Lupus Science & Medicine
proteinuria levels. Notably, in this study, 9 23% of patients with low-grade proteinuria, thought to be in complete or partial remission, still exhibited significant histological activity on repeat biopsies. This finding supports previous research highlighting a discordance between proteinuria and histological activity, 7 8 10 underscoring the limitations of relying solely on proteinuria as a biomarker. Although KBs are valuable for diagnosing 11 and, as has been shown more recently, for detecting ongoing histological activity in LN, 7–9 their invasive nature means that not all patients are suitable candidates for the procedure. Furthermore, per-protocol repeat KBs are still not considered standard practice in most centres. Thus, there is a need for more sensitive non-invasive biomarkers to detect subclinical kidney inflammation and improve disease monitoring following treatment of LN flares. Urinary biomarkers (UBs), including adiponectin, monocyte chemoattractant protein-1 (MCP-1), platelet factor 4 (PF4), soluble vascular cell adhesion molecule (sVCAM) and cluster of differentiation 163 (CD163), have been shown to effectively discriminate between active and quiescent LN, to correlate with histological activity on LN renal biopsies and to have the capacity to reflect clinical improvement. 12–14 In the current study, we investigated whether the levels of these five UBs measured 24 months after an LN flare were associated with subsequent LN flares and decline in kidney function, to determine their clinical utility as potential biomarkers to detect subclin- ical kidney inflammation.
Criteria for achieving PERR included: (1) achieving urinary protein excretion ≤ 700mg/day based on a 24-hour urine collection, (2) an eGFR ≥ 60 mL/min/1.73 m 2 and (3) no increase in immunosuppressive therapy for treatment failure. 18 The eGFR was calculated using the non-race-corrected Chronic Kidney Disease Epidemi- ology Collaboration (CKD-EPI) equation. 19 Assessed adverse long-term outcomes: (1) subsequent LN flare, defined as an increase in proteinuria to at least 1000 mg/day after achieving complete proteinuria response (urinary protein excretion <500 mg/day based on a 24-hour urine collection) or doubling of protein- uria if they reached a partial response ( ≥ 50% reduction in urinary protein to non-nephrotic levels) 20–23 and a change in therapy by the treating physician, and (2) 30% sustained decline in eGFR (recorded at two consecutive visits) after their 24-month urinary sample collection. UBs quantification All urine samples were centrifuged to remove cellular debris and frozen at −80°C. Each sample was then thawed, aliquoted and refrozen to prevent multiple freeze-thaw cycles. Aliquots were immediately thawed prior to use. UBs were analysed by ELISA using DuoSet Ancillary Reagent Kit 2 (Catalogue #DY08B, R&D Systems) together with the following DuoSet antibody (ab) sets (R&D Systems): PF4 (Catalogue #DY795), sVCAM (Catalogue #DY809), adiponectin (Catalogue #DY1065), MCP-1 (Catalogue #DY279) or CD163 (Catalogue #DY1607), following the manufacturer’s protocol. Urine samples were diluted with reagent diluent prior to plating at the following dilu- tions: PF4 (1:2-1:16), sVCAM; (1:128-1:512), adiponectin (1:4-1:256), MCP-1 (1:2-1:32), CD163 (1:2-1:32). ELISAs were run in duplicate, with their average biomarker level calculated through an ln-ln transformation with reference to their associated same-run standard curves and subse- quent correction for dilution factor. Any value above or below the linear range of the standard curve was rerun at a higher or lower dilution, respectively. Any value still below the linear range of the standard curve at the lowest dilution was given a value below the lower limit of the curve (this only applied to two patient samples in the PF4 assay). UB values were normalised to urine creatinine and expressed as pg per mmol of urine creatinine. Intra-assay coefficient of variation (CVs): adiponectin 3.21%, MCP-1 9.80%, PF4 3.45%, sVCAM 6.88% and CD163 9.87%; inter-assay CVs: adiponectin 7.05%, MCP-1 11.84%, PF4 20.10%, sVCAM 6.75% and CD163 10.94%. Statistical analyses Descriptive tables summarise the cohort characteristics. Categorical variables were reported as counts (%) and compared using Fisher’s exact test. Continuous variables were presented as median (IQR) and compared with the Mann-Whitney test. Spearman’s correlation assessed asso- ciations between UB levels and proteinuria. Four patients who did not have enough urine to process urinary creati- nine were excluded from this and subsequent analysis.
METHODS Patient selection
Patients were recruited from the University of Toronto Lupus Cohort between January 2005 and June 2014. At the clinic, patients are followed regularly (2–6 months). During each visit, clinical and laboratory data are collected using standardised protocols, and KBs are performed when clinically indicated. All participants fulfilled the revised 1997 American College of Rheumatology criteria (ACR), as well as the 2019 European Alliance of Associ- ations for Rheumatology/ACR classification criteria. 15 16 Study design Inclusion criteria: (1) LN flare defined as urinary protein excretion of ≥ 500 mg/day in a 24-hour urine collection that prompted an initiation or dose increase of gluco- corticoids and commencement/escalation/modification of immunosuppressive therapy; (2) preflare estimated glomerular filtration rate (eGFR) of ≥ 60 mL/min/1.73 m 2 ; and (3) available urine samples stored within 24±3 months from the onset of the LN flare in the LuNNET biobank. 12 17 Patients were excluded if they had ESRD at the time of the LN flare, defined as an eGFR <15 mL/ min/1.73 m 2 , were undergoing dialysis or had received a kidney transplant. The LN response to treatment was assessed on the date of the urine sample collection (24±3 months after the LN flare) using the PERR definition.
2
Baker R, et al . Lupus Science & Medicine 2026; 13 :e001724. doi:10.1136/lupus-2025-001724
Powered by FlippingBook