Abstract: Objectives: To gain insights into the pathogenesis of neuropsychiatric systemic lupus erythematosus (NPSLE) and identify potential drug targets through investigation of whole-blood human transcriptome.
Methods: We analysed differentially expressed genes in peripheral blood from active central nervous system (CNS) lupus (n = 26) and active non-neuropsychiatric SLE (n = 38) patients versus healthy controls (n = 497) from the European PRECISESADS project (NTC02890121). We further explored dysregulated gene modules in active CNS lupus and their correlation with serological markers. Lastly, we performed regulatory network and druggability analysis.
Results: Unsupervised weighted gene co-expression network analysis (WGCNA) revealed 23 dysregulated gene modules and two subgroups of active CNS lupus. The interferon gene module was prominently upregulated in subgroup 1, while the B cell, T cell, and cytotoxic/natural killer (NK) cell modules were downregulated. Subgroup 2 showed less marked dysregulation patterns. Subgroup 1 had lower estimated proportions of lymphoid cell subsets and proportionally more patients positive for anti-dsDNA antibodies compared to subgroup 2, pointing to molecularly distinct subgroups or misclassification of subgroup 2. In silico prediction algorithms demonstrated a greater anticipated response to anifrolumab, C3 inhibitors, and calcineurin inhibitors for patients in CNS lupus subgroup 1 compared with subgroup 2.
Conclusions: Gene dysregulation patterns related to innate and adaptive lymphoid immunity separated active CNS lupus patients into two distinct subgroups with differential anticipated response to type I interferon, C3, and calcineurin inhibition. Our study provides a conceptual framework for precision medicine in NPSLE and implications for overcoming the major clinical challenge of attributing neuropsychiatric features to SLE versus other causes.
