miRs: fine-tuning prognosis in CLL

In this issue of Blood, Stamatopoulos and colleagues present evidence that down-regulation of miR-29c and miR-223 in malignant B cells is an adverse prognostic indicator in CLL. The findings in this paper are remarkable in that they help refine disease stratification and prognosis.

Chronic lymphocytic leukemia (CLL) is a heterogeneous disease with a variable clinical course. Some patients start with an asymptomatic indolent disease that may or may not progress over their lifetime. Others present with an aggressive disease, requiring therapy soon after diagnosis.¹  Consequently, considerable effort has been focused on understanding molecular and clinical characteristics that are likely to predict the course of CLL. In past decades, high levels of specific serum markers (β2-microglobulin), lack of somatic hypermutations in the immunoglobulin-heavy-chain-variable-region (IgV_H), high levels of expression of the IgV_H surrogate markers (ZAP-70 mRNA and protein and lipoprotein lipase [LPL]),²  and adverse cytogenetics (11q- and 17p-) by fluorescent in situ hybridization were all found to be independent prognostic markers that predicted tumor progression and aggressive disease.¹ 

Recently, microRNAs (miRNAs), a new class of gene regulators, were discovered and found to play critical roles in CLL and other cancers.³  In addition to being causal in the pathogenesis of CLL,³  the expression levels of miRNA were found useful in predicting clinical behavior.^4,5  The first of these studies identified a unique miRNA signature that was differentially expressed in patients with varying IgV_H and ZAP70 kinase status.⁴  This study was followed by 2 others that generated largely nonoverlapping miRNA signatures when compared with the first study.^6,7  The disparity among their results probably reflects the different methodologies used to generate the miR profiles in each case and variation in the method of storage of patient samples. ^4,6,7  Nevertheless, each of these studies consistently identified 2 microRNAs, miR-223 and members of the miR-29 family, that were found to be down-regulated in CLL and associated with markers of aggressive disease^4,6  (see table).

This study by Stamatopoulos et al is the first to systematically evaluate the expression of miR-223 and miR-29c in relation to all clinically relevant prognostic factors in CLL.⁸  The authors wisely select for the investigation the 2 miRNAs identified as relevant in all major prior studies and find that the expression of miR-223 and miR-29c was decreased in patients that had a poor prognosis as defined by multiple parameters including Binet staging, measurements of tumor burden (soluble CD23, β2-microglobulin, lymphocyte doubling time), IgV_H status, ZAP70, LPL as well as cytogenetic abnormalities. Their study, although retrospective, indicated that low miR-223 and miR-29c was associated with inferior treatment free survival as well as a reduced overall survival. In addition, the authors devise a quantitative method for simultaneously determining the levels of miR-223, miR-29c, ZAP70, and lipoprotein lipase (LPL) in samples, allowing them to devise a polymerase chain reaction (PCR) score based on the expression of none to all of the above adverse prognostic markers within each sample. This technique allows for a restratification of CLL patients based on their combined PCR score and possibly offers a prognostic tool that is superior to individual prognostic markers used in isolation. These are remarkable findings on the path to useful marker discovery in clinical practice.

Future prospective studies that evaluate thousands of patients would be required to validate the prognostic value of miR-223 and miR-29c, and the value of the PCR tool in refining CLL risk stratification. Finally, standardization of serum/plasma assays for the accurate evaluation of miR-223 and miR-29c would aid in its development as a diagnostic marker for tumor progression in CLL.