Small trial hints at a new direction for epigenetic therapy

Epigenetic cancer therapies–drugs that reverse abnormalities in how DNA is packaged within the nucleus–clearly work for a while against some tumors, and some of these have been approved by FDA. Although they represent an important new class of therapies in hematological malignancies such as myelodysplasia or acute myeloid leukemia, their mechanism of action remains unknown. A number of possible mechanisms of action have been identified, including induction of cell-cycle arrest or inhibition of angiogenesis.

A recent study involving a subset of patients with acute myeloid leukemia or myelodysplasia who are taking part in a trial of two epigenetic drugs has unearthed a hitherto unknown, and very strong, effect that may expand their mechanisms of action. It appears that the epigenetic agents azacitidine and sodium valproate unmask the expression of potential tumor antigens, rendering the malignant cells vulnerable to attack by cytotoxic lymphocytes.

The drugs in the study represent both major forms of epigenetic agents: DNA methyltransferase (DNMT) inhibitors that prevent the biochemical masking and "silencing" of genes, and histone deacetylase (HDAC) inhibitors that have the effect of opening the DNA backbone so that genes can be expressed, activating their biological function. The two epigenetic processes are closely linked in normal biology. In cancer therapy, the theory is that they can reactivate tumor suppressor genes that have become silenced during the oncogenic process.

The larger study at the University of Birmingham in England was a non-randomized trial to test the safety and efficacy of four anti-leukemic drugs: Theophylline, all-trans retinoic acid, the DNMT inhibitor azacitidine and the HDAC inhibitor sodium valproate.  (A previous study at M.D. Anderson has already studied the latter three of these drugs.)

Knowing that the antigens used in some cancer vaccines are heavily masked by methylation on tumor cells, Oliver Goodyear, Charles Craddock, and coworkers pondered whether the last two drugs (the ones with epigenetic effects) might make them more visible to the immune system. Correlating antigen recognition by T cells with clinical response, they had a tantalizing revelation.

Their studies of the drugs' effects in leukemic cell lines focused their attention on MAGE (melanoma-associated antigen), a tumor suppressor gene that immune cells recognize on several solid tumor cell lines and on leukemic cells. Among 21 patients in the sub-study, only 15 were well enough to complete the analysis. Before drug treatment, only one patient had any detectable immune response to MAGE. But for 10 others, azacitadine and valproate induced a CD8+ cytotoxic lymphocyte response specific to MAGE. This response increased for all 11 patients during the course of therapy.

More impressive was the clinical response. Among these 11 patients, 8 showed major clinical responses that lasted from 2 to 27 months. Four showed complete remission. This was significantly higher than the overall response rate in the larger trial, the team observes in Blood.

So among all the unknowns about epigenetic therapy, add a few more: If DNMT and HDAC inhibitors have an immunological effect on tumors in their own right, for which cancers and patients is it effective? How does the duration of that response correlate with currently accepted regimens? Would these drugs also boost the effects of cancer vaccines?

This appears to be the first time an epigenetic treatment has produced a clinical response against a known tumor antigen. Although it's not (yet) directly proven, logic suggests that the epigenetic therapy enabled or at least boosted immune recognition of the malignant cells. Prospective studies are underway in Birmingham.