Objectives: Treatment of patients with systemic sclerosis (SSc) can be challenging because of clinical heterogeneity. Integration of genome-scale transcriptomic profiling for patients with SSc can provide insights on patient categorisation and novel drug targets.
Methods: A normalised compendium was created from 344 skin samples of 173 patients with SSc, covering an intersection of 17 424 genes from eight data sets. Differentially expressed genes (DEGs) identified by three independent methods were subjected to functional network analysis, where samples were grouped using non-negative matrix factorisation. Finally, we investigated the pathways and biomarkers associated with skin fibrosis using gene-set enrichment analysis.
Results: We identified 1089 upregulated DEGs, including 14 known genetic risk factors and five potential drug targets. Pathway-based subgrouping revealed four distinct clusters of patients with SSc with distinct activity signatures for SSc-relevant pathways. The inflammatory subtype was related to significant improvement in skin fibrosis at follow-up. The phosphoinositide-3-kinase-protein kinase B (PI3K-Akt) signalling pathway showed both the closest correlation and temporal pattern to skin fibrosis score. COMP, THBS1, THBS4, FN1, and TNC were leading-edge genes of the PI3K-Akt pathway in skin fibrogenesis.
Conclusions: Construction and analysis of normalised skin transcriptomic compendia can provide useful insights on pathway involvement by SSc subsets and discovering viable biomarkers for a skin fibrosis index. Particularly, the PI3K-Akt pathway and its leading players are promising therapeutic targets.
Keywords: fibrosis; gene expression profiles; systemic sclerosis; unsupervised clustering.
© Author(s) (or their employer(s)) 2019. No commercial re-use. See rights and permissions. Published by BMJ.