Memory-Like Natural Killer Cells for Patients With Advanced AML and MDS

Rizwan Romee, MD

Medical Oncologist

Siteman Cancer Center

Assistant Professor of Medicine

Washington University School of Medicine

St. Louis, MO

Acute myeloid leukemia (AML) and myelodysplastic syndromes (MDS) are the two most common hematologic malignancies of the myeloid lineage origin and both are more common in older patients. The median age of diagnosis for AML is around 67 years and has an estimated prevalence of 3.8 cases per 100,000. Unfortunately, less than half of AML patients are cured with current treatment approaches. The prognosis is particularly poor in older patients (≥60 years), in whom the success of achieving complete remission are significantly lower than younger patients and the long-term disease free-survival is often measured in single digits if patients are not eligible for stem cell transplantation. Furthermore, patients who are unable to achieve complete remission and or relapse after two or more chemotherapeutic regimens have extremely dismal outcomes, even with stem cell transplantation.

MDS is also more common in older patients and their long-term curative options are also limited if they are not eligible for stem cell transplantation. MDS patients without treatment often progress to acute myeloid leukemia. Hypomethylating agents, including azacytidine and decitabine, are the only class of medications that have demonstrated some clinical efficacy in these patients. However, patients with MDS who have failed and/or progressed while on hypomethylating agents have very limited treatment options.

Even with stem cell transplantation the outcomes are suboptimal in MDS patients with pretransplant high blast counts, underscoring an urgent need for developing more innovative treatment options for these patients.

Natural killer (NK) cells are innate immune lymphocytes important for host defense against pathogens and surveillance against malignant transformation. Early phase studies have demonstrated that allogeneic NK cells are a promising cellular therapy for leukemia. However, limited persistence and expansion of the adoptively transferred NK cells seen in previously reported studies remain some of the major limitations of the currently used strategies for NK cell—based immunotherapy protocols.

Recent work in the NK cell field has challenged their classification as members of the innate immune system, identifying memory-like properties of NK cells. This has included the identification of human cytokine-induced memory-like (CIML) NK cells by our group at Washington University School of Medicine, following preactivation with IL-12, IL-15, and IL-18. Human CIML NK cells exhibit enhanced cytokine and cytotoxic responses against myeloid leukemia target cells and they retain this enhanced functionality following transfer into immunodeficient NOD-SCID-gc-/- mice. Based on their enhanced antileukemia responses and ability to proliferate and expand after adoptive transfer, CIML NK cells are an attractive option to explore in relapsed/refractory AML and in MDS patients who have failed and/or progressed on hypomethylating agents.

We are currently enrolling patients on a firstin- human phase I study using CIML NK cells for advanced MDS and AML patients.

Eligibility: Patients (≥ 18 years) with one of the following diagnosis:

1. Refractory AML without complete remission (CR) after induction therapy (primary induction failure) or relapsed AML after obtaining a CR.

2. High-risk AML in complete remission (CR) and has either refused hematopoietic stem cell transplantation or is currently not eligible for hematopoietic stem cell transplantation or for whom hematopoietic stem cell transplantation is being reserved for later relapse.

3. MDS with excess blasts (>5%) and failed/progressed with the hypomethylating therapy.

Trial design and treatment strategy: It is a standard 3(+3) dose escalation design and includes 3 dose levels (0.5x106/kg, 1x106/kg, and all available CIML NK cells). Patients receive haploidentical donor CIML NK cells on day 0 after a nonmyeloblative regimen. CIML NK cells are generated from the peripheral blood NK cells donated by an HLA-haploidentical (“half matched”) related family member using a standard non-mobilized (without any growth factor or G-CSF) apheresis session. Peripheral blood NK cells are processed and activated with a cocktail of cytokines (IL-12, IL-15 and IL-18) over 12-16 hours to generate CIML NK cells at the Biologic Therapy Core (BTC) of the Siteman Cancer Center. Treatment schema is shown in the Figure.

Adoptively transferred CIML NK cells are tracked using HLA specific antibodies by flow cytometry. Patients are closely monitored for any adverse events and also undergo bone marrow biopsy at days 14 and 28 after their CIML NK cell infusion to document disease response. Patients who are able to achieve remission with CIML NK cells can proceed to a stem cell transplant and therefore, this treatment can be used as a bridge to transplant in eligible patients.

We have seen some encouraging results in the first few patients treated on this study. If found to be safe and with an efficacy signal in our current ongoing phase I study, we intend to initiate a larger phase II study in the near future. Efforts are also on to expand their usage to other hematologic malignancies, including multiple myeloma and lymphoma either alone or in combination with other novel immunotherapy approaches.

Further information about this novel protocol can be found at the Clinicaltrials.gov (Identifier: NCT01898783). Physicians who would like to potentially consider this study for their patients with AML or MDS can contact me at rromee@dom.wustl.edu or call my office number: (314)-454-8333.