bb2121 Continues to Show Potential in Multiple Myeloma

Noopur Raje, MD

Results from the multicenter phase I CRB-401 study evaluating the anti-BCMA chimeric antigen receptor T-cell therapy bb2121 in patients with relapsed/refractory multiple myeloma are highly encouraging, according to Noopur Raje, MD.

“As a proof-of-principle, first-in-human study, we have an active target—an active cellular therapy—with very little toxicity and pretty durable responses, which has been great,” said Raje, who is the lead study author of CRB-401.

In the trial, which accrued heavily pretreated patients with relapsed/refractory disease, bb2121 induced a median progression-free survival (PFS) of 11.8 months with a median duration of response of 10.8 months, according to the updated findings that were presented at the 2018 ASCO Annual Meeting.

The median follow-up was 194 days for patients treated with a >150 x 10⁶ dose at the March 29, 2018 data cutoff. Results showed that the objective response rate (ORR) was 95.5%. Additionally, the complete response (CR) or stringent CR rate was 50% and 36.4% had a very good partial response. In contrast, patients treated with an inactive dose (50 x 10⁶) had an ORR of 33.3% and a 1.9-month median duration of response.

Raje, director of the Center for Multiple Myeloma at Massachusetts General Hospital Cancer Center, also explained that bb2121 was found to be well tolerated. The rate of grade ≥3 cytokine release syndrome (CRS) was 5% and the rate of grade ≥3 neurotoxicity was 2%.

In November 2017, the FDA granted bb2121 a breakthrough therapy designation for previously treated patients with relapsed/refractory disease.

OncLive: What was the rationale for CRB-401?

In an interview with OncLive, Raje discussed the updated results from the CRB-401 trial and next steps with bb2121 in multiple myeloma.Raje: CAR T-cell therapy is clearly an exciting strategy not just in hematologic cancers, but a lot of solid tumors, where it is also being investigated. What we've done with the bb2121 product is use a CAR using a lentivirus vector, which recognizes a humanized BCMA. Along with that, for the costimulatory domain, we used 4-1BB. What we think is happening here is that, with the 4-1BB, we get a slow and consistent expansion of those T cells. Part of the reason for seeing less toxicity, we believe, is because of the 4-1BB costimulatory molecule.

Preclinically, we have studied BCMA in myeloma for a long time now. Most of us agree that BCMA is a good target in the treatment of [patients with] myeloma, and for us to be able to create and generate this CAR has been an exciting development going to a very targeted approach.

There were a couple of concerns prior to going into this study. Patients with myeloma can have circulating levels of BCMA and high BCMA levels. The question was, “Is that going to impact a BCMA-targeted strategy?” Preclinical work, at least, did not suggest that.

We also wanted to figure out whether BCMA-receptor density impacted whether a targeted agent such as a CAR T-cell therapy would work or not. Again, our preclinical data did not suggest that patients had to have a certain level of BCMA expression.

Having done all of this preclinical work, we then went forward and studied this product in patients. We did a very classic dose-escalation study. We started with 50 million cells and went all the way to 800 million cells. It was generally required that we collect patients’ lymphocytes— that's the leukapheresis process. It takes a few weeks to generate the CARs.

In that time frame from us collecting, and us giving back the leukapheresis or CAR T-cell product, we actually were allowed to give our patients bridging therapy. These are sick patients who need active myeloma therapy. The majority our patients were on some form of bridging therapy.

They then come into the hospital to get lymphodepletion, which can be either in-hospital or as an outpatient. It's very simple chemotherapy. I'd like to highlight that because there is a lot of confusion between this kind of an approach and autologous [stem cell] transplant, and they are very different.

The lymphodepletion chemotherapy we use is cyclophosphamide and fludarabine. These are at low doses and the treatment can happen on an outpatient basis. It's done over 3 days; then patients have a 2-day rest period, and we give them back this CAR T-cell product.

What is the safety profile with bb2121?

The CAR T-cell product is infused over 5 to 10 minutes, but that's the time when we actually keep people in hospital because 1 of the few things we worry about with CAR T cells, if they're really active, is certain toxicities. The 2 big toxicities which have been noted with cellular therapy have been neurotoxicity and CRS.What we have found with our bb2121 experience was that that the CRS signal was there in about 60% of patients; however, the CRS signal was fairly low at grade 1/2. It was very manageable. There were a couple of patients who had grade 3 CRS and not a lot of patients required the antidote to CRS, which is tocilizumab (Actemra). In general, we were very impressed by the safety signal.

As far as neurotoxicity is concerned, we were actually pleasantly surprised by the low signal. We did have 1 patient who had grade 4 neurotoxicity who we presented in great detail at the 2018 ASH Annual Meeting. That patient, fortunately for us, has recovered completely.

Going into this whole program, obviously, our concern when we take such sick patients was, is whether our strategy will be tolerated by patients in their 50s, 60s, and 70s in multiple myeloma when this therapy was typically designed for acute leukemia. To our surprise, it's been very well tolerated. We have had patients in their 70s on this treatment and the neurotoxicity signal was less than 30%. Again, this was something very reassuring; it was a good safety signal. There are other things that we need to keep in mind. Patients can get low [blood cell] counts, some of them because of the T-cell product, the lymphodepletion chemotherapy, and just the fact that they've had so much prior therapy for their myeloma.

The last thing, which is a good problem for all of us to have, is tumor lysis syndrome. We actually did not see that.

Can you discuss the efficacy findings?

Having said that, the 2018 ASCO Annual Meeting is the first time we've presented all 43 patients. What most patients have been used to hearing about is the 22 patients that we presented at the 2017 ASCO Annual Meeting. Now we have an experience on 43 patients; we have a longer follow-up, as well, going close to 1 year now. That is the longest follow-up of a CAR T-cell product, even compared with some of the lymphoma data.What we found was a 90%-plus response rate in the majority [of patients]. The other thing we did do out here was include patients with low BCMA expression. In our expansion cohort, patients with low BCMA expression were included and we, again, saw a 90%-plus response rate in that cohort, suggesting to us that it didn't matter whether you had high BCMA levels or low BCMA levels. You would still respond.

For the first time, we actually showed the PFS in all patients. It was 11.8 months. In this relapsed/refractory space, after an average of 8 or 9 lines of therapy, we have never seen this kind of data.

We honed in and looked at patients who were minimal residual disease—negative in their bone marrow and the PFS was 17.8 months. Again, this is an unprecedented PFS in a very sick patient population.

In general, we've been encouraged by this data. One, we were thrilled that it was safe. Two, what was quite striking was we took such refractory patients, took their own lymphocytes, and we were able to generate active T cells. To me, that speaks a lot because what can we achieve if we go a little earlier in treatment?

Is there a rationale to explore combination strategies to address relapse?

There are a lot of questions that come up with this strategy, obviously. We are seeing patients relapsing. The questions around that are, “Why are they relapsing? What is the mechanism of relapse? What is their T-cell repertoire?” All of those questions are ones we are trying to understand and address as we speak.Before we get into combination strategies, we need to understand why patients are relapsing. It's not a big surprise that they're relapsing because, again, these are end-stage patients. What we do with this is going to depend on what the cause of relapse is. Is it just that these are genetically poor-risk patients and the tumor cell is what is causing problems? In that case, my answer is going to be, “Don't wait until they get to their eighth relapse.” Do this maybe in their first or second relapse and you might see a better outcome.

Is there anything else about the results that surprised you?

We looked at the data for T-cell persistence and what we were able to see is, for up to 6 months, we were comfortably able to find CAR T cells. Therefore, the question would be, do you need to have circulating CARs to keep that disease under control? If that's the answer, then the question is not about combinations. The question is, “Can we reinfuse CARs? A lot of what we do going forward would be based on the understanding of the mechanisms of this resistance.”A couple of things that took me by surprise is how quickly these patients cleared their bone marrow. These are bone marrows that are 100% involved. Perhaps we don't do this in other studies, but what this study mandated was us doing a bone marrow on day 14 and at month 1. To see complete clearing of bone marrow there, was not something I was not expecting. Typically, when we use relapsed/refractory patients, we've seen responses happen slowly because they have very aggressive disease that is very resistant to treatment.

The other thing that I was pleasantly surprised by is that we weren't sure that when we went into a space like this that we would be able to generate active CARs, specifically in a myeloma patient population. This is an immunologically mediated disease where we're using a lot of drugs that kill all of the T cells and lymphocytes. The one thing that was quite encouraging was the fact that, on this trial, there was no eligibility for absolute lymphocyte count. Therefore, there were plenty of patients with low lymphocyte counts, and yet we were able to generate active CARs in 100% of patients. All of those were good surprises.

What is next with bb2121?

We would have liked [the response] to be a lot more durable, but that's something for us to figure out.The phase II study is already ongoing. It's going to be a very similar patient population so it's not as if we're going to move it into an earlier phase. If anything, the phase II study is going to have sicker patients. When we started this first-in-human study, it was a couple of years back and in the first dose-escalation arm, we didn't have a lot of daratumumab-exposed patients. In the expansion cohort, we made sure that everybody was daratumumab-exposed and we saw the same kind of response. In the phase II study one of the inclusion criteria is daratumumab exposure.

The concern for all of us has been, how can we scale up this technology? Given that we're doing this phase II trial—it's not at a single institution, it's multisite [trial]—and not only in the United States; it's also accruing in Europe. For us to be able to get to that level in myeloma is quite great.

Raje NS, Berdega JG, Lin BY, et al. bb2121 anti-BCMA CAR T-cell therapy in patients with relapsed/refractory multiple myeloma: updated results from a multicenter phase I study. J Clin Oncol. 2018;36 (suppl; abstr 8007).