SAN DIEGO – Anti-CD19 chimeric antigen receptor (CAR) T cells engineered with a “safety switch” yielded high rates of complete response and an acceptable toxicity profile in chemotherapy-resistant B cell acute lymphoblastic leukemia, according to a multicenter phase I/II trial.
Importantly, high tumor burden did not increase the risk of cytokine release syndrome, said Lung-Ji Chang, PhD, of Shenzhen (China) Genoimmune Medical Institute and the University of Florida in Gainesville. “This reliable, standardized CAR T-cell preparation protocol has now served more than 30 major medical centers in China,” he said at the annual meeting of the American Society of Hematology.
Anti-CD19 CAR T cells have shown dramatic potential for treating B-cell malignancies, but toxicities have been a concern. One potentially serious adverse reaction is cytokine release syndrome, in which patients develop marked rises in blood levels of several types of cytokines. Another problem is that anti-CD19 CAR T cells can trigger loss of CD19 B cells, ultimately leading to humoral deficiencies, Dr. Chang noted. Consequently, researchers have searched for ways to continue controlling the activity of CAR T cells even after infusing them into patients.
As part of that effort, Dr. Chang and his associates developed a standardized protocol for engineering next-generation anti-CD19 CAR T cells based on the established concept of a “safety switch.” After collecting T cells from patients with chemotherapy-resistant ALL, they used a lentiviral vector to transform them into CAR T cells with fusion proteins consisting of a proapoptotic molecule called caspase-9 that is linked to modified human FK506-binding proteins, or FKBP. The addition of iCaspase9-FKBP enables clinicians to induce CAR T cell apoptosis by treating patients with a synthetic dimerizer called AP1903.
Apoptosis occurs about 45 minutes after this drug is given, according to Dr. Chang. This “safety switch” also enables clinicians to eliminate anti-CD19 CAR T cells after tumor cells are eradicated so that patients can recover their humoral immunity. He and his associates further modified these anti-CD19 CAR T cells by introducing four intracellular signaling domains that are associated with T-cell activation, survival, and longevity, he said.
A total of 22 treatment centers helped test this approach in a phase I/II trial of 110 leukemia patients, about half of whom were children with a median age of 9 years. The median age of adults was 37 years, and the oldest patient was 70. Cancer types included Philadelphia chromosome–positive ALL, Philadelphia chromosome–negative ALL, and chronic myeloid leukemia with blast crisis. About a third of patients had bone marrow samples with at least 50% blasts, and a similar proportion had already undergone hematopoietic stem cell transplantation.
Cytokine release syndrome affected 86% of patients with low or no tumor burden, but only 53% of patients with bone marrow blasts exceeding 5%, Dr. Chang reported. He emphasized that patients with high tumor burden were no more likely to develop moderate or severe cytokine release syndrome than were patients with little or no tumor burden (P = .3). Furthermore, among 17 patients with more than 80% bone marrow involvement, only three developed grade 3-4 cytokine release syndrome, while eight developed grade 1 cytokine release syndrome.
A total of 96 patients (87%) had a complete response to this CAR T cell regimen, including 51 children and 45 adults, Dr. Chang reported. Median overall survival was 222 days (range, 23-1,041 days), and 60% of patients lived at least 400 days after treatment. Patients survived a median of 115 days without relapsing (range, 0-455 days), and 55% ultimately relapsed. Age did not appear to predict relapse, he noted.
Kaplan-Meier curves revealed no major differences in rates of overall survival (OS) between adults and children at 400-day data cutoff, Dr. Chang said. However, patients with more than 50% blast cells in their bone marrow had significantly lower rates of survival (P = .02) than did patients with less advanced ALL. A lower T-cell dose predicted lower survival in children (P = .04), but not in adults. Dr. Chang and his colleagues now dose patients of all ages with 106 cells per kilogram, he said.
Survival was significantly more likely when CAR T cell recipients went on to allogeneic hematopoietic stem cell transplantation (P = .0002) than otherwise. Based on the findings, Dr. Chang particularly recommends this approach for highly chemotherapy-resistant disease with a high tumor burden. Among patients who relapsed, repeating CAR T cell therapy led to better survival than administering combination chemotherapy-tyrosine kinase inhibitor therapy (P = .01).
These safety and efficacy results suggest that CAR T cell immunotherapy can benefit patients if they have very high-burden leukemia, Dr. Chang concluded. Patients outcomes remained consistent across centers due to a “highly standardized CAR T cell preparation profile,” he said.
Dr. Chang did not report funding sources. He reported having no relevant conflicts of interest.