Optimizing next-generation sequencing for genetic diagnosis in autosomal dominant polycystic kidney disease

Autosomal dominant polycystic kidney disease (ADPKD) affects 1:1000, causing 5% to 10% of kidney failure. The primary disease gene, PKD1, has 6 pseudogenes with 97% to 99% homology, a >12-kb transcript, high GC content, and polypyrimidine tracts. Although long-range polymerase chain reaction with Sanger sequencing has been the "gold-standard," next-generation sequencing (NGS) is increasingly used.

We performed exome sequencing (ES) on 203 ADPKD patients in 171 families from the Consortium for Radiologic Imaging Studies of Polycystic Kidney Disease cohort, with prior "gold-standard" results: n = 157 PKD1, n = 27 PKD2, n = 19 with no pathogenic variant detected. Clinical geneticists, blinded to genotype, reviewed ES data. We assessed effects of pipeline modifications, exome capture reagents, pseudogene alignment, and deeper ES or genome sequencing (GS) for unsolved cases.

Optimized ES identified 95.5% of defined PKD1 pathogenic variants, all PKD2 variants, and at least 9 of 19 "unsolved" cases. Standard pipelines on research-grade ES missed at least 22 PKD1 variants due to Genome Analysis Toolkit HardFiltering or alternative locus annotation. Higher-depth ES achieved 100% PKD1 variant detection. GS identified a balanced translocation t(1;16)(q31.1;p13.3).

NGS matches "gold-standard" sensitivity and considers additional disease genes. ES plus GS solve 96% of well-phenotyped ADPKD. We outline practical considerations for NGS on PKD1.