Cerebrospinal fluid antisuprabasin antibody had greater diagnostic accuracy for identifying neuropsychiatric systemic lupus erythematosus among patients with SLE presenting with neuropsychiatric symptoms than cerebrospinal fluid white blood cell count or serum antiribosomal P antibody in a single-center prospective consecutive diagnostic cohort study published in Lupus Science & Medicine.
Investigators consecutively screened 190 patients with systemic lupus erythematosus (SLE) who presented with new-onset neuropsychiatric or neurologic symptoms and underwent evaluation for suspected neuropsychiatric SLE at the emergency department of Renji Hospital in Shanghai, China, from January 2022 to October 2025. All included patients met the 2019 European League Against Rheumatism/American College of Rheumatology classification criteria for SLE. Following exclusions, 148 patients were included in the final cohort and underwent lumbar puncture.
An independent committee of 2 rheumatologists and an infectious disease specialist adjudicated final diagnoses using serologic and cerebrospinal fluid (CSF) testing, neuroimaging, treatment records, and 1 to 3 months of follow-up. A senior rheumatologist resolved disagreements. Committee members were blinded to antisuprabasin antibody results. Patients were classified as having neuropsychiatric SLE (NPSLE; n = 48), SLE with central nervous system infection (n = 59), or SLE with neuropsychiatric manifestations attributed to other causes (n = 41).
"[T]he practical diagnostic challenge is not to distinguish NPSLE from healthy controls, but to identify NPSLE among patients with SLE with neuropsychiatric symptoms and competing aetiologies," wrote lead author Fan Zhang of the Department of Emergency Medicine at Renji Hospital, Shanghai Jiao Tong University School of Medicine, and the Department of Internal Medicine at Pujiang Hospital, both in Shanghai, China, and colleagues.
Median CSF antisuprabasin antibody concentrations were 89 ng/mL among patients with NPSLE, 73 ng/mL among those with central nervous system infection, and 48 ng/mL among those with other causes, with statistically significant differences in all pairwise comparisons. Investigators combined the latter 2 groups as the non-NPSLE comparator group for diagnostic analyses.
For identifying NPSLE, CSF antisuprabasin antibody yielded an area under the receiver operating characteristic curve of 0.830, compared with 0.743 for CSF white blood cell count and 0.665 for serum antiribosomal P antibody. At the optimal cutoff of 76.36 ng/mL, CSF antisuprabasin antibody had 81% sensitivity and 74% specificity.
A model combining CSF antisuprabasin antibody, CSF white blood cell count, and serum antiribosomal P antibody yielded an area under the curve of 0.892, which was significantly higher than that of CSF antisuprabasin antibody alone.
In a secondary analysis of 42 patients with paired serum and CSF measurements, serum antisuprabasin antibody concentrations did not differ significantly among diagnostic groups, whereas CSF concentrations did. CSF antisuprabasin antibody concentrations were not significantly correlated with serum concentrations or the CSF/serum albumin quotient.
The investigators noted that the single-center design and limited cohort size may restrict generalizability of the proposed cutoff. Biomarker sampling was limited to the emergency setting, and the posttreatment trajectory of CSF antisuprabasin antibody remains unknown. The investigators noted that the 42-patient paired analysis lacked sufficient power to establish the biological origin of CSF antisuprabasin antibody. The study also did not include functional experiments to assess whether the antibody had direct biological activity in central nervous system inflammation.
The investigators called for multicenter longitudinal studies to validate the diagnostic threshold and evaluate biomarker changes over time.
Disclosures: The Shanghai Municipal Health Commission funded the study. The funding source had no role in study design, data collection, analysis, interpretation, manuscript preparation, or the decision to submit the study for publication. The investigators declared no competing interests.
Source: Lupus Science & Medicine
