Inhibitory and Proliferation-Inducing Effects of Valproic Acid on Leukemic Stem Cells from Different Acute Myeloid Leukemias.

The antileukemic activity of the histone deacetylase inhibitor valproic acid (VPA) in combination with all-trans retinoic acid and/or low-dose chemotherapy is currently evaluated in clinical trials for the treatment of acute myeloid leukemia (AML). Preliminary data demonstrate that only a small subset of patients responded to these VPA-containing regimens and none of them achieved a molecular remission (Bug et al., Cancer 2005, in press). Recently, it has been shown that the in vitro exposure of normal hematopoietic stem cells (HSC) to VPA significantly increases their proliferation and self-renewal potential. Given the fact that normal and leukemic stem cells are believed to share regulatory mechanisms of proliferation and self-renewal, we investigated whether VPA can also stimulate leukemic stem cells.

First we confirmed the expansion of normal HSC in vivo by treating C57BL/6J mice with VPA for two days. Here we report that orally administered VPA led to

• a significant increase in the cellularity of both bone marrow (BM) and spleen;

• a significant increase of Sca1+/c-Kit+ cell numbers in BM and spleen;

• a slight splenomegaly (0.11–0.42 g vs. 0.06–0.08 g in control mice);

• a significant acceleration of cell cycle progression in spleen cells.

Next we exposed highly enriched CD34+/CD38− cells (average purity of 84.2+/−1.6%) from the peripheral blood of 6 consecutive AML patients at first diagnosis to VPA (100 μg/ml) in presence of SCF, FL, TPO and IL-3 in liquid culture as well as in colony forming unit (CFU) assay. Chromosomal aberrations were detected in 5 out of 6 AML samples. Here we show that i.) the exposure to VPA inhibited differentiation and maintained a higher level of CD34+/CD38− cells in liquid culture at day 7 and 14 as compared to untreated controls in all samples; ii.) the majority of CD34+ cells were of leukemic origin as revealed by fluorescence in situ hybridization (FISH) analysis of 4 out of 5 AML samples with chromosomal aberrations; iii.) in one t(8;21)-positive AML sample VPA reduced the proportion of leukemic CD34+ cells to 60% and 17% on day 7 and 14, respectively, as compared to controls (92% and 86%); iv.) in one trisomy 8-positive AML sample VPA induced expansion of leukemic CD34+ cells; v.) the exposure to VPA did not only increase the number of CFUs, but also the cellularity of single colonies which exhibited an immature morphology and whose leukemic origin was confirmed by FISH analysis of one AML sample.

In conclusion, VPA was shown to have differential effects on CD34+/CD38- cells from untreated AML patients ranging from a selective inhibition of t(8;21)-positive cells, over a stabilization of the leukemic stem cell compartment, up to the amplification of trisomy 8-positive cells. Therefore, the inclusion of VPA into regimens for AML treatment has to be reconsidered.