Prognostic value of miR-155 in individuals with monoclonal B-cell lymphocytosis and patients with B chronic lymphocytic leukemia

Abstract

Noncoding RNAs play a pivotal role in the pathogenesis of chronic lymphocytic leukemia (CLL). We hypothesized that microRNAs (miRs) are involved in the transition from monoclonal B-cell lymphocytosis (MBL) to CLL and tested miR-15a/16-1 cluster, miR-21, and miR-155 expression in purified B cells of normal individuals, individuals with MBL, and patients with CLL. When we analyzed 224 samples from 2 independent training and validation cohorts, we found that miR-155 was overexpressed in B cells from individuals with MBL, and even more so in B cells from patients with CLL, when compared with B cells from normal individuals. Furthermore, we were able to identify miR-155 in circulating microvesicles from both individuals with MBL and patients with CLL. Next, to examine the prognostic role of miR-155, we measured its expression level in plasma samples collected before treatment initiation in 228 patients with CLL. We found significantly higher miR-155 expression levels in patients who failed to achieve a complete response compared with those who experienced complete response. Our findings support the use of cellular and plasma levels of miR-155 as biomarkers for the risk of progression in individuals with MBL, as well as to identify patients with CLL who may not respond well to therapy.

Figure 1

miR-155 levels in normal B cells, MBL cells, and CLL cells as measured by qRT-PCR. (A) miRNA expression in isolated and purified B cells from the training set (group A). (B) miRNA expression in isolated and purified B cells from the validation set (group B). Comparisons between MBL and Rai 0 for each set of samples are presented in panels C-D. We used 5S rRNA as a normalizer. Raw data for all samples are shown in supplemental Table 3. In panels A-B, the P values were calculated using the Kruskal-Wallis test.

Figure 2

Relationship between relative miR-155 expression levels in plasma and response to therapy in group C. (A) Relative expression levels of miR-155 in plasma were lower in patients treated with fludarabine, cyclophosphamide, and rituximab who experienced a complete response than in patients who experienced other treatment outcomes. (B) Relative expression levels of miR-155 in plasma according to type of response. Raw data for all plasma samples are shown in supplemental Table 4. CR, complete response; NPR, nodular partial response; NR, no response; PR, partial response. In panel B, the P value was calculated using the Kruskal-Wallis test.

Figure 3

Relationship between relative miR-155 expression levels in plasma and overall survival in patients in group C. (A) Kaplan-Meier survival curves according to relative expression levels of miR-155 in plasma (low, up to 56th percentile; high, above 56th percentile). (B) Correlation between plasma levels of miR-155 and serum B2M levels.

Figure 4

Relationship between relative miR-155 expression levels in plasma and response to treatment in patients in groups D and E. Relative expression levels of miR-155 in plasma were significantly higher in patients who achieved a complete response than in patients who experienced other types of clinical responses. (A) Patients who received fludarabine, cyclophosphamide, and rituximab as salvage therapy (group D). (B) Patients treated with lenalidomide (group E). The miRNA expression levels were unavailable for 7 patients (“Materials and methods”). Raw data for all plasma samples are shown in supplemental Table 4.