Treatment of acute myeloid leukemia with chimeric antigen receptor T cells

Adoptive T cell therapy has made major progress in the treatment of hematological malignancies. One particularly effective approach is the use of chimeric antigen receptors (CAR), which are single-chain monoclonal antibodies linked to the signaling machinery of the T-cell receptor and co-stimulatory molecules.  Several groups reported significant and lasting clinical responses to CAR therapy in CD19+ B-cell malignancies, including acute lymphoblastic leukemia and lymphoma. Thus, these therapies were subsequently FDA, EMA and Swissmedic approved in 2017 and 2018 for relapsed/refractory CD19+ acute lymphatic leukemia and relapsed/refractory aggressive lymphoma. A side-effect of these therapies is, naturally, that healthy CD19+ B-cells and their offspring immunoglobulin-producing cells are removed as well. While immunoglobulins can be substituted temporarily, B cells will subsequently re-grow from hematopoietic stem and progenitor cells (HSCs, HPCs, see figure on left side below).
In contrast to lymphoid malignancies, the situation in HSC diseases as acute myeloid leukemia (AML), myelodysplastic syndromes (MDS) as well as myeloproliferative neoplasias (MPN) is more complicated: all are clonal disorders of the hematopoietic stem and very early progenitor cells (HSCs/HPCs), caused by various genetic alterations, driving respective malignancies. Leukemia-initiating cells (LIC) are contained in the phenotypic HSC/HPC pool and carry many features of HSCs, i.e. can self-renew and give rise to a hierarchy of maturing blasts. While the proliferating mature blast pool is highly sensitive to chemotherapy, the more quiescent LICs are frequently relatively resistant, and are a source of frequent relapses.
Intensive chemotherapy can cure 30-40% of AML patients and allogeneic hematopoietic stem cell transplantation can increase cure rates by an additional 15-20%. However, these therapies are limited to clinically fit patients without significant comorbidities. As all HSC malignancies are mostly diseases of the elderly population, most patients will receive primarily palliative and non-curative treatments with thus far modest increase of life-expectancy of only months to few years.
We here propose that the best way to further success in AML (and other HSC disease) therapy is to radically eliminate LICs by targeted surface antigen attack and accept collateral HSC destruction (Step I), then eliminate respective effectors (Step II), and subsequently perform allogeneic HCT (Step III) to sustain long-term hematopoiesis.
We hypothesize that this approach has the potential to be successful for AML patients, irrespective of the genetic heterogeneity of the diseases. Currently, no antigen has been identified that would specifically and exclusively mark the LIC (versus HSCs), although the feasibility of AML CAR T-cell trials has already been established and some AML membrane antigens are at an early stage of evaluation as CAR T-cell targets (e.g. CD33, CD123).  
We aim to create a versatile platform for the generation of CAR T cells directed against a variety of known and novel leukemic and hematopoietic stem cell antigens. We will test CAR T-cell therapy with respect to tumor eradication and safety in vitro and in vivo mice carrying grafted human AML cell lines and primary patient AML (PDX models).
2017-2018, we have achieved the following goals:

  • Development of lentiviral vector based CAR T cells directed against human CD117 (c-Kit, the receptor for stem cell factor)
  • We have shown high efficacy of the anti-CD117 CAR T-cells against
    o    human AML cell lines (Kasumi-1 and CD117 engineered HL-60 cells) in vitro
    o    human primary HSCs in vitro
    o    human primary AML cells in vitro
    o    human primary AML cells in vivo in PDX models
  • We are currently testing efficacy of anti-CD117 CAR T-cells against
    o    human AML cell lines (CD117 engineered HL-60 cells carrying in addition luciferase) in vivo
  • Further, we are currently testing CAR T-cell in vivo depletion strategies to allow subsequent safe and effective HSCT by using
    o    an antibody-based depletion method (RQR8, Rituximab/anti-CD20 mAb depletion)
    o    inducible caspase 9 transduction and pharmacologic depletion
    o    mRNA transfection for temporal CAR expression on T-cells


  • Development of further lentiviral vector based CARs against AML/LIC antigens (ongoing: anti-CD123; -CLL1, -CD135)
  • Development of dual CAR carrying T-cells (combination of e.g. anti-CD117 with anti-CD123 CAR)
  • Development of CAR T-cells where intracellular signaling only becomes active upon binding of both CARs (to increase specificity and reduce potential toxicity)
  • Development of bi- or multi-specific T-cell engager (BiTE) antibodies to allow MHC independent, endogenous/host T-cell activation in order to substitute for CAR T-cell generation and transfer, e.g. targeting CD3 and one of the above indicated surface antigens (CD117, CD123, CLL1, CD135)
  • Testing of toxicity in non-human primates (if feasible by antibody cross-reactivity; not part of this fellow application)

Expected output: We expect that our project will pave the way to a clinical trial, implementing one or several CAR T-cells (or antibody-constructs) simultaneously or sequentially in patients with refractory or relapsed AML/MDS/MPN.