Cytokine profiling in active and quiescent SLE reveals distinct patient subpopulations

Objectives: To investigate patterns of serum cytokines in patients with active and stable SLE and to determine how they relate to clinical phenotype.
Methods: Serum levels of 10 cytokines were measured retrospectively in a cohort of SLE patients and healthy controls using a high-sensitivity multiplex bead array. Disease activity was determined using the SLEDAI-2K and BILAG-2004 indices. Logistic regression models were used to determine the association between cytokine levels and active SLE. Principal component analysis (PCA) and cluster analysis was then used to identify subgroups of patients on the basis of cytokine levels.
Results: Serum levels of CXCL10 and CXCL13 were significantly higher in SLE patients compared to healthy controls. Two cytokines (PTX3 and CXCL10) were significantly higher in patients with active disease after adjustment for potential confounding factors. Measurement of 4 cytokines (CXCL10, IL-10, IL-21, and PTX3) significantly improved the performance of a model to identify patients with clinically active disease. Cluster analysis revealed that the patients formed 3 distinct groups, characterised by higher levels of levels of IFNα and BLyS (Group 1), increased CXCL10 and CXCL13 (Group 2) or low levels of cytokines (Group 3). Group 2 had significantly lower serum complement and higher anti-dsDNA antibodies and increased prevalence of inflammatory arthritis.
Conclusions: Multiplex analysis has identified a serum cytokine signature for active SLE. Within the SLE population distinct cytokine subgroups were identified, with differing clinical and immunological phenotypes which appeared stable over time. Assessment of cytokine profiles may reveal unique insights into disease heterogeneity.