Idiopathic inflammatory myopathies, signified by distinctive peripheral cytokines, chemokines and the TNF family members B-cell activating factor and a proliferation inducing ligand

Abstract

Objective: Serum cytokines play an important role in the pathogenesis of myositis by initiating and perpetuating various cellular and humoral autoimmune processes. The aim of the present study was to describe a broad spectrum of T- and B-cell cytokines, growth factors and chemokines in patients with idiopathic inflammatory myopathies (IIMs) and healthy individuals.

Methods: A protein array system, denoted as multiplex cytokine assay was utilized to measure simultaneously the levels of 24 circulating cytokines, including B-cell activating factor (BAFF) and a proliferation inducing ligand (APRIL) of patients with IIMs and healthy individuals. Additionally, correlational clustering and discriminant function analysis (DFA), two multivariate, supervised analysis methods were employed to identify a subset of biomarkers in order to describe potential functional interrelationships among these pathological cytokines.

Results: Univariate analysis demonstrated that a complex set of immune and inflammatory modulating cytokines are significantly up-regulated in patients with IIMs relative to unaffected controls including IL-10, IL-13, IFN-α, epidermal growth factor (EGF), VEGF, fibroblast growth factor (FGF), CCL3 [macrophage inflammatory protein (MIP-1α)], CCL4 (MIP-1β) and CCL11 (eotaxin), whereas G-CSF was significantly reduced in IIM patients. Correlational clustering was able to discriminate between, and hence sub-classify patients with IIMs. DFA identified EGF, IFN-α, VEGF, CCL3 (MIP-1α) and IL-12p40, as analytes with the strongest discriminatory power among various myositis patients and controls.

Conclusions: Our findings suggest that these factors modulate myositis pathology and help to identify differences between subsets of the disease.

Fig. 1

Distinct cellular cytotoxic, humoral and chemotactic patterns identified in IIMs. Levels of 24 serum biomarkers were measured simultaneously using a biometric multiplex assay from serum of subsets of patients with IIMs, and unaffected control individuals. (A) Cellular cytotoxic patterns identified elevations of IL-1β, IL-7, IL-12, IL-17, IFN-α in DM, IL-17, GM-CSF, IFN-α, IL-1Rα, IL-7 in PM, IL-1β, IL-12, GM-CSF, IFN-α and IL-1Rα in JDM, and IL-1β, IL-17, GM-CSF, IFN-α, IL-1Rα in CAM. (B) Humoral patterns identified elevations of IL-6 and IL-10 in DM, and IL-10 in PM, JDM and CAM. (C) Chemotactic patterns identified elevations of CCL2, CCL3, CCL4, CCL11, CXCL8 and CXCL10 in DM, CCL11, CCL3 and CCL4 in PM, CCL11, CCL3, CCL4 and CXCL10 in JDM, and CCL11, CCL3, CCL4 and CXCL10 in CAM. Patients with overlap diseases had a mixed-diverse cytokine profile with significant elevations of IL-1β, IL-12, GM-CSF, IL-1Rα, IL-7, IL-6, IL-10, IL-13, CCL3, CCL4 and CXCL10. Cytokines significantly elevated in groups relative to controls are denoted with an asterisk (*P < 0.05 and **P < 0.001).

Fig. 2

Systemic cytokine profiles correlate with BAFF and APRIL in IIMs. To evaluate the linear relationship between BAFF, APRIL and cytokines, correlation of BAFF and APRIL with systemic cytokine profiles was assessed by correlation analysis. (A) Serum BAFF was highly correlated with IL-12, and moderately correlated with ESR, IL-1Rα, CXCL10 and IL-7. (B) Serum APRIL was highly correlated with IL-6, CXCL10 and RF. (C) To cluster the correlational profiles into meaningful structures, a hierarchical clustering was performed to obtain clear separation of variables, in which BAFF, APRIL, CXCL10, CCL2, CCL3, CCL4 and IL-1Rα depicted a tight cluster, when compared with other variables.

Fig. 3

PCA identifies diagnostic and prognostic potential of the modelling cytokine profiles in idiopathic inflammatory myopathies. PCA increases the ability to identify uncorrelated features and interpret and analyse the profiles by reducing dimensionality. (A) PCA was performed on the training set and the test data applied to this model had an area under the curve of 0.75, indicating the remarkable discriminatory power of the model. (B) In order to assess the prognostic power of the model identified by the PCA, we investigated the role of PCA modulatory cytokines in the therapeutic response of a patient with active JDM. The data, which were rescaled so that the value of 0 characterizes the level of healthy controls to create a cytokine activity score, were plotted over time during treatment to assess patient response. Changes for PC3 correlated well with clinical response as early as 2 weeks of therapy, and therefore indicate the prognostic potential of the model in IIM.

Fig. 4

Novel disease-specific patterns in IIMs. DFA was used to identify subset of cytokines whose expression values can be linearly combined in an equation, denoted a root, whose overall value is distinct for a given characterized group. Results of DFA are visualized here on a multidimensional plot, with class discrimination power represented by distance among groups. The variables identified by the DFA will continue to be included in the model, as long as the respective F-values for those variables are larger than the standard threshold and is then plotted in three dimensions to visually represent the relative differences in cytokines among the distinct populations. Using this form of multivariate analysis, subsets of IIM clustered into unique positions on a 3D plot, distinctly further away from controls, suggesting the diagnostic potential of DFA in IIM.