Circulating exosomal microRNAs in acquired aplastic anemia and myelodysplastic syndromes

Abstract

Exosomal microRNAs modulate cancer cell metabolism and the immune response. Specific exosomal microRNAs have been reported to be reliable biomarkers of several solid and hematologic malignancies. We examined the possible diagnostic and prognostic values of exosomal microRNAs in two human bone marrow failure diseases: aplastic anemia and myelodysplastic syndromes. After screening 372 microRNAs in a discovery set (n=42) of plasma exosome samples, we constructed a customized PCR plate, including 42 microRNAs, for validation in a larger cohort (n=99). We identified 25 differentially expressed exosomal microRNAs uniquely or frequently present in aplastic anemia and/or myelodysplastic syndromes. These microRNAs could be related to intracellular functions, such as metabolism, cell survival, and proliferation. Clinical parameters and progression-free survival were correlated to microRNA expression levels in aplastic anemia and myelodysplastic syndrome patients before and after six months of immunosuppressive therapy. One microRNA, mir-126-5p, was negatively correlated with a response to therapy in aplastic anemia: patients with higher relative expression of miR-126-5p at diagnosis had the shortest progression-free survival compared to those with lower or normal levels. Our findings suggest utility of exosomal microRNAs in the differential diagnosis of bone marrow failure syndromes. (Registered at clinicaltrials.gov identifiers: 00260689, 00604201, 00378534, 01623167, 00001620, 00001397, 00217594).

Trial registration: ClinicalTrials.gov NCT00260689 NCT00604201 NCT00378534 NCT01623167 NCT00001620 NCT00001397 NCT00217594.

Copyright© 2018 Ferrata Storti Foundation.

Figures

Figure 1.

miRNAs in severe aplastic anemia (SAA) and myelodysplastic syndromes (MDS) patients compared to healthy controls (HC) in the discovery set. Principal component analysis was performed to compare 384 exosomal miRNA expression levels: SAA versus HC (A), MDS versus HC (B), and MDS versus AA (C). Results are shown using volcano plots and tables. In volcano plots, x- and y-axes show estimated expression difference measured in Log2(FC) and the significance of the expression difference measured in −Log10(P-value), respectively. In the plots, horizontal and vertical lines indicate cut-off of significance (P<0.05) and expression levels greater than ±1.5-fold regulation (FR), respectively. For each comparison, miRNAs with ±1.5 FR and P<0.1 are displayed in correspondent tables in which P<0.05 is highlighted in bold.

Figure 2.

Validation of miRNA signatures in severe aplastic anemia (SAA) and myelodysplastic syndromes (MDS) patients. Principal component analysis was employed to compare the 48 miRNA expression levels in the validation set: SAA versus healthy controls (HC) (A), MDS versus HC (B), and MDS versus SAA (C). These results are shown with volcano plots in a similar manner as described in Figure 1. (D) Hierarchical clustering visualizes the 48 exosomal miRNAs in SAA, MDS, SAA-responders, SAA-non-responders, and HC. A red-green color scale indicates normalized miRNA expression levels (red: maximum; green: minimum).

Figure 3.

Exosomal miRNAs in severe aplastic anemia (SAA) and myelodysplastic syndromes (MDS) patients. Relative expression levels of the 48 miRNAs were calculated as Log2FC and shown for each group [SAA, MDS, and healthy controls (HC)]. (A) 7 exosomal miRNAs differentially expressed in both SAA and MDS compared to HC. (B) Receiver operating characteristic (ROC) curves for 4 miRNAs in AA and MDS. The ROC curve of the miRNA panel was generated based on the predicted probability for each patient and using the healthy group as a control. P<0.05 was considered statistically significant. AUC: area under the curve; CI: Confidence Interval.

Figure 4.

Exosomal miRNAs and their correlation with prognosis in severe aplastic anemia (SAA) responders. Relative expression levels were calculated as Log2FC for all 48 miRNAs in SAA patients and compared before and after immunosuppressive therapies (IST). (A) Only miR-4651 and miR-126-5p were significantly decreased after treatment in SAA-responders. (B) For these 2 exosomal miRNAs, receiver operating characteristic (ROC) curves were generated as described in Figure 3 using a healthy group as a control. (C) Relative expression of differentially expressed exosomal miRNAs at diagnosis and after treatment in responders (R) and non-responders (NR) of SAA patients were compared to healthy controls (HC) using one-way ANOVA with Kruskal-Wallis and the two-stage linear step-up procedure of Benjamini, Krieger and Yekutieli tests. Data are shown as mean+Standard Deviation (SD). (D) After calculation of progression-free survival (PFS), patients were divided into three groups according to miRNA relative expression shown as Log2FC at diagnosis of selected miRNAs. P<0.05 was considered statistically significant. AUC: area under the curve; FR: fold regulation.

Figure 5.

Pathway analysis using differentially expressed exosomal miRNAs. (A) VENNY (an interactive tool for comparing lists with Venn Diagrams) was used to find common or unique miRNAs among severe aplastic anemia (SAA), myelodysplastic syndromes (MDS), and SAA-responder patients. miRNAs classified into individual groups are listed accordingly. Red: increased exosomal miRNAs; blue: decreased exosomal miRNAs; miR-3200-3p is shown in black because of different expression profiles between SAA (down-regulated) and MDS (up-regulated). Predicted targeted genes of miRNAs exclusively expressed in SAA or MDS were used for pathway analysis by IPA software. Top 10 pathways in SAA (B) and the top 20 in MDS (C) are shown. (D) Venn diagram shows the number of unique or common pathways in SAA and MDS and a list of the 15 common signaling pathways.