Abstract

Introduction: Cytokine-induced killer ( CIK) cells are polyclonal non-MHC restricted T cells with potent cytotoxicity against acute myeloid leukemia (AML) cells in vitro. We have established culture of CIK cells with GMP compliance and infused into patients with various haematological malignancies. These include group 1, as adjuvant therapy post autologous transplant for acute myeloid leukaemia (AML), group 2, in untreated disease and group 3, for relapse post allogeneic transplant (NCT 00460694, NCT 00394381). Patients, Methods and Results: A total of 39 CIK cultures was produced over a 2 year period which resulted in 65 infusions given to 21 patients. We have demonstrated that it is feasible to expand CIK in large scale culture from both patients and allogeneic stem cell donors. The CD3+CD56+ subset expanded a median of 42.7 fold from 1.3% (0.2–5.3%) to 31.1% (10.4–76.9%) post culture for CIK derived from patients’s leukapheresis product, which is comparable with that derived from healthy haemopoietic stem cell donors. The cytotoxicity of these CIK against a panel of allogeneic AML targets showed variable potency (0% to 69%), with a median of 38%. Self limiting fever was the only infusion related side effect.

Patients In group 1 received an autologous transplant as consolidation for AML followed by adjuvant infusions of CIK cells., These were successfully cultured for 9 of 11 patients and infused into all 9 in aliquots of between 1–4 infusions. Follow up is short and a comparison against historical autologous transplant results are similar. Group 2 consisted of patients with overt leukemia who have failed or are unfit for chemotherapy. All 4 had CIK cells successfully cultured from a product containing variable % of leukemic cells, 3 of the patients who survived to receive CIK infusion did not have any response. However one of them had an incidental regression of basal cell carcinoma after 2 infusions.

In group 3, 6 patients who relapsed after an allogeneic transplant received allogeneic CIK after failing donor lymphocyte infusions (DLI). Another 3 patients had CIK generated from their own leukapheresis product due to donor unavailability (1 post cord blood and 2 post MUD transplant). Amongst the 8 patients who have received CIK infusion, two with overt refractory relapse (1 AML and 1 Hodgkin’s disease) did not respond to 3 and 4 infusions respectively. Four patients (2 AML and 2 ALL) had CIK infusion post salvage chemotherapy and therefore remission could not be entirely attributed to CIK infusion. Two patients had measurable responses attributable solely to CIK infusion. One was a post-transplant relapsed T cell ALL refractory to 5 different salvage regimen and repeated DLI. A marrow remission was achieved after one further Gemcitabin/Mitoxantrone salvage chemotherapy followed by CIK infusion. Marrow remission was maintained for 10 months with 4–6 weekly infusion of CIK while extramedullary (EM) disease progressed, suggesting control of marrow leukemia by CIK while leukemia evolution manifested at EM sites, known to be less susceptible to immunotherapy. A second patient had refractory Hodgkin’s disease in the lungs and vertebrae. A partial reduction in the size of lung nodules was achieved after the second CIK infusion but this was not sustainable. The dose of allo- CIK ranged from 10 – 200 million CD34/kg given as a step-wise increment for each patient. Three patients developed acute GVHD, one grade II liver, one grade II skin andgut, and a third patient had grade I upper gut GVHD, at doses of 50, 100 and 10 million CD3/kg respectively. All responded promptly to prednisolone at 1mg/kg.

Conclusion: We have shown that CIK infusion is feasible and safe in both the autologous and allogeneic setting, and GVHD that occurs is easily controlled. It is unlikely that CIK is effective against a large tumour burden. Its efficacy as an adjuvant therapy to eradicate minimal residual disease requires larger patient numbers and longer follow up. For allogeneic transplant, CIK culture has an additional advantage of expanding unrelated donor cells where availability is a problem by harvesting donor cells from patients for expansion. Further numbers are needed to compare against unmanipulated DLI in terms of efficacy and reduced GVHD severity.

Disclosures:No relevant conflicts of interest to declare.

Author notes

Corresponding author