Abstract: FR-PO752
Plasma Metabolomics in Steroid-Sensitive and Steroid-Resistant Nephrotic Syndrome
Session Information
- Clinical/Diagnostic Renal Pathology and Lab Medicine - II
November 03, 2017 | Location: Hall H, Morial Convention Center
Abstract Time: 10:00 AM - 10:00 AM
Category: Glomerular
- 1004 Clinical/Diagnostic Renal Pathology and Lab Medicine
Authors
- Gooding, Jessica, RTI International, Durham, North Carolina, United States
- Agrawal, Shipra, Nationwide Children's Hospital , Columbus, Ohio, United States
- Mcritchie, Susan, University of North Carolina at Chapel Hill, Kannapolis, North Carolina, United States
- Acuff, Zachery James, RTI International, Durham, North Carolina, United States
- Smoyer, William E., Nationwide Children's Hospital , Columbus, Ohio, United States
- Sumner, Susan Jenkins, University of North Carolina Chapel Hill, Chapel Hill, North Carolina, United States
Background
Nephrotic syndrome (NS) is a common kidney disease in children. Steroids are the primary therapy; however, they are ineffective in ~20% cases. Children with steroid-resistant NS (SRNS) fail to enter remission after prolonged steroid treatment, and are at high risk for steroid-induced side effects as well as progression of disease to end-stage renal disease (ESRD) within five years. This study aimed to discover markers of steroid-resistance that could be used to predict SRNS at presentation and develop a mechanistic definition of SRNS.
Methods
Citrate plasmas (n=86) were collected from 30 steroid-sensitive NS and 15 steroid-resistant NS patients at presentation (prior to steroid therapy) and after an average of 7 weeks of steroid treatment. Broad spectrum 1HNMR data was acquired, binned, and concentration fit. Multivariate analyses and hypothesis testing were used to determine the metabolites that best differentiated the phenotypic groups,and logistic regression using a stepwise variable selection method were used model the odds of steroid resistance at presentation.
Results
Treatment effects were observed between paired presentation and follow-up steroid-sensitive (SSNS) samples and between follow-up SSNS and SRNS samples. Metabolites affected by treatment included lipoproteins, adipate, tyrosine, valine, alanine, glutamine, glucose, pyruvate and creatine. After controlling for age, the step-wise logistic regression model selected glutamine (OR= 1.01; 0.99-1.02 95% CI). A similar model with children age >3 only, indicated that children with increased levels of malonate (OR=0.94; 0.89-1.00 95% CI) had an increased odds of responding to treatment.
Conclusion
Known effects of corticosteroid treatment were observed providing a proof-of-concept. The observed metabolic signature supports previous hypotheses that the proximal tubule is involved in the pathology of SRNS as reflected in circulating metabolites of renal gluconeogenesis. After controlling for age, logistic regression suggests that malonate concentration may be a potential biomarker for identifying SRNS at presentation.
Funding
- NIDDK Support