Transcript:Krishna V. Komanduri, MD: It is difficult if you have a very high-risk group of patients, and it’s one thing if you know that at 3 months, the person who has NCR is likely to be NCR at 1 year or 2 years, right? I think that that makes the decision not to do something, like an allogeneic transplant, very easy. But now that we’re starting to see fallouts, one of the things that we have done—again, hoping for long-term remissions—is, at times, watch patients, and we have seen relapses. Now, those relapses have sometimes been responsive to a second CAR T-cell infusion. At times, they’ve been limited in disease, even in patients who have bulky disease and we could get them to a CR, or near a CR, and then have taken these patients to allogeneic transplant. But the notion of CAR T-cell therapies as a bridge versus as an independent therapy is really a quite different process, I think emotionally for the patient and, as we view it, even in the health system.

Leo I. Gordon, MD: Right, I think so. That’s a major question right now that I think is going to evolve: is this a bridge? Allogeneic transplant is still a curative treatment for a significant number of patients. Not all, and it’s not for everybody because of the potential toxicity, but as Michael alluded to earlier, this is with nonablative allogeneic transplants. We can treat people into their 70s, and if they’ve got a donor available, it isn’t so much a problem. We can estimate now that 70% of patients ought to have a donor when you combine haploidentical-matched siblings or haploidentical-matched unrelated.

Krishna V. Komanduri, MD: I would go further. I would actually say nearly every patient has a donor.

Leo I. Gordon, MD: Yes, okay. I think that’s not the issue, but the issue is, will they tolerate the transplant? Will they tolerate the CAR-T? Is the CAR-T curative? Is it only a bridge to transplant? I don’t know the answer to that.

Krishna V. Komanduri, MD: Yes. And I want to make a point about transplant-eligible versus ineligible. We are now not using nonmyeloablative, as Michael talked about with FCR regimens in low-grade lymphoma. We can’t do that in aggressive lymphomas; you have to have more cytoreduction. So, we use regimens that are ablative; that is, if you gave this chemotherapy, you would not get hematopoietic recovery but they are reduced in their intensity. We typically would use fludarabine combined with melphalan at 140 m2, and that’s a pretty well-tolerated regimen. I think early 100-day mortality in a sibling population is under 5%, and we are doing allogeneic transplants in AML patients and in lymphoma patients up to the age of 75.

Now, above the age of 70, we look for clear fitness, but we really don’t exclude many patients under the age of 70. I think that’s important for community physicians because many community physicians think—we now estimate, in the work that I’m doing with the American Study for Blood Marrow Transplantation—that probably no more than 15% of acute myeloid leukemia patients are getting allogeneic transplants if you look at, for example, insurance databases. And probably no more than 30% of myeloma patients, despite the fact that we’ve had data for 20 years that suggested that autotransplant should be there.

So, I think that there is a lack of understanding, and for the community physician who is wondering whether there’s a role, I would say send the patient to us. We’re not going to do a transplant in somebody where we think that we’re going to do more harm than good. And often patients and their referring physicians are surprised that a patient is transplant-eligible because the transplant eligibility has changed dramatically in the last 20 years.

John Byrd, MD: One of the exciting things that’s going to come from the CAR T cells is—and Leo alluded to it with the checkpoint inhibitors, and there’s several abstracts at this meeting and papers that have been published—that small molecules can restore T-cell health. And ibrutinib has the ability to do that and to make CAR T cells work well, work better in preclinical models. And now this is being tested in studies. But it still gets that critical question: is this going to be curative or not? And until we get the answer, it’s going to be hard to see, doing this approach, until patients are at a more advanced stage in their disease, don’t you think?

Krishna V. Komanduri, MD: I agree. And I think that another significant point is really the cost of these therapies, especially if they are not curative therapies, and they are then applied in sequence with other therapies that have significant cost to the system, right? I think the financial notion of CAR T cells, as a curative one-time therapy, is quite different if you then now say that the patient needs an allogeneic transplant and if that allogeneic transplant, in the future, might consist of maintenance therapy, right? So, I think that these are really societal issues that we haven’t fully addressed, but clearly, we’ll have to have the information going forward.

I want to note a couple of things about the evolution of CAR T-cell therapy. There is a Kite study that’s looking at checkpoint inhibition in the lymphoma setting—where, again, the rate of neurotoxicity appears to be lower of CD19 CAR T cells—in combination with atezolizumab to try to take the brakes off of those cells. And Carl June presented some preliminary preclinical data that are going to lead to a human trial where they’re actually using CRISPR/Cas9 gene-editing technologies to get rid of PD-1 on the surface of T cells and also get rid of HLA class I, which might be the first step toward an off-the-shelf CAR T cell that would come from another donor. Therefore, if it doesn’t have class I, then it’s less likely to get rejected.

I think that there’s a lot we clearly don’t even know where they need to be applied or who’s going to get cytokine release syndrome or how to optimally manage the neurotoxicity, but I do think that we will learn more about this. There are some interesting biomarker studies that are going to be presented, and certainly the Juno and Novartis groups are, along with others, really doing very detailed and thoughtful analyses of the correlates of toxicities, both cytokine release syndrome and neurotoxicity. And I do think that we are likely to get to the point where we can either predict which patients may have unacceptable rates of neurotoxicity or be able to intervene. Mike Jenson has data, for example, in Seattle suggesting that very low doses of steroids, given preemptively, can be associated with decreased rates of cytokine release syndrome and neurotoxicity without compromising the efficacy of those cells. So, these are all things that, to me, give me lots of hope, but there will be steps forward and backward on the way.

Transcript Edited for Clarity