. The JULIET study as well as the TRANSCEND study both allowed for bridging therapies.

David Maloney, MD, PhD: But again, I think the main reason for that was because the production, timing was not—the beauty of ZUMA-1 was that the drug was there in 16, 17 days in a very reliable fashion. It’s really important that you can get this drug for your patients when you need it. We all know that if you have many failures, you may or may not have another chance to even get these patients to treatment.

Jason Westin, MD: In the JULIET study, it was closer to 50 days for the infusion time frame.

David Maloney, MD, PhD: Right. Initially for sure. Are there any other characteristics of the patient that are going to predict who is going to have a good outcome?

Max Topp, MD: Tumor volume is really very crucial. Slicing the patient into quartiles. It’s very obvious that the patient with the lowest tumor volume has the best response. This patient has the best long-term outcome and lowest rate of severe cytokine release syndrome or neurotoxicity. On the other side, the guys who come in with very high tumor volume are the ones...

David Maloney, MD, PhD: Fred, that came out of your data for the ZUMA-1 trial. That’s a very interesting observation. If you got those people on that lowest quartile, they did remarkably well—with, like, 67% 1-year PFS [progression-free survival]. Let me just take it 1 step further. Does that mean that if we pound on people before we give them CAR T cells and get them into whatever disease range, the outcomes are going to be better?

Frederick Locke, MD: I’m not sure about that. Again, back to the real-world evidence, we showed that bridging therapy doesn’t really correlate to improved outcomes. Hitting people with more therapies that aren’t working to reduce tumor burden may not be helpful. I think the interesting finding is that there may be an effector-to-target ratio when it comes to axicabtagene ciloleucel and durable responses to that therapy in large B-cell lymphoma. In fact, the quality of the CAR T cells probably matters as well. So you have to have quality CAR T cells and enough to overcome the amount of tumor. Hitting somebody with more chemotherapy may not reduce the tumor burden, and it may make the T cells worse if you do it before collection.

Jason Westin, MD: Even if you do reduce the tumor burden, it may be the tumor burden is a surrogate for other factors. You may have more aggressive disease because you have a big tumor burden. It’s not that the burden itself is necessarily the variable but why you have a big tumor burden.

Max Topp, MD: I would also like to stress that it is also T cell fitness. We do know that the patients who actually have high-peak expansion area under the curve are those patients with the best disease control. It may be an avenue of actually enhancing that expansion potential with other compounds in those patients receiving a CAR T cell product to maybe overcome this really detrimental situation.

Caron Jacobson, MD: Rather than pounding the patients with chemotherapy or pushing lots of bridging therapy, there’s 1 other strategy we could take. The NCCN [National Comprehensive Cancer Network] Guidelines don’t necessarily have us repeatedly scanning our large cell lymphoma patients in remission to catch an early relapse. Historically, finding an early relapse hasn’t led to improved outcomes. But this data would suggest that if you can find someone early with low disease burden, if it isn’t related to some other inherent aspect of their disease, treating them earlier may actually impact efficacy. And so we may be looking at a paradigm shift in terms of thinking about how we should survey these patients.

Max Topp, MD: I think the data that are emerging looking at MRD [minimal residual disease] or tumor DNA is actually where we should go.

Caron Jacobson, MD: Yes.

Max Topp, MD: And we see these patients who are at risk, so we just take them in a little earlier. We scan him earlier, 8 weeks, and then until we see clear evidence of disease.

Caron Jacobson, MD: Yeah, there are some really interesting data using the CAPP-Seq [cancer personalized profiling by deep sequencing] technology that Jason talked about before in large-cell lymphoma after frontline chemoimmunotherapy. It was very predictive for relapse. And so you’re exactly right. That may be the most sensitive way, rather than scanning people all the time to do this.

Transcript Edited for Clarity