Background/Purpose: Previous studies have described three patterns of disease activity over time in systemic lupus erythematosus (SLE), namely long quiescence, relapsing remitting, and persistently active. However, they all enrolled prevalent patients, many of whom in the late stages of the disease. As such, the patterns of disease course since diagnosis are not known. The aim of the present study was to assess the impact of these patterns on damage accrual, flare and mortality in an inception cohort.

Methods: Inception patients of our long-term longitudinal cohort (enrolled within 18 months of diagnosis), with at least 10 years of follow-up and no time interval >18 months between consecutive visits, were investigated. Prolonged remission (PR) was defined as a clinical SLEDAI-2K=0 [serology (anti-dsDNA antibodies and C3/C4 levels) excluded], achieved within five years since enrolment and maintained for ≥10 years after that. Relapsing-remitting (RR) pattern was defined based on ≥2 remission periods (one remission period equals two consecutive visits with a clinical SLEDAI-2K=0), while patients with no remission were categorized as persistently active (PA). Patients with only one remission period (“hybrids”) were re-allocated to the RR and PA groups according to the lower quartile of the time spent in remission for the RR patients (3.5 years). Multivariable analysis was performed using baseline demographic, clinical, immunological and therapeutic variables as well as disease-related variables at two years to identify predictors for disease course.  Descriptive and regression analyses were used to compare groups regarding cumulative damage at 10 years, mortality and flare rate beyond 10 years (median follow-up 17.5 years).

Results: Of 267 patients who fulfilled the inclusion criteria, 27 (10.1%) achieved prolonged remission, 180 (67.4%) RR and 25 (9.4%) PA disease. Thirty-five patients (13.1%) had only one remission period (“hybrids”). At enrollment, there were no significant differences in demographic, clinical, immunological and therapeutic characteristics among the PR, RR and PA groups. Multinomial regression analysis for the identification of predictors for group membership showed Black race [OR=2.78, 95%CI=1.05-7.31, p=0.039] and higher adjusted mean SLEDAI-2K over the first two years [OR=1.21, 95%CI=1.11-1.32, p<0.0001] to be associated with a more severe disease course. Outcomes at 10 years and beyond are given in Table 1.

Re-allocation of the hybrids into the RR and PA groups (17 and 18 patients respectively) provided similar results.

Conclusion: Black race and more severe disease over the first two years were associated with a worse disease course over time. Disease course was, in turn, associated with a gradual increase in damage accrual (from PR to RR to PA) as well as flare rate. Differences in mortality did not reach statistical significance.