Prevalence and characteristics of genetic disease in adult kidney stone formers

ABSTRACT

Background

Molecular mechanisms of kidney stone formation remain unknown in most patients. Previous studies have shown a high heritability of nephrolithiasis, but data on the prevalence and characteristics of genetic disease in unselected adults with nephrolithiasis are lacking. This study was conducted to fill this important knowledge gap.

Methods

We performed whole exome sequencing in 787 participants in the Bern Kidney Stone Registry, an unselected cohort of adults with one or more past kidney stone episodes [kidney stone formers (KSFs)] and 114 non-kidney stone formers (NKSFs). An exome-based panel of 34 established nephrolithiasis genes was analysed and variants assessed according to American College of Medical Genetics and Genomics criteria. Pathogenic (P) or likely pathogenic (LP) variants were considered diagnostic.

Results

The mean age of KSFs was 47 ± 15 years and 18% were first-time KSFs. A Mendelian kidney stone disease was present in 2.9% (23/787) of KSFs. The most common genetic diagnoses were cystinuria (SLC3A1, SLC7A9; n = 13), vitamin D-24 hydroxylase deficiency (CYP24A1; n = 5) and primary hyperoxaluria (AGXT, GRHPR, HOGA1; n = 3). Of the KSFs, 8.1% (64/787) were monoallelic for LP/P variants predisposing to nephrolithiasis, most frequently in SLC34A1/A3 or SLC9A3R1 (n = 37), CLDN16 (n = 8) and CYP24A1 (n = 8). KSFs with Mendelian disease had a lower age at the first stone event (30 ± 14 versus 36 ± 14 years; P = .003), were more likely to have cystine stones (23.4% versus 1.4%) and less likely to have calcium oxalate monohydrates stones (31.9% versus 52.5%) compared with KSFs without a genetic diagnosis. The phenotype of KSFs with variants predisposing to nephrolithiasis was subtle and showed significant overlap with KSFs without diagnostic variants. In NKSFs, no Mendelian disease was detected and LP/P variants were significantly less prevalent compared with KSFs (1.8% versus 8.1%).

Conclusion

Mendelian disease is uncommon in unselected adult KSFs, yet variants predisposing to nephrolithiasis are significantly enriched in adult KSFs.

Graphical Abstract

Graphical Abstract.

KEY LEARNING POINTS.

What was known:

• Kidney stone formation is strongly influenced by genetic factors (positive family history in 30–60%) and >30 Mendelian forms of kidney stone disease have been described, explaining 6.8–29.4% of stones in selected populations.

• However, the diagnostic utility of whole exome sequencing in unselected adults with nephrolithiasis has not been established.

• Current strategies to manage this common and relapsing disease are inadequate.

This study adds:

• Mendelian disease in unselected adults with nephrolithiasis is rare and far less common than previously reported.

• Genetic variants predisposing to kidney stone formation are significantly enriched in individuals with nephrolithiasis, yet the associated phenotypes display a significant overlap with individuals without diagnostic variants.

• While many Mendelian forms of nephrolithiasis can be detected with thorough phenotypic analysis, genetic testing can ameliorate the diagnosis in adult stone formers, where phenotypes are often less pronounced.

Potential impact:

• Whole exome sequencing allows accurate assessment of Mendelian disease and identification of predisposing variants in adult individuals with nephrolithiasis.

• Our study highlights the potential of genetic testing to direct patients to personalized therapies and clinical trials.

INTRODUCTION

Nephrolithiasis is a common global healthcare problem [1]. Kidney stones recur frequently and cause substantial morbidity, reduced quality of life and enormous cost [2–4]. A comprehensive phenotypic examination is paramount for the detection of urinary abnormalities and provides an efficient way for the identification of underlying Mendelian disease. Still, in many adult patients with kidney stones, the molecular pathogenesis of clinical traits associated with kidney stone formation (e.g. hypercalciuria) remains unknown. Consequently, undifferentiated dietary and pharmacological preventive measures are initiated, but many patients continue to form stones [5, 6]. Current strategies in managing this very common and debilitating disease are clearly inadequate and is an unmet need for novel diagnostic and therapeutic approaches.

Kidney stone formation is strongly influenced by genetic factors: a positive family history is present in 30–60% of individuals with kidney stones [7], and both twin [8] and genealogy [9] studies revealed a high heritability of nephrolithiasis. More than 30 Mendelian (also called monogenic) causes of nephrolithiasis have been described thus far [10]. Recent studies have also highlighted the importance of intermediate effect size/incomplete penetrance variants in increasing the risk for nephrolithiasis [11–14]. Identification of patients with Mendelian forms of nephrolithiasis remains a major challenge and no clear consensus exists on clinical parameters to guide genetic testing. Clinical phenotypes can be diagnostic for many Mendelian diseases—if phenotypic screening is diligently performed—and usually prompt personalized genetic testing. Yet, many cases of genetic forms of nephrolithiasis are missed and incorrectly labelled as ‘idiopathic’.

Whole exome sequencing (WES) is widely applied as a diagnostic tool for rare diseases and the detection of pathogenic variants in cancer [15, 16]. In contrast, the diagnostic utility of WES has not been established for most constitutional disorders, such as nephrolithiasis. In small, selected cohorts of early-onset or familial nephrolithiasis, a Mendelian cause was identified in 6.8–29.4% of cases [10, 17–19]. However, the frequency and phenotypic spectrum of genetic disease in sporadic adult-onset nephrolithiasis, by far the most common type encountered in clinical routine, is unknown. To address these important knowledge gaps, we analysed a WES-based gene panel in a deeply phenotyped, little selected European cohort of 787 adult kidney stone formers (KSFs) and in 114 non-kidney stone formers (NKSFs).

MATERIALS AND METHODS

Study cohort

The study was conducted with participants enrolled in the Bern Kidney Stone Registry (BKSR), an observational cohort of adult KSFs described previously and in the SI [20–23]. Inclusion criteria for the BKSR are written informed consent, age ≥18 years and one or more past kidney stone episodes.

Inclusion criteria for NKSFs were written informed consent, age ≥18 years and no history of past kidney stones and no evidence of asymptomatic nephrolithiasis or nephrocalcinosis on ultrasound at enrolment.

Whole exome sequencing and variant calling

Isolation of genomic DNA, exome capture, high-throughput sequencing and bioinformatic analysis including joint variant calling using the Genome Analysis Toolkit version 3.8 (GATK) according to the GATK best practices recommendations [24, 25] and annotation using Ensembl Variant Effect Predictor Release 106 (https://useast.ensembl.org/info/docs/tools/vep/index.html) [26] was performed using established methods, described in more detail in the Supplementary Methods. After filtering for predicted consequences and Genome Aggregation Database (gnomAD; https://gnomad.broadinstitute.org/) minor allele frequency (<1% in all populations) [27], variants in 34 genes [10, 17–19, 28] previously implicated in Mendelian kidney stone disease were examined (Fig. 1, lower panel). Further variant stratification was performed according to the recommendations of the American College of Medical Genetics and Genomics (ACMG) [29] after manual review of phenotype data. Only variants classified as likely pathogenic (LP) or pathogenic (P) applying the ACMG criteria were defined as diagnostic variants leading to a genetic diagnosis. Separation into Mendelian or LP/P variants predisposing to nephrolithiasis was conducted depending on the evidence available from this study and/or from published case–control studies, described in more detail in the SI [11, 12, 30, 31] (Fig. 1 and Supplementary Methods).

Figure 1:

(upper panel) Flowchart of patient inclusion, exclusion and genetic analysis in the BKSR. Data from 1152 individuals recruited into the BKSR were analysed. After exclusion of individuals without genetic information available, declined consent or multiple inclusions, the analysed cohort consisted of 901 individuals (787 KSFs and 114 NKSFs). Before inclusion in the BKSR, NKSFs underwent ultrasound imaging to exclude nephrolithiasis (NL) and/or nephrocalcinosis (NC). Genetic analysis was performed in a standardized prioritization pathway in 34 known kidney stone genes. N: number of individuals; genetic diagnosis: likely pathogenic or pathogenic variant according to ACMG criteria; LP/P variant: monoallelic LP/P variant, predisposing to nephrolithiasis. (lower panel) Kidney stone disease gene panel used for genetic analysis. Panel of 34 known nephrolithiasis genes used with their inheritance mode, as accepted for classification as ‘Mendelian disease’ in this article, grouped by phenotypes. AD: autosomal dominant; AR: autosomal recessive; XLR: X-linked recessive; MODY: maturity-onset diabetes of the young; FRTS: Fanconi renal tubular syndrome.

Statistical analysis

Continuous variables were reported as medians with 25th–75th percentiles [interquartile range (IQR)] or means with standard deviations (SDs) and categorical variables were reported as counts with percentages, as appropriate. Different statistical methods were employed to analyse differences between independent groups, depending on the nature of the variables involved. The Mann–Whitney U test was used for non-normally distributed continuous variables, while the Student's t-test was applied for normally distributed variables. For categorical variables, the Fisher's exact test was used. Findings from these analyses were summarized and presented as descriptive tables. Statistical tests were two-sided and a P-value <.05 was considered statistically significant. Statistical analyses were performed using Stata version 16 (StataCorp, College Station, TX, USA). Pie charts and dot plots were generated with GraphPad Prism version 8.4.3 (GraphPad Software, San Diego, CA, USA).

RESULTS

Study cohort

We performed WES in 901 individuals, including 787 KSFs and 114 NKSFs (Fig. 1). Clinical characteristics of the study cohort, separated as KSFs and NKSFs, are shown in Table 1. The mean age was 46.6 ± 14.5 years in KSFs and 42.4 ± 14.9 years in NKSFs. The majority of KSFs were male [71.8% (n = 565)], but the percentage of men was lower in NKSFs [54% (n = 61)]. Consistent with previous reports, KSFs were significantly more likely to report a positive family history of kidney stones compared with NKSFs (43.6% versus 6.8%; Table 1) [32, 33], and lumbar spine bone mineral density (BMD) was lower in KSFs compared with NKSFs (Table 1) [34, 35].

Table 1:

Characteristics of the study cohort.

Characteristics	KSFs (n = 787)	NKSFs (n = 112)a	P-value
Age (years), mean (SD)	46.64 (14.57)	42.41 (14.88)	.007
Male, n (%)	565 (71.8)	61 (54.0)	<.001
Body mass index (kg/m2), mean (SD)	26.83 (5.08)	26.34 (4.55)	.37
Hypertension, n (%)	255 (36.0)	28 (35.9)	1.00
Diabetes, n (%)	24 (3.1)	1 (1.1)	.50
Obesity (BMI ≥30 kg/m2), n (%)	168 (22.4)	14 (15.1)	.11
Hyperuricemia, n (%)	346 (44.0)	54 (47.8)	.48
Family history of kidney stone disease, n (%)	317 (43.6)	3 (6.8)	<.001
Kidney stone recurrence (>1 stone event), n (%)	614 (82.2)
Age at first stone event (years), mean (SD)	35.41 (13.85)
Total number of stone events, median (IQR)	3.00 (2.00–4.00)
Blood parameters
Total calcium (mmol/l), mean (SD)	2.36 (0.12)	2.30 (0.12)	<.001
Ionized calcium (mmol/l), median (IQR)	1.21 (1.19–1.23)	1.19 (1.18–1.22)	.002
Phosphorus (mmol/l), mean (SD)	1.01 (0.17)	1.01 (0.16)	.89
Magnesium (mmol/l), mean (SD)	0.83 (0.07)	0.86 (0.08)	<.001
Uric acid (μmol/l), mean (SD)	328.96 (169.56)	308.44 (71.51)	.26
Intact PTH (ng/l), median (IQR)	39.06 (30.00–51.00)	37.50 (30.50–48.00)	.56
25-OH vitamin D3 (nmol/l), median (IQR)	40.18 (27.00–59.00)	48.50 (32.72–72.44)	.017
eGFR creatinine (CKD-EPI 2009 equation, ml/min/1.73 m2), mean (SD)	94.18 (21.08)	96.31 (21.88)	.36
Urine parameters
Total urine volume (l/24 h), median (IQR)	1.98 (1.47–2.54)	2.10 (1.49–2.65)	.41
Urine pH, median (IQR)	5.94 (5.39–6.60)	6.00 (5.50–6.75)	.14
Urine sodium:creatinine ratio (mmol/mmol/24 h), median (IQR)	13.62 (10.97–16.98)	13.76 (9.92–16.21)	.37
Urine potassium:creatinine ratio (mmol/mmol/24 h), median (IQR)	4.70 (3.77–5.82)	5.56 (4.43–6.55)	<.001
Urine uric acid:creatinine ratio (mmol/mmol/24 h), mean (SD)	0.61 (0.10)	0.38 (0.27)	<.001
Urine calcium:creatinine ratio (mmol/mmol/24 h), median (IQR)	0.43 (0.30–0.58)	0.33 (0.21–0.44)	<.001
Urine magnesium:creatinine ratio (mmol/mmol/24 h), median (IQR)	0.30 (0.24–0.38)	0.32 (0.27–0.42)	.059
Urine citrate:creatinine ratio (mmol/mmol/24 h), median (IQR)	0.19 (0.13–0.27)	0.24 (0.18–0.30)	.003
Urine phosphate:creatinine ratio (mmol/mmol/24 h), mean (SD)	2.18 (0.54)	2.24 (0.57)	.41
Urine oxalate:creatinine ratio (mmol/mmol/24 h), median (IQR)	0.03 (0.02–0.04)	0.03 (0.02–0.05)	.062
Stone phenotypes, n (%)
Calcium oxalate dihydrate	73 (12.9)
Calcium oxalate monohydrate	287 (50.8)
Calcium phosphate	130 (23.0)
Uric acid	38 (6.7)
Cystine	18 (3.2)
Struvite	15 (2.7)
BMD, mean (SD)
Lumbar spine (g/cm2)	1.02 (0.14)	1.06 (0.13)	.011
Lumbar spine T-score	−0.57 (1.26)	−0.11 (1.20)	.005
Lumbar spine Z-score	−0.11 (1.38)	0.29 (1.29)	.024
Femoral neck (g/cm2)	0.84 (0.13)	0.86 (0.12)	.21
Femoral neck T-score	−0.55 (1.02)	−0.31 (0.95)	.076
Femoral neck Z-score	0.11 (0.99)	0.24 (0.86)	.30

^aThe NKSFs with LP/P variants (#5948, #5500) were removed for phenotypic analyses.

Characteristics are indicated separately for participants with and without kidney stone disease.

eGFR: estimated glomerular filtration rate.

Genetic analysis

Among 787 unrelated, unselected adult KSFs, we discovered in 23 patients (2.9%) with a Mendelian form of nephrolithiasis, i.e. pathogenic (P) or likely pathogenic (LP) variants in genes associated with Mendelian forms of nephrolithiasis respecting the relevant mode of inheritance (autosomal dominant and/or autosomal recessive and X-linked recessive) (Table 2, Fig. 1, Fig. 2). Mendelian disease was detected in 9 of 34 analysed genes (n = number of patients): AGXT (n = 1), ATP6V1B1 (n = 1), CYP24A1 (n = 5), GRHPR (n = 1), HOGA1 (n = 1), SLC3A1 (n = 9), SLC4A1 (n = 1), SLC7A9 (n = 4) and SLC12A1 (n = 1).

Table 2:

Genotype and phenotype of patients with Mendelian diagnosis.

ID	Gene	Inheritance	Allelic state	Nucleotide change	Amino acid change	ACMG class	ACMG criteria	AF gnoMAD (allel/tot/hom) gnoMAD (NFE)	Acc. no. dbSNP	Sex	Age at first stone (years)	Past stone events, n	Stone composition analysis (%)	Family history	Key phenotype	Clinical diagnosis before WES	Genetic diagnosis after WES
AGXT—Primary Hyperoxaluria Type 1 (PH1)
5011	AGXT	AR	hom	c.466G>A	p.Gly156Arg	P	PS1, PM5, PM1, PP2, PM2, PP3, PP5	7/226.862/0 5/101.706/0	NM_000030.3 rs121908530	M	4	8	1) 80 COM, 20 COD 2) 80 COM, 20 COD	N	CaOx NL, normal kidney function Hyperoxaluria (1100)	PH1	PH1
ATP6V1B1–Distal renal tubular acidosis with sensorineural hearing loss/deafness (AR dRTA)
5775	ATP6V1B1	AR	hom	c.242T>C	p.Leu81Pro	P	PP3s, PM2, PP5	1/250.636/0 0/113.172/0	NM_001692.4 rs121964880	M	N/A	>5	100 CA	N	CaP NL, dRTA, medullary NC, sensorineural deafness	AR-dRTA	AR-dRTA
CYP24A1–(Infantile) HC/NL, 1,25-OH vitamin D-24 hydroxylase deficiency
5033	CYP24A1	AR	hom	c.1186C>T	p.Arg396Trp	P	PM2, PM5, PP3m, PP5vs	167/251.228/1 124/113.676/1	NM_000782.5 rs114368325	F	24	1	N/A	Y	NL, osteoporosis, single kidney cyst. Hypercalcemia (total 2.64; ionized 1.4) Normal PTH (47) Normal 25-Vit D (83) Normal 1,25-Vit D (105) Hypercalciuria (7.7)	Idiopathic, infantile HC, NL, s/p parathyreo-idectomy	(Infantile) HC/NL
5052	CYP24A1	AR	hom	c.1186C>T	p.Arg396Trp	P	PM2, PM5, PP3m, PP5vs	167/251.228/1 124/113.676/1	NM_000782.5 rs114368325	M	N/A	1	N/A	Y	NL, Nephrocalcinosis, multiple kidney cysts, CKD stage 4, Osteopenia, Chondrocalcinosis. Hypercalcemia (total 2.8; ionized 1.41) Low PTH (4) Low 25-Vit D (16) Normal 1,25-Vit D (112) Hypercalciuria (7.2)	Infantile HC/NL, CKD IV° due to NC, s/p parathyreo-idectomy	(Infantile) HC/NL
5152	CYP24A1	AR	het	c.1186C>T	p.Arg396Trp	P	PM2, PM5, PP3m, PP5vs	167/251.228/1 124/113.676/1	NM_000782.5 rs114368325	M	26	2	N/A	N	NL, multiple kidney cysts. Normocalcemia (total 2.4; ionized 1.26) Low PTH (10) Normal 25-Vit D (87) Normal 1,25-Vit D (177) Hypercalciuria (19)	Idiopathic NL, polycystic kidney disease	(Infantile) HC/NL
			het	c.428_430 del	p.Glu143del	P	PM2, BS2, PM4, PP5vs	133/251.266/1 119/113.708/1	rs777676129
5334	CYP24A1 (NB* also CLDN16 monoallelic)	AR	het	c.1226T>C	p.Leu409Ser	LP	PM2, PP5vs	188/251.066/0 153/113.656/0	NM_000782.5 rs6068812	M	21	3	1) 90 COM, 10 COD; 2) 80 COM, 20 COD	Y	CaOx NL, osteopenia, multiple kidney cysts. Normal plasma calcium (total 2.48; ionized 1.27) Normal plasma magnesium (0.72) Low PTH (9) Normal 25-Vit D (59) Normal 1,25-Vit D (93) Hypercalciuria (14.3)	Idiopathic NL, polycystic kidney disease	(Infantile) HC/NL
			het	c.400T>G	p.Trp134Gly	LP	PM2, PP3m, PP5s	4/251.146/0 3/113.702/0	rs1170841548
5794	CYP24A1	AR	hom	c.428_430del	p.Glu143del	P	PM2, BS2, PM4, PP5vs	133/251.266/1 119/113.708/1	NM_000782.5 rs777676129	M	61	2	N/A	N	NL, NC, multiple kidney cysts, incomplete dRTA Normal plasma calcium (total 2.49; ionized 1.25) Low PTH (12) Low 25-Vit D (26) Normal 1,25-Vit D (73) Hypercalciuria (10.2)	Medullary sponge kidney	(Infantile) HC/NL
GRHPR—Primary hyperoxaluria, type 2 (PH2)
5432	GRHPR	AR	hom	c.103del	p.Asp35ThrfsTer11	P	PVS1, PM2, PP5vs	59/248.776/0 57/111.550/0	NM_012203.2 rs80356708	M	28	8	1) 80 COM, 20 COD 2): 90 COM, 10 COD 3): 90 COM, 10 COD	Y	CaOx NL, CKD stage 3 Hyperoxaluria (1796) Low urine glycolate/high urine glycerate Hypocitraturia (1.3)	PH2, start dialysis at age 57	PH2
HOGA1 –Primary hyperoxaluria, type 3 (PH3)
5595	HOGA1	AR	hom	c.700 + 5G>T	p.?	P	PM2, PP3m, PP5vs	312/251.412/1 239/113.698/1	NM_138413.4 rs185803104	M	36	5	80 COM, 20 COD	N	CaOx NL, CKD stage 2 Normal urine calcium (2.88) Normal urine oxalate random diet (428–463), high urine oxalate low Ca/Na diet (1090)	Idiopathic NL	PH3
SLC3A1–Cystinuria type A
5030	SLC3A1	AR	het	c.1400T>C	p.Met467Thr	P	PM1, PP2, PM2, PM5, PP3, PP5vs, PP4	627/251.156/4 479/113.498/3	NM_000341.4 rs121912691	F	18	5	80 Cystine, 20 CA	Y	Cystine NL, CKD stage 3, osteoporosis. Urine cystine (115-254 mmol/mol Creatinine) Norm: <30)	Cystinuria	Cystinuria type A
			het	c.1617 + 1G>A	p.?	LP	PVS1s, PM2, PP4	4/250.972/0	rs558461213
5092	SLC3A1	AR	het	c.851A>G	p.Asp284Gly	LP	PP3s, PM2, PP2, PP4	NF/NF	NM_000341.4 –	M	17	1	100 Cystine	Y	Cystine NL, Urine cystine (263 mmol/mol Creatinine) Norm: <30)	Cystinuria	Cystinuria type A
			het	c.1617 + 5G>A	p.?	LP	PS4m, PM2, PP3m, PP4	NF/NF	–
5094	SLC3A1	AR	het	c.1400T>C	p.Met467Thr	P	PM1, PP2, PM2, PM5, PP3, PP5vs, PP4	627/251.156/4 479/113.498/3	NM_000341.4 rs121912691	M	54	5	1) 90 CA 10 COM 2) 100 cystine	Y	Cystine NL, CKD stage 2, kidney cysts Urine cystine (247 mmol/mol Creatinine) Norm: <30)	Cystinuria	Cystinuria type A
			het	c.809G>A	p.Arg270Gln	LP	PP3s, PM2, PP2, PP4	2/251.430/0 0/113.720/0	rs142358712
5156	SLC3A1	AR	hom	c.1400T>C	p.Met467Thr	P	PM1, PP2, PM2, PM5, PP3, PP5vs, PP4	627/251.156/4 479/113.498/3	NM_000341.4 rs121912691	M	22	5	100 cystine	Y	Cystine NL Urine cystine (193-235 mmol/mol Creatinine) Norm: <30)	Cystinuria	Cystinuria type A
5162	SLC3A1	AR	het	c.1400T>C	p.Met467Thr	P	PM1, PP2, PM2, PM5, PP3, PP5vs, PP4	627/251.156/4 479/113.498/3	NM_000341.4 rs121912691	F	24	5	100 cystine	N	Cystine NL, CKD stage 3 Urine cystine (267-939 mmol/mol Creatinine) Norm: <30)	Cystinuria	Cystinuria type A
			het	c.787A>C	p.Ser263Arg	LP	PP3m, PM2, PP2, PP4	NF/NF	–
5168	SLC3A1	AR	het	c.851A>G	p.Asp284Gly	LP	PP3s, PM2, PP2, PP4	NF/NF	NM_000341.4 –	F	18	5	1) 100 cystine 2) 100 cystine 3) 90 cystine, 10 CA	Y	Cystine NL Urine cystine (213-435 mmol/mol Creatinine) Norm: <30)	Cystinuria	Cystinuria type A
			het	c.1617 + 5G>A	p.?	LP	PS4m, PM2, PP3, PP4	NF/NF	–
5479	SLC3A1	AR	het	c.1400T>C	p.Met467Thr	P	PM1, PP2, PM2, PM5, PP3, PP5vs, PP4	627/251.156/4 479/113.498/3	NM_000341.4 rs121912691	F	22	3	1) 100 cystine 2) 100 cystine	Y	Cystine NL, CKD stage 2 Urine cystine (191-232 mmol/mol Creatinine) Norm: <30)	Cystinuria	Cystinuria type A
			het	c.1094G>T	p.Arg365Leu	P	PM1, PP2, PM2, PM5, PP3m, PP5, PP4	2/251.376/0 1/113.678/0	rs567478582
	SLC7A9	AD/AR	het	C.544G>A	p.Ala182Thr	P	PM2, PM1supp, PP2, PP5vs	727/282810/2 504/113.736/2	NM_014270.5 rs79389353				See above		See above
5288	SLC3A1	AR	hom	c.647C>T	p.Thr216Met	LP	PP3s, PM2, PP2, PP5, PP4	23/251.414/0 15/113.710/0	NM_000341.4 rs369641941	F	19	5	1–5) 100 cystine	N	Cystine NL, CKD stage 3 Urine cystine (291-306 mmol/mol Creatinine) Norm: <30)	Cystinuria	Cystinuria type A
5578	SLC3A1	AR	hom	c.833T>C	p.Phe278Ser	LP	PP3s, PM2, PP2, PP4	1/251.432/0 1/113.722/0	NM_000341.4 rs762218116	M	20	18	N/A	N	Cystine NL, CKD stage 3 Urine cystine (145-239 mmol/mol Creatinine) Norm: <30)	Cystinuria	Cystinuria type A
SLC7A9–Cystinuria, type B
5038	SLC7A9	AR	het	c.1316A>G	p.Tyr439Cys	LP	PM2, PP3m, PP2, PP4	NF/NF	NM_014270.5-	F	17	28	1) 60 cystine, 40 CA 2) 100 cystine	N	Cystine NL, CKD stage 3 Urine cystine (301 mmol/mol Creatinine) Norm: <30)	Cystinuria	Cystinuria, type B
			het	c.775G>A	p.Gly259Arg	LP	PP3m, PM2, PP2, PP5, PP4	2/251.260/0 2/113.626/0	rs121908483
5160	SLC7A9	AR	hom	c.313G>A	p.Gly105Arg	P	PM1, PP2, PM5supp, PP3m, PP5, PP4	96/251.028/1 64/113.436/1	NM_014270.5 rs121908480	M	1	4	100 cystine	Y	Cystine NL Urine cystine (286 mmol/mol Creatinine) Norm: <30)	Cystinuria	Cystinuria, type B
5412	SLC7A9	AR	hom	c.313G>A	p.Gly105Arg	P	PM1, PP2, PM5supp, PP3m, PP5, PP4	96/251.028/1 64/113.436/1	NM_014270.5 rs121908480	M	44	5	100 cystine	N	Cystine NL Urine cystine (223 mmol/mol Creatinine) Norm: <30)	Cystinuria	Cystinuria, type B
5716	SLC7A9	AR	hom	c.313G>A	p.Gly105Arg	P	PM1, PP2, PM5supp, PP3m, PP5, PP4	96/251.028/1 64/113.436/1	NM_014270.5 rs121908480	F	16	2	100 cystine	Y	Cystine NL Urine cystine (273 mmol/mol Creatinine) Norm: <30)	Cystinuria	Cystinuria, type B
SLC12A1–Bartter syndrome, type 2
5769	SLC12A1	AR	het	c.1493C>T	p.Ala498Val	VUS-LP	PP3s, PM2, PP4	4/152.154/0 0/112.752/0	NM_000338.3 rs1366101480	F	22	1	1) 100 CA 2) 20 COD, 20 Struvite, 60 CA 3) 20 COD, 20 Struvite, 60 CA	N	CaP/CaOx NL, NC, CKD stage 2 Low Plasma K (3.3) High plasma bicarbonate (28) High urine pH (7.0) High PTH (88) Low 25-Vit D (22) Normal 1,25-Vit D (114) Hypercalciuria (5.31) Hypocitraturia (1.8) Normal urine oxalate (219)	Idiopathic NL and NC, CKD, incomplete dRTA	Bartter syndrome, type 2
			het	c.1878G>A	p.Trp626Ter	LP	PVS1, PM2, PP4	1/251.116/0 1/113.462/0	rs768765027

Acc.No.: RefSeq accession number; ACMG: American C;llege of Medical Genetic:, AF: allele frequency; AR: autosomal recessive; AD: autosomal dominant; dbSNP: reference SNP number of variant; het: heterozygous; hom: homozygous; LP: likely pathogenic, P: pathogenic; tot: total; Novel: mutation detected for the first time in this study/not previously described; NS: nonsense; family history: N = no, Y = yes; M: male; F: female; mo: months; NL: nephrolithiasis; NC: nephrocalcinosis; dRTA: distal renal tubular acidosis; CaOx: calcium oxalate; COM: calcium oxalate monohydrate; COD: calcium oxalate dihydrate; CaP: calcium phosphate; CA: carbonate apatite; Brushite: calcium hydrogen phosphate dihydrate; MAP: magnesium ammonium phosphate (=Struvite); UA: uric acid.

Normal/pathologic values laboratory parameters:

Plasma calcium (mmol/l): 2.15–2.5; hypercalciuria (mmol/24 h): >5.

Plasma magnesium (mmol/l): 0.66–1.07; hyperoxaluria (µmol/24 h): >500.

Plasma phosphate (mmol/l): 0.81–1.45; hypocitraturia (mmol/24 h): <1.65.

PTH (pg/ml): 15–65.

Urine magnesium (mmol/24 h): 3–5.

25-vitamin D (nmol/l): 50–135.

Urine phosphate (mmol/24 h): 13–32.

1,25-vit D (pmol/l): 48–190.

Figure 2:

Overall yield of genetic diagnoses: (A) diagnostic yield (Mendelian versus LP/P variants predisposing to nephrolithiasis) in KSFs; (B) diagnostic yield in NKSFs; (C) overview of Mendelian diagnoses in KSFs, sorted by phenotype groups; (D) overview of LP/P variants predisposing to nephrolithiasis in KSFs, sorted by phenotype groups.

A total of 66 individuals (8.4%), 64 (8.1%) of which were KSFs, presented with LP/P variants not fulfilling our stringent criteria for Mendelian disease, but predisposing for nephrolithiasis (Supplementary Table S1). LP/P variants were detected in 9 of 34 genes (n = number of patients): ADCY10 (n = 1), CASR (n = 4), CLDN16 (n = 8), CYP24A1 (n = 8), SLC4A1 (n = 1), SLC7A9 (n = 3), SLC34A1 (n = 15), SLC34A3 (n = 17) and SLC9A3R1 (n = 8). Of these variants, 18% (12/66) were novel and previously unreported in ClinVar or HGMD (Supplementary Table S2).

Prior to WES, a Mendelian form of nephrolithiasis was known or suspected in 18 of 23 individuals (78%) with a post-WES Mendelian diagnosis, despite each case being reviewed by a kidney stone expert. Therefore, in 5 of 23 individuals (22%) with Mendelian disease [i.e. 5/787 (0.6% of total KSFs)], genetic analysis established a new or corrected a suspected a priori diagnosis. The diagnostic yield was similar between first-time and recurrent stone formers (Supplementary Table S3) and between men and women (Table 3, Supplementary Table S4).

Table 3:

Characteristics of stone formers with and without pathogenic gene variants.

Characteristics	No genetic diagnosis (n = 700)	LP/P variants (n = 64)	Mendelian disease (n = 23)	P-value
Age (years), mean (SD)	47.18 (14.30)	43.73 (15.78)	39.13 (15.83)	.009
Male, n (%)	502 (71.7)	50 (78.1)	14 (60.9)	.27
Body mass index (kg/m2), mean (SD)	26.94 (5.12)	26.56 (4.91)	24.42 (3.15)	.074
Hypertension, n (%)	223 (35.4)	23 (41.1)	9 (40.9)	.57
Diabetes, n (%)	23 (3.4)	1 (1.6)	0 (0.0)	.86
Obesity (BMI ≥30 kg/m2), n (%)	152 (22.8)	15 (24.2)	1 (4.8)	.12
Hyperuricemia, n (%)	306 (43.7)	28 (43.8)	12 (52.2)	.75
Family history of kidney stone disease, n (%)	278 (43.1)	29 (49.2)	9 (40.9)	.66
Kidney stone recurrence (>1 stone event), n (%)	547 (82.5)	49 (79.0)	18 (81.8)	.73
Age at first stone event (years), mean (SD)	35.96 (13.76)	33.89 (13.47)	23.48 (12.29)	<.001
Total number of stone events, median (IQR)	3.00 (2.00–4.00)	3.00 (2.00–4.00)	5.00 (2.00–5.00)	.14
Blood parameters
Total calcium (mmol/l), mean (SD)	2.35 (0.12)	2.38 (0.12)	2.40 (0.11)	.075
Ionized calcium (mmol/l), median (IQR)	1.21 (1.19–1.23)	1.22 (1.19–1.25)	1.23 (1.21–1.25)	.17
Phosphorus (mmol/l), mean (SD)	1.01 (0.17)	0.96 (0.14)	1.07 (0.13)	.037
Magnesium (mmol/l), mean (SD)	0.83 (0.07)	0.84 (0.07)	0.81 (0.07)	.40
Uric acid (μmol/l), mean (SD)	328.87 (177.47)	325.73 (74.03)	342.85 (78.56)	.93
Intact PTH (ng/l), median (IQR)	40.00 (30.00–52.00)	35.60 (27.80–44.00)	36.00 (22.00–55.00)	.069
25-OH vitamin D3 (nmol/l), median (IQR)	40.00 (27.00–59.00)	45.00 (29.50–60.00)	32.45 (22.00–73.00)	.71
eGFR creatinine (CKD-EPI 2009 equation; ml/1.73 m2)	96.42 (81.53–109.15)	99.10 (83.07–110.36)	93.52 (71.42–97.96)	.28
Urine parameters
Total urine volume (l/24 h), median (IQR)	1.96 (1.47–2.50)	2.05 (1.37–2.59)	2.54 (2.38–3.30)	<.001
Urine pH, median (IQR)	5.89 (5.35–6.52)	6.29 (5.60–6.70)	6.82 (6.63–7.16)	<.001
Urine sodium:creatinine ratio (mmol/mmol/24 h), median (IQR)	13.58 (10.97–16.79)	12.39 (10.70–17.20)	18.78 (15.27–20.41)	<.001
Urine potassium:creatinine ratio (mmol/mmol/24 h), median (IQR)	4.69 (3.75–5.78)	4.26 (3.71–5.56)	6.52 (5.18–7.57)	<.001
Urine uric acid:creatinine ratio (mmol/mmol/24 h), mean (SD)	0.62 (0.10)	0.59 (0.10)	0.60 (0.14)	.19
Urine calcium:creatinine ratio (mmol/mmol/24 h), median (IQR)	0.43 (0.30–0.58)	0.47 (0.36–0.60)	0.40 (0.24–0.53)	.13
Urine magnesium:creatinine ratio (mmol/mmol/24 h), median (IQR)	0.30 (0.24–0.38)	0.28 (0.23–0.35)	0.38 (0.28–0.47)	.039
Urine citrate:creatinine ratio (mmol/mmol/24 h), median (IQR)	0.19 (0.13–0.27)	0.20 (0.14–0.27)	0.22 (0.13–0.31)	.72
Urine phosphate:creatinine ratio (mmol/mmol/24 h), mean (SD)	2.19 (0.55)	2.15 (0.47)	2.15 (0.46)	.82
Urine oxalate:creatinine ratio (mmol/mmol/24 h), median (IQR)	0.03 (0.02–0.04)	0.02 (0.01–0.04)	0.04 (0.02–0.05)	.12
Urine cystine:creatinine ratio (mmol/mol/24 h), median (IQR)	4.00 (3.00–5.00)	4.00 (3.00–6.00)	204.00 (3.00–263.00)	<.001
Stone phenotypes, n (%)
Calcium oxalate dihydrate	66 (13.1)	7 (15.9)	0 (0.0)	.24
Calcium oxalate monohydrate	264 (52.4)	19 (43.2)	4 (23.5)	.035
Calcium phosphate	115 (22.8)	13 (29.5)	2 (11.8)	.35
Uric acid	37 (7.3)	1 (2.3)	0 (0.0)	.40
Cystine	7 (1.4)	0 (0.0)	11 (64.7)	<.001
Struvite	12 (2.4)	3 (6.8)	0 (0.0)	.17
BMD, mean (SD)
Lumbar spine (g/cm2)	1.01 (0.14)	1.00 (0.11)	1.11 (0.13)	.022
Lumbar spine T-score	−0.58 (1.28)	−0.71 (1.01)	0.32 (1.05)	.022
Lumbar spine Z-score	−0.11 (1.40)	−0.28 (1.14)	0.64 (1.15)	.086
Femoral neck (g/cm2)	0.83 (0.13)	0.84 (0.10)	0.89 (0.21)	.37
Femoral neck T-score	−0.57 (1.02)	−0.48 (0.87)	−0.14 (1.50)	.26
Femoral neck Z-score	0.11 (1.00)	0.09 (0.79)	0.34 (1.39)	.67

Characteristics are indicated separately for stone formers with Mendelian disease and with/without LP/P predisposing gene variants.

eGFR: estimated glomerular filtration rate.

Genotype–phenotype correlation

The mean age of the first kidney stone event (23.5 ± 12.3 versus 35.9 ± 13.8 years; P < .001) and the mean age of presentation for metabolic workup (39 ± 15.8 versus 47.2 ± 14.3 years; P = .009) were lower in KSFs with a Mendelian diagnosis compared with KSFs without, with intermediate results for patients with LP/P variants predisposing for nephrolithiasis (Table 3, Supplementary Fig. S3). Furthermore, cystine stones were significantly more common (64.7% versus 1.4%) and calcium oxalate monohydrate stones less common (23.5% versus 52.5%) in KSFs with a Mendelian diagnosis compared with KSFs with LP/P variants or without a genetic diagnosis, in line with higher urinary cystine and urine pH in KSFs with a Mendelian diagnosis (Table 3, Supplementary Fig. S2). The prevalence of a positive family history of kidney stone disease was similar in all three groups (43.1% versus 49.2% versus 46.0%). Additionally, KSFs with Mendelian disease had more past kidney stone events [median 5 (IQR 2–5) versus 3 (2–4); P = 0.052] than KSFs with LP/P variants. In contrast, KSFs with LP/P variants had significantly lower plasma phosphate (Table 3). Lumbar spine BMD was higher in KSFs with Mendelian disease compared with KSFs without genetic disease or LP/P variants predisposing for nephrolithiasis (Table 3, Supplementary Fig. S6).

Mendelian disease

The most common Mendelian diagnosis was cystinuria (n = 13) due to biallelic diagnostic variants in SLC3A1 (type A cystinuria; n = 9) or SLC7A9 (type B cystinuria; n = 4). Urine cystine excretion, where available, was increased in all KSFs with a genetic diagnosis of cystinuria. One patient had biallelic diagnostic variants in SLC3A1 and a monoallelic LP/P variant in SLC7A9. KSFs with biallelic diagnostic SLC3A1 or SLC7A9 variants all had cystine stones.

Five patients had biallelic diagnostic variants in CYP24A1, which encodes the vitamin D–inactivating enzyme vitamin D 24-hydroxylase. Interestingly, all KSFs with biallelic variants in CYP24A1 (but no KSFs with a monoallelic CYP24A1 variant) displayed cystic kidney disease, as reported previously [31, 36]. Common misdiagnoses for patients with biallelic variants in CYP24A1 (n = 3) were polycystic kidney disease or medullary sponge kidney. Two patients with biallelic CYP24A1 variants underwent parathyroidectomy for suspected primary hyperparathyroidism (PHPT), without phenotype alleviation postoperatively.

We further identified three patients with primary hyperoxaluria (PH) in our cohort. The diagnosis was already established in the two patients with primary hyperoxaluria type 1 (PH1) and primary hyperoxaluria type 2 (PH2). The diagnosis was previously unknown in the patient with primary hyperoxaluria type 3 (PH3) due to a homozygous pathogenic HOGA1 variant. This patient had his first kidney stone episode at age 36. Metabolic workup at age of 67 showed CKD stage 2 and a urine oxalate in the upper normal range on free-choice diet, but a strong increase after an instructed 1-week diet low in calcium and sodium.

No Mendelian form of nephrolithiasis was detected in the 114 NKSFs included in this study (Fig. 1, Fig. 2, Supplementary Table S5).

LP/P variants predisposing to nephrolithiasis

The most common identified genetic predisposition to nephrolithiasis was renal phosphate wasting due to monoallelic variants in the genes encoding the renal sodium–phosphate co-transporters NaPi-2a and NaPi-2c [SLC34A1 (n = 15) and SLC34A3 (n = 18)] or in the gene SLC9A3R1, encoding the regulatory interaction protein for NaPi-2a/2c, NHERF (n = 7). LP/P variants in SLC34A1 and SLC34A3, but not in SLC9A3R1, were enriched in KSFs compared with controls (NKSFs and gnomAD, Supplementary Table S5). KSFs with LP/P variants in SLC34A1/A3 or SLC9A3R1 had similar plasma phosphate and renal tubular maximum reabsorption rate of phosphate/glomerular filtration rate (TmP/GFR) compared with KSFs without a genetic diagnosis, but urine calcium in individuals with LP/P variants in SLC34A3 was higher compared with KSFs without a genetic diagnosis (Supplementary Fig. S4).

We also identified eight patients carrying the same previously described monoallelic LP/P variant (c.458A>G, p.Asn153Ser) in CLDN16, encoding the tight junction protein claudin-16. Biallelic pathogenic variants in CLDN16 cause familial hypomagnesemia with hypercalciuria and nephrocalcinosis [37], and an increased prevalence of nephrolithiasis has been observed in monoallelic carriers of pathogenic CLDN16 variants [30, 38]. Individuals with the monoallelic CLDN16 variant presented with hypercalciuria and calcium oxalate stones but normal renal function and plasma magnesium. The identified CLDN16 variant was more prevalent in KSFs versus NKSFs and versus all LP/P-variants in CLDN16 in gnomAD non-Finnish European (NFE) (Supplementary Table S5).

Additionally, eight patients carried monoallelic LP/P variants in CYP24A1, and the respective variants were five to eight times more prevalent in KSFs versus NKSFs and versus gnomAD NFE (Supplementary Table S5). Overall (i.e. considering the prevalence of all variants), CYP24A1 variants were not enriched in our cohort compared with gnomAD NFE (Supplementary Table S5). KSFs with mono- and biallelic CYP24A1 variants displayed hypercalciuria, but hypercalcemia and/or suppressed parathyroid hormone (PTH) or nephrocalcinosis were only present in KSFs with biallelic variants. Urine calcium was higher in KSFs with biallelic variants, but KSFs with monoallelic variants had a higher number of past stone events (Supplementary Fig. S5). BMD and the prevalence of osteoporosis/osteopenia were similar in patients with genetic variants predisposing to urinary phosphate or calcium wasting compared with KSFs without a genetic diagnosis (Supplementary Fig. S6 and Supplementary Table S7).

We also identified six patients with monoallelic LP/P variants in SLC7A9. Risk variants were either previously described to cause autosomal dominant cystinuria [c.544G>A p.(Ala182Thr)] [39–42] or were enriched in KSFs versus NKSFs and gnomAD [c.313G>A p.(Gly105Arg)]. In contrast to biallelic variants, the six KSFs with monoallelic SLC7A9 variants presented with calcium-containing kidney stones without cystine content, similar to previously reported cases [43–45]. KSFs with monoallelic SLC7A9 variants had elevated urine cystine, albeit lower levels compared with KSFs with biallelic variants in SLC3A1/SLC7A9. No significant differences in stone number or age at first stone event were detected between the two groups of patients (Supplementary Fig. S5). In seven individuals with cystine-containing kidney stones and two individuals with increased urine cystine, no diagnostic variants in SLC3A1 or SLC7A9 could be identified (Supplementary Table S6).

Four KSFs had LP/P variants in the gene encoding the calcium-sensing receptor (CASR), which is associated with familial hypocalciuric hypercalcaemia type 1 (FHH1) or autosomal dominant hypocalcaemia. The frequency of CASR LP/P variants in KSFs was significantly higher than LP/P variants in all gnomAD NFE participants (Supplementary Table S5). The phenotype of these patients was variable: two patients presented with hypercalciuria but normal plasma calcium, phosphate and PTH; one patient had hypercalcemia with a low-normal PTH and low urine calcium; and another patient presented with elevated PTH, hypercalcemia, pronounced hypophosphatemia, hypercalciuria and calcium oxalate dihydrate stones, compatible with the diagnosis of PHPT. Parathyroidectomy led to a complete normalization of the phenotype.

In the 114 NKSFs included in this study, exome sequencing revealed an LP/P variant in two individuals, corresponding to a prevalence of variants predisposing to nephrolithiasis of 1.8% (compared with 8.13% for KSFs) (Fig. 1, Fig. 2, Supplementary Table S5).

DISCUSSION

In this exome-based targeted panel study in a large, unselected European cohort of 787 adult KSFs, we detected a Mendelian kidney stone disease in 23 patients (2.9%) using stringent diagnostic criteria. This diagnostic yield is significantly lower compared with previous studies conducted in selected groups of KSFs [10, 17, 46], yet the fraction of additional individuals (8.13%) with LP/P variants predisposing to nephrolithiasis is substantial, especially when considering the broad inclusion criteria, the high prevalence of the disease and the potential rate of false negatives due to stringent molecular genetic diagnosis criteria and technical limitations of exome sequencing, such as an inability to detect deep intronic or difficulty in reliably calling copy number variants. In fact, if Mendelian disease and variants predisposing to nephrolithiasis are combined, the overall diagnostic yield is 11% (87/787 individuals), comparable to neurometabolic disorders or cancer, where exome sequencing is routinely used [16, 47, 48]. Interestingly, the diagnostic yield was not different between first-time and recurrent stone formers and did not differ when stratified by sex. In NKSFs, no Mendelian kidney stone disease was detected, and LP/P variants in nephrolithiasis genes were significantly less prevalent compared with KSFs (1.8% versus 8.1%).

Cystinuria due to biallelic diagnostic variants in SLC3A1 or SLC7A9 was the most common Mendelian disease in our cohort (57%; n = 13/23 individuals with Mendelian disease), and an additional 6 individuals carried monoallelic LP/P variants in SLC7A9. We confirmed a high prevalence of CKD in KSFs with cystinuria [49], yet this applied only to KSFs with biallelic diagnostic variants. We were unable to provide a genetic diagnosis in nine KSFs with a cystinuria phenotype, including seven individuals with cystine-containing kidney stones and two individuals with elevated urinary cystine excretion. While this could be at least partially due to technical limitations, such as the inability to detect deep intronic variants or missed copy number variations [39], it may also indicate that the genetic architecture of cystinuria is not yet completely deciphered. Of note, the prevalence of phenotypic cystinuria (including patients with elevated urinary cysteine but no cysteine stones) in the BKSR [22/787 KSFs (2.8%)] was only marginally higher than described for adult stone formers (1–2%), but significantly lower than in paediatric cohorts (6–10%) [50, 51].

Of the LP/P variants detected predisposing to nephrolithiasis, by far the most common [63% (n = 40/64)] were variants in SLC34A1/3 or SLC9A3R1, highlighting the pathophysiological importance of renal phosphate loss in the increasing risk for kidney stone formation [12, 52]. However, the phenotype of KSFs with LP/P variants in these genes on a free-choice diet was very subtle and mostly indistinguishable from KSFs without predisposing variants. Future studies need to determine if the phenotype of KSFs with LP/P variants in these genes can be unmasked by dietary interventions (e.g. by a low phosphate diet) for diagnostic purposes and mitigated for prevention of future stone events (e.g. by phosphate supplementation).

We only detected one KSF with a monoallelic LP/P variant in ADCY10, indicating that variants in ADCY10 are not contributing significantly to the overall nephrolithiasis risk, at least in individuals of European descent [43, 53]. Further, we identified four patients with CASR LP/P variants in our cohort with very variable phenotypes. Atypical characteristics such as chondrocalcinosis or kidney stones have been described in patients with FHH1 but complicate the separation from PHPT [54–56]. Yet, separation is of critical importance: parathyroidectomy is the first-line treatment in PHPT, but surgery is ineffective in FHH1. Additional complexity arises from the fact that PHPT and FHH1 can coexist, as observed in patient 6023 [57, 58]. Interestingly, the diagnostic rate in uric acid stone formers was very low, with no Mendelian disease and only one monoallelic LP/P variant (in ADCY10) found in 38 uric acid stone formers, suggesting no significant contribution of Mendelian disease in uric acid stone disease.

Genetic diagnoses in KSFs can have important prognostic implications and therapeutic consequences. However, while the pathophysiology is well delineated, the effectiveness of therapeutic interventions has not been studied systematically in most forms of Mendelian nephrolithiasis. The widespread use of genetic testing will facilitate the inclusion of KSF with Mendelian disease and LP/P variants predisposing to nephrolithiasis in prospective registries and clinical trials to study disease evolution and evaluate tailored therapeutic interventions. Currently KSFs are only subjected to genetic testing if a specific disease is strongly suspected. While clinical signs suggesting an inherited disorder have been proposed [59], Mendelian disease was missed or misclassified in 5 of 23 (22%) KSFs in our cohort, despite individual review of each case by a kidney stone expert. The situation is even far more difficult in KSFs with LP/P variants predisposing to nephrolithiasis: the associated phenotypes were discrete and variable, with a large phenotype overlap with KSFs without LP/P variants. But the low prevalence of LP/P variants detected in NKSFs suggests a rather high penetrance of LP/P variants predisposing to nephrolithiasis in adults, with the limitation of a small NKSF sample size. Together, these results preclude the definition of clear phenotypic criteria to reliably prioritize genetic testing in adult KSFs, with the notable exception of Mendelian forms of nephrolithiasis with clear and unequivocal phenotypes. In addition, LP/P variants present with their own challenges in clinics and considerations for clinical reporting of LP/P variants and risk alleles have recently been suggested [14].

Strengths of our study include very minor patient pre-selection with broad inclusion criteria, the large sample size, availability of ethnically matched non-stone-forming controls and the detailed phenotype. Our study also has limitations, such as the inability to detect deep intronic and copy number variants, lack of family recruitment for segregation analyses, monocentric recruitment, overrepresentation of men and limited ethnic diversity.

In conclusion, our study shows that Mendelian disease is far less common in adult KSFs than previously suggested, but simultaneously reveals a high prevalence of LP/P variants predisposing to nephrolithiasis in adult KSFs. Our study also highlights the potential of genetic testing in unselected adult KSFs for accurate assessment of Mendelian disease and identification of variants predisposing to nephrolithiasis to direct patients to tailored therapies and clinical trials.

FUNDING

M.A.A. was supported by a grant for protected research time by the InselGruppe. E.G.O was supported by Postdoc Mobility Stipendien of the Swiss National Science Foundation (grants P2ZHP3_195181 and P500PB_206851) and Kidney Research UK (grant Paed_RP_001_20180925). D.G.F. was supported by the Swiss National Science Foundation (grants 31003A_172974 and 33IC30_166785/1) and the Swiss National Centre of Competence in Research Kidney.CH. J.A.S. is supported by the Northern Counties Kidney Research Fund (22/01), Kidney Research UK (Paed_RP_001_20180925), LifeArc and the Medical Research Council (MR/Y007808/1).

AUTHORS’ CONTRIBUTIONS

D.G.F., A.S. and J.A.S. conceptualized the study. D.G.F. and M.A.A. acquired financial support. M.A.A., E.G.O. and D.G.F. designed the data analysis plans. M.A.A., E.G.O., M.B., A.S. and D.G.F. performed data analysis and data interpretation. M.A.A. wrote the first draft of the manuscript. All authors contributed to discussion and editing of the text and approved the final version of the manuscript.

DATA AVAILABILITY STATEMENT

The data supporting the findings of this study are available from the corresponding author upon request.

CONFLICT OF INTEREST STATEMENT

The authors declare no competing interests.