Caution in using the granulocyte colony-stimulating factor (G-CSF) has been urged because it may stimulate proliferation of leukemic cells [1]. In contrast, few reports suggest that G-CSF causes the growth of nonmyeloid tumor cells. We describe a patient with acute myelogenous leukemia (AML) and monoclonal gammopathy of undetermined significance treated with G-CSF for neutropenia after induction therapy against AML.

A 69-year-old Japanese man referred to our hospital in December 1988 was diagnosed as having AML (M2 according to the French-American-British criteria) associated with IgG -type monoclonal gammopathy. Laboratory investigation showed a hemoglobin of 10.3 g/dL, a leukocyte count of 189 x 10⁹/L with 70% blasts, a platelet count of 33 x 10⁹/L, and an IgG of 2767 mg/dL. Bence-Jones protein was not detected, nor did the patient have renal insufficiency or skeletal lesions. A bone marrow aspirate revealed 58% blasts and fewer than 5% plasma cells. Flow cytometry analysis showed surface -, -, and IgG-positive cells at levels of 1%, 1%, and 2%, respectively. Two courses of induction therapy for AML were given, consisting of behenoyl-ara C, daunorubicin, 6-mercaptopurine, prednisolone, and vincristine. Marked neutropenia continued for 3 months; bone marrow aspirate showed hypocellularity and few blasts. The patient was treated with recombinant G-CSF, filgrastim prepared by Kirin (Tokyo, Japan). G-CSF was injected intravenously in an escalating dosage from 3 µg/kg to 12 µg/kg for 40 days. Three days after G-CSF therapy was begun, the serum IgG increased to 3653 mg/dL. Surface -, -, and IgG-positive cell levels were 2%, 3%, and 6%, respectively, in the bone marrow. Ten weeks later, bone marrow surface - and IgG-positive cell levels increased to 43% and 55%, respectively, while the surface -positive cells did not increase. Thereafter, serum IgG and bone marrow surface - and IgG-positive cells decreased, with stable bone marrow blasts. In May 1990, the blasts increased to 30%, and chemotherapy was started. However, the patient died of sepsis.

In this case, a transient increase of both IgG -type M-protein and clonal surface IgG -positive lymphocytes was related to the administration of G-CSF. Endogenous G-CSF has been reported in the short-term culture of bone marrow cells from patients with myeloma; this cytokine is said to control the endogenous production of IL-6 by the bone-marrow, a growth factor of myeloma cells [2]. Klein and colleagues [3] described a 5-day injection of G-CSF into a patient with myeloma and marked leukopenia that corrected the leukopenia but increased the percentage of myeloma cells in the S-phase eightfold and increased the percentage of myeloma cells in the bone marrow twofold. Taken together, we postulate that G-CSF could have resulted in the progression of a myeloma-like syndrome [4].