Whole Exome Sequencing of Patients with Steroid-Resistant Nephrotic Syndrome

Abstract

Background and objectives: Steroid-resistant nephrotic syndrome overwhelmingly progresses to ESRD. More than 30 monogenic genes have been identified to cause steroid-resistant nephrotic syndrome. We previously detected causative mutations using targeted panel sequencing in 30% of patients with steroid-resistant nephrotic syndrome. Panel sequencing has a number of limitations when compared with whole exome sequencing. We employed whole exome sequencing to detect monogenic causes of steroid-resistant nephrotic syndrome in an international cohort of 300 families.

Design, setting, participants, & measurements: Three hundred thirty-five individuals with steroid-resistant nephrotic syndrome from 300 families were recruited from April of 1998 to June of 2016. Age of onset was restricted to <25 years of age. Exome data were evaluated for 33 known monogenic steroid-resistant nephrotic syndrome genes.

Results: In 74 of 300 families (25%), we identified a causative mutation in one of 20 genes known to cause steroid-resistant nephrotic syndrome. In 11 families (3.7%), we detected a mutation in a gene that causes a phenocopy of steroid-resistant nephrotic syndrome. This is consistent with our previously published identification of mutations using a panel approach. We detected a causative mutation in a known steroid-resistant nephrotic syndrome gene in 38% of consanguineous families and in 13% of nonconsanguineous families, and 48% of children with congenital nephrotic syndrome. A total of 68 different mutations were detected in 20 of 33 steroid-resistant nephrotic syndrome genes. Fifteen of these mutations were novel. NPHS1, PLCE1, NPHS2, and SMARCAL1 were the most common genes in which we detected a mutation. In another 28% of families, we detected mutations in one or more candidate genes for steroid-resistant nephrotic syndrome.

Conclusions: Whole exome sequencing is a sensitive approach toward diagnosis of monogenic causes of steroid-resistant nephrotic syndrome. A molecular genetic diagnosis of steroid-resistant nephrotic syndrome may have important consequences for the management of treatment and kidney transplantation in steroid-resistant nephrotic syndrome.

Keywords: Child; Exome; Humans; Kidney Failure, Chronic; Mutation; Nephrosis, congenital; Phenotype; Renal Insufficiency, Chronic; genetic renal disease; kidney transplantation; molecular genetics; nephrotic syndrome; pediatric.

Graphical abstract

Figure 1.

In 74 of 300 (25%) families with steroid-resistant nephrotic syndrome, a causative mutation was detected in one of 20 genes known to cause steroid-resistant nephrotic syndrome (shades of blue). In 3.7% of families, a mutation was found in genes causing a kidney disease that may represent phenocopies of steroid-resistant nephrotic syndrome (orange). In 28% of families, one or more potential novel candidate genes were identified (red). In 44% of families, no causative mutations or candidate genes were detected. SRNS, steroid-resistant nephrotic syndrome.

Figure 2.

A causative mutation in a steroid-resistant nephrotic syndrome gene, phenocopy gene, or novel candidate gene is detected in a greater proportion of consanguineous families compared to non-consanguineous families. We detected a causative mutation in 38% of consanguineous families and 13% of nonconsanguineous families. Through homozygosity mapping and a recessive hypothesis, we were able to identify potentially causative mutations in 42% of consanguineous families. Potential causative mutations in novel candidate genes were detected in nonconsanguineous families by evaluating for overlapping genes in siblings. Percentages>10% are rounded to the nearest whole number. SRNS, steroid-resistant nephrotic syndrome.

Figure 3.

After dividing the 335 individuals from 300 families with steroid-resistant nephrotic syndrome by gene identification status (steroid-resistant nephrotic syndrome gene, phenocopy gene, no mutation detected) and sorting by age and sex, the median age of individuals with a mutation detected in a steroid resistant nephrotic syndrome gene is significantly lower than the median age of individuals with a mutation detected in a phenocopy gene or individuals with no mutation detected. (A) Families in whom a causative mutation in a known steroid-resistant nephrotic syndrome gene (blue) or phenocopy gene (orange) was detected as compared with those families where no causative mutation was detected (gray). Bars and numbers at end of bars represent number of affected individuals in each category, divided into those with a causative mutation detected in a steroid-resistant nephrotic syndrome gene (blue), those with a causative mutation detected in a phenocopy gene (orange), and those without a causative mutation detected (gray). Percentages at end of each bar reflect the same three categories. Percentages>10% are rounded to the nearest whole number. (B) Median age of onset in patients with a causative mutation detected in a steroid-resistant nephrotic syndrome gene was 1.7 years versus 4 years in those without a mutation detected (range, 0–24 years). For those with a causative mutation detected in a steroid-resistant nephrotic syndrome gene, the range was 0–21 years. Mann–Whitney U test P<0.01. Median age of individuals with a phenocopy mutation detected was 4 years (range, 0.3–16), which was not statistically significant. Data of the characteristics of the steroid-resistant nephrotic syndrome cohort compared with the subcohort of those individuals with a causative mutation detected in a steroid-resistant nephrotic syndrome gene or phenocopy gene are given in Supplemental Table 3. SRNS, steroid-resistant nephrotic syndrome.

Figure 4.

Gene identification in a steroid-resistant nephrotic syndrome gene occurs in a statistically significant greater proportion of homozygous families (homozygosity >100 Mb) when compared to non-homozygous families (homozygosity <100 Mb) and in a statistically significant greater proportion of consanguineous families when compared to non-consanguineous families. Families in whom causative mutations in a known steroid-resistant nephrotic syndrome gene (blue) or a phenocopy gene (orange) were detected, compared with those families in whom no causative mutation was detected (gray). Bars and numbers at end of bars represent total number of families in each category, divided into those families with a causative mutation detected (blue), those families with a causative mutation detected in a phenocopy gene (orange), and those families without a causative mutation detected (gray). Percentages at end of each bar reflect the same three categories. Percentages>10% are rounded to the nearest whole number. Rate of detection of a causative mutation in a steroid-resistant nephrotic syndrome gene did not vary with number of affected individuals per family. Number of affected individuals per family did not have a statistically significant difference between one affected individual per family versus two affected individuals, or between one affected individual and ≥3 individuals. Mutation detection rate in a steroid-resistant nephrotic syndrome gene was significantly higher in those families that were reported clinically as consanguineous or had homozygosity on mapping >100 Mbp than those that were nonconsanguineous or had homozygosity<100 Mbp on mapping (two-sided chi-squared test P<0.001 for each condition). Data of the characteristics of the steroid-resistant nephrotic syndrome cohort compared with the subcohort of those families with a causative mutation detected in a steroid-resistant nephrotic syndrome gene or phenocopy gene are given in Supplemental Table 4. *Statistically significant. SRNS, steroid-resistant nephrotic syndrome.