Article Text

AB0156 Quantitative Proteomic Approaches for Identifying Urinary Biomarkers in Lupus Nephritis
  1. V.G. Anania1,
  2. M. Cascino2,
  3. L.L. Dragone2,
  4. C. Looney1,
  5. J. Rae1,
  6. P. Liu3,
  7. W. Sandoval3,
  8. T.R. Katsumoto2,
  9. M. Neighbors1,
  10. A. Morimoto1,
  11. A.E. Herman1,
  12. W.R. Mathews1
  1. 1Biomarker Development
  2. 2Clinical Science
  3. 3Research, Genentech, South San Francisco, United States


Background Lupus nephritis (LN) is a severe clinical manifestation of systemic lupus erythematosus (SLE) associated with significant morbidity and mortality. Assessment of severity and activity of renal involvement in SLE requires a kidney biopsy, an invasive procedure with limited prognostic value. Despite years of research, a need remains for proximal, non-invasive biomarkers to help inform treatment decisions and to monitor disease activity and progression. Recent retrospective studies have revealed that baseline urine protein-creatinine ratio (UPCR) and reduction in UPCR during induction therapy are significantly associated with renal outcome1. Although proteinuria is highly correlated with disease progression in LN, the composition of the urinary proteome of LN patients remains poorly characterized.

Objectives To better characterize the urinary LN proteome, we performed a preliminary proteomics study comprised of three complimentary discovery proteomic methods to identify urinary biomarkers of LN.

Methods Urine samples from three female patients with biopsy confirmed LN, high proteinuria (>3 mg/mmol, and low serum complement C3 (<80 mg/dL) were compared to age- and gender- matched healthy controls. These samples were profiled using three mass spectrometry-based methods: 2D SDS-PAGE fractionation, a chemical labeling approach using tandem mass tags, and a label free data-independent acquisition (DIA) method.

Results Using these combined approaches >2600 proteins were identified, 290 of which are up-regulated >2-fold in LN samples compared to healthy controls. While the chemical labeling approach enabled identification of more total proteins (2,598 with chemical labeling vs. 919 with DIA, 1% False Discovery Rate), the DIA approach outperformed the chemical labeling approach in identification of proteins significantly up-regulated in LN samples (52 with chemical labeling vs. 173 with DIA).

Conclusions These results suggest that DIA-based approaches are less biased towards high abundance analytes and therefore potentially more suitable for proteomic profiling of biological matrices with a broad dynamic range like urine. Furthermore, candidate biomarkers identified using the DIA method are easily adapted into a targeted, multiplexed mass spectrometry assay suitable for absolute quantitation of candidate biomarkers in a clinical trial. Results from this study will be used to inform longitudinal and interventional studies focused on understanding the biological implications of these candidate biomarkers and to direct development of novel tools to evaluate disease progression and treatment efficacy of current and future LN therapeutics.

  1. Dall'Era et al. Lupus Sci Med. 2015.

Disclosure of Interest None declared

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.