Background/Purpose:

Localized scleroderma (LS) is a fibrotic autoimmune disease of the skin and underlying tissues which can lead to disfiguring sequlea, especially in childhood-onset.  Untreated active disease is associated with long-term damage and disability; therefore accurate assessment of the active disease state is imperative, and a distinct activity biomarker would assist clinicians to intervene before disease progression. The purpose of this study was to investigate a wide variety of cyto/chemokines, specifically within disease transition from active to inactive phases in individual patients, which correlate to established clinical measures of disease activity in LS.

Methods:

Plasma from 70 juvenile LS (jLS) patients with clinical data and longitudinal samples (average of 6 years follow-up), including at least one active and one inactive specimen (272 samples total), were evaluated by microarray to study a 60-analyte cyto/chemokine panel. This included the study of T_H1, T_H2, and T_H17 associated cytokines with an additional IFN-γ panel that emphasized CXCL9, CXCL11, and MIP-3β chemokines. Wilcoxon Signed-Ranks Test (p<0.05) was used to compare cyto/chemokines between active and inactive disease subsets. Next, to focus on clinically significant cytokines, additional analyses included spearman’s correlation of analytes with the modified Localized Skin Severity Index (mLoSSI), a validated active disease variable, followed by multiple regression analyses of the significantly correlated cytokines using R language.

Results:

When dichotomizing the LS samples into active vs. inactive subjects, many reoccurring cytokines were significantly elevated (Table 1 – left column). Correlation of the 60 analyte panel resulted in moderate to strong correlation of IFN-γ and T_H1-like associated cyto/chemokines with the mLoSSI, including CXCL9 (MIG), MIP-3β, and IL-10 (Table 1 – right column). Of those that correlated strongly with the mLoSSI, IL-10 and CXCL9 were statistically significant predictors in multiple linear regression analyses (p = 0.002 and 0.001, respectively), with an adjusted R-squared of 0.71; meaning the change of IL-10 and CXCL9 are related to the change in mLoSSI, signifying change in disease activity status.

Conclusion:

This is a unique examination of circulating cytokines in jLS and further correlates the biological activity to clinical disease activity measures used for disease monitoring, the mLoSSI. The correlated chemokines indicate a strong IFN-γ associated environment. CXCL9 is a predominant IFN-γ inducible chemokine and is part of the IFN-γ family that contributes to the activation of the M1 phenotype and recruitment of T_H1 cells, both of which induce further inflammatory cytokine response. Further analysis using the best-fit model is underway with the overall goal of finding the strongest core set of biomarkers that highly predicts the mLoSSI in LS patients, signifying degree of disease activity.