Toxicity Management for CAR-T in Hematologic Malignancies

Transcript:Krishna V. Komanduri, MD: I would agree with what Fred said. We are taking a similar approach. We have four trials open and six trials coming in ALL and non-Hodgkin’s lymphoma, and now there are a couple of solid tumors at Sylvester Comprehensive Cancer Center. When it comes to solid tumors there are physicians that we’re not always frequently interacting with. We are having a joint conference that happens every couple of weeks where we’re discussing the pipeline for these patients. In our service, those other physicians aren’t directly attending on a day-to-day basis, and that was really their choice. They wanted us to get called at 3 in the morning about the decision of whether to give atezolizumab or when to give steroids. Obviously, all of you have centers where this is happening—the models differ from place to place. What is common is that there has to be involvement of a cellular apheresis center, a cell therapy laboratory, high-level nursing, and the kinds of sophisticated services that are typically found in transplant centers. It’s not to say that it can’t happen outside of a transplant center, but I think you have to have those resources. And, whether the transplant physicians are doing it, you often see the transplant floors utilized, because that’s naturally part of what we do.

Andre Goy, MD, MS: I want to ask you a question about this. There are a lot of strategies for trying to control the toxicity of CAR-T therapy because we don’t want to give high-dose steroids, as you mentioned. There’s an approach where we’re using a suicide gene to try to block the proliferation. So, what do you think of this?

Krishna V. Komanduri, MD: This is really exciting. And I think really the JCAR017 trial is the only one that has a negative regulator built in. The other trials that are happening right now don’t have that. One thing we don’t know is what degree of expansion is actually needed. We talked about this issue of persistence and it seems intellectually necessary that persistence is needed. But, we also know from the NCI data that there are patients who have had durable responses with no persistence. So, it may be that you need a big spike to actually reduce that high burden of disease. It could be that if you abort it with a suicide vector that you’re actually doing a disservice. I think that the concept is important. Whether we give steroids or use a suicide vector that actually deletes the cells to prevent that toxicity from happening, we have to think about that. But, I think we’re really early in understanding when we can successfully abort and how to balance the toxicities from the clinical effects.

Andre Goy, MD, MS: There’s a little bit of a debate about the durability of those CAR-T cells, and they need to be around for a long time. But, I think everyone would agree now, and here at this table, that you need to have an amplification in the beginning. And then, patients who have no toxicity, no fevers, and no neurology co-manifestation typically don’t have a response. Another point I wanted to make about immunotherapy, including checkpoint inhibitors, is the fact that for our audience, the restaging studies are different. When you look at the PET scan, very often the imaging is worse. And I think this is important to not be concerned, because there is a huge inflammatory reaction. So, we sometimes, on checkpoint inhibitors or also in CAR-T cells, see the first imaging that is worse.

Frederick L. Locke, MD: Yes, and I think there’s another strategy besides that suicide switch. If we could administer a drug alongside these newer constructs of these CAR-T cells that could dial up or down the number of T cells or their activity, that would be very useful rather than to terminate them completely. We don’t know about persistence. We don’t know how important persistence is and, in fact, some of the newer trials are looking in more sensitive ways to find the CAR-T cells within the peripheral blood. You don’t need a lot of memory T cells to hang out in order to re-expand and cause immune response. So, it’s still unclear whether you need persistence that we’re not detecting or whether we’re completely curing these diseases like ALL and lymphoma.

Krishna V. Komanduri, MD: And in studies of CMV (cytomegalovirus) immunotherapy, which seems like a different world, Paul Moss, in England, showed that infusing just 40,000 CMV-specific T cells could control refractory viral infections that have failed multiple antiviral agents. We know a small number of cells can do great good. I think to get to this issue of which strategies to use and when do we need to dial it up and when do we need to dial it down, is in these biomarker studies, both of efficacy. Is it the peak response? Is it persistence, and, also, of the toxicity? Can we predict, for example, by the basis of an inflammatory marker, like CRP (C-reactive protein), or something that’s in the CSF? Can we then say, ‘Okay, this is the individual in which they hit a threshold which now we have to shut it off, whether it’s by giving steroids or giving other agents, or by using a suicide vector’? I think we need that correlative data before we know how to tweak the response to maximize the good effects without seeing the bad ones.

Frederick L. Locke, MD: I think it’s also important to mention that, although early on there were reports that steroids would reduce the number of these cells, it’s possible that if the peak number are quite high before steroids are given, that steroids won’t affect the responses. So, we don’t fully know whether steroids affect the response rates in these treatments. We have to safely administer these treatments.

Andre Goy, MD, MS: It’s very early. But, I think we should mention for our audience the data in myeloma, where immunotherapy has not been really exciting, so far, except maybe the combination of checkpoint and lenalidomide, as you mentioned before, Fred. The deal in myeloma is that there’s a handful of patients who achieved a remarkable response after failing all kinds of therapies.

Krishna V. Komanduri, MD: That’s the NY-ESO-1-specific strategy. But, at least I heard, again, anecdotally that more recently that many of those patients are progressing. So, I think that it’s early, and we have to find…

Andre Goy, MD, MS: Proof of concept, yes.

Krishna V. Komanduri, MD: Yes.

Transcript Edited for Clarity