Abstract
OBJECTIVES: Systemic lupus erythematosus (SLE) is a complex autoimmune disease. Significant morbidity and early mortality necessitate early intervention. This study harnessed SLE-associated immune dysregulation to create a Lupus Classification Risk Index (LCRII) and Lupus Disease Activity Immune Index (LDAII) that identified individuals at risk for SLE classification and disease activity.
METHODS: The LCRII was developed from 84 military personnel who developed classified SLE (≥4 American College of Rheumatology criteria) versus matched healthy controls, which was confirmed in 56 lupus blood relatives who developed SLE versus 154 matched unaffected relatives and 77 unrelated controls. The LDAII was informed by SLE patient visits with low (n=132) or active (n=179) disease and 48 matched controls. Data from blood samples assessed for circulating SLE-associated autoantibody specificities and soluble immune mediators informed the LCRII and LDAII. Random forest modelling guided the selection of informative analytes.
RESULTS: An LCRII informed by 32 or 17 log-transformed/standardised mediators, weighted by their correlation to SLE-associated autoantibodies, differentiated pre-SLE individuals before reaching disease classification (area under the curve (AUC) ≥0.79, p<0.0001; effect size ≥1.1), even before the appearance of clinical criteria (AUC ≥0.74, p<0.0001; effect size ≥0.9). The LCRII-32, LCRII-17 and select mediators, MCP-3/CCL7, TNFRII, stem cell factor (SCF), IL-1α, IP-10/CXCL10 and TGF-β differentiated renal and serositis classification criteria (p<0.05). An LDAII informed by 26 or 13 log-transformed/standardised mediators, weighted by their correlation to SLE-associated autoantibodies or disease activity (hybrid Systemic Lupus Erythematosus Disease Activity Index; hSLEDAI), differentiated SLE patients with low (hSLEDAI <4) or active (hSLEDAI ≥4) disease (AUC >0.6, p ≤0.002, effect size ≥0.4), including clinical/serologic active versus quiescent disease (AUC ≥0.7, p<0.0001, effect size ≥0.6). The LDAII-26, LDAII-13 and select mediators MCP-1/CCL2, TNFRII, SCF, IL-2Rα, IL-10 and TGF-β differentiated renal and serositis manifestations.
CONCLUSIONS: We have conceptualised two immune mediator-informed indexes, the LCRII that predicts SLE from months to years before clinical presentation, and the LDAII that analogously predicts active disease in SLE to distinguish patients who would benefit from early intervention.