Getting to the root of a hard-to-treat childhood leukemia

Giving chromosomes their structure and shape, strands of DNA, shown in gray, are coiled around histones, depicted as spheres. Scott Armstrong thinks that drugs that block a particular histone methylation pathway could be the key to treating a rare but devastating childhood leukemia. (Courtesy Eric Smith/DFCI)

In the 40 years of the war on cancer, there is probably no greater success story than that of childhood leukemias. Once nearly uniformly fatal, some forms of acute lymphoblastic (ALL) and acute myeloid (AML) leukemias can now be cured in 80 or even 90 percent of cases.

The prognosis for the remaining 10 to 20 percent is not as good, especially if the cancer involves a reshuffling or rearrangement of the mixed lineage leukemia (MLL) gene. “We still only achieve about 50 percent success in treating these MLL-rearranged leukemias,” according to Scott Armstrong, a pediatric oncologist at Children’s and Dana-Farber Cancer Institute. “We need to find better ways of caring for these patients.”

Armstrong and his colleagues may have just given patients with MLL-rearranged leukemias a leg up by finding and exploiting a core epigenetic vulnerability in this type of cancer.

“The oncology field is very excited about epigenetic inhibition right now,” says Armstrong. “What we’ve done is show that we can block one of these enzymes and get very specific anti-tumor activity in a previously very hard-to-treat disease.”

Let’s step back for a moment and lay some groundwork: In 2001, Armstrong and his mentor, Dana-Farber’s late Stanley Korsmeyer, found that leukemias with reshuffled MLL genes had gene expression patterns quite distinct from ALL and AML leukemias. In these MLL cancers – which account for some 10 percent of all children and adults diagnosed with ALL or AML – a portion of chromosome 11 (where the MLL gene resides) breaks off and fuses with parts of other chromosomes to create new fusion proteins. The fusion proteins subvert the normal function of the MLL gene and activate a set of leukemia-causing genes.

Seven years later, Armstrong led a team that found that in MLL leukemia cells an unusually high proportion of histones – scaffolding proteins that help manage gene activation – near specific cancer-causing genes have small chemical tags called methyl groups attached to them. This kind of chemical modification, called methylation, is a well-known epigenetic mechanism that greatly influences how a cell behaves. In this case, the methylated histones drive the distinct gene expression program that fuels the cancer.

MLL (top) and Dot1l conspire to wreak genomic havoc in MLL leukemias. (Emw/Wikimedia Commons)

In that same study, Armstrong’s team also found that the MLL fusion proteins call on an enzyme called Dot1l to do their dirty work of methylating the histones, making the enzyme a prime target for therapeutic development. The case is even stronger now thanks to a pair of Cancer Cell papers published this week. In the first paper, Armstrong, with Kathrin Bernt and Andrew Kung of Children’s and Dana-Farber, found that they could make the cancerous gene expression and methylation patterns in MLL cells go away by knocking out Dot1l, confirming that the enzyme’s methylating activity lies at the root of the genomic havoc seen in MLL-rearranged cells.

Says Armstrong, “We now know that these leukemias fully rely on this enzyme and the methylation pattern it generates in order to persist and grow.” Which is good news: “While methylation tags on histones are very difficult to manipulate directly,” he continues, “Dot1l is much easier to target therapeutically.”

The second Cancer Cell paper, which included collaborators at a biotechnology company called Epizyme, takes the first paper’s findings a step further by showing that a small molecule called EPZ004777, which interferes with Dot1l, does indeed eliminate the abnormal methylation pattern in MLL cells. In culture, the molecule caused MLL-rearranged leukemia cells to die off in about two weeks’ time; it also prevented mice injected with MLL leukemia cells from developing cancer.

“We have known for a while that MLL leukemias arise from widespread alterations not in the genetic code itself, but in the structure of the DNA and the proteins associated with it,” Armstrong says in a recent press release. “We now show that these epigenetic changes indeed turn on cancer-promoting genes within white blood cells, and ultimately cause the leukemia.

“Even more importantly,” he concludes, “we show that we can reverse the process.”

Bernt KM, Zhu N, Sinha AU, Vempati S, Faber J, Krivtsov AV, Feng Z, Punt N, Daigle A, Bullinger L, Pollock RM, Richon VM, Kung AL, & Armstrong SA (2011). MLL-Rearranged Leukemia Is Dependent on Aberrant H3K79 Methylation by DOT1L. Cancer Cell, 20 (1), 66-78 PMID: 21741597

Daigle SR, Olhava EJ, Therkelsen CA, Majer CR, Sneeringer CJ, Song J, Johnston LD, Scott MP, Smith JJ, Xiao Y, Jin L, Kuntz KW, Chesworth R, Moyer MP, Bernt KM, Tseng JC, Kung AL, Armstrong SA, Copeland RA, Richon VM, & Pollock RM (2011). Selective Killing of Mixed Lineage Leukemia Cells by a Potent Small-Molecule DOT1L Inhibitor. Cancer Cell, 20 (1), 53-65 PMID: 21741596