Krishna V. Komanduri, MD: I want to talk a little bit, since we have now the axi-cel approval, other products that are in the pipeline. David, the JCAR017 product, obviously both the Fred Hutchinson Cancer Center and Memorial Sloan Kettering Cancer Center contributed to the design of the trials that led to the JUNO trials that are ongoing. Do you want to talk a little bit about how JCAR017 may be similar or different from the commercial axi-cel product?

David Maloney, MD, PhD: Yes. There are quite a bit of differences between JCAR017 and axi-cel. This was really based on studies we had done with a product called JCAR014, which is really the same construct. The cells are produced in a slightly different fashion. That’s what JCAR017 is. JCAR017 is now in phase II trials through the dose expansion phase and is going into its cohort, hopefully then to be submitted obviously to the FDA. This is a different construct. It’s based on 4-1BB rather than CD28 in the costimulatory domain, which leads to some differences. And also, the approach that JUNO has taken is based on our data at the Hutch, which was to put this into separate cell populations.

So, most of the CARs that have been given to date just collect T cells from the patient and expand them and put the CAR in and give them back in the right number. And in our studies at the Hutch by Stan Riddell’s group really showed that the types of T cells you put into this matter. So, what we would actually do is select CD4 cells, CD8 cells, make 2 CARs, and then combine them in a 50/50 mixture and give them back. And when we did those in our studies, we showed that there was a dose response and a dose-toxicity relationship. And being able to dial this in, we think we have a better safety profile, at least with our earlier studies. This was then transferred into the JCAR017, which is now also in a multicenter trial and being reported at this meeting. And Dr. Abramson has reported very encouraging activities, very similar to what we have seen in the other studies: 70+% overall response rate, some 50+% CR rates in similar-type populations.

What seems to be somewhat intriguing is that the rate of severe CRS and neurotoxicity appears to be quite a bit lower with the 4-1BB products compared to CD28. And we’re not quite sure whether that’s because of the cell population or just the actual construct of the CAR. And I think Steve’s approach using Novartis tisagenlecleucel in lymphoma also kind of mirrors that where the toxicities come on a little bit later and maybe slightly less severe. But, obviously, we really need 2-year to 5-year follow-up. Follow-up is very short, and we need to see where those relapses continue to occur or whether people are in a stable remission.

Stephen J. Schuster, MD: It’s kind of dangerous right now to compare one product to another product. I think the patient populations were slightly different in the studies, and I think we need more time. But it’s exciting that we have 3 products that work.

Krishna V. Komanduri, MD: I want to comment just a little bit on that. We’ve studied viral-specific T cells in my laboratory and have done so for a couple of decades. About 20 years ago, when we first had the first methods of looking at single-cell responses to viruses, we recognized that individuals who had more CD4 cells also had more CD8 cells and that those ratios matter. Now again, I think the basis and preclinical data suggesting that the combination of measured populations of these cells might lead to better persistence and better efficacy. I think that they still do need to be compared because even an un-engineered product has both CD4 and CD8 cells, and this is an important question for the field.

David Maloney, MD, PhD: It’s absolutely an important question. But the question is, by using a combination, can you then dial down the dose to get to a safer population? But there’s no direct comparison yet between these approaches, and I think that’s really key. But it’s an exciting part of the field. Fast forward 2 or 5 years where we have 3 CARs approved for large cell lymphoma. Then what’s going to be driving their use in the clinic—and I suspect it will be efficacy—both short-term and long-term, it will be toxicity, or lack of, and it will be cost. And so, those things eventually will have to start playing in to how we look at these products.

Michael Pulsipher, MD: And it’s obvious that those questions are going to be foremost because almost all of the efficacy data are very similar between the products right now.

David Maloney, MD, PhD: It seems to be.

Transcript Edited for Clarity