Background/Purpose: The most pathogenic responses against self and foreign antigens that lead to systemic autoimmune diseases and infections, respectively, overall engage similar immunologic components. Consequently, there are no molecular markers that distinguish between anti-self and anti-foreign responses. Our aim was to test the hypothesis that whole blood RNA sequencing may distinguish between diverse systemic autoimmune diseases and various infections.

Methods: For comparison we used consecutive peripheral blood RNA sequencing data from the Gene Expression Omnibus database published until December 2021, encompassing 584 patients with various systemic autoimmune diseases and 615 patients with viral, bacterial and parasitic infections.

Results: Data were used as input to our novel state-of-the-art preprocessing pipeline based on gene read counts ranking. With this and additional Machine Learning approaches, we created a model that discriminates between the two groups of pathologies with 98% accuracy. This high discriminating capacity was externally validated in additional datasets comprising of samples not included in the models’ training.
Further analyses using the most informative gene expression profiles revealed that ‘autophagy’, ‘DNA damage response’, ‘RAP-1 signalling’ and ‘NOTCH signalling’ were among the top discriminating pathways. 
Moreover, we identified a set of 24 genes, including the autoimmunity related RPL7, TLK2 and ANK2, whose relative RNA expression levels specifically distinguish between systemic autoimmunity and infection with 88% accuracy.

Conclusion: RNA expression levels of specific genes in peripheral blood distinguish between anti-self and anti-foreign responses, thereby providing a biomarker for the differential diagnosis of the corresponding pathologies in patients presenting with an inflammatory disorder. In addition, this analysis may provide new mechanistic understanding of the systemic autoimmune response.

« Back to ACR Convergence 2024

ACR Meeting Abstracts - https://acrabstracts.org/abstract/a-blood-transcriptome-signature-distinguishes-between-diverse-systemic-autoimmune-diseases-and-infection/