Leo I. Gordon
In the late 1990s, Eshhar and colleagues were credited with reengineering autologous T cells to function more effectively at targeting and killing tumor cells. Over the past two decades, this concept has materialized with the development of various second-generation chimeric antigen receptor T-cell products comprising a single-chain variant fragment linked by a CD8-alpha hinge/transmembrane sequence to an intracellular co-stimulatory domain, (either CD28 or 4-1BB), and a CD3-zeta signaling domain.
In the last 2 years, two such constructs have been approved by the FDA for patients with relapsed or refractory diffuse large B-cell lymphoma that have failed two lines of therapy specifically CD28 construct axicabatagene ciloleucel (Yescarta; Kite/Gilead) and 4-1BB construct tisagenlecleucel (Kymriah; Novartis). These constructs have shown great promise with long-term follow-up of over 18 months demonstrating durable responses in approximately 40% of patients.
Despite therapeutic efficacy of CAR T-cell therapies, associated toxicities including cytokine release syndrome (CRS) and neurotoxicity are major concerns and pose barriers to broad therapeutic application. Notably, CRS primarily driven by IL-6 has been reported for axicabatagene ciloleucel and tisagenlecleucel in pivotal trials at rates of 93% and 58%, respectively.
Alternatively, the underlying mechanism for neurotoxicity remains poorly understood, although IL-1 may be responsible; neurotoxicity rates reported for these constructs were 64% for axicabatagene ciloleucel and 21% for tisagenlecleucel. Differences in toxicity profiles for these constructs have been attributed in part to differences in their co-stimulatory molecules, although disease-related factors also likely compound the problem.
Ying and colleagues have generated a new anti-CD19 CAR molecule, representing a modification of the tisagenlecleucel prototype with an alteration in the CD8-alpha hinge/transmembrane sequence, and they suggest that this construct delivers therapeutic effect while mitigating toxicity. The researchers reported on their preclinical data and phase 1 clinical trial of CD19-BBz(86) CAR T-cell therapy in patients with relapsed B-cell lymphoma.
Twenty-six patients were enrolled in the trial and 25 were treated: 7 patients had follicular lymphoma (grade 2 to 3a), 1 patient had transformed follicular lymphoma and 17 patients had DLBCL or high-grade B-cell lymphoma. Twenty-four patients had advanced-stage disease.
Patients received lympho-depleting chemotherapy (fludarabine 25 mg/m2 and cyclophosphamide 250 mg/m2 on days -3, -2 and -1) followed by CD19-BBz(86) CAR T cells on day 0 administered as a split dose over 3 days (30% on day 0, 30% on day 1 and 40% on day 2).
Patients were monitored in the hospital only during this time; the rest of follow-up was conducted in the outpatient setting, with weekly visits for the first month and examinations every 2 to 3 months thereafter. The overall response rate was 60% (15 of 25 patients) and CR occurred in 28% (7/25) of patients; however, for 11 patients treated at a higher dose of 2 X 108 or greater, 54.5% (6 of 11 patients) achieved a CR. The median duration of response for these 6 patients who achieved a CR was more than 181 days.
Seven patients (28%) incurred grade 1 CRS and there were no cases of neurotoxicity. Interestingly, in contrast to previous reports with other constructs, serum cytokines measured in 11 patients including IL-6, TNF-alpha, IFN-gamma and CRP remained at basal or low levels, with the exception of higher serum levels of granzyme A and B found in responders compared with patients with progressive disease. Ying and colleagues did demonstrate CD19-BBz(86) CAR T-cell proliferation and differentiation into memory cells in vivo, with persistence identified for more than 300 days in one patient.
Based on these findings, it is difficult to discern how the CD19-BBz(86) CAR T-cell construct measures up to tisagenlecleucel as a prototype and other second-generation constructs. The sample size for this trial was small, with varying doses of CAR T cells administered, and it included a heterogeneous group of patients with relatively short follow-up. Higher CR rates of 71% and 40% have been reported for patients with relapsed or refractory follicular lymphoma and DLBCL treated with tisagenlecleucel, respectively, with median durability of response not reached in the DLBCL group with 19.3 months of follow-up. By comparison, the efficacy of CD19-BBz(86) CAR T cells seemingly falls short.
Rates of toxicity for CD19-BBz(86) CAR T cells do appear to compare favorably with those reported for tisagenlecleucel and axicabatagene ciloleucel. However, this could merely be a function of split dosing, the inclusion of younger patients and those with follicular lymphoma, as well as poor capture of true toxicity rates with low-intensity follow-up after CAR T-cell infusion. Additionally, it is unclear whether the CD19-BBz(86) CAR T-cell construct offers any safety gains over lisocabtagene maraleucel, yet another 4-1BB construct that is close to FDA approval.
To the strength of this publication, Ying and colleagues did include preclinical data from a mouse model with measures of cytokines post-tisagenlecleucel prototype and CD19-BBz(86) CAR T-cell infusion; cytokine levels were demonstrably lower with the latter. Collectively with the demonstration of basal or low levels of cytokines in patients treated with CD19-BBz(86) CAR T cells, these findings support the claim for less toxicity (believed to be cytokine driven) with the novel construct. As such, this publication does provide some guidance on potential strategies for toxicity mitigation. Ultimately, future efforts need to continue focusing on enhancing both the efficacy and safety of CAR T cells to ensure measurable gains for patients.
Abramson JS, et al. J Clin Oncol. 2018;doi:10.1200/JCO.2018.36.15_suppl.7505.
Eshhar Z, et al. Proc Natl Acad Sci U S A. 1993;doi:10.1073/pnas.90.2.720.
Locke FL, et al. Lancet Oncol. 2019;doi:10.1016/S1470-2045(18)30864-7.
Norelli M, et al. Nat Med. 2018;doi:10.1038/s41591-018-0036-4.
Schuster SJ, et al. N Engl J Med. 2019;doi:10.1056/NEJMoa1804980.
Schuster SJ, et al. N Engl J Med. 2017;doi:10.1056/NEJMoa1708566.
Reem Karmali, MD, MS, and Leo I. Gordon, MD, FACP
Robert H. Lurie Comprehensive Cancer Center; Northwestern University Feinberg School of Medicine
Disclosures: Gordon reports patent-pending nanoparticles in cancer treatment; advisory boards for Bayer, Celgene/Juno and Gilead; and research funding from Celgene/Juno. Karmali reports being on the speakers bureau for AstraZeneca and Gilead/Kite; advisory boards for Celgene/Juno, Gilead/Kite and Seattle Genetics; and research funding from Bristol-Myers Squibb, Celgene/Juno, Gilead/Kite and Takeda.