Epigenetic therapies for acute myeloid leukemias : pre-clinical validation and study of molecular mechanisms

Epigenetic therapies of cancer have been intensely studied, because of the high frequency of epigenetic alterations found in tumor cells. Epigenetic modulators, such as histone deacetylases (HDAC) and DNA methyltransferase (DNMT), can be targeted by specific drugs with the intent to revert the epigenetic alterations induced by tumor progression. Although a few DNMT inhibitors have been approved for the therapy of some specific hematopoietic diseases, most of the results of clinical trials on other types of cancer have given limited results. The need for a better understanding of the molecular mechanism(s) underlying sensitivity to epigenetic therapies is therefore very strong. For obvious reasons, the use of preclinical models is mandatory to achieve this deeper understanding, to be translated to better clinical trials. We therefore used one form of acute myeloid leukemias (APL) as a preclinical model for epigenetic therapies. The choice of APL is due to the fact that the APL-associated fusion protein PML/RARα is known to induce epigenetic alterations in APL blasts by recruiting HDACs and DNMTs, making APL blasts obvious candidates for this kind of drugs. Monotherapies of either HDACi (VPA) or DNMTi (DAC) induced increased survival in APL mice and their combination were able to further prolong survival of APL mice, demonstrating the importance of hitting multiple targets of the epigenetic “Silencing Loop”. Both the studied epigenetic drugs also demonstrated an enhancement of the therapeutic effect of differentiating therapy with All Trans Retinoic Acid (ATRA). We also tried to identify the molecular mechanisms involved in anti-leukemic activity. Both VPA and DAC demonstrated an induction of the apoptotic response in leukemic cells through the upregulation of members of the Death Receptor pathways. When treating with VPA or DAC, we also observed some modifications in chromatin structure and DNA methylation pattern in a specific CpG rich region within the TRAIL promoter. These results suggest a partial direct epigenetic mechanism underlying TRAIL induction by DAC and VPA. We were able to confirm the results obtained in the APL model in an unrelated form of AML derived from the expression of AML1/ETO fusion protein, suggesting that a substantial fraction of AML patients could benefit from epigenetic treatments, and further reinforcing our view that the APL model represents a paradigm for the study of epigenetic drugs.