Abstract: FR-PO323
Frequent Genetic Variants in Autosomal Dominant Tubulointerstitial Kidney Disease
Session Information
- Cystic Kidney Diseases - II
November 03, 2017 | Location: Hall H, Morial Convention Center
Abstract Time: 10:00 AM - 10:00 AM
Category: Genetic Diseases of the Kidney
- 801 Cystic Kidney Diseases
Authors
- Olinger, Eric Gregory, University of Zurich, Zurich, Switzerland
- Schaeffer, Celine, San Raffaele Scientific Institute, MILAN, Italy
- Kidd, Kendrah O., Wake Forest University School of Medicine, Winston-Salem, North Carolina, United States
- Fuster, Daniel G., University Hospital of Bern, Bern, Switzerland
- Kistler, Andreas D., Cantonal Hospital Frauenfeld, Frauenfeld, Switzerland
- Sayer, John, Institute of Genetic Medicine, Newcastle, United Kingdom
- Bleyer, Anthony J., Wake Forest University School of Medicine, Winston-Salem, North Carolina, United States
- Rampoldi, Luca, San Raffaele Scientific Institute, MILAN, Italy
- Devuyst, Olivier, University of Zurich, Zurich, Switzerland
Background
An increasing number of purported pathogenic genetic variants are detected as relatively common in exome data from the general population, suggesting previous misclassification. Mutations in UMOD cause autosomal dominant tubulointerstitial kidney disease (ADTKD) that is rare (1-10/1’000’000), well indexed and has genetic features (private mutations, complete penetrance) making it paradigmatically suited to address the evolving distinction between normal genetic variation and pathogenic variants.
Methods
Reported UMOD mutations from our ADTKD registry were retrospectively compared to sequencing data from gnomAD (http://gnomad.broadinstitute.org). Matching variants were tested by cellular studies and segregation analysis.
Results
Based on then-available filtering strategies, 107 UMOD variants were reported as pathogenic in 181 families from our ADTKD registry. Two of them are reported in gnomAD: p.T62P (AF 0.00034) and p.T469M (0.00070). An allele frequency of ~5x10-8 would be expected based on the disease prevalence and the number of reported mutations. The p.T62P and p.T469M variants have been previously linked to ADTKD and they were detected respectively in 9 and 2 unrelated patients with CKD, gout and positive family history in our registry. T62 lacks any evolutionary constraint and trained classifiers (eg. PolyPhen-2) predict a benign to possibly damaging phenotype for p.T62P. T469 is a conserved residue and p.T469M is predicted to be damaging. Expression studies in kidney cell lines reveal an extremely mild or no trafficking defect for p.T62P and p.T469M, as opposed to the well-established mutation p.C150S. Furthermore, the p.T62P variant does not segregate with disease, including several aged carriers with absent kidney disease.
Conclusion
The high frequency of 2 imputed variants in UMOD led us to reevaluate the genetic diagnosis of several ADTKD families in our registry and to definitively infirm the pathogenicity of UMOD p.T62P. With evolving sequencing data, careful curation of implausibly common variants in Mendelian diseases other than ADTKD is warranted.