Quick progress was seen with a novel class of agents, chimeric antigen receptor T-cell therapies, in the setting of diffuse large B-cell lymphoma—a swift jump from early phase clinical trials to FDA-approved products.
Anas Younes, MD
Quick progress was seen with a novel class of agents, chimeric antigen receptor (CAR) T-cell therapies, in the setting of diffuse large B-cell lymphoma (DLBCL) a swift jump from early phase clinical trials to FDA-approved products.
The speedy arrival of CAR T-cell therapies for DLBCL was preceded by several setbacks in clinical trials, which were attempting to improve on standard R-CHOP (rituximab [Rituxan] with cyclophosphamide, doxorubicin, vincristine, and prednisone) therapy with added agents or changes in dose intensity, said Anas Younes, MD, during a presentation at the 2nd Annual Live Medical Crossfire®: Hematologic Malignancies.
The unmet need in relapsed/refractory DLBCL was further enhanced by an announcement on July 11, 2018, that the phase III PHOENIX trial combining ibrutinib (Imbruvica) with R-CHOP failed to meet its primary endpoint of improvement in event-free survival, despite very promising phase II findings. Data from the PHOENIX trial will be presented at the 2018 American Society of Hematology Annual Meeting, Younes said.
Adding to these unsuccessful strategies, findings from the CORAL and SCHOLAR-1 trials further illuminated the high unmet need for patients with refractory DLBCL.1,2 In each of these studies, the long-term overall survival (OS) rate was just 15% to 20% for patients relapsing within 12 months of stem cell transplant or with refractory disease.
“The past several years of research had not taken us anywhere, and then the CAR T cells showed up,” said Younes, chief of the Lymphoma Service at Memorial Sloan Kettering Cancer Center and program chair of the crossfire meeting. “Now we have 2 FDA-approved products, with a third one on the way. These are not identical products, but there are similarities between them.”
Sattva S. Neelapu, MD
Discussion during the meeting explored the advancements with CAR T-cell therapies for patients with DLBCL and ongoing explorations of the impact of the 2 approvals in this setting.
CAR T-Cell Therapies Emerge
In October 2017, axicabtagene ciloleucel (axi-cel; Yescarta) became the first CAR T-cell therapy approved for patients with relapsed or refractory non-Hodgkin lymphoma, based on findings from the phase II ZUMA-1 study.3The best objective response rate (ORR) achieved with the therapy was 82%, and the best complete remission (CR) rate was 54%. With a follow-up of 12 months, the durable ORR was 42%, and the durable CR rate was 40% (TABLE).
Tisagenlecleucel (Kymriah) joined axi-cel in May 2018, with an approval for large B-cell lymphoma that was based on the phase II JULIET study. In an update from this study,4the best ORR was 52%, and the best CR rate was 40%. Based on the presentation at the 2018 European Hematology Association Annual Meeting, Younes postulated that the durable ORR at 12 months was likely 34%, and the durable CR rate was 29% (TABLE).
Another CAR T-cell therapy in development, lisocabtagene maraleucel (liso-cel; JCAR017), has also shown promise in a phase II study, which will likely lead to FDA approval, Younes noted. In the phase II TRANSCEND study at the dose level being explored for FDA submission,5the best ORR was 80%, and the best CR rate was 59% among patients in the core population with DLBCL. At 6 months, the durable ORR was 47%, and the durable CR rate was 41% (TABLE).
“This has been a paradigm shift. These therapies are definitely going to be moving forward to earlier lines of therapy,” said discussant Sattva S. Neelapu, MD, the director of laboratory and translational research in the Department of Lymphoma/Myeloma at The University of Texas MD Anderson Cancer Center. “Besides DLBCL, I think these therapies are also going to be approved for other B-cell malignancies, like mantle cell lymphoma.”
David G. Maloney, MD, PhD
The median progression-free survival (PFS) with axi-cel was 5.8 months, and the 12-month OS rate was 59%. The JULIET study showed a median PFS of 2.9 months for tisagenlecleucel and a 12-month OS rate of 49%. These data points were not yet available for liso-cel.
“When you look at the data, you cannot compare them head-to-head, because there are differences in the eligibility criteria in terms of refractory and relapsed and so forth, but it is good to know the data,” Younes said. “The median PFS with these agents is not high, up to about 5 months, but there’s a tail on the curve. Even though median PFS is not very long, there [are] a handful of patients who are doing very well and are probably curable.”
The most concerning adverse events (AEs) with the CAR T-cell therapies were cytokine release syndrome (CRS) and neurotoxicity (NT). For axi-cel, grade ≥3 CRS occurred in 13% of patients, and grade ≥3 NT was seen in 31% of patients, with steroids and antiinterleukin-6 therapies needed for 29% and 45% of patients, respectively. AEs leading to death occurred in 3 patients.
For those treated with tisagenlecleucel, the rates of grade ≥3 CRS and NT were 22% and 12%, with 11% and 15% requiring tocilizumab or steroids. For liso-cel, grade ≥3 CRS was experienced by just 1% of patients, and 13% had grade ≥3 NT. Steroids were required for 21% of patients, and 15% received tocilizumab.
In addition, different AE grading systems were used across the studies. The Lee criteria were used for CRS grading in ZUMA-1 and TRANSCEND, whereas a University of Pennsylvania scaling model was used in the JULIET study. All trials used the Common Terminology Criteria for Adverse Events for NT grading.
“These are really exciting times. Anyone who has seen a CAR T-cell patient respond with their tumor melting away in less than 4 weeks can see that this is truly astounding,” said discussant David Maloney, MD, PhD. “We’re learning a lot, in terms of CRS and neurotoxicity; we know that some patients are going to have substantial toxicity.”
With the knowledge gained in the refractory setting, Maloney, Neelapu, and Stephen J. Schuster, MD, who also discussed the findings, all agreed that CAR T-cell therapy would likely move toward the frontline setting.
“Any new therapy is going to start in refractory patients. [Although] we can get responses in completely refractory patients, it will likely be better if you can do this at an earlier time,” said Maloney, medical director of cellular immunotherapy at Fred Hutchinson Cancer Research Center and a professor of medicine at the University of Washington. “We’re going to see these start moving to a place where we think about them before people have failed several lines of salvage.”
Which CAR Is Best?
Although the agents represent a significant advance, many questions remain regarding their use. Moreover, in the absence of head-to-head trials, treatment selection is a challenge. Each of the studies had variations in patient populations, and the construct of each agent is unique, making cross-study comparisons a challenge. Despite these obstacles, some characteristics have emerged to help guide therapy, the experts noted.
The typical criteria for deciding between effective treatment options is to look first at the efficacy, Neelapu noted; if efficacy is equal, an analysis of the toxicity and logistics comes next. The presenters noted that the longest follow-up currently available was for axi-cel, providing the highest certainty for its clinical efficacy; however, in their experience, axi-cel also seemed to elicit the highest rate of AEs.
Patient characteristics in each trial could also help determine which agent to use in a case-by-case scenario, Maloney said. In the JULIET study, some patients had received prior allogeneic stem cell transplant prior to tisagenlecleucel, suggesting that this agent works in this setting. Additionally, JULIET included children and those with central nervous system (CNS) involvement, suggesting it may be ideal in this population.
“Having these data could help with reimbursement,” Maloney said. “If you have prior allogeneic [transplant] and CNS involvement, my selection goes toward tisagenlecleucel, right now.”
In cases where a distinct advantage is not apparent, it may boil down to familiarity and comfort with the available data.
Costs and Patient Selection
In an informal poll of the audience, which consisted of approximately 75 healthcare providers, few individuals said they had already used or referred a patient to receive CAR T-cell therapy. Questions tended to center on predictors of response given the high costs associated with each therapy.
Although early predictors of response have been identified, there is not 1 clear winner, Neelapu said. Responses have been noted across most patient subgroups. Maloney noted that in some early studies with JCAR014, pre-lymphodepletion serum lactate dehydrogenase was associated with better response and outcomes. Tumor burden can be used to predict responses, he added.
In long-term findings for 19-28z CAR T cells,6 patients with low disease burden (<5% bone marrow blasts) had superior OS and event-free survival compared with patients with high disease burden. The median OS was 20.1 months in the low disease burden group compared with 12.4 months for those with high burden (P= .02). Additionally, patients with high disease burden had more AEs compared with those in the low group.
Another remaining obstacle: the price tag exceeds $373,000. At this point, CMS has decided to reimburse a set rate for CAR T-cell infusion, which does not cover the total costs involved in the procedure. CAR T-cell infusions require hospitalization and other associated costs, which often bring the grand total for these therapies to nearly $1 million.
Along these lines, Maloney noted that he has treated all patients in the outpatient setting when using liso-cel, given its comparatively mild AE profile. This approach avoids the added costs of a 2-week hospital stay. Schuster, director of the Lymphoma Program at the University of Pennsylvania, said that he has also started using tisagenlecleucel in the outpatient setting, without major issues.
Both tisagenlecleucel and liso-cel use a 4-1BB costimulatory domain, whereas axi-cel has a CD28 domain. This differentiator could be key in the ability to treat in the outpatient setting, the panelists suggested. Moreover, liso-cel potentially holds further safety advantages via a fixed 1:1 ratio of CD4:CD8 cells. Maloney felt this added feature helped make the safety profile more predictable.
Although not yet in regulatory review, cohorts from the TRANSCEND study will be submitted to the FDA once mature. Several factors, especially the specific FDA label and outpatient treatment potential, will dictate how this agent is used. Additionally, the list price for the T-cell therapy along with CMS reimbursement will play a major role.