Abstract: SA-OR045
Amniotic Fluid Peptide Biomarkers for In Utero Prediction of Postnatal Renal Function in CAKUT
Session Information
- Pediatric Nephrology and Developmental Biology
November 04, 2017 | Location: Room 285, Morial Convention Center
Abstract Time: 05:18 PM - 05:30 PM
Category: Developmental Biology and Inherited Kidney Diseases
- 403 Pediatric Nephrology
Authors
- Klein, Julie, Inserm U1048, Toulouse, France
- Buffin-Meyer, Benedicte, Inserm U1048, Toulouse, France
- Boizard, Franck, Inserm U1048, Toulouse, France
- Magalhães, Pedro, Mosaiques Diagnostics GmbH, Hannover, Germany
- Zürbig, Petra, Mosaiques Diagnostics GmbH, Hannover, Germany
- Breuil, Benjamin, Inserm U1048, Toulouse, France
- Levtchenko, Elena N., University Hospitals Leuven, Herent, Belgium
- Hindryckx, An, University Hospitals Leuven, Herent, Belgium
- Nadia, Lounis, Hopital des Enfants, Toulouse, France
- Auriol, Francoise, Hopital des Enfants, Toulouse, France
- Bascands, Jean-loup, Inserm U1048, Toulouse, France
- Decramer, Stéphane, Hopital des Enfants, Toulouse, France
- Schanstra, Joost, Inserm U1048, Toulouse, France
Group or Team Name
- Bioman-consortium
Background
Clinical management of fetuses with bilateral CAKUT is hampered by the lack of methods able to predict evolution towards kidney failure. Here we explored the amniotic fluid (AF) peptidome of 152 bilateral CAKUT pregnancies in order to identify biomarkers predictive of disease progression.
Methods
Using capillary electrophoresis coupled to mass spectrometry, comparison of the AF peptidome from 32 fetuses with normal postnatal renal function at 2 years and 18 fetuses with early renal failure allowed the identification of 59 differentially abundant peptides, including fragments from extracellular matrix proteins, osteopontin, proSAAS or thymosin beta-4.
Results
Modelling the 59 peptides into a random forest classifier combined with clinical features (AF volume and the age of the fetus at the time of sampling) predicted the renal outcome at 2 years in a separate validation cohort of 68 CAKUT fetuses with 89% sensitivity, 98% specificity and a positive likelihood ratio of 44. Next, we used the classifier to discriminate 34 CAKUT fetuses subjected to termination of pregnancy (TOP) but where fetopathology analysis either failed to demonstrate severe renal damage or was absent or inconclusive. The classifier scores suggested that TOP without severe renal damage could have been avoided in 80% of the cases. In addition, the classifier predicted 55% of absent/inconclusive fetopathology as having severe postnatal outcome and 45% with normal function. Taking into account likelihood ratio and sensitivity, we can speculate that approximately 40% of these fetuses with non-interpretable fetopathology were subjected to TOP while most likely evolving to normal postnatal renal function.
Conclusion
We believe that identification of the 59 AF peptide biomarkers is a significant step forward for antenatal prediction of the postnatal renal function outcome in CAKUT fetuses and should be of great help for early prenatal counsellingand improved clinical management of CAKUT pregnancies, hence alleviating the psychological burden imposed on the parents. [Equal contribution JK and BBM].
Funding
- Private Foundation Support