A Pilot Study of Continuous Infusion Ara-C in Combination with rhG-CSF in Relapsed Childhood Acute Myeloid Leukemia

Relapse in acute myeloid leukemia (AML) following intensive chemotherapy bears a bad prognosis. We treated 18 children with relapsed AML on two separate protocols that included continuous infusion (CI) of cytosine arabinoside (ara-C) (total dose 4gr-6gr/m2) over 96120 hours. In an attempt to increase the fraction of blasts in S-phase and render them more sensitive to cell-cycle specific agents such as ara-C, 10 patients received Smcg/kg rhG-CSF twice daily beginning 48 hours before and continuing through the duration of the CI ara-C (POG #9192 study). The percentage of cells is S phase before and after G-CSF administration was determined. In a second group of patients (n = 8) who received ara-C alone, endogenous concentrations of G-CSF and serial blood counts were measured (St Jude's R4 study). The rationale of the St Jude's R4 was to optimize the schedule of the second course of ara-C at a time when the patient's endogenous G-CSF concentration was increased and thus maximize the percent of cells captured in S phase.Four out of 8 patients receiving CI ara-C alone and 4 out of 10 patients receiving CI ara-C with rhG-CSF achieved a complete remission (CR) after 1 cycle of therapy. Four patients in CR underwent marrow transplantation (2 allogeneic and 2 autologous). Cell cycle analysis of blast cells cultured in vitro with or without G-CSF showed a two fold increase in the percentage of cells in S phase (P = 0.03) whereas cells obtained from patients before and after G-CSF administration showed no difference in cell cycling. Correlation between G-CSF concentrations and ANC showed a negative association indicating that the regulatory mechanisms for G-CSF production remained intact. In our relatively small series, CI ara-C achieved a CR rate of 44% with rhG-CSF having no effect on the remission rate. Although in vitro rhG-CSF increased the percentage of blasts in S phase significantly, in vivo effects were not observed. Larger studies with combinations of different hematopoietic growth factors and cell-cycle active drugs are needed to evaluate the role of these cytokines in the therapy of recurrent AML.