A durable complete response was achieved in a high proportion of adults with refractory B-cell malignancies who received CD19+ chimeric antigen receptor (CAR) T-cells made up of a defined 1:1 ratio of CD8+ and CD4+ cells.
The data were presented at ASCO 2016 by Cameron John Turtle, MBBS, PhD, FRACP, FRCPA, Assistant Member, Clinical Research Division, Fred Hutchinson Cancer Research Center, Seattle, WA.
“Preclinical studies had established that a defined composition of CD8 T central memory–derived CAR T-cells and CD4+-derived T-cells provides optimal potency in xenogeneic models,” said Dr Turtle.
In this phase 1/2 study, 90 patients with refractory or resistant acute lymphocytic leukemia (ALL), non-Hodgkin lymphoma (NHL), or chronic lymphocytic leukemia (CLL) underwent leukapheresis to produce autologous CAR T-cells. Lymphodepleting chemotherapy was followed 2 to 4 days later by infusion with 1 of 3 CAR T-cell dose levels (2 × 105 cells/kg, 2 × 106 cells/kg, or 2 × 107 cells/kg).
The highest complete response rate was achieved in patients with ALL. A total of 36 patients with ALL received treatment, and allogeneic stem-cell transplant had failed in 33%. Of the 34 who completed a response assessment, 32 (94%) had a complete response by bone marrow flow cytometry, and 2 had minimal residual disease. In vivo CAR T-cell expansion and toxicity correlated with bone marrow blast count and the infused dose of defined composition CAR T-cells.
CAR T-cell expansion and persistence were dramatically increased in a subset of patients who underwent lymphodepletion with cyclophosphamide and fludarabine compared with those who did not receive fludarabine. “In addition to improving the CAR T-cell kinetics in vivo, we also saw a marked improvement in disease-free survival in the patients who receive cyclophosphamide/fludarabine depletion,” said Dr Turtle. At 18 months, a disease-free survival rate of approximately 60% was observed in the patients who received optimized lymphodepletion compared with <10% of those who received cyclophosphamide only.
Overall, 41 patients with NHL received treatment in the study, 66% of whom had received ≥4 previous lines of therapy.
The objective response rate was 80% among patients with NHL who received 2 × 105 cells/kg of CAR T-cell infusion with cyclophosphamide/fludarabine lymphodepletion (16 of 20 patients), 50% of whom achieved a complete response. With this preferred regimen, the median progression-free survival (PFS) and overall survival for the patients who achieved a complete response had not yet been reached.
A total of 13 patients with CLL received treatment, most of whom had high-risk disease, defined by deletion 17p and/or complex karyotype. All patients had received previous therapy with ibrutinib therapy: 54% had ibrutinib-refractory disease and 15% were intolerant of ibrutinib. The objective response rate was 91% in patients with cyclophosphamide/fludarabine depletion, and 45% of them achieved a complete response.
“One of the observations from the CLL patients was the high rate of marrow clearance in these individuals,” Dr Turtle said. Of 11 patients, 10 cleared disease from the bone marrow with CAR T-cell therapy. The index clone was not detected in all 4 patients in whom immunoglobulin H deep sequencing of the bone marrow was performed.
In the patients with CLL and optimized lymphodepletion, the median PFS had not yet been reached.
The proportion of patients with high-grade cytokine release syndrome was 26% for those with ALL who received the preferred lymphodepletion and risk-adapted CAR T-cell dosing regimen, 10% for patients with NHL who received preferred lymphodepletion with level 2 CAR T-cell dosing, and 23% for patients with CLL. Overall, 26%, 10%, and 3% of patients in these 3 groups, respectively, had neurotoxicity.
“Although the toxicities of CAR T-cells can be significant, they are, in general, manageable,” said Dr Turtle. The evidence is the number of days patients spent in the hospital for their entire treatment cycle, he said, which ranged from 5 to 8 days.