Abstract: PO1524
Using Whole-Exome Sequencing to Identify PKD1 and PKD2 in 50,000 UK Biobank Participants
Session Information
- Cystic Kidney Diseases: Mechanisms, Genetics, and Treatment
October 22, 2020 | Location: On-Demand
Abstract Time: 10:00 AM - 12:00 PM
Category: Genetic Diseases of the Kidneys
- 1001 Genetic Diseases of the Kidneys: Cystic
Authors
- Sukcharoen, Kittiya, Royal Devon and Exeter NHS Foundation Trust, Exeter, Devon, United Kingdom
- Bingham, Coralie, Royal Devon and Exeter NHS Foundation Trust, Exeter, Devon, United Kingdom
- Gale, Daniel P., University College London, London, London, United Kingdom
- Weedon, Michael, University of Exeter, Exeter, Devon, United Kingdom
- Oram, Richard A., University of Exeter, Exeter, Devon, United Kingdom
Background
Studies have demonstrated that genetic testing using whole-exome sequencing (WES) detects undiagnosed monogenic kidney disease in up to 2% patients with predefined clinical phenotypes and/or strong family histories of kidney disease. We aimed to take advantage of newly available datasets with WES and medical information on 50,000 people from UK Biobank (UKBB) to identify PKD1 and PKD2 variants in a sample not selected for kidney disease, to compare their phenotypic features to people with ICD 10 codes for PKD in UKBB.
Methods
We analysed data from the subset of 50,000 individuals from UK Biobank (n=500,000) who have had WES data released. We looked for mutations in PKD1 and PKD2. Our primary analysis involved looking for a subset of mutations, protein-truncating variants, that had a very high likelihood of being disease-causing. We performed standard quality control which included visual inspection and assessing individual mutation on genome databases.
Results
We found 53 protein truncating variants (44 in PKD1 and 9 in PKD2). The Average age for those with mutations was 57, the same as the UKBB population. We excluded 33 variants on the basis that they were either very common in GNOMAD therefore unlikely to be pathogenic or did not pass visual inspection on IGV plot. This left 20 likely pathogenic mutations (13 PKD1 and 7 PKD2). An ICD 10 code for PKD on hospital records was found in 8 of those with mutations. The 8 individuals with mutations and a PKD ICD 10 code had a more severe phenotype; 7/8 (88%) were hypertensive compared with 6/12 (50%) in those with mutations but without a PKD ICD10 code. Their renal function was worse (63% v 15% had CKD, eGFR 53 v 80, p=0.01) and 1 individual received a renal transplant.
Conclusion
We were able to find disease causing mutations in PKD1 and PKD2 and link this to phenotype in UKBB. People with protein truncating mutations and hospital codes for PKD had independent evidence of kidney disease however those without an ICD 10 code of PKD could either have milder undiagnosed PKD, or non-pathogenic mutations. The genetic complexity of PKD1 and 2, and the difficulty of ascertaining mutations with exome sequencing means that further work needs to be done to see if prevalence of PKD, and in particular undiagnosed mutations, could be assessed using WES from the complete UKBB dataset when available.