Background/Purpose: Lupus nephritis (LN) is a major determinant of morbidity and mortality in Systemic Lupus Erythematosus (SLE).  Variability in clinical course, underlying renal injury, and response to treatment pose therapeutic challenges. Management of LN would be served by the discovery of biomarkers that accurately reflect changes in disease activity, aiding in the prompt identification of flares and evaluation of response to therapy.  Renal biopsy is the most reliable way to determine the extent and nature of renal injury, with serologic changes (anti-dsDNA antibodies) or measures of renal dysfunction (proteinuria) faltering in the diagnosis of impending flares and/or assessment of therapeutic response.  We used a proteomics approach to identify potential urinary biomarkers associated with LN.

Methods: Urine was obtained from 60 LN patients within 2 weeks of biopsy, 25 active non-LN SLE patients, and 24 controls.  The mean age and proportion of females (83-88%) was similar in the 3 groups. 128 distinct analytes were quantified by Luminex and normalized by scaling to urinary creatinine levels.  Data was analyzed by hierarchical clustering using divisive analysis (DIANA), linear modeling, and non-parametric statistics, with appropriate corrections for multiple comparisons.

Results: LN and non-LN SLE patients had comparable SLEDAI-2K scores (13.8±7.6 and 11.1±3.9, respectively), with the majority of the SLEDAI-2K in LN patients arising from the renal indices (renal SLEDAI = 8±4.3). The distribution of the renal biopsy classes (ISN-RPS) for the LN patients was: I – 1; II-3; III or III/V-12; IV or IV/V-32; V-9; VI-2; TIN-1.  The mean biopsy activity and chronicity scores biopsy were 6.68 (range 0-19) and 3.13 (range 0-10), respectively.  Following hierarchical clustering, significant clustering was seen for LN as compared to non-LN SLE patients and healthy.  Linear modeling was used to determine the urinary proteins whose abundance differed significantly between disease states (SLE vs healthy control) and between the presence or absence of LN (active LN vs active non-LN).  There were 9 analytes that differed significantly (q value <0.01) between SLE patients and controls and 42 between LN and non-LN, of which 37 differed only in LN patients as compared to active non-LN patients.   A number of proteins, not previously proposed as urinary LN biomarkers (e.g., MMP-2, TIMP-1), and known candidate LN biomarkers (e.g., adiponectin), were identified, with several of the novel biomarkers showing an enhanced ability to discriminate between LN and non-LN patients over potential biomarkers reported in the literature. Ten proteins were found to significantly correlate with the activity score on renal biopsy (q < 0.01), 5 of which (IP-10, vWF, adiponectin, IL-16 and PAI-1) strongly discriminated between active proliferative and non-proliferative/chronic renal lesions.

Conclusion: Using a proteomics approach we identified promising urinary biomarkers that correlated with the presence of active renal disease and/or renal biopsy changes. Experiments to assess the ability of these candidate biomarkers to fluctuate over time and predict clinically significant changes in renal disease activity and outcome are ongoing.

---

« Back to ACR/ARHP Annual Meeting 2014

ACR Meeting Abstracts - https://acrabstracts.org/abstract/identification-of-urinary-biomarkers-for-lupus-nephritis/