DNA methylation patterns predict treatment response in MDS patients

DNA methylation patterns present at diagnosis could help predict which patients with myelodysplastic syndromes (MDS) will respond to azacitidine. That's the key finding of a new multicenter study involving researchers at Sylvester Comprehensive Cancer Center, part of the University of Miami Miller School of Medicine. Published in the Journal of Clinical Investigation, the study found that DNA methylation was more useful than gene activity alone in predicting treatment response, offering a potential path toward more personalized care for patients with MDS.

For patients with MDS, choosing a treatment comes with a stubborn challenge: Doctors cannot reliably predict which patients will respond to treatment.

The new study suggests that biological features present at diagnosis could help physicians identify patients most likely to benefit from therapy and points to a possible way forward. By examining DNA methylation – so-called epigenetic changes – alongside gene expression, mutations and clinical information, researchers developed a model that could predict response to azacitidine, a primary therapy used for MDS.

"Our goal is to get closer to answering one of the most important questions for a patient at diagnosis: Which treatment is most likely to work for me?" said Maria "Ken" Figueroa, M.D., Sylvester researcher, associate director of translational research and the study's co-lead. "These findings suggest that the epigenetic features of the disease could provide information we have not been able to capture with conventional approaches."

One of the Holy Grails of treating cancer is to determine whether or not a therapy will work before we even get started. While further research will be needed before these findings can guide routine care, this is a step in the right direction. But this finding offers a step toward more informed treatment decisions."

Mikkael Sekeres, M.D., M.S., chief and professor of hematology at Sylvester and co-lead of the study

MDS is a group of disorders in which bone marrow does not produce healthy blood cells normally. A hallmark is altered DNA methylation: changes to chemical marks on DNA that help turn genes on or off without changing the DNA sequence.

Drugs known as DNA methyltransferase inhibitors are used to treat MDS, but reliable biomarkers of treatment response have been lacking.

Researchers examined DNA mutations, chemical marks on DNA and gene activity in 98 patients with MDS. They grouped patients by their DNA methylation patterns, developed computer models to predict azacitidine response and tested the models in an independent patient group.

 "We sought to capture a fuller picture of the disease and understand which features could help distinguish patients who respond to azacitidine," said first author Qin Yang, Ph.D., assistant scientist at Miller School.

The analysis identified seven groups with distinct biological features. In one, referred to as Cluster VI, which was characterized by having relatively few DNA mutations, 71% of patients responded to azacitidine.

DNA methylation was more useful than gene activity alone for predicting treatment response. Importantly, some of the strongest predictive signals came from regulatory regions beyond the parts of DNA researchers have traditionally focused on.

"These findings give us a new lens for understanding why some patients benefit from treatment," said Figueroa. "That opens an opportunity to think more broadly about how we use epigenetic information to understand MDS and, eventually, to guide care."

While there is more work ahead, studies like this provide those in the field a reason to be hopeful, noted Sekeres.

"Our goal is to turn a better understanding of MDS into better guidance for patients, so they can face treatment with greater confidence in the path ahead," he said.