American Society of Hematology

MicroRNAs in the Plasma: Not miRly a New Biomarker

Erin HertleIn, PhD
John C. Byrd, MD

Published on: July 01, 2011

Drs. Hertlein and Byrd indicated no relevant conflicts of interest. 

Moussay E, Wang K, Cho JH, et al. MicroRNA as biomarkers and regulators in B-cell chronic lymphocytic leukemia. Proc Natl Acad Sci USA. 2011;108:6573-6578.

Perhaps the most important factors in the treatment of chronic lymphocytic leukemia (CLL) are deciding when to initiate treatment and which treatment is best for a particular patient. The typical watch-and-wait approach dictates that treatment is not initiated until symptoms develop. This is mainly due to the risk of adverse side effects of common therapies, particularly when patients may not become symptomatic for many years following the initial diagnosis. However, if predicted correctly, aggressive cases of CLL would most definitely benefit from immediate treatment.

Many prognostic factors that have been determined to predict time-to-treatment or disease severity, such as Zap70 expression, are difficult to reproducibly measure. Immunoglobulin heavy chain mutational status and select cytogenetic markers can also classify disease and predict severity and are therefore very important diagnostic parameters. However, the detection of these abnormalities can be cost- and labor-intensive and is therefore not performed on a routine basis at most institutions. More accessible circulating markers, such as CD23, thymidine kinase, and β2-microglobulin, have not been as extensively validated as reliable markers. However, the ability to use plasma markers to predict disease remains a desirable objective.

The recent study by Moussay and colleagues from Luxembourg explores the use of microRNAs (miRNAs), particularly miRNAs circulating in the plasma of CLL patients, to predict CLL disease status. These investigators found that many miRNAs are overexpressed in CLL patient samples compared to normal volunteers, not only in the cells but in the plasma compartment as well. The authors found that plasma levels of miR-195 and miR-20a are able to reproducibly distinguish CLL samples from normal controls, and the level of miR-20a correlates with disease severity. In addition, combined analysis of several plasma miRs, with particular emphasis on miR-29a, is able to distinguish between CLL cases that are positive or negative for Zap70 expression.

Interestingly, several miRNAs (mir-135*, miR-451, and miR-486-5P) show an unusual pattern of regulation, in that they are expressed at lower levels in the actual CLL cells compared to the serum. This suggests that the approaches described above, which focus only on the CLL cells, may not paint an accurate picture of the miRNA profile in patient samples. This is because they neglect to take into account miRNAs that are rapidly exported from the cell or miRNAs that are produced by other cells in the tumor microenvironment.

The overall importance of this study is not only that miRNAs can be used as prognostic factors to accurately predict disease state, but also that the composition of miRNAs in the plasma of patients might actually be a more accurate diagnostic tool. In addition, the development of recent technologies that allows analysis of the entire miRNome in a single assay without the need for RNA extraction or cDNA synthesis make the detection of plasma miRNAs simple and cost effective, which makes this much more attractive as a clinical diagnostic tool.

back to top