Summary
A Japanese medRxiv preprint reports clinically concordant genomic diagnoses in 10 of 108 sequenced patients who began dialysis by age 50, after clinically apparent polycystic kidney disease was excluded. Nine diagnoses had pathogenic or likely pathogenic genetic support.
A Japanese medRxiv preprint reports that whole-genome sequencing, interpreted alongside clinical findings, identified clinically concordant diagnoses in 10 of 108 people who began maintenance dialysis at age 50 or younger. The analysis excluded patients with clinically apparent polycystic kidney disease (PKD) before sequencing; the findings come from a voluntary, nonrandomly enrolled group, not a representative sample of all young people on dialysis.
What the genomic analysis found
The researchers first identified 355 patients who had started dialysis by age 50 among 1,139 people receiving maintenance dialysis at participating Japanese facilities. After excluding 17 with clinically apparent PKD, 338 were eligible; 108 underwent whole-genome sequencing. The analysis covered 465 nuclear candidate genes, mitochondrial DNA, and copy-number and structural variants. These are different types of genetic changes, including changes in the number or arrangement of DNA segments.
Using clinical and genetic evidence together, the team classified 10 participants as having a diagnosis: seven renal and three systemic or metabolic. That is 9.3% of the sequenced group, with a 95% confidence interval of 5.1–16.2%. Nine diagnoses (8.3%; 95% confidence interval, 4.4–15.1%) were supported by a pathogenic or likely pathogenic genetic finding. These classifications indicate that the evidence supported a disease-causing role for the relevant variant.
The remaining diagnosis was clinical Dent disease, associated with a CLCN5 variant of uncertain significance. The researchers included it on the basis of the participant’s clinical presentation, rather than counting it among diagnoses supported by a pathogenic or likely pathogenic variant.
Diagnoses appeared among participants whose existing clinical categories included diabetic kidney disease, nephrosclerosis, chronic glomerulonephritis, kidney failure of unknown cause, and other categories. This suggests that genetic findings can add a possible underlying diagnosis even when a clinical label is already in place; the reported results do not establish how often such findings change treatment or patient outcomes.
Interpreting the findings
A monogenic diagnosis points to a condition associated with a change in a single gene. The study’s use of both sequencing and clinical assessment matters: a genetic variant must be interpreted in the context of a person’s features, and a variant of uncertain significance is not equivalent to a pathogenic finding.
In a post hoc analysis that excluded three participants with low target coverage, the integrated diagnostic proportion was 9 of 105 (8.6%); the pathogenic/likely pathogenic-supported proportion was 8 of 105 (7.6%). Coverage refers to how much of the targeted DNA was adequately read. Variation in coverage can affect how confidently a negative sequencing result is interpreted.
The work is a preprint, and its diagnostic yield comes from 108 people who chose to enroll from 338 eligible patients. The authors therefore identify nonrandom enrollment and variable sequencing coverage as limits on generalisability and on interpreting negative results. They also included Gitelman syndrome as a clinically manifest renal disorder while noting that its contribution to kidney failure was uncertain.