Reprogramming cancer cells to treat an aggressive type of leukemia

Though AML is a genetically heterogeneous disease, all its subtypes share a common feature: impaired differentiation of myeloid progenitor cells in the bone marrow. This differentiation block results in the accumulation of immature precursors of these cells within the bone marrow and circulation, ultimately impairing normal processes by which blood cells are replenished (hematopoiesis) and other essential biological functions.

Researchers co-led by Ludwig Oxford's Yang Shi and Amir Hosseini -- as well as Abhinav Dhall at Shi's laboratory at Harvard Medical School and colleagues at the University of Pennsylvania and the University of Helsinki -- report in the current issue of Nature a potentially new combination therapy that, in preclinical studies, treats some AML subtypes by dismantling that barrier to differentiation in two mechanistically distinct ways.

"The drug combination we have identified works by activating genes that drive cell differentiation while suppressing genes that promote cell proliferation and cancer growth," said Shi.

The inability of cellular precursors to differentiate into mature myeloid cells, a defining characteristic of AML, has long pointed to a strategy for treating the cancer: the possibility of employing drugs to circumvent that developmental blockade. Indeed, a subtype of AML -- acute promyelocytic leukemia (APL) -- is already treated this way using a pair of drugs (all-trans retinoic acid and arsenic trioxide) that shove APL cells down the differentiation process. The combination cures about 95% APL cases, but there remains a pressing need to identify similar strategies for the treatment of other AML patients.

One way to circumvent the differentiation blockade is by targeting the dysfunctional gene expression programs that drive the phenomenon in leukemic stem cells. Such changes are induced by the aberrant activity of enzymes that chemically modify DNA and its histone protein packaging to regulate gene expression. One such "epigenetic" enzyme, LSD1 -- discovered in 2004 by Shi and his colleagues and shown to erase methyl groups that are tacked on to histones -- is expressed at high levels in AML cells and known to help maintain leukemic stem cells.

"While LSD1 inhibitors have been developed and shown to induce differentiation in AML stem cells, they've had limited success in clinical studies owing to their toxicity when used alone," said Hosseini. "To limit that toxicity, we thought we'd try to identify other drugs that could synergize with LSD1 inhibitors to overcome the differentiation arrest and suppress the proliferation of cancer cells."

Using mouse leukemic cells, the researchers screened multiple molecules for such synergies, eventually settling on an inhibitor of the GSK3α/β enzyme. The GSK3 inhibitor is already being evaluated as a cancer drug in clinical trials and is well tolerated by patients. When combined with a low dose of the LSD1 inhibitor, the GSK3 inhibitor induced differentiation in laboratory cultures of multiple subtypes of AML and suppressed cell proliferation.

Hosseini, Shi and colleagues then demonstrated that the treatment induced differentiation of leukemic cells, inhibited their proliferation and extended survival of mice engrafted with human AML cells. Further, their experiments indicated that the drug combination selectively targets leukemic cells -- not healthy hematopoietic ones -- lowering the risk of toxicity in patients.

"We are also encouraged by the observation that the gene expression signature induced in leukemic cells by this combination therapy correlates with that observed in the cancer cells of AML patients who live relatively longer," said Hosseini.

The researchers describe the molecular mechanisms by which combination therapy re-wires gene-expression programs to suppress the stem cell-like traits of leukemic cells that drive the cancer and to promote differentiation, which may have important therapeutic implications for other cancers that are driven by the overactivation of WNT signaling pathway.

"Our findings provide compelling evidence to support the testing of this combination therapy in AML patients, especially since both of the inhibitors involved are not only available but have been developed for human use and are currently being evaluated in the clinical trials," said Shi.

This study was supported by Ludwig Cancer Research, the National Institutes of Health, the Research Council of Finland, the Cancer Foundation Finland, the Sigrid Jusélius Foundation, the National Institute for Health Research, the Oxford Biomedical Research Centre and Cancer Research UK.

In addition to his Ludwig post, Yang Shi is a Professor in the Nuffield Department of Medicine at the University of Oxford.