Abstract Background: Despite improved treatments, overall survival of patients with acute myeloid leukemia (AML) remains poor, especially for those with high-risk features or relapsed/refractory (r/r) disease, for whom hematopoietic stem cell transplantation (HSCT) is the best therapeutic option. Although chimeric antigen receptor (CAR) T-cells have successfully eradicated some lymphoid malignancies, they have been less successful as therapy for AML, partly due to difficulty identifying an ideal target antigen. C-type lectin 1 (CLL-1, CD371) is expressed at variable levels on myeloid blasts and leukemic stem cells. Thus, CLL-1 directed CAR T-cells can selectively target leukemic progenitor cells as well as AML blasts, whilst sparing normal tissues. We developed a second-generation CLL-1-specific CAR with CD28 costimulatory endodomain, and designed a phase 1 study (CARMEN; NCT04219163) to evaluate the safety of escalating doses of CLL-1.CAR T-cells in patients with r/r AML, as a bridge to allogeneic HSCT. Methods: CLL-1.CAR T-cells were generated from autologous PBMCs using gammaretroviral transduction. The length of cGMP manufacturing was 7-9 days with additional two weeks for release testing. Patients up to 75 years of age with ≥30% CLL-1 expression on AML blasts were eligible. Prior to CAR T infusion, all patients were transplant-eligible, yet unable to proceed due to residual disease. All patients had been heavily pretreated (median 3; range 2 -7 prior lines of therapy); 6 with prior allo-HSCT. Patients received lymphodepleting chemotherapy with cyclophosphamide and fludarabine followed by a single dose of CLL-1.CAR T-cells at dose level (DL)1: 1x107, DL2: 3x107, or DL3: 1x108 cells/m2. Results: To date, we have treated a total of 14 patients (12 adults and 2 adolescents; median age 46y range: 12-70 y) with CLL-1+ r/r AML (median expression 81%; range 52-99%) across 3 DLs. One patient had isolated extramedullary disease, and all other patients had active marrow disease with median blasts pre-lymphodepletion 21.5% (range 0-85%). Eight patients received CLL-1.CAR T-cells on DL1. One dose limiting toxicity (DLT) occurred and 2 patients were not evaluable for DLT due to progressive disease and receipt of other therapies prior to the end of the DLT monitoring period, and were replaced. Three patients were treated on each, DL2 and DL3. In all patients, CAR T-cells reached peak expansion in peripheral blood by week 2 post-infusion followed by a gradual contraction. Cytokine release syndrome (CRS) occurred in 12/14 patients: grade 1 (n=9), grade 2 (n=1), and grade 3 (n=2). Two patients developed immune effector cell associated neurotoxicity (ICANS); max grade 3. Additionally, 2 patients on DL1 developed immune effector cell associated hemophagocytic lymphohistiocytosis-like syndrome (IEC-HS), which manifested as hyperferritinemia, hypofibrinogenemia, and hypertriglyceridemia. Cytopenias were the commonest grade 3/4 adverse events. Infectious complications included bacteremia (n=5), viremia (n=3), viral upper respiratory infection (n=1) and fungemia (n=1). Two patients (1 DL1 and 1 DL3) achieved complete response with incomplete count recovery and 1 patient (DL1) achieved morphologic leukemia free status. Two patients (both treated on DL1) underwent HSCT with 1 patient still alive, in remission 2 years post T-cell infusion. One patient had a partial response, 6 had stable disease, 3 had progressive disease, and one (DL1) was not evaluable due to death from hemorrhagic shock possibly related to IEC-HS or disseminated intravascular coagulation related to refractory leukemia/infection. This was deemed a dose-limiting toxicity; no other deaths suspected to be related to the CLL-1.CAR T-cells occurred. Conclusions: CLL-1.CAR T-cells following lymphodepletion were tolerated in adolescent and adult patients, up to a dose level of 1x108 cells/m2with a low incidence of severe CRS or ICANS. However, IEC-HS may be a more pervasive toxicity requiring vigilant monitoring. There was no evidence towards increased toxicity at higher dose levels or in the 2 adolescents. CLL-1.CAR T-cells induced morphologic remissions sufficient to enable consolidative HSCT in 2 patients. Given the small number of patients treated, a dose expansion phase is being considered to further evaluate treatment efficacy once the maximum tolerated dose is determined.
Wang et al. (Mon,) studied this question.