Ara-C Treatment of Acute Myeloid Leukemia Does Not Lead to Prolonged Enrichment of Stem Cells or a Cell Cycle Arrest

Abstract 2178

The major therapeutic barrier in acute myeloid leukemia is chemotherapy resistance. Although most patients will respond to treatment with chemotherapy over 50% of responders will relapse and eventually die of disease. Many hypotheses have been proposed to explain chemotherapy resistance but none of these have lead to new therapies or a complete understanding of the molecular mechanisms of AML chemotherapy resistance. In order to develop an improved understanding of chemotherapy resistance in AML, we have developed a chemotherapy model of human AML in NSG mice. Mice are engrafted with primary AML samples from patients seen at diagnosis or relapse of disease or who demonstrated primary chemotherapy resistance. After demonstrating AML engraftment, mice are treated with cytosine arabinoside (Ara-C) given IP daily for 5 days as a single agent at 10mg/kg daily, 30mg/kg daily, and 60 mg/kg daily which correlates with human dosing. In >75% of mice treated with 10 mg/kg of Ara-C there is a cytoreductive effect at 2 weeks post-treatment with relapse at 4 weeks post-treatment. In all mice treated with 30 mg/kg of Ara-C there was a cytoreductive effect at 2 weeks post-treatment with relapse at 4 weeks post-treatment. In all mice treated with 60 mg/kg of Ara-C there was a cytoreductive effect at 2 weeks post-treatment with relapse 4–13 weeks post-treatment when relapse occurred, demonstrating that there is a dose response relationship in the model in terms of nadir leukemic burden and time to peripheral blood relapse. Two weeks after treatment, there was up to a 50-fold decrease in total AML cell burden in the peripheral blood of mice treated with 30 mg/kg and and up to a 70-fold decrease in the peripheral blood of mice treated with 60 mg/kg of Ara-C. We found no enhancement in quiescent or G0 cells after chemotherapy treatment. We did however, in 1/5 samples tested see a change in phenotype after chemotherapy treatment with an increase in the total number of CD34+38+ cells with a concomitant decrease in CD34-38+ cells. In all other samples tested, there was no change in phenotype after chemotherapy treatment. These results are in contrast to recent studies using a 1 gm/kg dose of Ara-C and analysis at Day 3 following a single treatment. These results have implications for understanding the physiologic response to Ara-C at different doses. We are currently analyzing mRNA expression arrays of AML without or with Ara-C treatment to identify novel mechanisms of chemotherapy resistance. Taken together, this model provides a novel approach for development of new therapies in AML.