Introduction: Systemic sclerosis (SSc) is a connective tissue disease characterized by dysregulated immunity and fibrosis of the skin and internal organs. SSc-related interstitial lung disease (SSc-ILD) occurs in over half of patients with SSc and is the leading cause of disease specific mortality. Predicting which patients will develop SSc-ILD is difficult. Treatments are often delayed and have variable efficacy. We hypothesize that the non-invasive skin transcriptome can serve as an intermediate phenotype to study the development of lung function impairment.

Methods: The Gene Expression Omnibus repository was searched using the terms scleroderma, expression profiling by array and Homo Sapiens. Two datasets were selected. Weighted gene correlation analysis (WGCNA) was used to determine clusters of highly correlated genes (modules). Modules with significant overlap between datasets were identified and pathway analysis was performed.

Results: In the first dataset, WGCNA identified 13 modules correlated with skin score, and 7 modules correlated with skin score and lung function in the second dataset. Using genes overlapping these significant modules, 10 groups of overlapping modules were identified. Pathway analysis of these modules was enriched for cytokine-cytokine receptor interaction, extra cellular matrix receptor interaction and chemokine signaling.

Conclusion: Skin transcriptomic profiles of two distinct datasets identified multiple gene networks associated with severity of skin involvement and lung function in SSc. These non-invasive signatures have the potential to identify patients at risk of lung disease and may define pathways for novel pharmacologic targets.