Immunosuppressants can modify the transcriptomic signature in patients with systemic lupus erythematosus

Abstract

Background/Aims Transcriptomics has the potential to revolutionise the way we approach complex diseases such as systemic lupus erythematosus (SLE), which is both clinically and biologically heterogeneous. We aimed to explore whether deconvolution of whole blood transcriptomic data could predict differences in immune cell proportions between active SLE patients, and whether these are attributable to the SLE phenotype or medication use.

Methods Patients with active SLE were recruited from the BILAG-Biologics Registry (BILAG-BR) prior to change in therapy. Disease activity was measured using the BILAG-2004 Index. Whole blood RNA-sequencing was conducted at baseline and FPKM-normalised data was deconvoluted using CIBERSORTx with the LM22 signature matrix used as reference. The effects of five different drugs (azathioprine, cyclophosphamide, hydroxychloroquine, methotrexate and mycophenolate mofetil [MMF]) were explored. Predicted cell frequencies were compared between active disease domains (defined by BILAG A or B score) and concomitant immunosuppressant use using Mann-Whitney U tests with Benjamini-Hochberg correction (significance at p < 0.05).

Results We recruited 109 patients of whom 104 (95.4%) were female, with a median (IQR) age of 38 years (29-49) years, disease duration of 10 (6.5, 16.5) years. Patients had active disease with a median SLEDAI score of 8 (4-14). For patients taking MMF (53/109, 46.8%), there was a lower proportion of macrophages M0 (0.435% vs 1.391%, p = 0.001), naïve CD4 T cells (0.961% vs 2.251%, p = 0.002) and regulatory T cells (1.858% vs 3.574%, p = 0.007), and a higher proportion of memory activated CD4 T cells (1.826% vs 1.113%, p = 0.015), when compared to patients not taking mycophenolate mofetil. MMF use was associated with statistically significant differences in predicted immune cell (macrophages m0, naïve CD4 T cells, regulatory T cells and memory activated CD4 T cells) frequency after adjusting for age, gender, ethnicity, disease duration, renal disease and steroid use. No differences were observed for the other immunosuppressants or between patients with active or inactive disease in each of the nine organ domains.

Conclusion The use of immunosuppressants can significantly modify the transcriptomic signature in patients with SLE affecting key immune cell type populations differently, highlighting the need to adjust for background medication use in future studies, particularly in those using whole blood transcriptomics. It would also be valuable to evaluate the differences identified in this study to determine whether they can be used to monitor or predict treatment response.