Treatment nonresponders in lupus nephritis characteristically exhibit persistent type I interferon signalling in monocytes: a longitudinal single-cell transcriptomic analysis.

OBJECTIVES

We aimed to characterise the dynamic transcriptional changes associated with treatment responses in lupus nephritis (LN) through longitudinal single-cell RNA sequencing analysis of peripheral blood mononuclear cells (PBMCs) during standard-of-care induction therapy.

METHODS

We leveraged the Korean Unlimited multi-Dimensional Omics research in Systemic lupus erythematosus cohort, comprising patients with biopsy-proven active proliferative LN. Single-cell RNA sequencing of PBMCs was conducted at baseline and subsequently at 3, 6, and 12 months following initiation of therapy (n = 10). For validation purposes, bulk RNA sequencing of monocytes was performed in an independent patient group at baseline and at 3 months (n = 13). Renal response was classified as complete response or nonresponse at 12 months according to predefined criteria.

RESULTS

In single-cell RNA sequencing analysis of patients with LN, the myeloid cell population-particularly classical and intermediate monocytes-demonstrated the highest number of differentially expressed genes following induction therapy. Weighted gene coexpression network analysis identified distinct gene modules closely associated with treatment responses. Complete responders exhibited progressive suppression of type I interferon (IFN-I) signalling, whereas nonresponders maintained persistent IFN-I-driven gene expression characterised by sustained inflammatory features. Bulk RNA sequencing of monocytes from an independent group of patients with LN confirmed that 6 IFN-I-responsive genes (IRF7, ISG15, LY6E, IFI44, IFI44L, and IFI6) were significantly downregulated by 3 months in complete responders, but not in nonresponders. This gene signature correlated with both proteinuria and disease activity scores.

CONCLUSIONS

This study demonstrates that persistent IFN-I-driven gene expression in monocytes characterises treatment resistance in LN. Our findings suggest that early transcriptional profiling may enable timely identification of nonresponders and warrant further investigation in larger, independent cohorts.