Plasma cell-free DNA genome-wide methylation profiling enables detection and activity assessment in systemic lupus...

Cell-free DNA (cfDNA) has emerged as a promising non-invasive biomarker in precision medicine and is increasingly recognized as relevant to relevance in systemic lupus erythematosus (SLE), where elevated plasma cfDNA levels have been consistently observed. However, the cfDNA methylation pattern in SLE is unclear.

Differentially expressed genes (DEGs) and differentially methylated probes (DMPs) related to SLE were obtained from the Gene Expression Omnibus (GEO) database. Through multi-omics approaches, the immune microenvironment in renal tissues, whole blood, and peripheral blood mononuclear cells from SLE patients was integrated and analyzed. Applying whole-genome bisulfite sequencing technique, we evaluated the epigenomic features of plasma cfDNA methylation in patients with lupus nephritis (LN). The neural network deep-learning approach was employed to construct a cfDNA-based lupus methylation diagnostic model, which was subsequently validated in two independent cohorts. Finally, the hub genes of the model were screened as potential biomarkers for clinical application in LN patients.

This study found that SLE occurred in abnormal immune microenvironments in multiple sample types, including renal tissue, whole blood, peripheral blood mononuclear cells, and plasma cfDNA, accompanied by abnormal activation of the interferon-related signaling pathway and antiviral response. CfDNA methylation levels were reduced in patients with LN compared to healthy controls, particularly at gene promoter regions. By integrating 13 datasets associated with SLE, and combining them with cfDNA methylation sequencing data, we have screened 163 conserved dysfunctional methylated regions that are related to SLE. The model constructed based on the identified differentially methylated regions demonstrated excellent diagnostic performance for SLE with an area under the curve (AUC) of 0.987, 0.84 in GSE82218 and GSE96879, respectively. Moreover, the model scores differed significantly across disease activity levels (Normal vs. SLE, p<0.0001; low vs. high activity, p<0.01), indicating its potential utility in distinguishing disease states and assessing disease severity.

Multi-omics analyses indicate that frequently aberrant methylation sites are correlated with immune-related pathways, disease onset and progression. The analysis of cfDNA methylation profiles can serve as a diagnostic tool for distinguishing disease states and assessing disease severity in SLE.