In part 1 of this video series, Bijal Shah, MD, Assistant Member, Moffitt Cancer Center, Tampa, Florida, shares results from the phase 2 ZUMA-3 study of KTE-X19, a novel anti-CD19 chimeric antigen receptor (CAR) T-cell therapy, in relapsed/refractory B-cell acute lymphoblastic leukemia (R/R B-ALL), the data of which were presented at the 2021 American Society of Clinical Oncology (ASCO) Annual Meeting.

Transcript

Hi, my name is Bijal Shah. I'm an assistant member at the Moffitt Cancer Center in Tampa, Florida.

In 2021, we're still suffering with low survivals in ALL. When we talk about what we can anticipate, from the age 20 and up, the median survival is only around 38 percent. This is based on the most recent CR data cut.

That hasn't really budged much in the better part of 20-plus years. When you look at the outcomes in relapse B-ALL it's far worse—you can take that number and cut it in half right off the top. That's assuming you get a complete remission in first salvage, you're able to make it to the transplant.

You'll realize that potential survival, we're talking about going down to the next line of therapy where we're talking about 5 to 10 percent long-term survival at best. When we do an even deeper dive into those numbers, those benefits, those survival numbers that I'm talking about, were really only for the youngest patients.

We weren't seeing the same benefit when we started going beyond the adolescent, young adult age into the 30s and 40s and beyond. Even with blinatumomab, even with inotuzumab, there still is an unmet medical need in ALL.

This has all compounded even further by the fact that some of these patients will have already had an allotransplant as part of their original course of therapy. Whether that's chemotherapy, followed by transplant consolidation, or chemotherapy, then blinatumomab in the setting of MRD followed by transplant, or whatever it may be. We're not talking about a group of patients with this disease where we can apply a one-size-fits-all formula. It's always going to be patient-specific. It's driven by not just treatment factors, it's driven by disease-related factors.

What does a p53 mutation mean in the context of ALL in terms of success of therapy, and again, ultimately, expectations associated with therapy? That's something we're recognizing, of course, more and more.

The bottom line is we wanted to develop CAR T-cell therapy in adults because there still remains a huge unmet medical need for this patient population, especially in the relapse setting.

The ZUMA-3 Trial, which we presented at ASCO, was a study of KTE-X19 in the US, it also goes by the name of Brexu-Cel. The KTE-X19 is a little bit different than Axi-Cel, which we currently give for diffuse large B-cell lymphoma. It's more similar to the product we give for mantle cell lymphoma, the FDA-approved product. It undergoes a process of T-cell isolation first, and then once we isolate those T-cells, we do the manufacturing. The manufacturing is going to create a product that utilizes the CD28 costimulatory domain.

This is different and distinct from now the other CAR-T product that we have in the pediatric space for relapsed ALL, tisagenlecleucel. This is the one from Novartis. That product from Novartis uses a 4-1BB costimulatory domain.

The first thing that differentiates this study is it's the first multicenter study for relapsed ALL using CAR-T cell therapy in adults, but also utilizing a CAR-T cell therapy that's dependent on the CD28 costimulatory domain.

The trial anticipated high-burden disease. When we did our phase I studies, what a lot of people don't know is we required not just a blast threshold of 5 percent when the study started, we mandated that they had to have at least 25 percent blast.

In fact, most of the patients came on with marrow replacement by their leukemia when they enrolled. We studied 3 different dose cohorts, and we went through a lot of effort, not just in terms of identifying the best dose, but also in developing toxicity guidelines so that we could develop a CAR-T cell product that would be safe within the adult space.

I don't want to be overly colorful with my words, but I think that's why we succeeded with this trial, in terms of not just meeting our primary endpoint of overall response, but in terms of being able to safely deliver a CAR-T cell therapy in adults, we put a lot of work into this in the phase I. That's what ultimately allowed us to come into the phase II and to say, "We don't care what the tumor burden is, as long as it's over 5 percent blasts, no MRD." In fact, one of the patients actually had to come off, because they were enrolled with fewer than 5 percent blasts in the bone marrow.

We did require CD19 expression, and that's because, one, the CAR T targets CD19.

The second part of it is, we had allowed patients who had prior blinatumomab to come into this study. We had no way of controlling for patients who were resistant to blinatumomab by virtue of low or absent CD19.

We had to specify, one, have any therapy you want before you come on. Broadly, you're allowed to have blinatumomab. You're allowed to have inotuzumab. You're allowed to have transplant. You're allowed to have the standard armamentarium of therapies for B-ALL and come onto this study, because we're able to show, not by central lab assessment, by local laboratory assessment, that these patients remained CD19-positive.

Now, I said I would comment a little bit about what wasn't allowed.

We had seen previously that patients who received clofarabine and similar therapies, we just couldn't collect and manufacture a T-cell product. That was one regimen that I know was not permitted as a prior therapy before coming onto this study.

In terms of the absolute lymphocyte cutoff, it was low—100 was the ALC cutoff. It might have been 200, but it was in that same ballpark. With that, I will say there were some manufacturing failures. It was a small percentage—8 percent across the whole study.

It is reflective of the fact that we're coming in with patients who have, again, replacement of their bone marrow by leukemia. The majority of what's circulating in the blood is leukemia, and now, we're trying to collect T-cells and successfully manufacturing them in that setting.

In terms of other elements of the study design that I think are important, we did prescribe the bridging therapy options. Once we collect the T-cells, it was only about a 2-week turnaround to get from collection to successful manufacture. That was fast for those patients where we successfully manufactured.

The challenge, though, is they're coming in with 75 percent blasts, high white counts, whatever it may be, and we can't just watch those patients for that 2-week block of time, or at least, many of us didn't feel comfortable with that.

The bridging therapy options ranged from as simple as a pump maintenance to as intense as fludarabine and cytarabine for hyper-CVAD-based inductions. We again do outline these bridging therapy options in the phase I manuscript published in "Blood" a few months ago. Really talked broadly about intense bridging therapy options and less-intense bridging therapy options.

Whether that made a difference in terms of the ultimate clinical outcome, meaning success of getting to CAR-T cell therapy and those kinds of things, I think those are analyses that we're still doing. I would refer folks to a paper by the Memorial Sloan Kettering Group—this was also recently published in "Blood"—looking at this concept of how intensively do you bridge a patient, and what can you expect by virtue of doing so?

As it turns out, it probably does more harm than good. You probably end up with fevers and other complications that delay the cell infusion, or as we saw in the case of ZUMA-3, a cohort of patients never made it to the infusion. We enrolled 71 patients, and we ultimately were able to treat 55. The leading reason why we couldn't treat patients were infections. For the patients we treated, as I alluded to earlier, we did hit our primary endpoint. We achieved a CR and CRI rate of a little over 70 percent.

What was remarkable here is the true CR rate. That means a complete remission with hematologic recovery was over 50 percent. This is not common in multiply relapsed B-ALL, so this was a big deal.

Independent of whether there's a CR or CRI by flow, we used a central lab for flow cytometry for MRD assessment. Essentially, all were MRD-negative. There was one patient where we didn't have any information.

This translated into an amazing DoR, RFS, and overall survival. The duration of remission here for the CR/CRI patients was almost 13 months. If we focus on those patients who achieved a true CR, it was upwards of 15 months. It was quite outstanding.

For patients, almost half had already had prior allo and prior blinatumomab. Almost a quarter had had prior inotuzumab, so a heavily pretreated population. The blast burden when they came in was 65 percent.

This was what I called real ALL. This was not cosmetic. This was significant in terms of the types of patients that we enrolled to be able to see these kinds of results. The median RFS was around 12 months for the whole group.

I have to be clear. When we say RFS in this study, it's going to be more similar to how people think of event-free survival. Meaning, in this context, we looked at the remission, essentially, relapse-free survival independent of response.

We saw that, going back to the DoR, when we censored for transplant, subsequent therapies, the outcome was the same. We didn't see any meaningful change. Around 30 percent of the patients were still in ongoing CR at the date of cutoff who hadn't had a transplant.

That's also remarkable, because I think one of the questions which we're still going to be dealing with for a while, is what is the role of consolidation transplant after CAR-T cell immunotherapy?

In terms of the median overall survival, it was 18.2 months, and it had not yet been reached in those who were complete responders.vThis is incredible, because our expectations in this same population are certainly much lower, where we anticipate the long-term survival in multiply relapsed ALL around 5 to 10 percent.

I hinted at the safety a little bit. We did see CRS and neurotoxicity that was grade 3 and above in about a quarter of the patients. If you look back to our phase I experience, it was a little bit higher, and that was because we were still learning.

In fact, we talk a lot about in the manuscript how we intervened with tocilizumab and steroids of lower grades of CRS and limited the use of tocilizumab to CRS, rather than using it in the context of neurotoxicity. I think that helped out tremendously in terms of improving the safety profile.

There was one person who did die from cerebral herniation due to brain edema. I'll be honest. I still am struck by this case. It was devastating. I'm also struck at how rare it generally occurs. I was in an advisory board meeting just a few days ago, and I said, "We've treated over 500 patients with CAR-T cell immunotherapy at our cancer center. I don't think we've seen a single case."

Trying to understand these very, very rare toxicities is extraordinarily challenging. I'm never going to be able to say, "This is how we prevent a similar case like this from happening in the future." All I can say is it seems to happen during that cytokine release phase, meaning the earlier inflammatory phase, where the T-cells are robustly growing and doing their thing.

Somehow, in that context, we're seeing these cases emerge, but it is something we'll have to pay close attention to, and as more cases emerge, hopefully develop some guidance to both manage and ultimately prevent it.