For patients with relapsed/refractory acute myeloid leukemia (R/R AML), prognosis is often poor. This is a first-in-human study examining the use of compound chimeric antigen receptor (cCAR) T-cells for AML. Because the genomic signature of AML cells is heterogeneous, single CAR-based therapies may not be effective. As such, this study was designed to target leukemia stem cells that express CLL1—and comprise a small population of cells that play an important role in disease progression and relapse—and bulk proliferating AML cells (blasts) that express CD33 antigen in the majority of AML patients. The study was performed in China, and its goal was to evaluate both efficacy and toxicity related to CLL1-CD33 cCAR for the treatment of patients with R/R AML. Another goal of the study was to explore the probability of reducing the intensity of conditioning of allogeneic hematopoietic stem-cell transplantation (allo-HSCT) after cCAR therapy.
Investigators generated a CLL1-CD33 cCAR with 2 complete CAR constructs that were connected with porcine teschovirus-1 2A peptide, a cleavable linker. Eight were manufactured from autologous cells, and 1 was manufactured from a human leukocyte antigen–matched sibling donor. Nine enrolled patients received conditioning of fludarabine (30 mg/m2 daily) and cyclophosphamide (300 mg/m2 daily). CAR T-cells were given with a single or split dose by a dose escalation at 1-3 ×106/kg.
The median age was 32 years (range, 6-48 years). Seven of the 9 patients had de novo AML, 1 (a 6-year-old girl) had juvenile myelomonocytic leukemia-transformed AML, and 1 had chronic myeloid leukemia in accelerated phase. Median bone marrow blasts count before treatment was 47%. Dose level 1 (1 × 106/kg) was given to 4 patients, dose level 2 (2 × 106/kg) was given to 3 patients, and dose level 3 (3 × 106/kg) was given to 2 patients.
Eight patients experienced cytokine release syndrome (CRS); grade 1, 2, and 3 CRS was experienced by 3, 3, and 2 patients, respectively. One patient experienced grade 1 neurotoxicity, and 3 patients experienced grade 3 neurotoxicity. Both CRS and neurotoxicity resolved following treatment. Grade 4 pancytopenia was experienced by all 9 patients; 5 had a mild increase in liver enzyme (1 with a mild increase in bilirubin). Coagulation disorder and diarrhea were experienced by 4 patients each, while skin rash and renal insufficiency were documented in 1 patient each. Sepsis, pneumonia, and fungal infection were seen in 3, 3, and 2 patients, respectively. All adverse events resolved after treatment. CRS and neurotoxicity were reduced by early use of steroids in later enrolled patients.
Within 4 weeks of CAR T-cell infusion, 7 patients were minimal residual disease (MRD)-negative (evaluated by flow cytometry) and 2 patients had no response. One of the patients with no response was CD33+/CLL1–, confirming the importance of the CLL1 target for treatment. Of the 7 patients who achieved MRD-negative status, 6 proceeded to allo-HSCT; 1 received standard myeloablative conditioning (busulfan [Bu] 3.2 mg/kg daily for 4 days) and 5 received a lower intensity regimen (Bu 3.2 mg/kg for 1-2 days). Five patients engrafted with a persistent full chimerism, and 1 died of sepsis 6 days after allo-HSCT.
This study suggests that CLL1-CD33 cCAR has favorable efficacy and manageable toxicity in R/R AML patients and may provide an opportunity to perform reduced intensity or nonmyeloablative conditioning in preparation for allo-HSCT. A phase 1 trial is now underway.
Reference
Liu F, Zhang H, Sun L, et al. First-in-Human CLL1-CD33 Compound CAR (CCAR) T Cell Therapy in Relapsed and Refractory Acute Myeloid Leukemia. Presented at: 25th European Hematology Association Congress Virtual; June 11-21, 2020. Abstract S149.