Background/Purpose: Unpredictability is a major challenge in systemic lupus erythematosus (SLE). Routinely used clinical and laboratory parameters fail to predict the risk of and time to flare. We hypothesise that gene signatures may better predict flaring and might thus have merit in disease monitoring.

Methods: Eligible for this analysis were SLE patients from the European multicenter PRECISESADS project (NCT02890121) with available transcriptome data and long-term follow-up including disease flares. The analysis was restricted to patients with quiescent disease at baseline, defined as a clinical SLEDAI-2K score < 6. Flare was defined as any increase in disease activity resulting in a change of therapy. For each patient Reactome pathways according to the Functional Analysis of Individual Microarray Expression (FAIME) algorithm, were calculated. Time-dependent analysis for interval censored data was conducted with parametric models correcting for relevant confounding covariates associated with flares and individual regression weights. Q value < 0.05 and a hazard ratio (HR) >1.5 for causative pathways or < 0.667 for protective ones were considered significant.

Results: Long-term data were available in 131 patients, including 85 with quiescent disease at baseline. At the time of blood sampling, those patients had a mean (SD) age of 47.5 (13.9) years and a mean disease duration of 15.7 (9.7) years, and they were mostly women (n=84, 98.8%). The mean clinical SLEDAI-2K was 1.5 (1.5). At baseline, 59 (69.4%) patients were treated with hydroxychloroquine, 25 (29%) with synthetic immunosuppressants, and 36 (42.3%) with glucocorticoids at a mean daily dose of 2.3 (0.5) mg of prednisone equivalent. After a mean observation time of 6.9 (2.6) years, flares occurred in 30 patients (35%).Among those 30 patients, the first flare was developed after a mean time of 3.0 (2.0) years, and the non-cumulative count of those flares per domain was 18 articular, 9 cutaneous, 5 constitutional, 4 haematological, 3 vascular, and 2 renal. Overall, 1265 Reactome pathways were explored, of those 83 and 48 pathways were significantly associated with increased or reduced flaring hazards, respectively (Figure 1).Flaring patients had a reduced capability of DNA repair, increased DNA damage due to impaired telomere function, an increased activity interferon-related pathways, an increased activity of the complement system, an increased inflammasome, reduced CTLA4 and CD28 inhibitory mechanisms, a disrupted circadian clock and several metabolic alterations.

Conclusion: Our findings reveal that specific pathway deregulations linked to SLE pathogenesis may herald the occurrence of flares in quiescent patients. Impaired DNA repair, increased interferon signaling, complement activation, inflammasome upregulation, and reduced CTLA4/CD28 inhibitory mechanisms suggest a predisposition to immune dysregulation preceding clinical flares. These insights highlight potential molecular predictors of flaring and suggest that targeted immunomodulation or specific interventions may be warranted in selected patients to prevent flares and mitigate the risk of long-term damage accrual, ultimately improving disease monitoring strategies in SLE.

Figure 1

Heatmap

Survival Clusters

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/transcriptome-analysis-of-quiescent-sle-cases-uncovers-dysregulated-pathways-associated-with-disease-flares/