Update Sustains Liso-Cel as Strong Contender in CAR T-Cell Space

Jeremy S. Abramson, MD, MMSc

The CD19-directed chimeric antigen receptor (CAR) T-cell therapy lisocabtagene maraleucel (JCAR017; liso-cel) is showing promising response rates in patients with high-risk diffuse large B-cell lymphoma (DLBCL).

Updated findings from the TRANSCEND trial presented at the 2018 ASCO Annual Meeting showed a complete response (CR) rate of 46% (95% CI, 30%-63%) at 6 months, as well as an ongoing objective response rate of 49% (95% CI, 32%-66%). This phase I, multicenter trial also showed durable responses across poor-risk DLBCL subgroups, including those who are chemorefractory.

The median overall survival (OS) had not been reached among patients who achieved a CR at the pivotal dose. The 12-month OS rate in these patients was 89% (95% CI, 72%-96%). Among patients who achieved a partial response, the median OS was 10.3 months (95% CI, 6.8-not evaluable) and the 1-year OS rate was 33% (95% CI, 9%-60%).

OncLive: What is the background of the TRANSCEND trial?

In an interview with OncLive during ASCO, lead investigator Jeremy S. Abramson, MD, MMSc, clinical director, Center for Lymphoma, Massachusetts General Hospital, discussed the latest data for liso-cel and its potential to become the third FDA-approved CD19-targeted CAR T-cell therapy.Abramson: The TRANSCEND trial is a pivotal trial looking at the anti-CD19 CAR T-cell product JCAR017, now known as lisocabtagene maraleucel, in relapsed/refractory aggressive B-cell lymphomas. Liso-cel targets CD19 using a 4-1BB costimulatory domain, and signals CD3 zeta. We separated the CD4 and CD8 cells, then they were separately transduced and expanded, and then administered back to patients in a 1:1 ratio. What that results in is a fixed precise dose of CD4 and CD8 CAR-positive T cells administered to every patient. We began with a dose-finding component of the study, where we looked at 3 dose levels. [We evaluated] a single-dose level of 50 million cells administered at a single dose, a double-dose level of the same dose of 50 million cells administered 2 weeks apart, and then a dose level of 100 million cells as a single dose.

What are the updated findings with this study?

Based on the findings in that component of the study, we expanded into 2 dose-expansion cohorts, which was both dose level 1 and dose level 2, each administered in a single dose. Based on the results of that, we have expanded into a pivotal cohort of the trial.At the 2018 ASCO Annual Meeting, I updated the results from the dose-finding and dose-escalation cohorts—the first 2 phases of the study. We have 102 patients who have been treated with liso-cel. All patients have now reached the 6-month time point or beyond, so we are really dealing with maturing data at this point. We are finding really exciting results. All patients have had at least 2 prior lines of therapy. The median prior line is 3, and some patients received up to 8. These are high-risk patients—70% of them have chemorefractory disease, 20% have double-hit lymphoma, about half of patients have never achieved a CR to prior therapies, and 40% have had a prior stem cell transplant, most of which were autologous. This compared favorably to any other conventional therapy that one would give these patients.

The most important question is durability of remission in large cell lymphoma. When you look 6 months later, we are seeing a high proportion of durable remissions. In fact, 40% are in ongoing response 6 months later.

We then evaluated our core DLBCL population, which is a population that we are now accruing in the pivotal component of the study. Those patients have relapsed/refractory DLBCL, transformed follicular lymphoma, or double-hit or triple-hit lymphoma. It is a narrower set of patients in that cohort. In that subset of patients, the median number of prior therapies is 3 and 70% are chemorefractory. In that population of patients, treated at our second dose level, about half of patients are in ongoing remission 6 months later. We are seeing a very encouraging signal of durable responses in high-risk DLBCL.

Of course, the other big question when it comes to a CAR T-cell product is toxicity. As we know, the adverse events of special interest are cytokine release syndrome (CRS) and neurologic toxicity. We are seeing low and manageable rates of both of these toxicities. CRS has only been seen in 37% of patients on the study, and only a single patient had severe CRS. Neurologic toxicity has been similarly low at 23%, with 13% of patients having severe CRS. Virtually all of these toxicities have been reversible, and there have been no deaths related to either CRS or neurologic toxicity. We have seen relatively low overall use of rescue medications—anti-cytokine therapy with tocilizumab (Actemra) was seen in 17% of subjects, predominantly for grade 2 CRS. Corticosteroids, primarily with dexamethasone, were used in 21% of patients.

What are the next steps?

Based on these data, we are seeing that with maturing results, there are encouraging durable responses and manageable low rates of toxicity in this high-risk population of DLBCL. The major next step for our study is completing the evaluation of the pivotal cohort. The pivotal population in this relapsed/refractory DLBCL has actually completed accrual. We are eagerly awaiting those data in the not-too-distant future. We hope that these data will reflect what we have already seen already in this ongoing trial, and warrant bringing this drug to the marketplace.

Do you foresee liso-cel being moved up to earlier lines of therapy?

The next steps beyond that are bringing it to a broader population of patients with relapsed/refractory DLBCL. We are getting ready for the TRANSFORM trial. This is a randomized trial, which is going to compare liso-cel with a standard relapsed/refractory approach of salvage chemotherapy and autologous stem cell transplant at first relapse in DLBCL. These are specifically going to target high-risk patients at the time of their initial relapse, and patients who relapsed within 1 year of their initial R-CHOP. Those patients have a fairly low chance of durable response to the conventional therapy and stem cell transplant. Absolutely. The next likely move forward is second-line therapy right after first relapse. That means, for the younger, fitter, transplant-eligible patients, replacing transplants via this [method] and other randomized trials looking to accomplish this with similar CAR T-cell products in this space.

Will we get to a point where physicians have to choose between multiple CAR T-cell products? Would patients be able to move from one to another?

The other question is, “What about those patients who aren't candidates for an autologous transplant—therefore not candidates for those trials?” As we know, most patients with DLBCL are older, with a median age of late 60s. These are often patients who are ill with medical comorbities. If we look at the real-world population of DLBCL at relapse, many of these patients won't be candidates for transplant. Those are patients who we desperately need better therapies for. Thinking about first relapse after initial R-CHOP therapy for nontransplant-eligible candidates is absolutely part of the next steps, as well. Ultimately, I personally predict that CAR T cells will move into the second-line setting for nearly all patients with relapsed/refractory DLBCL in the future. These are important questions. We currently have 2 FDA-approved products for relapsed/refractory DLBCL in the third-line setting—axicabtagene ciloleucel (axi-cel; Yescarta) and tisagenlecleucel (Kymriah). If liso-cel gets approved, we will have 3 anti-CD19 CAR T cells in the same indication. How does one choose between them? Evolving maturity of the data with the 3 products are going to help guide that decision. The number one most important thing is to cure more patients. The evolving efficacy is going to be vital.

Axi-cel has reported a 42% ongoing response rate at 1 year. Both liso-cel and tisagenlecleucel have only reported data out to 6 months, thus far. We need to see mature data at the 1-year mark and beyond, because the benchmark has been set by axi-cel.

The next important question is going to be toxicity. There do appear to be differences across these products in terms of the toxicity rates and severity that they induce. Axi-cel is an anti-CD28 costimulated CAR T-cell product that is different from both tisagenlecleucel and liso-cel, which use 41bb. These lead to very different kinetics in terms of their expansion within the patient. Axi-cel produces an earlier, more rapid and robust expansion of these CAR T cells in the patient, and that leads to an earlier and more robust CRS and neurologic toxicity, as well. They have reported a 93% incidence of CRS, for example. Tisagenlecleucel has reported a lower but still substantial rate of CRS in about two-thirds of patients. We [with liso-cel] have now reported a rate of 37%.

Is there anything else you would like to add?

There appear to be differences across these products. If we have multiple products that look similarly effective in this high-risk population, we are going to be choosing them based on the toxicities. That will also allow us to expand the number of patients we are giving this to. It might be that a less toxic product can be given to a frailer patient, where a more toxic product would be potentially riskier or more dangerous. That is where the landscape is going to unfold, but we need maturing data before we make that call.This is an unprecedented time where we have multiple anti-CD19 CAR T-cell products available in a truly unmet medical need. This is transformative therapy that is clearly saving lives, where we previously had no life-saving therapy. A rising tide lifts all boats, and having multiple of these products being evaluated is helping patients. This will only expand as we go forward.

Abramson JS, Gordon LI, Palomba ML, et al. Updated safety and long term clinical outcomes in TRANSCEND NHL 001, pivotal trial of lisocabtagene maraleucel (JCAR017) in R/R aggressive NHL. J Clin Oncol. 2018;36 (suppl; abstr 7505).