Background/Purpose: Patients with childhood-onset systemic lupus erythematosus (cSLE) are at high risk for neuropsychiatric involvement as disease onset typically occurs during the critical period of adolescent brain development. Previous work has linked serum brain injury markers to systemic inflammation and cSLE disease features. However, the connection between these markers and brain volumetric changes is understudied. We aimed to i) compare serum brain injury markers and brain volumes between cSLE and controls; ii) evaluate the associations between brain injury markers and brain volume parameters; and iii) determine the associations within sub-analysis of the cSLE group.

Methods: This cross-sectional study included cSLE participants (ages 12–17) recruited from the Lupus Clinic at a Canadian tertiary children’s hospital (Jan 2020–Dec 2023) and age/sex-matched healthy controls. Serum brain injury markers (serum neurofilament light (sNFL), glial fibrillary acidic protein (GFAP), Tau) were quantified using Simoa Human Neurology 4–Plex B assay (Quanterix, Billerica, MA, USA). T1-weighted MRIs acquired at 3T from cSLE patients and controls were preprocessed with FreeSurfer to automatically segment grey matter, white matter, lateral ventricle, and total brain volumes. Wilcoxon rank-sum tests compared serum brain injury marker levels between both groups and ANCOVA assessed brain volume differences adjusted for age and estimated total intracranial volume (eTIV). Semi-partial Pearson correlations evaluated associations between brain injury markers (adjusted for age) and brain volumes (adjusted for eTIV) in the combined group (cSLE and controls), and in a sub-analysis of the cSLE group.

Results: 56 cSLE participants (mean age=15.1±1.8 years, 86% female) and 43 controls (mean age=15.1±1.7 years, 81% female) were included. GFAP (114.0 vs 74.3 pg/mL) and Tau (3.57 vs 2.58 pg/mL) serum levels were significantly higher in cSLE compared to controls (Figure 1). Adjusted total brain volume [F(1,95)=4.871, p=0.03] and grey matter volume [F(1,95)=5.571, p=0.020] were significantly lower in cSLE compared to controls; there were no differences for white matter and lateral ventricle volume. In the combined group, GFAP had a significant negative correlation with grey matter volume, while both sNFL and GFAP were positively associated with lateral ventricle volume (Table 1). No associations were found between Tau and brain volumes. In the cSLE sub-analysis correlations between brain injury markers and brain volumes paralleled the whole group analysis, although not statistically significant (Table 2).

Conclusion: We found elevated serum brain injury markers and reduced total brain and grey matter volumes in patients with cSLE compared to controls. The significant relationships between sNFL/GFAP and grey matter/lateral ventricle volumes in the cohort suggest that these markers are associated with cerebral atrophy. Future studies of larger and longitudinal cohorts will help further explore the relationship between brain inflammation, injury, structure and function.

Figure 1: Boxplots showing elevated serum brain injury marker levels in cSLE compared to controls.

Table 1: Relationship between Brain Injury Markers and Brain Volumes in Combined Cohort of cSLE Patients and Controls (n=99)

Table 2: Sub-Analysis of Relationship between Brain Injury Markers and Brain Volumes in cSLE only (n=56)

« Back to ACR Convergence 2025

ACR Meeting Abstracts - https://acrabstracts.org/abstract/investigating-the-relationship-between-brain-injury-markers-and-brain-volume-in-children-with-systemic-lupus-erythematosus-and-healthy-controls/