Aplastic anemia (AA) and myelodysplastic syndrome (MDS) are acquired hematologic diseases in which bone marrow failure predominates, leading to life-threatening anemia, neutropenia, and thrombocytopenia. Clinically, it may be difficult to distinguish AA from MDS, especially when blood count depression is moderate or marrow hypocellularity is variable. Many studies have focused on immune destruction of hematopoietic stem and progenitor cells in AA and cytogenetic abnormalities in MDS, but comprehensive studies on circulating cytokine profiles of patients with these disorders are scarce. Here, we used multiplex bead array assays (Luminex) to detect 31 cytokines in the plasma of 25 untreated AA patients, 35 MDS patients (including 6 RCMD [refractory cytopenia with multilineage dysplasia], 6 RA-RS [refractory anemia with ringed sideroblasts], 6 MDS-U [MDS, unclassified], 8 RAEB [refractory anemia with excess blasts], 5 MDS 5q- [MDS associated with isolated del(5q)], and 4 RA [refractory anemia]) and 36 healthy controls in order to understand the potential contributions of these cytokines to the pathophysiology of AA and MDS and to provide a possible assistance in differential diagnosis. The measurements were transformed using a natural log-transformation to reduce the skewness of the distributions, and pairwise t-tests were used to compare the means among AA, MDS and healthy controls. The results are summarized in Table 1. AA patients showed significantly higher levels of thrombopoietin (Tpo) and granulocyte colony-stimulating factor (G-CSF) but much lower levels of CD40 ligand (CD40L), Eotaxin, C-X-C chemokine 5 (CXCL5), chemokine (C-C motif) ligand 5 (CCL5), and CXCL11 than MDS patients or healthy controls. Both AA and MDS patients showed higher levels of IL-6 and CCL-3, but lower levels of epidermal growth factor (EGF) than controls. However, levels of vascular endothelial growth factor (VEGF), tumor necrosis factor-a (TNF-a), monocyte chemoattractant protein-1 (MCP-1), interleukin (IL)-1ra, hepatocyte growth factor (HGF), CCL4, interferon (INF)-g inducible protein-10 (IP-10), and IL-8 were higher in MDS only and not elevated in AA. There was no difference in Leptin among AA, MDS patients, and controls. INF-g, IL-1b, IL-2, IL-10, IL-4, IL-5, IL-12p70, IL-13, IL-17, fibroblast growth factor (FGF) basic, granulocyte-macrophage colony-stimulating factor (GM-CSF), and IL-1a were undetectable in most samples. We investigated alterations in CD40L, Eotaxin, CXCL5, and CCL5 in 9 AA patients retrospectively as their clinical status changed: levels of these cytokines positively correlated with hematologic recovery, indicating that they might be useful as prognostic markers of clinical outcome. In summary, we found both similarities and distinctions in cytokine profiles between AA and MDS. High concentrations of Tpo and G-CSF, in combination with low levels of CD40L, Eotaxin, CXCL5, CCL5 and CXCL11 may represent a cytokine signature for AA, while elevated MCP-1, IL-1ra, VEGF and TNF-a may be the cytokine signature for MDS. VEGF may reflect the tumor-associated angiogenic properties described in some cases of MDS. Our results suggest that plasma cytokine screening may be practically useful to distinguish between AA and MDS and also allow inferences concerning pathophysiology in marrow failure.

Table 1. Cytokine levels (log-transformed [pg/mL]) in the plasma of patients with AA, MDS and HC cytokine (Mean ± SE) p values

cytokine (Mean ± SE)p values
AAMDSHCAA vs HCMDS vs HCAA vs MDS
AA: aplastic anemia; MDS: myelodysplastic syndrome; HC: healthy controls; SE: standard error; NS: not significant, p > 0.05. 
G-CSF 4.63 ± 0.42 2.28 ± 0.25 0.89 ± 0.16 <0.0001 <0.0001 <0.0001 
Tpo 7.61 ± 0.07 6.36 ± 0.13 5.69 ± 0.05 <0.0001 <0.0001 <0.0001 
CD40L 5.80 ± 0.19 7.61 ± 0.18 7.60 ± 0.11 <0.0001 NS <0.0001 
Eotaxin 5.32 ± 0.16 6.49 ± 0.09 6.34 ± 0.14 <0.0001 NS <0.0001 
CXCL5 4.27 ± 0.23 5.91 ± 0.23 7.01 ± 0.13 <0.0001 0.0001 <0.0001 
CCL5 7.34 ± 0.21 9.19 ± 0.17 9.71 ± 0.06 <0.0001 0.0062 <0.0001 
CXCL11 4.64 ± 0.26 6.80 ± 0.12 5.55 ± 0.23 0.0132 <0.0001 <0.0001 
CCL3 1.67 ± 0.48 4.56 ± 0.16 0.45 ± 0.21 0.0273 <0.0001 <0.0001 
IL-6 1.47 ± 0.30 1.85 ± 0.14 0.52 ± 0.07 0.0044 <0.0001 NS 
EGF 2.35 ± 0.21 2.52 ± 0.31 3.53 ± 0.28 0.0013 0.0184 NS 
MCP-1 5.34 ± 0.14 5.70 ± 0.11 5.05 ± 0.08 NS <0.0001 0.0388 
IL-1ra 6.15 ± 0.12 7.12 ± 0.12 6.36 ± 0.08 NS <0.0001 <0.0001 
VEGF 2.10 ± 0.30 3.24 ± 0.17 2.50 ± 0.24 NS 0.0172 0.0033 
TNF-a 0.81 ± 0.29 2.26 ± 0.13 0.80 ± 0.14 NS <0.0001 0.0001 
IL-8 3.34 ± 0.38 3.21 ± 0.20 2.64 ± 0.16 NS 0.0307 NS 
CCL4 4.50 ± 0.40 4.31 ± 0.17 3.78 ± 0.13 NS 0.0154 NS 
HGF 5.18 ± 0.15 5.55 ± 0.12 4.99 ± 0.10 NS 0.0006 NS 
IP-10 4.50 ± 0.22 4.89 ± 0.15 4.15 ± 0.12 NS 0.0002 NS 
Leptin 9.12 ± 0.25 8.99 ± 0.20 9.46 ± 0.22 NS NS NS 
cytokine (Mean ± SE)p values
AAMDSHCAA vs HCMDS vs HCAA vs MDS
AA: aplastic anemia; MDS: myelodysplastic syndrome; HC: healthy controls; SE: standard error; NS: not significant, p > 0.05. 
G-CSF 4.63 ± 0.42 2.28 ± 0.25 0.89 ± 0.16 <0.0001 <0.0001 <0.0001 
Tpo 7.61 ± 0.07 6.36 ± 0.13 5.69 ± 0.05 <0.0001 <0.0001 <0.0001 
CD40L 5.80 ± 0.19 7.61 ± 0.18 7.60 ± 0.11 <0.0001 NS <0.0001 
Eotaxin 5.32 ± 0.16 6.49 ± 0.09 6.34 ± 0.14 <0.0001 NS <0.0001 
CXCL5 4.27 ± 0.23 5.91 ± 0.23 7.01 ± 0.13 <0.0001 0.0001 <0.0001 
CCL5 7.34 ± 0.21 9.19 ± 0.17 9.71 ± 0.06 <0.0001 0.0062 <0.0001 
CXCL11 4.64 ± 0.26 6.80 ± 0.12 5.55 ± 0.23 0.0132 <0.0001 <0.0001 
CCL3 1.67 ± 0.48 4.56 ± 0.16 0.45 ± 0.21 0.0273 <0.0001 <0.0001 
IL-6 1.47 ± 0.30 1.85 ± 0.14 0.52 ± 0.07 0.0044 <0.0001 NS 
EGF 2.35 ± 0.21 2.52 ± 0.31 3.53 ± 0.28 0.0013 0.0184 NS 
MCP-1 5.34 ± 0.14 5.70 ± 0.11 5.05 ± 0.08 NS <0.0001 0.0388 
IL-1ra 6.15 ± 0.12 7.12 ± 0.12 6.36 ± 0.08 NS <0.0001 <0.0001 
VEGF 2.10 ± 0.30 3.24 ± 0.17 2.50 ± 0.24 NS 0.0172 0.0033 
TNF-a 0.81 ± 0.29 2.26 ± 0.13 0.80 ± 0.14 NS <0.0001 0.0001 
IL-8 3.34 ± 0.38 3.21 ± 0.20 2.64 ± 0.16 NS 0.0307 NS 
CCL4 4.50 ± 0.40 4.31 ± 0.17 3.78 ± 0.13 NS 0.0154 NS 
HGF 5.18 ± 0.15 5.55 ± 0.12 4.99 ± 0.10 NS 0.0006 NS 
IP-10 4.50 ± 0.22 4.89 ± 0.15 4.15 ± 0.12 NS 0.0002 NS 
Leptin 9.12 ± 0.25 8.99 ± 0.20 9.46 ± 0.22 NS NS NS 

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author

Sign in via your Institution