Jennifer Brudno, MD, Discusses Anti-CD19 CAR-T Therapy for B-Cell Lymphoma

Jennifer Brudno, MD, Associate Research Physician, National Cancer Institute, Bethesda, Maryland, discusses findings from a clinical trial on the safety and feasibility of anti-CD19 CAR-T therapy for patients with B-cell lymphoma.

Transcript

My name is Jennifer Brudno. I'm an associate research physician at the National Cancer Institute in Bethesda, Maryland.

We designed an initial phase I clinical trial here at the NCI that was looking at a CAR construct that was developed by my mentor, Dr. James Kochenderfer, and Dr. Rosenberg here in the branch.

That CAR T cell study was very exciting. 22 patients were treated who had very treatment refractory non-Hodgkin lymphoma with several prior therapies. The complete remission rate was 55 percent, which was great in that population. A small number of patients seems to be better than historical controls.

Unfortunately, about half of the patients had pretty severe neurologic toxicities that we grade as 3 or 4, so more severe end neurologic toxicities. Fortunately, this was reversible in all of the patients, but that clearly is a barrier to using CAR T cell therapy.

That construct eventually was used (and is) now the commercial product axi-cel, which is now available commercially and many patients have received. And for that CAR T cell product very similarly since it's using the same CAR construct, neurologic toxicity has been a big barrier.

The Kochenderfer lab had been in pre-clinical studies looking at variations of the CAR construct. Can we make it less immunogenic and potentially less toxic? A CAR was designed that was meant to meet those criteria.

The Kochenderfer lab developed a new CAR which pre-clinically looked pretty exciting. Rather than having a CD28 hinge and transmembrane region, it had a CD8 hinge and transmembrane region. The costimulatory domain, just like the axi-cel product, was CD28.

Another structural difference is that the single chain Fv, single chain variable region, was made of a fully human protein rather than the murine derived protein FMC63, which is again what is used in most commercially available CAR T cell products and in the axi-cel product.

We treated 20 patients using this new CAR construct with these different structural components as I described. The complete remission rate was very similar. It was 55 percent, also, in a heavily pre-treated high-risk cohort of non-Hodgkin lymphoma patients.

The surprising thing was that the neurologic toxicity rate was very low. Only 1 patient out of 20, so 5 percent, had grade 3 or 4 neurologic toxicity, which was readily reversible with fairly low doses of corticosteroids.

That was the major difference between the clinical trials. The reason for this we believe is lower cytokine levels that is produced when these CAR T cells are infused into the patients. The peak cytokine levels that we measure in the blood appears to be lower than those in the patients who were treated in the phase I trial of this. I'll call it the FMC63 construct, but the original CAR construct. Which again was eventually developed into the YESCARTA commercial CAR T cell product.

Specifically, granzyme A, granzyme B, interfere on gamma, IL-10, these are all proteins that in the original trial seemed to correspond with neurologic toxicity. Those levels were much lower in our clinical trial.

Fortunately, we did not see corresponding decrease in efficacy. It seemed like—again, very low numbers—but it seems like our efficacy is very similar to the original phase I trial. This shows that changing structural components of the CAR can potentially decrease your toxicity, but efficacy can remain intact.

This is a very early phase clinical trial, so it's a very low number of patients, but it shows us the idea that as we develop newer CAR T cell products, which eventually become commercial CAR T cell products, that potentially can be less toxic, and patients can require potentially less monitoring. The dream is having CAR T cell therapy be safe enough to be performed as an outpatient, or be performed at smaller centers.

We want CAR T cell therapy since it's so active to be more accessible. Right now, it's patients very often, especially with the axi-cel product, have in-patient admissions. They sometimes need to get immuno-suppressive therapy such as corticosteroids, which can lead to infections. If they have certain types of CAR T cell toxicities, they can have long durations of time in the hospital.

The real-world application is as we make versions 2.0 and 3.0 of our current CAR T cell therapy and may become used in commercial practice, it will be just much easier for our patients to receive them.

In the Kochenderfer lab they have developed a bicistronic CAR. That means it targets 2 things. It will target CD19 and CD20 and it will use this—we call it the Hu19—but the Hu19 construct for the CD19 target. Then we have a new CAR that we're using for the CD20 target.

We anticipate a trial of that opening for non-Hodgkin's lymphoma sometime in 2022.