Systemic autoimmune diseases (SADs) are chronic inflammatory conditions with autoimmune aetiology and many common clinical features, hampering diagnosis and adequate treatment decisions. Finding new treatments or applying the existing ones in a more effective way is especially hard in SADs due to the heterogeneity of molecular mechanisms within the same disease class. Based on this premise, the first step towards establishing a precision medicine strategy for SADs is to reclassify these conditions at the molecular level, which might result in a more homogenous stratification in terms of pathological molecular pathways. SADs have a multi-factorial predisposition, where the interplay between genetic, epigenetic and environmental conditions is essential in the pathogenesis of the diseases. Thus, in order to capture as many aspects as possible the molecular stratification is performed using multiple layers of information (e.g. genome, transcriptome, methylome or metabolome). Among all available molecular levels of information, the most informative in terms of functionality and dimensionality are transcriptome and methylome, reflecting different aspects of regulation and environmental influence, respectively, giving a wide view of the molecular background.
We performed an unsupervised integrative clustering analysis to classify SADs patients into subtypes based on genome-wide trancriptome and methylome profiling of ~800 cases distributed across 7 different clinical entities (rheumatoid arthritis, systemic lupus erythematosus, systemic sclerosis, primary Sjögren ́s syndrome, primary antiphospholipid antibody syndrome, mixed connective tissue disease and undifferentiated connective tissue disease) and ~200 healthy individuals. We propose a new molecular reclassification of the diseases and will discuss this during my presentation.