Evaluating CAR T Therapies in Lymphoma

John Leonard, MD: Cyrus, we have a number of different constructs. They sometimes feel to me, and our cell therapists do the CAR [chimeric antigen receptor] T cells here, so I don’t directly administer them, although I have lots of discussions with them about it. But axi-cel [axicabtagene ciloleucel], tisa-cel [tisagenlecleucel], liso-cel [lisocabtagene maraleucel], JULIET, ZUMA, TRANSCEND, we’ve got all these names and alphabet soup, and it’s hard to keep track of all of them. To some degree, the referring doctor, if they’re not administering it, maybe they don’t need to know the details of this. But I think it’s helpful to have a general sense, these are more similar than they are different, clearly, but there are some differences. What are some of the key differences people should know about? I’ll ask you, Cyrus, about the agents, and then I’ll ask you, Loretta, about the studies out there, and how you think about comparing and contrasting them a little, knowing that again, they’re more similar than different, I would say. So, Cyrus.

Cyrus Khan, MD: Sure. You’re right, it’s very difficult to remember all these names, obviously, but thefirst 2 drugs were [axicabtagene ciloleucel] and then tisagenlecleucel from JULIET, and the most recent one was the lisocabtagene maraleucel that was approved. I think the main difference is the construct. The axicabtagene ciloleuce has the anti-CD28…domain, and the other one has the 4-1BB construct, that’s the main difference. I think as far as efficacy is concerned, and the duration of response and the PFS [progression-free survival], they’re more or less the same. One of the differences that was seen in one of the more recent abstracts, and we’ll talk about that, would be the toxicity profile. I think with the axicabtagene ciloleucel, you can see a little bit more of the CRS [cytokine release syndrome], perhaps a little more of the neurotoxicity also. In the others, not so much. Also, how long the primary CAR T cells remain engaged in the bloodstream also is longer in the other 2 constructs, rather than the axicabtagene ciloleucel. Whether that means anything as far as prolonged response is concerned, nobody knows, but longer-term studies have shown that they probably don’t mean anything. These are some of the differences, and obviously, manufacturing time is very similar between all of them. I think how we choose at our center is what we can get to the quickest. I think with the COVID-19 pandemic, it’s been rather tough getting slots for patients, especially for axicabtagene ciloleucel, and the other ones also. It depends on where we can get a spot, and how I can be comfortable using any of these CAR T-cell constructs for any of our patients until we get more data in which one is better in each patient, depending on those patients’ other features.

John Leonard, MD: Great. Loretta, as you look at the studies, and obviously trying to compare across studies is difficult, but it seems to me like there might be some differences in adverse effects, there might be some differences in the percentage of patients who are enrolled and then get treated. What’s your big picture take on that, and how has that affected your plans and recommendations beyond? Or is it like Cyrus, you get what you can get, and whatever logistically works, you move ahead with?

Loretta Nastoupil, MD: One thing I’d say, we participated in all 3 of the pivotal phase 2 studies, so I have experience from the prospective studies, and then I have commercial experience with all 3, so it’s a luxury of working at The University of Texas MD Anderson Cancer Center and being agnostic to a particular construct. I’ll highlight the key differences in the study population because you can’t compare across. The studies were done differently, for instance, ZUMA-1 studying axicabtagene ciloleucel, this was in a refractory patient population, so that was defined as patients who had failed 2 prior chemotherapies or had progressed within 12 months of an autologous stem cell transplant. The JULIET trial did enroll relapsed or refractory patients. There was a longer time from enrollment to cell infusion with tisagenlecleucel, so as a result, maybe a little bit less refractory patient population. There was clearly a longer time of about 56 days from enrollment to cell infusion. They also allowed for bridging, and they excluded primary mediastinal patients. Then the TRANSCEND trial, which is lisocabtagene maraleucel, had the broadest patient population, so they allowed for ECOG performance status of 2, for instance, an ejection fraction as low as 40%, creatinine clearance as low as about 30 mL/min, and they also included transformed disease from other indolent histologies, such as marginal zone, Richter syndrome, and a history of CNS [central nervous system disease]. The baseline characteristics are clearly different. The percentage of patients who were successfully manufactured and infused were different. It was highest among the axicabtagene ciloleucel patients, probably speaking to the fact it’s about a week shorter time in terms of manufacturing. There was also different grading applied in terms of toxicity, and different management applied.

So, you cannot compare across these studies, which is why at ASCO [American Society of Clinical Oncology annual meeting] this year there were retrospective analyses trying to control for baseline characteristics comparing tisagenlecleucel to lisocabtagene maraleucel in terms of efficacy outcomes. Then Jordan Gauthier, MD, from Seattle, is involved in a multicenter retrospective study looking at the acute toxicity and how it varies across the constructs. I do think that the toxicity is highest with axicabtagene ciloleucel. There are higher rates of grade 3 or higher neurotoxicity, higher likelihood of ending up in the ICU [intensive care unit], nearly all have CRS within 2 days. Again, it is administered inpatient. For those reasons you might have centers that steer away from axicabtagene ciloleucel, but there are also very high successful rates of manufacturing. That comment you heard Cyrus made, we do feel like sometimes we have 1 shot at this, and so we want to take our best shot. That’s why at our center we’re still prescribing axicabtagene ciloleucel at higher rates than the others. I think the real question is if lisocabtagene maraleucel can manufacture because they have probably the most favorable toxicity profile. At the end of the day that might be a deciding factor in terms of health care resource utilization.

John Leonard, MD: David, what’s your take on this? There were data at ASCO suggesting that the product might impact the CRS and ICANS [immune effector cell-associated neurotoxicity syndrome] severity, and that seems to be an issue that comes up a bit. What’s you’re take on that? My understanding is that the criteria are not always easy to line up, or at least historically haven’t been, because probably this wasn’t an entity or an issue or a scale going back several years. What’s your take on that when you think about these products?

David Rizzieri, MD: I think Loretta outlined it quite well. I think it’s really difficult to try to compare across studies and no good way to do it. I do think we have a sense of certain constructs giving slightly more adverse effects, where we’re continuing to keep the patients inpatient, but we’re looking at doing more and more outpatient with the new approval as well. That makes a difference for cost and support for the patients. We’re going to be trending toward choosing the agents with the least toxicities per the studies and publications likely in that regard. That’s probably going to play a main role as long as the efficacy seems to be similar. Of course, these are all first-generation products, right? I think in a few years hopefully we’ll be at the next step of even better agents. That’s is our initial take anyway.

TRANSCRIPT EDITED FOR CLARITY