Abstract SCI-30

The FDA has approved four epigenetic therapies for oncology: Vidaza™ (azacitidine, 2004), Dacogen® (decitabine, 5-aza-2′-deoxycytidine, 2006), Zolinza™ (vorinostat, 2006), and Istodax® (romidepsin, 2009); and many more are in clinical and preclinical development. While we know the enzymatic targets of these agents, we still do not fully understand the true mechanism of action of these agents. Ostensibly, both classes of agents work to inhibit enzymes that silence genes with tumor suppressor function: DNA methyltransferases (DNMTs) and histone deacetylases (HDACs). Reduction of hypermethylation has been documented following inhibition of DNMT by 5-aza-2-deoxycytidine and 5-azacytidine. Significant clinical activity including durable responses and improvement in survival initially led to approval of these agents in myelodysplastic syndrome (MDS). Although demethylation has been confirmed following treatment with these agents, particularly in the low dose, chronic schedule developed for MDS, the extent of demethylation does not correlate with response, suggesting other factors are important. The genes critical for response are still under study – several known to be hypermethylated in MDS have been intensively studied, including the p15INK4B and the fragile histidine triad (FHIT) genes. There are many parallels for the HDAC inhibitors, which also inhibit enzyme activity and promote gene transcription. FDA approvals for vorinostat and romidepsin are based on clinical activity in cutaneous T-cell lymphoma (CTCL) including durable and complete responses. Activity in peripheral T-cell lymphomas has also been observed in clinical trials with romidepsin and other HDAC inhibitors. Increased global histone acetylation has been noted in clinical samples; however, there is no correlation of this marker with response to vorinostat. Romidepsin, on the other hand, requires activation and our clinical data have suggested a correlation between persistent histone acetylation and clinical response. The gene alterations critical for clinical activity still require definition. Although HDAC inhibitors reliably induce the CDK inhibitor p21, overall about 5% of genes assayed are upregulated by HDAC inhibitors in array studies, and an equal number downregulated. Notably, multiple cellular proteins are also affected by acetylation; some of them potentially responsible for the activity of the HDAC inhibitors. This may depend upon cellular context but in some cell types is likely to be as important if not more important than their effects on gene transcription. T-cell lymphomas in more than one-third of patients show sensitivity to the HDAC inhibitors, and it is not known which of the myriad activities is critical for the activity in the T-cell lymphomas. What we do know is that in a subset of patients, responses can be very good, and of very long duration. Critical needs for epigenetic therapies include a more precise determination of mechanisms of action, so that we may identify the subsets of patients who will have disease response. We need to define mechanisms of resistance that emerge after treatment fails if we hope to prevent that outcome. We need effective combinations with epigenetic therapies to increase their activity in the diseases currently known to be susceptible, but also to extend their range of activity. Epigenetic therapies – originally identified as differentiating agents – are unique among anticancer therapies in having the potential to actually reset the oncogenic phenotype. A significant research effort is needed to harness this potential.

Disclosures:

Bates:Gloucester Pharmaceuticals: Research Funding. Off Label Use: romidepsin efficacy in peripheral T cell lymphoma is experimental but efficacy data are available.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution