The association between lupus serology and disease outcomes: A systematic literature review to inform the treat-to-target approach in systemic lupus erythematosus.

Introduction: Serological markers such as anti-double stranded (ds)DNA antibodies and complement fractions C3/C4, are integral components of disease activity assessment in patients with systemic lupus erythematosus (SLE). However, it remains uncertain whether treatment should aim at restoration of serological abnormalities.

Objectives: To analyze and critically appraise the literature on the prognostic impact of active lupus serology despite clinical disease quiescence.

Methods: A systematic literature review was performed in PubMed and EMBASE using the PICOT(S) (population, index, comparator, outcome(s), timing, setting) system to identify studies evaluating the association of serum anti-dsDNA, C3 and C4 levels assessed at the time of clinical remission or during the disease course, against the risk for impending flares and organ damage. Risk of bias was determined by the Quality in Prognosis Studies and ROB2 tools for observational and randomized controlled studies, respectively.

Results: Fifty-three studies were eligible, the majority having moderate (70.6%) or high (11.8%) risk of bias and not adequately controlling for possible confounders. C3 hypocomplementemia during stable/inactive disease was associated with increased risk (2.0 to 3.8-fold) for subsequent flare in three out of seven relevant studies. Three out of four studies reported a significant effect of C4 hypocomplementemia on flare risk, including one study in lupus nephritis (likelihood ratio-positive 12.0). An increased incidence of flares (2.0 to 2.8-fold) was reported in 11 out of 16 studies assessing the prognostic effect of high anti-dsDNA, and similarly, the majority of studies yielded significant relationships with renal flares. Six studies examined the effect of combined (rather than individual) serological activity, confirming the increased risk (2.0 to 2.7-fold) for relapses. No consistent association was found with organ damage.

Conclusion: Notwithstanding the heterogeneity and risk of bias, existing evidence indicates a modest association between abnormal serology and risk for flare in patients with stable/inactive SLE. These findings provide limited support for inclusion of serology in the treat-to-target approach but rationalize to further investigate their prognostic implications especially in lupus nephritis.