OBJECTIVE: Persistence of leukemic cells after induction therapy has been shown to correlate with poor survival in acute myeloid leukemia (AML). In this study, we tested if human mesenchymal stromal cells (hMSCs) have protective effects on leukemic cells undergoing chemotherapy. METHODS: Persistent disease was used as marker to identify cases with therapy-resistant leukemic cells in 95 patients with AML. Immunophenotyping, cell cycle, and apoptosis assays were assessed by flow cytometry. AML coculture studies were performed with hMSC of healthy donors. RESULTS: Samples from patients with persistent disease had increased fractions of CD34+ CD38- and quiescent leukemic cells. Comparison of sample series collected at time points of diagnosis and blast persistence showed a relative therapy resistance of quiescent leukemic cells. Consistent with these observations, relapsed disease always displayed higher proportions of quiescent cells compared to samples of first diagnosis suggesting that quiescence is an important therapy escape mechanism of resistant cells. Co-culture studies demonstrated that hMSC protect leukemic cells from the effect of AraC treatment by enriching for quiescent cells, mimicking the effects observed in patients. This effect was even detectable when no direct stromal contact was established. CONCLUSIONS: Our data suggest that hMSC contribute to quiescence and therapy resistance of persistent AML cells.
OBJECTIVE: Persistence of leukemic cells after induction therapy has been shown to correlate with poor survival in acute myeloid leukemia (AML). In this study, we tested if human mesenchymal stromal cells (hMSCs) have protective effects on leukemic cells undergoing chemotherapy. METHODS: Persistent disease was used as marker to identify cases with therapy-resistant leukemic cells in 95 patients with AML. Immunophenotyping, cell cycle, and apoptosis assays were assessed by flow cytometry. AML coculture studies were performed with hMSC of healthy donors. RESULTS: Samples from patients with persistent disease had increased fractions of CD34+ CD38- and quiescent leukemic cells. Comparison of sample series collected at time points of diagnosis and blast persistence showed a relative therapy resistance of quiescent leukemic cells. Consistent with these observations, relapsed disease always displayed higher proportions of quiescent cells compared to samples of first diagnosis suggesting that quiescence is an important therapy escape mechanism of resistant cells. Co-culture studies demonstrated that hMSC protect leukemic cells from the effect of AraC treatment by enriching for quiescent cells, mimicking the effects observed in patients. This effect was even detectable when no direct stromal contact was established. CONCLUSIONS: Our data suggest that hMSC contribute to quiescence and therapy resistance of persistent AML cells.