Search

We have developed culture conditions for the efficient expansion of cytotoxic effector cells from peripheral blood mononuclear cells (PBMNCs) by the timed addition of interferon-γ (IFN-γ), interleukin-2 (IL-2), and the monoclonal antibody (MoAb) OKT3. These cells, termed cytokine-induced killer (CIK) cells, are composed primarily of T cells, and the population of cells with the greatest cytotoxic activity is an otherwise rare population of CD3+CD56+ cells that expand dramatically under these culture conditions. CIK cells were expanded from PBMNCs from 13 patients with chronic myeloid leukemia (CML). These cultures contained a variable number of T cells at the start of the culture (median 44%, range 1% to 64%), yet after 21 to 28 days of culture, virtually all of the cells were CD3+ T cells (median 97%, range 90% to 99%). The CD3+CD56+subset of cells expanded significantly (median 25-fold, range 2.2- to 525-fold). CIK cells from all patients showed cytotoxicity against the tumor cell lines OCI-LY8 and K562. In four patients the expanded CIK cells suppressed colony growth of autologous CML blast cells and myeloid progenitor cells. Allogeneic CIK cells from normal donors also suppressed CML colony growth but did not inhibit growth of normal hematopoietic colonies. Twelve of the 13 cultures were exclusively composed of Philadelphia (Ph)-negative cells and one culture had 1 out of 20 Ph-positive metaphases after 4 weeks in culture. Intracellular cytokine production was assayed by fluorescence-activated cell sorter (FACS), and the expanded T-cell cultures produced IL-2, IFN-γ, and tumor necrosis factor- (TNF-), but not IL-4. Both the CD4+ and CD8+ subsets secreted this cytokine profile. To test the in vivo activity of the expanded CIK cells, CML was engrafted into severe combined immunodeficiency disease (SCID) mice using matrigel. After 4 weeks, 4 × 107autologous CIK cells were injected intravenously by tail vein injection into groups of mice, and the animals were sacrificed after a total of 18 weeks. Bcr-abl was detected in the bone marrow or spleen of 5 out of 6 control mice and only 2 out of 13 mice who received the autologous CIK cells (P = .02). In an additional series of animals, the mice did not engraft with CML but instead developed large human Epstein-Barr virus–associated lymphomas by 12 weeks. The mice who received autologous CIK cells at 4 weeks had either no tumor (5) or small tumors (5), whereas all 10 mice that received CIK cells at week 8 developed lymphomas; however, these were not as large as in the 10 control mice who did not receive CIK cells (P = .03). This study shows that CIK cells, which are Ph chromosome–negative, can be expanded from patients with CML and have potent in vitro and in vivo efficacy against autologous tumor cells. © 1998 by The American Society of Hematology.

Chronic myeloid leukemia (CML) is a clonal disorder of primitive hematopoietic stem cells characterized by a reciprocal translocation between chromosomes 9 and 22. Animal models of CML would be useful to study the biology and potential therapies in this disease. Mice with severe combined immunodeficiency (SCID) which will accept human xenografts have been useful in the study of a variety of human malignancies. CML has been difficult to establish in SCID mice possibly due to the lack of a functioning human stroma and relevant cytokines. To facilitate engraftment we injected cells in matrigel which is a soluble extract of basement membranes; is liquid below 22°C and gels at 37°C. CD34+ myeloid blast crisis cells (2×106) were mixed in matrigel and injected subcutaneously into 10 SCID mice. All mice developed large tumours which spread to the mouse BM and spleen. However the percentage of human cells in the mouse BM and spleen was variable and ranged from 1 to 50 per cent. In contrast chronic phase (CP) CML cells mixed in matrigel did not form subcutaneous tumours and spread to the BM and spleen was detectable by PCR and not macroscopically. Groups of mice were injected with matrigel containing 1–20×107 MNC (2–20×105 CD34+ cells) from five patients with CP CMP. Bcr-abl sequences were detected by RT-PCR in the peripheral blood (PB) of 38/84 (45 per cent) mice at 3–10 weeks following injection of the CML cells but rarely at later time points. In addition, 33/75 (44 per cent) of mice sacrificed between 7 and 35 weeks following injection of CML cells were bcr/abl positive in the bone marrow and 17/70 (24 per cent) were positive in the spleen. Bcr-abl positive human CFU-GM colonies were also cultured from the murine bone marrow of several mice indicating that hematopoietic progenitor cells were able to migrate from the matrigel and engraft in murine hematopoietic organs. Engraftment of CP-CML was more successful in mice given higher numbers of CD34+ cells. Histological examination revealed that myeloid cells grow locally in the matrigel for several weeks, during which time the matrigel is infiltrated by blood vessels which may allow for the migration of CML progenitors to the murine bone marrow. This model system may be useful for studying the role of immunotherapy after allogeneic and autologous bone marrow transplantation. Copyright © 1998 John Wiley & Sons, Ltd.

BACKGROUND Retrospective studies suggest that dose intensity is an important determinant of outcome in the treatment of patients with a variety of malignant diseases such as breast cancer, ovarian cancer, and lymphoma. Unfortunately, these results have not been clearly substantiated in prospective randomized trials. One problem with these studies may be that the degree of dose escalation is not sufficient to result in an improved outcome because the chemotherapy doses are limited by hematopoietic toxicity. In an attempt to deliver more dose-intensive therapy, the feasibility of the administration of multiple cycles of high dose chemotherapy with hematopoietic progenitor cell and growth factor support was investigated in patients with advanced malignancies. METHODS Nineteen patients with metastatic breast cancer and six patients with refractory non-Hodgkin's lymphoma were initially treated with etoposide (VP-16) (2 gm/m2) and granulocyte-colony stimulating factor (G-CSF). Peripheral blood hematopoietic progenitor cells were collected by leukapheresis and cryopreserved as the patients' leukocyte counts recovered from the nadir induced by VP-16. Patients were then treated with four cycles of mitoxantrone (18 mg/m2), thiotepa (150-200 mg/m2) and cyclophosphamide (4500-5000 mg/m2) as a 48-72 hour continuous infusion followed by infusion of one-quarter of their progenitor cells 48 hours later. All patients also received G-CSF (5 micrograms/kg/day) until engraftment. RESULTS A total of 88 of a planned 100 cycles of therapy were administered to these 25 patients. The median time to recovery of an absolute neutrophil count of 500/microliters or greater was 13-14 days (range, 7-18 days) and time to recovery of a platelet count of 20,000/microliters or greater was 13-14 days (range, 7-16 days) after the initiation of each cycle of chemotherapy. The median number of platelet transfusions required after each cycle was 2-3 (range, 0-18 transfusions) and the number of erythrocyte transfusions was 4 (range, 0-10). The most common toxicity was diarrhea. Prophylactic intravenous antibiotics were administered to avoid fever with neutropenia. Two patients developed interstitial pneumonitis and one patient died. One heavily pretreated patient failed to engraft after the first cycle. Reversible veno-occlusive disease of the liver developed in one patient after the fourth cycle of therapy. Four patients progressed while on therapy. Eight patients were disease free and 13 patients had a partial response or had a positive bone scan as the only evidence of disease at the completion of therapy. Seven patients, two with lymphoma and five with breast cancer (28%), remain progression free with a median follow-up of 24.7 months (range, 17-28 months). CONCLUSIONS Support with hematopoietic progenitor cells and growth factors allows the timely administration of repetitive cycles of high dose chemotherapy, resulting in a significant increase in dose intensity with acceptable toxicity.

High-dose chemotherapy with or without radiotherapy followed by autologous transplantation of hematopoietic progenitor cells is an effective treatment for patients with high-risk or relapsed non- Hodgkin's lymphoma. Chemotherapy and/or hematopoietic growth factors have been used to mobilize progenitor cells in the peripheral blood for transplantation. However, the mobilized blood cell products have been found to be frequently contaminated with tumor cells, and techniques have not been developed to purge tumor cells from these products. In addition, the minimum number of hematopoietic progenitor cells required for engraftment has not yet been fully elucidated. We treated 21 patients with a single infusion of cyclophosphamide (4 g/m2) followed by daily administration of granulocyte colony-stimulating factor (G-CSF). After recovery of the white blood cell count, a single 3-hour apheresis collection was performed. The apheresis product was then applied to a discontinuous Percoll gradient. The low-density fractions resulting from this separation procedure were enriched for CD34+ progenitor cells (total cell yield, 19.5%; CD34+ cell recovery, 81.2%). These enriched cellular products were treated with a panel of anti-B cell or anti-T cell monoclonal antibodies and complement in an effort to remove residual tumor cells. After treatment of the patient with myeloablative therapies, the enriched and purged cells were reinfused. Hematologic recovery was rapid, with median neutrophil engraftment in 10 days [absolute neutrophil count (ANC), greater than 0.5 x 10(9)/L] and 11 days (ANC, greater than 1.0 x 10(9)/L). Median platelet transfusion independence required 13 days. The rapidity of multilineage engraftment correlated with the number of CD34+ cells per kilogram that were infused. Patients who received more than 2 x 10(6) CD34+ cells per kilogram had rapid hematologic engraftment, whereas those patients transplanted with less than 2 x 10(6) CD34+ cells per kilogram had slower platelet recovery. Modeling studies using a lymphoma cell line with a t(14; 18) chromosomal translocation demonstrated the successful removal of tumor cells assayed using the polymerase chain reaction (PCR) after the processing and purging. Four of the 21 patients had PCR-detectable lymphoma cells in the bone marrow and peripheral blood; however, the enriched and purged blood products reinfused in all four did not contain detectable tumor cells.(ABSTRACT TRUNCATED AT 400 WORDS)