New Technique Finds Vulnerabilities in Deadly Leukemias

Using a powerful new experimental system, researchers at Columbia University have characterized the genetic changes that drive the development of the most aggressive form of leukemia. The project, described in the current issue of Blood, lays the groundwork for a deeper understanding of this poorly understood cancer, and suggests novel strategies for treating it.  

“Human cells have 23 pairs of chromosomes, but most solid cancers don’t,” says Sergei Doulatov, PhD, associate professor of physiology and cellular biophysics at Columbia and a member of the Herbert Irving Comprehensive Cancer Center (HICCC). While common in solid tumors, aberrant chromosome number, a condition called aneuploidy, is less common in blood cancers. “But there’s this subset of leukemias called AML-CK (acute myeloid leukemia with complex karyotype) that do have aneuploidy, very much like solid tumors, and are very genomically unstable,” says Doulatov. AML-CK is also aggressive and exceptionally difficult to treat.  

It’s also been hard to study. Most patient samples come from advanced cases, where the cells already carry multiple gene mutations and chromosome changes, so researchers didn’t know which were essential to AML-CK development and which were side effects of the cells’ runaway growth. But after an extensive search, Philip Creamer, an associate research scientist in Doulatov’s lab, identified a clone from one patient’s bone marrow that contained just one chromosomal abnormality. “For this particular case, we have the first … mutation that happened in the patient, [which] suggested this clone was in the patient’s marrow before progressing to the complex karyotype stage,” says Creamer. 

Collaborating with researchers at Columbia and several other institutions, Creamer, an associate research scientist in the Doulatov lab, devised a new system for dissecting AML-CK’s molecular biology. The approach uses induced pluripotent stem cells, allowing the team to derive multiple cell types that all share a common, unique genetic origin from the patient they came from.  

“Most patients with this type of leukemia have what’s called therapy related leukemia, where prior chemotherapy for solid cancer acts as a trigger [for] these clones, which is what we’ve modeled in the dish,” says Doulatov. In this system, cells that lacked a chromosome piece called 5q underwent a dramatic progression to aneuploidy after treatment with a genome destabilizing compound. Cells with 5q intact did not. 

The 5q deletion is a well-known feature of AML-CK and has long been suspected to be a key step in the condition’s pathogenesis, so the team’s initial result wasn’t surprising. However, being able to give a petri dish of cells a case of the disease is a powerful new capability. “With this work, we now have the first experimental model to dissect many hypotheses about how this disease originates ,” says Creamer. 

If all of these cancers have a shared Achilles heel that you can go after, then that presents another way for us to think about attacking [them] that we hadn’t thought about before.

Using a battery of sophisticated single-cell analytical techniques, the scientists then probed the genetic programs involved in AML-CK progression. “We’re able to say which clones are present at different time points, who is competing with who, who has the highest fitness, but also what’s their gene expression profile?” says Creamer. The investigators then compared those gene expression profiles with data from other patient samples and found a common gene expression signature shared across AML-CK patients.

Researchers had previously assumed that after its initial 5q loss, each case of AML-CK followed a unique path to malignancy, which would make drug development nigh impossible. The new work overturns that pessimistic view. “If all of these cancers have a shared Achilles heel that you can go after, then that presents another way for us to think about attacking [them] that we hadn’t thought about before,” says Doulatov.