Background/Purpose: Cognitive impairment is common in childhood-onset systemic lupus erythematosus (cSLE), particularly in domains like executive function and attention. However, attributing cognitive difficulties to brain inflammation—such as interferon (IFN)-mediated processes in cSLE—remains challenging. Neuroglial proteins are potential biomarkers of brain injury and cognitive impairment in cSLE. S100B is a calcium-binding signaling protein produced by astrocytes that reflects acute brain injury. Elevated levels of S100B and related S100 proteins have been reported in patients with neuropsychiatric lupus (NPSLE) and cognitive impairment but remain understudied in pediatric populations. We compared serum S100B levels between children with cSLE and controls, and investigated its relationship with cSLE disease features, serum-IFN levels, and cognitive functioning.

Methods: Cross-sectional data were prospectively collected from cSLE participants (ages 12-17 years) meeting ACR classification criteria for SLE, recruited from the Lupus Clinic at a Canadian tertiary children’s hospital from Jan 2020–Dec 2023 and age-, sex-matched healthy controls. S100B, IFN-α and IFN-γ serum levels were quantified using their respective Simoa assays (Quanterix, Billerca, MA, USA). Disease features included disease activity (SLEDAI-2K), damage (SLICC damage index, SDI >0), glucocorticoid (GC) dose at study visit, and cumulative GC exposure (prednisone-equivalent). Assessment of cognitive function domains included: cognitive inhibition and mental flexibility using Delis-Kaplan Executive Function System Color Word Interference Test (inhibition, inhibition/switching), working memory using Digit Span from the Wechsler Adult Intelligence Scale IV/Wechsler Intelligence Scale for Children V, and attention using Conners Continuous Performance Test 3 (omissions, commissions, hit reaction time). Wilcoxon rank sum test compared S100B levels; Spearman correlations tested associations.

Results: 49 cSLE participants (mean age=15.2±1.8 years, 90% female) and 40 controls (mean age=15.3±1.7 years, 80% female) were included. For cSLE, median disease duration was 25.1 months (IQR 13.3-46.2), median SLEDAI-2K was 2.0 (IQR 2.0-4.0), 10% had disease damage, 41% were using glucocorticoids at study visit, and median cumulative GC exposure was 2.3 grams (IQR 0.6-7.0). One patient had a NPSLE diagnosis. There were no group differences in serum S100B levels between cSLE and controls (median 1714.9 vs 2369.4 ng/mL; Figure 1). In cSLE, S100B showed positive but non-significant correlations with SDI and current GC dose (Table 1). Significant inverse associations were observed between S100B and both IFN-α and IFN-γ (Table 1). No correlations were found between S100B and cognitive test performance (Table 2).

Conclusion: In cSLE, serum S100B was not associated with cognitive functioning and did not differ from controls, contrasting prior findings in adult SLE. Inverse associations with type I and II IFNs suggest a potential downregulation of this protein in IFN-driven inflammation, perhaps during the early stages of brain inflammation. Future studies will examine these relationships longitudinally in larger cSLE cohorts.

Figure 1: Boxplot showing comparison of serum S100B level between cSLE and controls.

Table 1: Relationship between S100B, Disease Characteristics, and Interferons in cSLE (n=49)

Table 2: Correlations between S100B and Cognitive Measures in cSLE (n=49)

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/s100b-in-childhood-onset-systemic-lupus-erythematosus-associations-with-disease-features-interferon-levels-and-cognitive-functioning/