Molecular Signature May Predict Therapy Response in CMML Patients

A distinct epigenetic DNA signature may distinguish those with chronic myelomonocytic leukemia (CMML) that respond to treatment with decitabine (Dacogen/DAC), an injectable drug that targets DNA methyltransferase. These specific DNA methylation signatures may be able to predict whether or not a patient would respond to the drug at diagnosis.

The study is published in the March 30, 2015 issue of The Journal of Clinical Investigation.  

Both CMML and myelodysplastic syndromes have mutations in genes that encode enzymes that control DNA methylation (an epigenetic signature on DNA), resulting in aberrant DNA methylation in patients with these hematologic malignancies. A current treatment approach is to inhibit the enzymes that add a methyl group to DNA, known as DNA methyltransferases. One DNA methyltransferase inhibitor is decitabine, an infused cytidine nucleoside analog. While DNA methyltransferase inhibitors (DMTis) are approved for the treatment of CMML, not all patients respond, and it is not currently possible to predict which patient will respond to treatment. 

Currently, only about half of patients treated with decitabine respond as indicated by a hematological improvement. Furthermore, as much as 6 months of therapy is needed to know whether a patient has responded to decitabine.

To identify potential biomarkers of response, Kristen Meldi, PhD, a postdoctoral fellow at the University of Michigan Medical School in the department of pathology, Ann Arbor, Mich., and colleagues examined gene expression, DNA methylation patterns, and mutations using next generation sequencing in 40 CMML patients; 20 who were responsive and 20 who were resistant to decitabine. The authors used bone marrow samples from the patients at the time of their diagnosis and after decitabine treatment.

Somatic mutations did not differ between responders or those who were resistant to treatment, but the authors identified 167 regions with differing DNA methylation patterns at baseline that could distinguish between responders and nonresponders.

The differing methylation regions were predominantly located in nonpromoter regions, and overlapped with DNA regulatory elements called distal regulatory enhancers. Using the DNA methylation patterns detected, Meldi and colleagues developed an epigenetic classifier that could accurately predict decitabine response at diagnosis and prior to therapy initiation.

An additional transcriptional analysis--genome-wise--also predicted differences in gene expression between responders and nonresponders. Genes found to be upregulated in those CMML patients who responded to decitabine included cell cycle regulating genes.

While the precise mechanism of action of decitabine is still disputed, the agent is known to be incorporated into DNA during the replication phase of the cell cycle, covalently trapping DNA methyltransferases and targeting these enzymes for degradation. This drug can also cause DNA damage.

The work suggests that DNA methylation signatures may be a way to determine the type of therapy a CMML patient should receive. But further studies, using larger independent cohorts, are needed to better understand whether the signature identified can indeed predict responsiveness to decitabine. Additionally, studies are also needed to test whether this signature can also predict response to decitabine for myelodysplastic syndrome patients.

Still, the "present study demonstrates not only that DNA methylation differences exist between patients with different responses to DAC, but that these DNA methylation differences are sufficiently robust to be harnessed for use in the clinic as accurate classifiers," concluded the study authors.