Panelists: Mark R. Litzow, MD, The Mayo Clinic; Jae Park, MD, Memorial Sloan Kettering Cancer Center; Bijal Shah, MD H. Lee Moffitt Cancer Center & Research Institute; Anthony Stein Gehr Family Center for Leukemia Research
Bijal D. Shah, MD: But I told you I was going to be difficult, so now I am going to be difficult.
Jae Park, MD: Bring it on.
Bijal D. Shah, MD: So we talk a lot about toxicity with CAR T cells. We’ve talked a lot about neurotoxicity and the challenges there. We talked a lot about CRS [cytokine release syndrome] and the challenges there. And I’m going to ask you a question, knowing full well that you’ve had a lot of experience with this. How often do you think that’s really going to…are we going to see patients pass away as a consequence of toxicity, which is the ultimate endpoint? Because, if these are transient, reversible toxicities—if they’re manageable toxicities—then how worried do we need to be when we have an 80-year-old in front of us or even a 75-year-old in front of us with ALL with—I’m going to make something up—40% blasts?
Jae Park, MD: I mean, as well reported, there have been fatal incidences due to just toxicity, both for the CRS—cytokine release syndrome—usually with a concomitant infection, [so] that we know the risk is higher for these adult patients. Neurotoxicity—there have also been fatal incidences, usually due to cerebral edema, but even…encephalopathic disease can lead into it. These are rare events, but, having said that, these are real events that happen, and these are the ones that we worry [about]. I think that’s for the other reasons. CRS, to me, to some degree easily is more predictable and more manageable…and there are more modifications you may do. If somebody is having high fever even before the T cells go in, [having] uncontrolled infection, [that] may not be the best person to infuse. You may want to wait a little bit if you’re able to before infusing the T cells.
Neurotoxicity is a little bit difficult in the sense that if we see a patient with a grade 3 neurotoxicity, which is nonverbal, kind of at the time, the difficult thing is, are they going to stop there in grade 3 and just got better grade 2 and 1 and then completely reverse, or are they going to progress to grade 4 and hopefully not to grade 5, but if that’s going to be the event…When you’re in front of the patient at the time of toxicity, I think these are the difficult moments to decide. There’s really no good way to predict what is the stop.
So, before treating…As you said, if I have an 80-year-old person…there are some predictive markers or associative markers that correlate with toxicity, disease burden, and so forth. If I have a high risk, I’m going to have to really think hard about doing this. So that’s all the reasons the prevention approach will be better, because there’s really no very good management of neurotoxicity, other than steroid. We all do it and they work, but—I mean, we think they work. They get better, but is it because of steroid or is it natural course of the neurotoxicity kind of resolved?
Bijal D. Shah, MD: So the reason I’m being difficult is…I will say that the mortality from CAR T across trials, particularly—and, again, it’s never good to average across trials, but when you look at the average mortality, we’re talking around 3%. We’re talking about in line with autologous stem cell transplant, and if this is someone that I would auto, then usually this is someone that I would think about CAR T-cell therapy for. And [although] there are rare events, I’ve become much more comfortable moving patients into the CAR-T space, just by virtue of that very profound CR [complete response]. And, again, [that] can be significant but transient toxicity.
Mark R. Litzow, MD: So, Jae, one of the questions that I think is [on] many of our minds is [about] the high response rates and the durability of some of the responses that we’re seeing with CAR T-cell therapy: Where does transplant fit in for those patients? Should we be transplanting all those patients, none of them? How do you think about that?
Jae Park, MD: I think this is a really challenging question. Easy answer is, we need more data. But the 1 thing to consider is that I don’t think every patient needs to go to transplant. I do believe there are a subset of patients who do not need to go to transplant after CAR T cells. The challenging part is, who are these patients? Really, the important thing is…, we have to examine each clinical trial data separately on its own. So it really depends on what patient population, what type of cell therapy they received, and what have been the outcomes of those particular [studies]? And it’s impossible, and we shouldn’t be really extrapolating data from the one and then [applying] it to another, assuming it’s the same. So I think all these factors kind of put into it.
[From] our own data with our own 19-28z [zeta] CAR T cells…, we do know that there is a subset of the patients who do better even without the transplant. These are low-disease-burden patients. High-burden patients did well…and then, whether they got the transplant or not, unfortunately, they did poorly. But the low-burden patients, less than 5% essentially MRD [minimal residual disease], the positive patients, they tend to do well even without the transplant. So that [is] the decision that I would be making if I’m treating some patients with a 19-28z. And with the other type of trials, we have to look at that.
There have been some adult ALL trials that have been presented so far, even at this year’s ASH [American Society of Hematology], as well—the 2018 ASH. And then there are some other clinical trials looking at [patients who] automatically went to transplant, had to go to transplant if they had early relapse. In that data set, if you look at them…, the 1 question is, why are they relapsing so quickly? But in the…
Mark R. Litzow, MD: Is this relapse after CAR T?
Jae Park, MD: Relapse after CAR T-cell therapy. The data [were] presented earlier looking at that. So they were arguing from that particular study from the 2 Chinese groups; they were transplanting almost everybody—patients—because their relapse rates were high. I think that makes sense for that particular group of patients and particular CAR T product, because they clearly have their own data to suggest that these remissions are not durable. But just because of that, that also doesn’t mean that everybody should be going to transplant after CAR T-cell therapy. So I think that’s the point: that we have to look at individual kind of…data to look at…
And then—so many other variable factors in not only the relapse but also the nonrelapse-related mortality after transplant, too. If somebody already had a transplant, a second transplant is really feasible, or if we think it’s going to benefit the patient. But, if it’s a young person who never had a transplant, is it right, or should we not think about the transplant? So in all [these] data, we need more time to kind of digest and then really fully interpret the data. But kind of my major take is that we have to look at individual data, [the] clinical trial, and try to understand that a little bit more and not…apply 1 [set of] data to another and generalize kind of all the data together.
Anthony S. Stein, MD: I think in this ASH meeting, the data from [the] Seattle group…actually broke down the data into patients who had CAR T cells and then went to their first transplant versus those patients who got CAR T-cells after having relapsed from a prior transplant and then went to a second transplant. And I think this is the first paper that I’ve seen with patients where the trial has been broken down, whether patients are receiving a first versus a second transplant. Now, in their abstract, they showed that for second transplant, there was no benefit. For first transplant, there seemed to be a benefit. And then the other question is, is there some biomarker like B-cell aplasia—when you lose, if you don’t get that or you lose it, is that the time to do a transplant?
Jae Park, MD: Yeah. These also all depend on the CAR T construct there, too. The other question to ask is, what about the persistence? I think…just because you see them doesn’t mean they’re not going to relapse also. It doesn’t guarantee, but you may just relapse with a 19-negative one. So we also have to look at persistence with the B-cell aplasia to see whether these are functional T-cell persistence.
If you lose B-cell aplasia, I think some of the data coming from the 4-1BB from the CHOP [Children’s Hospital of Philadelphia] and UPenn [University of Pennyslvania] in ALL studies—you know, these are patients that may be at higher risk of a relapse and therefore can attempt a transplant. And whether it’s going to be true for the 19-28z, which we haven’t looked at as extensively to see whether it means the same for the 19-28z, which has different kinetics, less persistence, how reliable is that as a biomarker—I think we do worry, monitor those patients much more closely, if not sending them to transplant.
So I think that definitely helps us [in] making that decision. But is it [an] automatic sign to go to transplant now? I think these are very difficult questions at this time. But I think which disease state they receive the CAR T-cells for and whether they had a prior transplant or not and then what is anticipated or estimated transplant-related mortality—it’s complicated but complex decisions that [everyone] needs to kind of take part in.