The Exploration of Non-Invasive Biomarkers for Lupus Nephritis

Abstract

Background

The current gold standard for Lupus Nephritis (LN) disease diagnosis involves evaluation of a kidney biopsy. Significant damage to the kidney must occur prior to seeing changes in proteinuria, GFR, etc. to justify a kidney biopsy, which cannot be routinely taken due to its invasive nature. We assessed the utilization of the Luminex platform for multiplex analysis of urine proteomics to develop algorithms to predict NIH Activity (NIH-AI) and Chronicity (NIH-CI) Indices. The utilization of urine tumstatin concentration as a single analyte marker for disease was also evaluated.

Methods

Urine and serum samples from healthy, systemic lupus erythematosus (SLE), LN, and non LN renal disease biopsy control (BC) patients were tested on a large Luminex multiplex. Stepwise regression and single variable regression results were combined to limit the possible candidate urine biomarkers. A multivariate logistic regression model (MLRM) was applied to the remaining markers. Tumstatin urine concentrations in healthy, SLE, LN, and BC patients were measured using Nordic Biosciences A/S (Herlev, Denmark) proprietary ELISA and normalized with creatinine concentrations. Immunohistochemistry (IHC) was performed on baseline LN and BC kidney biopsies for collagen IV, collagen IVα3, MPO/CD163, and MMP9.

Results

Data from 280 markers were evaluated and four markers (CD163, ferritin, KIM-1, and antileukoproteinase) were identified (P

Conclusion

Novel combinations of urine markers were used to predict activity and chronicity (high and low categories) with high accuracy in LN patients. These algorithms could be used to track disease progression without the need for biopsy. The potential use of tumstatin as a single urinary biomarker for early SLE or LN diagnosis/activity is supported by the findings of this study and further evaluation is warranted.