Undetermined Kidney Disease in Adults — Genetic Approach

Adult chronic kidney disease without an established cause (CKDx): indications, strategy and interpretation of genetic diagnosis, with a focus on monogenic disorders and the APOL1 risk allele.

Literature and regulatory information updated: 6 November 2026.

0. Initial Reference Table

Undetermined kidney disease is not a single disease but a clinical category of exclusion. The first table summarises the category and the second the genetic entities that most often explain adult chronic kidney disease (CKD) initially without a cause.

Table abbreviations: AD, autosomal dominant; AR, autosomal recessive; XL, X-linked; ADTKD, autosomal dominant tubulointerstitial kidney disease; C3G, C3 glomerulopathy; CAKUT, congenital anomalies of the kidney and urinary tract; CNV, copy-number variant; ESKD, end-stage kidney disease; ADPKD and ARPKD, autosomal dominant and autosomal recessive polycystic kidney disease; FSGS, focal segmental glomerulosclerosis; LVH, left ventricular hypertrophy; MANE, Matched Annotation from NCBI and EMBL-EBI reference transcript; GBM, glomerular basement membrane; EM, electron microscopy; MLPA, multiplex ligation-dependent probe amplification; MODY, maturity-onset diabetes of the young; NPH, nephronophthisis; OR, odds ratio; BP, blood pressure; PH1, primary hyperoxaluria type 1; MRI, magnetic resonance imaging; aHUS, atypical haemolytic uraemic syndrome; SRNS, steroid-resistant nephrotic syndrome; eGFR, estimated glomerular filtration rate; VNTR, variable number tandem repeat; VUS, variant of uncertain significance.

The Orphanet codes of the main entities are ORPHA:63 (Alport syndrome), ORPHA:730 (ADPKD), ORPHA:731 (ARPKD), ORPHA:34149 (ADTKD), ORPHA:656 (hereditary SRNS), ORPHA:655 (nephronophthisis), ORPHA:324 (Fabry disease), ORPHA:329918 (C3 glomerulopathy), ORPHA:416 (primary hyperoxaluria), ORPHA:213 (cystinosis) and ORPHA:1652 (Dent disease) [20–30].

1. Introduction

1.1. From "nephropathy of unknown cause" to CKDx

CKD affects about 10% of the world's adult population, and a constant proportion of patients reach dialysis or transplantation without an aetiological diagnosis [3]. In France, according to the REIN registry (Réseau Épidémiologie et Information en Néphrologie), nephropathy of unknown cause is the third cause of stage 5 chronic kidney failure, after diabetes and arterial hypertension, and kidney replacement therapy registries and CKD cohorts classify 15 to 25% of adults as having kidney disease of undetermined origin [1]. The consensus of the ERA (European Renal Association) Genes & Kidney Working Group estimates that at least 16 to 20% of CKD remains unexplained and stresses that this fraction is underestimated, because many patients receive a "diagnosis of convenience", in particular that of hypertensive or vascular nephropathy [2]. Diabetes and hypertension may be comorbidities rather than the cause: diabetic or hypertensive nephropathy requires a concordant set of arguments (duration and severity of the disease, diabetic retinopathy, proteinuria profile, course and, when available, histology), and in their absence the disease should be considered of undetermined origin [1].

The term CKDx (Chronic Kidney Disease of uneXplained cause) was proposed to distinguish this situation from CKDu (chronic kidney disease of unknown etiology), a term used in the international literature for the endemic CKD of heat-exposed agricultural communities, such as Mesoamerican nephropathy and the forms seen in Sri Lanka and India, defined after exclusion of diabetes, long-standing hypertension and known glomerulopathies [1, 2]. CKDx is a diagnosis of exclusion relative to the healthcare system in which the patient was assessed: the same patient may be CKDx in a setting without access to biopsy or sequencing and receive a precise diagnosis in another [2].

1.2. Historical milestones of the diseases that make up genetic CKDx

The entities that today explain most genetic CKDx were described as familial syndromes long before the genes were identified. Guthrie described in 1902 a family with 12 individuals with recurrent haematuria; Hurst documented in 1923 the progression to uraemia in the same family; and Alport, in 1927, added deafness and observed that affected men died of uraemia whereas women lived to old age, which gave his name to the association of hereditary nephropathy and deafness [31]. Barker et al. identified in 1990 the first structural alterations of COL4A5 in three Utah families with the X-linked form [31]. In polycystic disease, Dalgaard's Danish study (1957) showed that autosomal dominant polycystic kidney disease (ADPKD) is one of the most frequent genetic diseases, with about 1 in 1000 individuals affected, and the European Polycystic Kidney Disease Consortium isolated the PKD1 gene in 1994 from the breakpoint of a t(16;22) translocation in an affected family [32].

The tubulointerstitial forms were recognised as familial juvenile hyperuricaemic nephropathy and as medullary cystic disease. Duncan and Dixon (1960) and Rosenbloom et al. (1967) described families with early hyperuricaemia, gout and kidney failure without distinctive histological findings, with father-to-son transmission [33]. Hart et al. identified in 2002 heterozygous variants in exon 4 of UMOD in four families, three with a clinical diagnosis of hyperuricaemic nephropathy and one with medullary cystic disease [33]. Živná et al. identified in 2009 a 3 bp deletion in REN in a four-generation Belgian family with dominant hyperuricaemic nephropathy and anaemia [34]. Kirby et al. described in 2013, in six families, the insertion of a cytosine into a stretch of seven cytosines within one unit of the long (1.5–5.0 kb), GC-rich MUC1 VNTR (variable number tandem repeat), which explains why this disease escaped sequencing techniques for years [35]. Bolar et al. associated in 2016 SEC61A1 missense variants with progressive CKD with congenital anaemia and intrauterine and postnatal growth restriction [36].

Other milestones complete the picture. Horikawa et al. (1997) associated variants in HNF1B (then TCF2) with MODY (maturity-onset diabetes of the young) with non-specific nephropathy in a Japanese family, and Bingham et al. (2001) showed that hypoplastic glomerulocystic kidney disease belongs to the same spectrum [37]. Boute et al. identified in 2000 the NPHS2 variants in families with steroid-resistant nephrotic syndrome (SRNS) [38]. Genovese et al., in 2010, studying African Americans with focal segmental glomerulosclerosis (FSGS) and with end-stage kidney disease (ESKD), located in APOL1 the two variants G1 and G2, subject to positive selection because they confer resistance to Trypanosoma, and changed the interpretation of "nephroangiosclerosis" in populations of African ancestry [39, 40]. Fabry disease, described in 1898 by Fabry and by Anderson, is the paradigmatic treatable phenocopy [41].

1.3. Sequencing as a diagnostic tool in adults

Lower costs, greater analytical capacity and shorter turnaround times have integrated genetics into the diagnostic pathway of adults with undetermined kidney disease [1]. Groopman et al. published in 2019 the exome analysis of 3315 patients with CKD from two North American cohorts, 91.6% older than 21 years and 35.6% of self-reported non-European ancestry: a genetic diagnosis was obtained in 9.3% (307/3315), with 66 different monogenic diseases, and in 17.1% (48/281) of patients with nephropathy of unknown origin [8]. In the same year, Connaughton et al. described the monogenic causes of CKD in Irish adults selected for genetic suspicion [42]. The KDIGO (Kidney Disease: Improving Global Outcomes) conference on genetics in CKD, held in March 2021, and the 2022 ERA and ERKNet recommendations systematised the indications and interpretation of testing [3, 43]. In France, the exome-first strategy was assessed prospectively at Tenon hospital (Paris), with 24% of diagnoses in 538 adults [4], and combined with biopsy in Marseille [5, 44]. The ERA consensus on CKDx was published in 2025 [2] and the PNDS (French National Diagnosis and Care Protocol) of the Haute Autorité de Santé (HAS), coordinated by the MARHEA and MaReGE reference centres under the aegis of the ORKiD network, was published on 6 October 2026 [1].

1.4. Nomenclature and reclassifications

The KDIGO conference recommended two-part designations combining the condition and the gene (for example, ADTKD-UMOD), and recognised a continuum between monogenic and complex disease, with no upper age limit for monogenic disease [3]. The 2015 KDIGO consensus created the designation ADTKD (autosomal dominant tubulointerstitial kidney disease) and abandoned the terms familial juvenile hyperuricaemic nephropathy and medullary cystic kidney disease types 1 and 2, because medullary cysts are not characteristic [11, 12, 45]. The unified classification of collagen IV α345 diseases and the Alport syndrome testing recommendations treat COL4A3/COL4A4 heterozygotes as autosomal dominant Alport syndrome, and GeneReviews records the strong consensus to abandon "benign familial haematuria", because the risk of ESKD is higher than in the general population [9, 46, 47].

The ERA consensus lists the low-certainty labels that mask CKDx: histological labels (FSGS, membranoproliferative glomerulonephritis, chronic interstitial nephritis, thrombotic microangiopathy, thin basement membrane nephropathy, nephroangiosclerosis, tubulointerstitial disease, crystalline nephropathy), clinical labels ("benign" familial haematuria, atrophic or hypoplastic kidneys, chronic glomerulonephritis, hypertensive or vascular nephropathy, nephrocalcinosis, hyperuricaemic nephropathy, CAKUT — congenital anomalies of the kidney and urinary tract) and genetic labels (variant of uncertain significance, VUS) [2]. The PNDS adds that FSGS, chronic tubulointerstitial nephropathy (CTIN), interstitial fibrosis with tubular atrophy and "benign nephroangiosclerosis" describe a lesion or a pattern of kidney response, not an aetiology [1]. To make the depth of investigation visible, the ERA proposes that the CKDx diagnosis include the CGA classification (cause, GFR, albuminuria) and an "h/g" extension, in which "h" indicates histology and "g" genetics, with "+" and the year when the test was performed and "−" when it was not, for example "CKDx G4 A3 (h−g+(2024))", also indicating the genetic modality (panel, exome-based panel, exome, genome) and whether the biopsy included electron microscopy (EM) [2].

1.5. Scope definitions

This article covers the genetic diagnostic approach to adults with CKD without an established cause and the subcategories defined by the PNDS: kidney involvement labelled as "nephroangiosclerosis" with warning criteria; non-immunological glomerulopathies (FSGS without immune deposits and persistent microscopic haematuria with glomerular basement membrane changes); CTIN without an evident cause; atypical cystic nephropathies; and undiagnosed syndromic kidney involvement [1]. APOL1-associated kidney disease is treated in more detail, because the PNDS devotes a section to it in the absence of a dedicated PNDS [1]. The treatment and follow-up of each identified disease are not developed here: the PNDS is limited, by definition, to the pathway up to diagnosis [1].

2. Molecular Genetics and Pathophysiology

2.1. Genetic architecture of adult CKDx

More than 600 monogenic kidney disease genes have been described; monogenic diseases explain up to 50% of non-diabetic CKD in paediatric cohorts and about 30% in adults, and the observational heritability of CKD lies between 25 and 44%, higher in populations of African ancestry [3]. The KDIGO conference described a continuum ranging from high-penetrance Mendelian diseases, through risk alleles of intermediate effect (APOL1) and heterozygous variants of reduced penetrance (for example, COL4A3/COL4A4), to polygenic predisposition, which does not yet have clinical use [3]. In adult CKDx, genetic heterogeneity and the frequency of phenocopies, that is, diseases whose presentation mimics another clinical category, justify the shift from a phenotype-centred approach to a genome-wide analysis [1]. In the French cohort of Doreille et al., 47 monogenic diseases explained the 127 diagnoses, and eight of them (17%) accounted for 52% of diagnoses: type IV collagen nephropathies, ciliopathies, PKD1, PKHD1, TTC21B, UMOD and PAX2 [4].

The categories of genetic disease follow, with some exceptions, the dominant histological pattern. In the series of Robert et al. of 52 adults with biopsy-proven CKDx, the concordance between the dominant histological lesion and the category of the identified genetic disease was 92.3% [1, 44]. The gene groups correspond to these categories: glomerular basement membranes, podocyte, tubule and interstitium, cilium and cysts, development of the kidney and urinary tract, complement and endothelium, metabolism and storage, and risk alleles.

2.2. Collagen IV α3α4α5 (COL4A3, COL4A4, COL4A5)

The α3, α4 and α5 chains of collagen IV associate through their carboxy-terminal NC1 domains into helical heterotrimers that form polygonal networks in the basement membranes of the glomerulus, cochlea and eye [9]. In the absence of one of the chains, the glomerular basement membrane (GBM) retains fetal isoforms, which are more susceptible to proteolysis, leading to thinning, lamellation and glomerulosclerosis [48]. COL4A5 (Xq22.3; MANE NM_033380.3) explains 80–85% of classic Alport syndrome, with 85–90% of variants detected by sequencing and 10–15% by deletion/duplication analysis; COL4A3 (2q36.3; NM_000091.5) and COL4A4 (2q36.3; NM_000092.5) explain 12–15% and 5–8%, with about 98% sequence variants and recessive (biallelic) or dominant (heterozygous) inheritance [9, 49]. Digenic COL4A3/COL4A4 inheritance exists, with an intermediate phenotype [9, 48]. In males with the X-linked form, rearrangements, nonsense and frameshift variants lead to ESKD before age 20 in 50% of cases, splicing variants in about 65% before age 30, and missense variants in 30% before age 30 and 50% by age 40 [9]. 5' deletions spanning COL4A5 and the first exons of COL4A6, with a breakpoint in intron 2 of COL4A6, cause diffuse leiomyomatosis with 100% penetrance, and contiguous Xq22.3 deletions including ACSL4 cause AMME syndrome (Alport, intellectual disability, midface hypoplasia and elliptocytosis) [9, 48].

The frequency of heterozygotes has become central to adult CKDx. gnomAD analysis estimated predicted pathogenic COL4A3 or COL4A4 variants in 1 in 106 people (0.94%) and COL4A5 variants in at least 1 in 2320 [9]. The absolute risk of ESKD of a COL4A3/COL4A4 heterozygote was estimated well below 3%, reflecting reduced penetrance and variable expressivity [9]. In the French cohort of Pagniez et al., with 97 patients from 82 families with the autosomal dominant form (COL4A3 47, COL4A4 50; 70.1% women), 28.7% reached ESKD during follow-up, with no difference between the two genes (p=0.28) or between truncating and non-truncating variants, which the PNDS supporting document uses to state that this form is not benign [50]. Coexistence with another glomerulopathy is frequent: a heterozygous carrier with a presentation disproportionate to the usual phenotype should be investigated for an associated glomerular disease before the whole phenotype is attributed to the variant [1].

2.3. Podocyte and slit diaphragm

Slit diaphragm and podocyte cytoskeleton genes cause SRNS and FSGS. SRNS has a genetic cause in about 30% of childhood-onset forms and in 10–15% of adult-onset forms, with about 60 genes described [14]. In non-syndromic adult forms, INF2 (14q32.33; NM_022489.4; dominant; ~3% in adolescents and adults), TRPC6 (11q22.1; NM_004621.6; dominant; ~3%) and COL4A3/COL4A4/COL4A5 (10–30%, especially with onset from the second decade) are the main causes, followed by ACTN4 (~1%), ANLN, ARHGAP24 and CD2AP (<1%), COQ8B (3–5%), LMX1B (3–5%) and WT1 (10–20%, also congenital and infantile) [14, 49]. NPHS2 (1q25.2; NM_014625.4) causes 20–30% of non-syndromic SRNS in childhood and also adult-onset forms, and NPHS1 (19q13.12; NM_004646.4) 10–20%, up to 50% in congenital nephrotic syndrome [14]. The NPHS2 variant NM_014625.4:c.686G>A p.(Arg229Gln) has a carrier frequency of 3% in the multi-ethnic population, 7% in Finland and 0.01% in East Asia [14, 51]. Among TRPC6 variants associated with FSGS, more than 85% slow the calcium-dependent inactivation of the channel, which constitutes gain of function [52]. For INF2, 17% of 54 families with dominant FSGS had variants, versus 1 in 84 sporadic cases, and 12 of 16 patients with intermediate Charcot-Marie-Tooth disease and FSGS without PMP22 or MPZ variants had an INF2 variant [52]. Syndromic genes with apparently isolated kidney expression in adults, such as COL4A3/4/5, COQ8B, CRB2, INF2, LMX1B and WT1, often reveal extrarenal manifestations only after the genetic result [14].

2.4. Tubule and interstitium: ADTKD and HNF1B-related disease

ADTKDs are endoplasmic reticulum storage diseases. ADTKD-UMOD results from defective maturation and retention of mutant uromodulin in the endoplasmic reticulum of thick ascending limb cells, with PAS-positive deposits [11]. UMOD (16p12.3; MANE NM_003361.4) was associated with the disease by Hart et al., who described heterozygous variants in exon 4 [33]; sequencing detects more than 95% and no causal deletions or duplications have been described; the p.Val93_Gly97delinsAASC allele is associated with less gout and later ESKD, and GeneReviews records a possible founder effect for the variant designated C744G [11]. In ADTKD-MUC1, the variant creates a frameshift protein (MUC1fs), whose accumulation causes the disease; in 99% of patients the variant lies in the VNTR, and in 95% it is the duplication of a cytosine in a stretch of seven cytosines, designated 27dupC, which conventional panels, exome and short-read genome do not detect [12, 35]. The ERA consensus describes this variant as NM_001204286.1:c.428dupC in the reference repeat unit [2]. MUC1 (1q22) has no MANE transcript, because of its repetitive structure [49]. In REN (1q32.1; NM_000537.4), signal peptide variants alter the targeting and co-translational translocation of preprorenin into the endoplasmic reticulum, with reticulum stress, cytosolic accumulation of non-glycosylated preprorenin and accelerated autophagy, and sequencing detects all the variants described [53]. In SEC61A1 (3q21.3; NM_013336.4), the p.Thr185Ala and p.Val67Gly variants described by Bolar et al. alter the selectivity and permeability of the translocon pore; the same gene also causes common variable immunodeficiency (OMIM 620670) and severe congenital neutropenia, three distinct allelic entities [36, 52]. DNAJB11 produces a mixed phenotype, with small bilateral cysts without nephromegaly and interstitial fibrosis, described as atypical ADTKD [10, 11].

HNF1B (17q12; NM_000458.4) encodes a transcription factor for the development of the kidney, pancreas, liver and genital tract, and is dosage-sensitive (haploinsufficiency with ClinGen score 3) [52]. More than half of HNF1B alterations are whole-gene deletions, which almost always correspond to the recurrent 1.2 Mb deletion at 17q12, including HNF1B and at least seven other genes [13, 43, 52]. Identification of an intragenic variant excludes the deletion syndrome and establishes HNF1B-related disease [13]. Neurodevelopmental disorders occur in about 25% of deletion carriers and in 6.8% of intragenic variant carriers, and ESKD during follow-up in 51% and 78%, respectively [13]. De novo variants are frequent in severe dominant forms such as this, which explains many negative family histories [1].

2.5. Primary cilium: polycystic disease and nephronophthisis

Polycystins 1 and 2 form a functional complex, with a cryo-electron microscopy structure of three polycystin 2 molecules for one polycystin 1, located in the primary cilium, with a probable sensory or receptor function [10]. ADPKD involves PKD1 (16p13.3; MANE NM_001009944.3) in about 78% of cases (sequence ~97%, deletions and duplications ~3%), PKD2 (4q22.1; NM_000297.4) in about 15%, IFT140 in 1–2% and ALG5, ALG9, DNAJB11 and GANAB in less than 0.5% each, with about 5% left without an identified cause [10, 49]. Cyst formation is focal and results from a clonal somatic second hit [54]. The presence of homologous PKD1 pseudogenes requires specific test design, and the exome may not detect all PKD1 variants [10]. The hypomorphic PKD1 variant p.Arg3277Cys produces few or no cysts in the heterozygous state but, in biallelic combination, can cause prenatal-onset disease resembling ARPKD; digenic PKD1+PKD2 inheritance aggravates the disease, and COL4A1, PKHD1, PRKCSH, SEC63 and HNF1B act as modifiers [10]. Mosaicism has been documented in at least 1% of patients [10]. Monoallelic IFT140 (NM_014714.4), described in 66 individuals from 38 families, causes few large cysts and limited kidney failure [10, 55]; DNAJB11 (NM_016306.6), in 77 individuals from 27 families, led to ESKD in 32 between ages 55 and 89, median 75 [10, 56].

Nephronophthisis (NPH) is a recessive ciliopathy; the homozygous 290 kb NPHP1 deletion explains about 25% of cases and 20–40% of juvenile cases and requires MLPA (multiplex ligation-dependent probe amplification) or a SNP array [15, 43]. Snoek et al. showed that the NPHP1 deletion also causes adult-onset ESKD, and the oldest patient described started kidney replacement therapy at age 61 [15, 57]. TTC21B, WDR19 and other ciliary genes combine NPH, liver fibrosis and skeletal abnormalities, and XPNPEP3 and SLC41A1 give NPH-like phenotypes [1, 15].

2.6. Development of the kidney and urinary tract (CAKUT)

CAKUT has a prevalence of 3–6 per 1000 live births, accounts for about 20% of anomalies detected prenatally and is a probably underestimated cause of CKDx in adults [43]. Copy-number variants (CNVs) explain up to 16% of hypodysplasias, with 45 genomic disorders at 37 loci in 4.1% of 2824 cases, plus 2% novel CNVs; monogenic forms explain 10–15%, with PAX2, HNF1B and EYA1 being the most frequent genes [43]. In a study of 522 children with renal agenesis, hypoplasia or dysplasia, microarray detected an unsuspected genomic imbalance in 10.5%, mainly at the 17q12 and 22q11.2 loci, half inherited from an apparently unaffected parent; in another series, pathogenic CNVs were present in 8% of children with CKD of all causes [58]. Renal coloboma syndrome (PAX2) and the branchio-oto-renal spectrum (EYA1) exemplify genes with subtle extrarenal signs that guide the analysis [1, 58].

2.7. Complement, endothelium and metabolism

Genetic atypical HUS (aHUS) has dominant inheritance with reduced penetrance (CFH, CD46, CFI, C3, CFB, THBD, CFHR5, VTN) or recessive inheritance (DGKE), and CFH (21–25%), CD46 (10%), C3 (6%) and CFI (6%) are the most frequent genes; CFH/CFHR1 hybrids (~3–5%) result from non-allelic homologous recombination and are not detected by conventional short-read sequencing [19]. The homozygous CFHR3/CFHR1 deletion (26.5%) and the CFH-H3 and CD46-GGAAC haplotypes are modifiers [19]. Pathogenic or likely pathogenic variants are present in 50–60% of aHUS and in 15–20% of immune-complex membranoproliferative glomerulonephritis and C3 glomerulopathy (C3G) [43]. Defects of intracellular cobalamin metabolism (MMACHC, cblC) cause thrombotic microangiopathy with raised homocysteine and methylmalonic acid and may present in adulthood [1]. Fabry disease (GLA, Xq22.1; NM_000169.3) results from α-galactosidase A deficiency, with accumulation of globotriaosylceramide (Gb3) and its derivative lyso-Gb3; the spectrum includes classic forms, with enzyme activity below 1% in males, and late-onset forms, with renal variants described in Japanese haemodialysis patients misdiagnosed with chronic glomerulonephritis and cardiac variants such as p.Asn215Ser [17]. Primary hyperoxaluria type 1 (AGXT, 2q37.3; NM_000030.3) results from deficiency of hepatic peroxisomal alanine-glyoxylate aminotransferase; types 2 (GRHPR) and 3 (HOGA1) are less severe and type 3 may be silent [18, 59]. APRT (adenine phosphoribosyltransferase) deficiency causes 2,8-dihydroxyadenine crystal nephropathy, with progressive CKD as the most frequent presentation and first symptoms in adulthood in half of the more than 400 patients described [60].

2.8. APOL1: risk allele with toxic gain of function

APOL1 (22q12.3; MANE NM_003661.4) encodes apolipoprotein L1, the active component of the trypanosome lytic factor of human serum; it is not a Mendelian gene and the mechanism of the risk alleles is a toxic cation-channel gain of function, not a deficiency [1, 52]. The G1 allele is a haplotype of two missense variants, NM_003661.4:c.1024A>G p.(Ser342Gly) (rs73885319) and c.1152T>G p.(Ile384Met) (rs60910145); G2 is the 6 bp deletion NM_003661.4:c.1164_1169del p.(Asn388_Tyr389del) (rs71785313) [1, 51, 61]. Both confer resistance to Trypanosoma brucei rhodesiense, which explains their high frequency in sub-Saharan African ancestry through positive selection, and they are almost absent in populations of European, Asian and Native American origin [1, 39]. In the original study, comparing 205 African Americans with biopsy-proven FSGS and 180 controls, the odds ratio (OR) was 10.5 (95% CI 6.0–18.4) for FSGS and 7.3 (95% CI 5.6–9.5) for ESKD, confirmed in 1030 ESKD cases and 1025 controls [39, 40]. Kopp et al. estimated an OR of 17 (95% CI 11–26) for FSGS and 29 (95% CI 13–68) for HIV-associated nephropathy with two risk alleles [52, 62]; the PNDS attributes the value of 17 to the Genovese study, and the difference arises from the two studies and not from a substantive contradiction [1].

The p.(Asn264Lys) variant, designated M1, NM_003661.4:c.792C>A (rs73885316), lies in the membrane-addressing domain, occludes the cation pore and neutralises G2 toxicity when on the same chromosome (G2-M1 haplotype); it does not protect in trans or in G1 carriers, and the G0-M1 haplotype is equivalent to G0 [1, 63]. ClinVar classifies it as benign/likely benign, as a modifier, and the G1/G2 haplotypes as a risk factor [61]. The strict recessive model was nuanced by a case-control study in Ghana and Nigeria with more than 4900 participants, in which about one third carried risk alleles and a single allele was associated with CKD (OR 1.18; 95% CI 1.04–1.33) and FSGS (adjusted OR 1.61; 95% CI 1.04–2.48) [1, 7]. Penetrance of the biallelic genotype is partial (lifetime risk of nephropathy of about 15%), attributed to the need for a second factor, such as viral infection, inflammation or interferon [1, 64].

2.9. Inheritance patterns, penetrance and modifiers in adults

Dominant transmission predominates in CKDx diagnosed in adulthood (heterozygous collagen IV, ADPKD, ADTKD, HNF1B, INF2), but consanguinity or common parental origin increases the probability of recessive disease (NPH, for example due to TTC21B, forms of SRNS, primary hyperoxaluria) [1]. Incomplete penetrance, variable expressivity (microalbuminuria, isolated haematuria, coloboma, subfertility, preauricular tag or cervical fistula as the only manifestations in a relative) and de novo variants explain a "silent" family history [1]. Some variants classified as pathogenic do not imply individual causality, because of reduced penetrance, as with heterozygous SLC34A1 variants, and the ERA consensus requires consistency with the inheritance pattern before attributing the phenotype [2]. The ClinGen working group for low-penetrance and risk alleles proposed a specific terminology for these alleles, distinct from Mendelian classification [65]. The same disease can have two inheritance patterns in the same gene (COL4A3/COL4A4 recessive and dominant; SEC61A1 with three distinct dominant phenotypes), and interpretation must consider the gene, the variant type and the phenotype together [9, 52].

3. Clinical Spectrum and Manifestations by System

3.1. What adult cohorts find

The yield of genetic testing in CKDx ranges from 11 to more than 30% and depends mainly on selection criteria; almost all studies in large CKD cohorts obtain at least 10%, whatever the presentation, and transplant candidates or recipients without histology reach about 40% [2]. The prior clinical diagnosis predicts the yield: the glomerular and tubulointerstitial categories yield more than diabetic nephropathy [3]. Monogenic diseases explain about 70% of ESKD in children and 10–15% in adults, and the genetic diagnosis reclassifies the original diagnosis in 10–22% of patients in large cohorts [43]. The table summarises the adult cohorts used by the PNDS and the ERA consensus, with the population and method of each; the figures are not comparable with one another without considering the inclusion criteria.

The French cohorts show that the yield increases in women and with a family history: in the 2024 series of Robert et al., it reached 56.8% when first- and second-degree relatives were affected, and 13 of 44 patients labelled with another nephropathy (29.5%) were phenocopies [5]. In the Tenon cohort, the diagnosis had a clinical implication considered important in 98% of those diagnosed, mainly through definition of the mode of transmission (78%) and family screening (64%) [4, 50]. A 2025 review by the same authors estimates that about 20% of CKDx is attributable to unrecognised Mendelian disease [68].

3.2. Hereditary glomerulopathies

Type IV collagen nephropathies. These are the most frequent category in the French CKDx cohorts [5]. In males with the X-linked form, macroscopic haematuria occurs in 67% (mean age 3 years), microscopic haematuria is constant (a male with persistently normal urinalysis at age 5 is very unlikely to have the disease), proteinuria never precedes haematuria and all develop it, about 30% with nephrotic syndrome; high-frequency sensorineural deafness affects 83% (mean age 11 years), hypertension appears in adolescence and the historical mean age at ESKD was 21 years, with a mean extension of 13 years under angiotensin-converting enzyme (ACE) inhibitors [48]. Dot-and-fleck retinopathy affects 85% of affected adult males, anterior lenticonus 25% (about 50% in the GeneReviews series), and posterior polymorphous corneal dystrophy is rare [9, 48]. In heterozygous females with the X-linked form, the course is variable: one third have macroscopic haematuria (mean age 9 years), 1–2% a male-like course, one third hypertension in middle age, and the lifetime risk of ESKD is 8–15%; some remain asymptomatic up to age 80 with microscopic haematuria [48]. The recessive form is indistinguishable from the X-linked form in males, and the dominant form has late and variable ESKD, with rare deafness and ocular changes [9, 48]. In adults, these diseases often present as FSGS, atypical "IgA nephropathy", "nephroangiosclerosis" or "benign familial haematuria" [1, 2].

Podocytopathies. Genetic SRNS is, by definition, resistant to glucocorticoids and corrected by transplantation: about 70% of non-genetic SRNS recur rapidly in the graft, whereas monogenic forms rarely do [14]. In adults, genetic forms of FSGS (NPHS1, NPHS2, INF2, TRPC6) often present with incomplete, steroid-resistant nephrotic syndrome [1]. INF2 disease begins in adolescence or adulthood, is associated with intermediate Charcot-Marie-Tooth disease type E (motor conduction velocity 25–45 m/s) and does not respond to corticosteroids or immunosuppressants, with a favourable course after transplantation [52]. COL4A3/4/5 variants explain 10–30% of non-syndromic SRNS, especially with onset from the second decade [14].

3.3. Tubulointerstitial nephropathies

ADTKD. The disease is slowly progressive from childhood or adolescence, with a bland urinary sediment and proteinuria below 250 mg/24 h (UMOD) or 500 mg/24 h (MUC1) until advanced stages, and normal-sized or small kidneys without characteristic cysts [11, 12]. In ADTKD-UMOD, hyperuricaemia is present in almost all and precedes the decline in GFR, gout affects 55% (median 28 years) and 8% of children before age 18, and ESKD occurs between the third and seventh decades, with variation within the same family (30 versus 70 years) [11]. In ADTKD-MUC1, CKD is the only manifestation: eGFR is below 90 in about 50% of children at age 15 and rarely below 60 before 18, the median age at ESKD is 46 years [69], few patients reach it before 20 and some do not reach it after 70; gout is not primary and occurred in 5 of 75 patients, and hyperuricaemia accompanies CKD (about 50% with CKD, 81% with ESKD) [12]. ADTKD-REN manifests in childhood with hypoproliferative anaemia (haemoglobin 9–11 g/dL, low erythropoietin), hyperkalaemia (above 5 mEq/L, up to 6.5, in about 50%), mild acidosis, low-normal BP with low renin and aldosterone and hyperuricaemia in 80%, and reversible acute kidney injury with dehydration or non-steroidal anti-inflammatory drugs (NSAIDs); in adults it progresses as tubulointerstitial CKD with gout [53]. ADTKD-SEC61A1 combines slow CKD with congenital anaemia, intrauterine and postnatal growth restriction, neutropenia and abscesses, and in founder families the kidneys may be small and dysplastic with cysts [36, 52]. Neither the MUC1 nor the UMOD form recurs in the graft [12, 52].

HNF1B-related disease. In carriers not selected for kidney disease, 87% (167/193) have kidney anomalies, and 35% develop them only in childhood or later; hypomagnesaemia affects 43% (46/107), MODY5 usually appears before age 25 (10–50) and is present in 37% (58/155), and there is Müllerian aplasia in women and anomalies of the vasa deferentia and seminal vesicles in men [1, 13]. The tubulointerstitial disease combines a concentrating defect, hyperuricaemia, hypomagnesaemia and hypokalaemia, with a bland sediment and minimal albuminuria; in adults, a fractional excretion of magnesium above 2% with normal function indicates tubular loss [13].

Adult nephronophthisis. NPH causes about 0.5% of adult ESKD; it manifests with polyuria and polydipsia, resistant anaemia and small to normal-sized hyperechoic kidneys, with poor corticomedullary differentiation and cysts at the corticomedullary junction in 50% at a late stage, and proteinuria and hypertension are late [15]. The isolated form has no extrarenal signs in 80–90% of patients [14].

3.4. Cystic diseases

In ADPKD, cysts are age-dependent and affect the kidney, liver, pancreas and spleen; prevalence is about 1 in 2000 in all population groups according to the Oxford Desk Reference and 1 in 1000 at birth according to GeneReviews, and the mean age at ESKD is 58.0 years for PKD1 and 74.8 for PKD2, with more than 50% of PKD2 patients having adequate function at age 70 [10, 54]. Truncating PKD1 variants are associated with ESKD at 55.6 years and non-truncating variants at 67.9 [10]. About 6% have an intracranial aneurysm, with rupture at a mean age of 41 and familial clustering, and up to 80% have liver cysts, more numerous and earlier in women [54]. In adults with CKDx, atypical cystic diseases present with late onset of kidney failure, an incomplete family picture, absence of a PKD1/PKD2 variant or multiple cysts in normal-sized or slightly enlarged kidneys, a phenotype suggestive of DNAJB11 or IFT140 [1, 10]. ADTKD-ALG5 is accompanied by non-enlarged kidneys and interstitial fibrosis with late ESKD, and GANAB produces mild cystic disease with hepatic predominance [10].

3.5. CAKUT in adults

Small hyperechoic kidneys with isolated CKD, without proteinuria or sediment abnormalities, suggest tubulointerstitial disease, ciliopathy or CAKUT [43]. Renal parenchymal defects have a higher prevalence of genetic causes and worse renal survival than isolated hydronephrosis, and the presence of extrarenal defects predicts worse renal survival [58]. Reflux nephropathy is an important cause of ESKD in adults, and vesicoureteric reflux affects about 1% of live births [58].

3.6. Phenocopies with targeted treatment

Fabry disease. Proteinuria usually appears in the fourth decade and kidney failure in the fifth, sometimes at age 21, with a concentrating defect, casts and birefringent "Maltese cross" lipid bodies in the urine [70]. Late-onset variants may have a predominantly renal phenotype, and the cardiac variant has mild to moderate proteinuria with normal kidney function for age and deposits limited to podocytes [17]. In a study of cryptogenic stroke, almost 5% of 342 men had a GLA variant [70]. Life expectancy in the Fabry registry was 58.2 years in men (versus 74.7 in the population) and 75.4 in women (versus 80.0) [17].

Thrombotic microangiopathy and C3 glomerulopathy. Mechanical haemolytic anaemia with thrombocytopenia points to complement (CFH, CFI, CFB, C3, CD46), DGKE, MMACHC and rarer genes (TTC21B, TSEN2, INF2, EXOSC, SOX18); malignant hypertension is a frequent cause of thrombotic microangiopathy in severe CKD, and the penetrance of complement variants is incomplete [1]. Macular drusen before age 50 may precede the kidney manifestations of C3 glomerulopathy and be attributed to macular degeneration [1].

Primary hyperoxaluria type 1 in adults. About 20% of patients are diagnosed in adulthood, mainly because of recurrent lithiasis; 20–50% are diagnosed only with advanced CKD or kidney failure, about 10% of adults only after recurrence in the graft (oxalate crystals on biopsy), often with graft loss, and some remain almost asymptomatic until the sixth decade [18].

Other metabolic and inflammatory phenocopies. LCAT (lecithin-cholesterol acyltransferase) deficiency combines glomerular proteinuria, kidney failure, corneal opacities and haemolytic anaemia, with undetectable or very low HDL cholesterol [1, 14]. Methylmalonic aciduria with homocystinuria cblC may present in adulthood with renal thrombotic microangiopathy [1]. The mitochondrial disease m.3243A>G (MT-TL1) combines diabetes, deafness, exercise intolerance and tubulointerstitial nephropathy, and blood heteroplasmy may be low [1, 2]. Muckle-Wells syndrome (NLRP3) may present as renal amyloidosis with recurrent fever and deafness [2].

3.7. APOL1-associated kidney disease

High-risk genotypes reach up to 50% in south-east Nigeria, 37% in central Ghana and 25% in south-west Nigeria, and about 13% of African Americans have two risk alleles [14, 64]. Among African Americans with two alleles, 15–20% develop CKD and 40–60% of primary FSGS occurs in this genotype [64]. The disease phenocopies "hypertensive nephropathy", "nephroangiosclerosis" and FSGS, includes collapsing glomerulopathy and HIV-associated nephropathy, and is associated with accelerated progression to ESKD (hazard ratio, HR, of 2.3) [1]. The lifetime risk of FSGS with two alleles was estimated at about 4%, and that of untreated HIV-associated nephropathy at about 50% [52]; the PNDS refers to a penetrance of about 15% for any nephropathy [1]. These values measure different outcomes and do not contradict each other.

3.8. Extrarenal manifestations by system

Extrarenal manifestations guide genetic analysis and, in adults, are often incomplete, subtle or attributed to other causes [1]. Hearing (high-frequency sensorineural deafness in collagen IV disease, MYH9, PAX2 and mitochondrial disease), the eye (lenticonus, retinopathy, optic disc coloboma, cornea verticillata, retinitis pigmentosa, drusen, retinal angiomas and hamartomas), the nervous system (INF2 neuropathy, early stroke in Fabry, COL4A1/COL4A2 and ADPKD, neurodevelopmental disorders in ciliopathies, HNF1B, Lowe syndrome and CNVs), metabolism (early gout, atypical diabetes, very low HDL), the genital tract (bicornuate uterus, agenesis of the vasa deferentia, primary amenorrhoea with FSGS in WT1-related Frasier syndrome), skin and appendages (angiokeratomas, dysplastic nails and absent patellae, angiofibromas, cutaneous leiomyomas, abnormal hair and teeth), the skeleton and the lung (pneumothorax in Birt-Hogg-Dubé, lymphangioleiomyomatosis in tuberous sclerosis complex) appear in the PNDS table of warning signs [1]. The haematological system provides its own clues: the macrothrombocytopenia of MYH9, the disproportionate anaemia of REN, the leucopenia of SEC61A1 and the polycythaemia of von Hippel-Lindau disease (VHL) [1, 14, 52].

3.9. Natural history and treated disease

The ages at ESKD in historical studies are not estimates for treated patients. In X-linked Alport syndrome, the historical mean age at ESKD of 21 years in males is extended by a mean of 13 years with ACE inhibitors [48]. In ADPKD, the gene-specific age-at-ESKD data come from cohorts predominantly not treated with tolvaptan [10]. In the treatable phenocopies, the natural history changes with targeted treatment; the life expectancy cited for Fabry disease comes from the 2009 analysis of the Fabry registry and does not separate treated and untreated patients [17]. A late genetic diagnosis, at ESKD, no longer alters the patient's renal history, but retains value for transplantation, donor selection and relatives [1].

4. Clinical Recognition and Warning Signs

4.1. Warning signs for the attending physician

The attending physician is in the front line for recognising the signs that, in unexplained chronic kidney failure, suggest a genetic cause [1]. The PNDS summary highlights four groups: a family history of CKD, dialysis or transplantation in a first- or second-degree relative, including early cardiovascular deaths that may have masked CKD; severe hypertension before age 40 or familial hypertension, which may be the initial sign of an underlying kidney disease; an atypical profile, that is, progressive kidney failure without diabetes or massive proteinuria (above 1 g/g) or with small kidneys (below 9 cm) from diagnosis; and extrarenal signs, such as sensorineural deafness, gout before age 30, liver cysts, ocular abnormalities or diabetes in a young person with a family history suggestive of monogenic MODY-type diabetes [1]. The KDIGO conference adds conditions with targeted treatment, consanguinity, congenital or cystic forms, SRNS and relatives of a patient with a known monogenic disease, especially potential donors [3].

4.2. History: duration of disease and early clues

History-taking is the most important step, especially when the patient is seen at an advanced stage [1]. The child health record and old medical documents should be consulted for: birth weight below 2500 g (and especially below 1500 g) or intrauterine growth restriction, which reduce nephron number and increase the risk of kidney failure in adulthood by 70%; prematurity, because nephrogenesis is completed only at 36 weeks of gestation and extreme prematurity may explain adult proteinuria or glomerulosclerosis labelled as undetermined; perinatal asphyxia or neonatal acute kidney injury, even if resolved; growth faltering, short stature or delayed puberty; BP measured from age 3; and extrarenal signs documented in childhood (hearing screening, ophthalmological abnormalities, polydactyly) [1]. Proteinuria or haematuria detected on routine examinations (occupational, school or military medicine, blood donation) 10 or 20 years earlier and never investigated are clues of prime importance [1]. The chronology of hypertension should be reconstructed, because early hypertension, long before kidney failure, is a sign of the nephropathy and not its cause [1].

The obstetric history systematically includes hypertensive disorders and proteinuria in pregnancy. In a recent cohort cited by the PNDS, pre-eclampsia was associated with a 2.5-fold higher risk of persistent albuminuria and a 2.2-fold higher risk of long-term reduction in GFR; in CKDx, it should be read as a physiological "stress test" that revealed glomerular or endothelial fragility and may be the first manifestation of a collagen IV disease (dominant or X-linked forms) or of an alternative complement pathway abnormality [1]. Terminations of pregnancy for medical reasons, fetal deaths and the results of fetopathological examinations complete the history [1].

4.3. Structured family history

The ERA consensus recommends a structured family history in all patients with CKD, covering first- and second-degree relatives, consanguinity, premature deaths, type of disease, age at start of kidney replacement therapy and extrarenal manifestations [2, 71]. The PNDS describes the technique: notify the patient in advance so that they can gather documents and information; ask member by member (parents, siblings, children), avoiding the generic question; use life markers ("has anyone had dialysis, a transplant, protein in the urine on a test, or wears a hearing aid?") instead of "kidney disease"; explain that the same disease may have different expressions in the family (for example, diabetes without metabolic syndrome in an uncle, cysts in an aunt and kidney failure in the patient, in a family with an HNF1B variant); and build a pedigree of at least three generations with the type of nephropathy, the age at kidney failure and at kidney replacement therapy, and extrarenal manifestations (deafness, eye, liver cysts, diabetes, gout, intracranial aneurysm) [1].

Three pitfalls mask heritability. The premature death of a parent or grandparent from an extrarenal cause (accident, myocardial infarction, stroke, cancer) may have concealed a late-onset disease such as ADPKD or dominant Alport syndrome [1]. Consanguinity or common parental origin (same village, endogamous population), even without known kinship, increases the probability of recessive disease [1]. A "silent" family history does not exclude hereditary disease, because of incomplete penetrance, mild expression or a de novo variant [1]. A history of cancer should be recorded for two reasons: kidney, skin or uterine cancers may indicate VHL, hereditary leiomyomatosis and renal cell cancer (HLRCC, FH) or Birt-Hogg-Dubé syndrome (FLCN), and a significant family oncological history prepares the information about possible incidental findings in cancer predisposition genes on exome or genome [1].

4.4. Targeted physical examination

The examination seeks a clinical signature: no isolated sign is pathognomonic, but the association of minor signs (for example, kidney cysts, polydactyly and obesity) guides the analysis [1]. On dialysis, oedema and hypertension depend on kidney replacement therapy and lose aetiological value; the examination then actively seeks subtle or misleading extrarenal signs [1]. An overall impression of "constitutional atypia" (facial dysmorphism, low-set ears, unusual gait or voice) and neurodevelopmental disorders, poorly assessed by the adult nephrologist, justify assessment by a clinical geneticist [1].

The table reproduces the PNDS warning signs and pitfalls (Table 1), with the main diagnostic orientations.

Source: HAS PNDS, Table 1 [1].

4.5. Settings with a high probability of a genetic cause

The probability of a molecular diagnosis is higher in early CKD (before age 40), with a family history, with unexplained tubulointerstitial involvement, atypical cysts or extrarenal manifestations [1]. In the Tenon cohort, a score predicting positivity was built from these elements [4]. In the 2024 series of Robert et al., the yield was higher in women (p=0.02) and with a family history (p<0.0001) [5]. In "hypertensive nephropathy" before age 45, the yield was 30% in patients without African ancestry and 7% in those of African ancestry, in whom APOL1 explains part of the disease [16]. A stable, isolated eGFR between 45 and 60 mL/min/1.73 m², without albuminuria or sediment abnormalities, in a patient older than 65–70 years, corresponds to renal senescence and not to CKDx with a genetic indication [1, 72].

4.6. Urgent situations

Some diseases revealed by the investigation of CKDx require immediate treatment, before the genetic result. In thrombotic microangiopathy with suspected aHUS, the ERA recommends performing genetic and biochemical complement screening in parallel and starting the complement inhibitor without delay while awaiting results [43]. In thrombotic microangiopathy with raised homocysteine, the European guideline on remethylation defects recommends parenteral hydroxocobalamin in the cblC defect [1, 73]. In primary hyperoxaluria, acute kidney injury due to bilateral obstruction is a form of presentation in adults [18]. In ADTKD-REN, dehydration and NSAIDs precipitate acute kidney injury and hyperkalaemia [53].

4.7. Newborn screening and targeted screening

There is no newborn screening for CKDx. Newborn screening programmes for Fabry disease in some US states and European countries revealed frequencies of 1 in 3000 to 1 in 9000 males, mainly of late-onset variants [17]. Targeted screening in dialysis and hypertrophic cardiomyopathy suggests that late-onset Fabry forms are more frequent than classic ones [17]. The KDIGO consensus on APOL1 considers the evidence insufficient for population screening or routine genotype testing [64].

5. Diagnosis: Criteria, Tests and Algorithms

5.1. CKDx criterion and recording the depth of investigation

CKDx is defined by the absence of a cause established with sufficient certainty after clinical, laboratory, imaging and, when indicated, histological evaluation, and the PNDS recommends using this designation whenever the data do not allow the disease to be reliably attributed to a precise cause [1]. The ERA consensus makes four general recommendations: pay attention to the "cause" component of the KDIGO CGA classification (cause, GFR category, albuminuria category); classify CKD without a defined cause as CKDx; recognise that low-certainty names, such as histological patterns, may lead to unnecessary therapies or the loss of useful ones; and adopt a cause-directed approach in all CKD, with a dedicated diagnostic re-evaluation clinic that returns to the original sources (reports, images, slides) [2]. The report should record the "h/g" extension with the year, the genetic modality and whether the biopsy included EM, for example "CKDx G3a A1 (h+g+(2024))" [2].

5.2. Minimum assessment before referral

Before referring the patient to a nephrologist, the minimum assessment includes creatinine with estimated GFR (eGFR) by the EKFC (European Kidney Function Consortium) or CKD-EPI (Chronic Kidney Disease Epidemiology Collaboration) equation, spot albumin/creatinine ratio (ACR), urine microscopy and culture, urine dipstick, bilateral renal ultrasound (size, cysts, echogenicity), previous creatinine values to calculate the eGFR slope and a simplified pedigree [1]. Referral is made in the first instance to the local nephrologist and, if a rare kidney disease, a complex situation or a lack of local experience is suspected, to a competence or reference centre of the ORKiD network [1]. The assessment is coordinated by the nephrologist, who may direct the patient to a nephrogenomics centre when there are criteria for genetic investigation [1].

5.3. Laboratory tests

Kidney function. Interpretation of tests depends on the stage: many markers are informative in early and moderate stages and lose aetiological value in ESKD and dialysis, where abnormalities are a non-specific consequence of uraemia [1]. The EKFC equation is recommended by the PNDS as a robust alternative in Europe, applicable from age 2 to 100, with continuity across the transition from paediatric to adult care and better precision at the extremes of age; the 2009 CKD-EPI remains widely used but has limitations between ages 18 and 25 [1, 74]. The 2021 CKD-EPI equations without a race coefficient are the North American reference equations [75]. The European study of Astley et al., with more than 1.5 million healthy individuals, established eGFR reference values by age and sex: GFR declines from age 40, at age 80 the 5th percentile lies at 45–50 mL/min/1.73 m², and it falls below 60 after age 70–75 [1, 72]. Cystatin C is strongly recommended to confirm the stage in sarcopenia or muscle wasting (creatinine overestimates function), in high muscle mass (it underestimates) and in unexplained discordance between creatinine and the clinical picture [1].

Guiding laboratory abnormalities. The table reproduces Table 2 of the PNDS, with the orientations and limits of each abnormality.

Source: HAS PNDS, Table 2 [1].

Urine. In the anuric patient on dialysis, old urine tests in the medical record allow the initial urinary syndrome to be characterised [1]. Quantification is by protein/creatinine ratio (PCR) or ACR, preferably on the first morning urine, which excludes orthostatic proteinuria of young adults and correlates better with 24-hour proteinuria; in the presence of nocturia, the result is valid but should be interpreted in the light of this fragmentation [1]. If albumin accounts for more than 60% of total proteinuria, the involvement is glomerular; if it accounts for less than 40–50%, it suggests low-molecular-weight proteinuria of proximal tubular origin (Dent disease, Fanconi syndrome, drug toxicity) [1]. Microscopy of the sediment distinguishes glomerular haematuria (dysmorphic red cells, red cell casts) from urological haematuria and shows white cell or granular casts of tubulointerstitial involvement [1]. Crystalluria requires a dedicated fresh sample, because crystallisation continues after collection [1]. In France, urine microscopy and culture follows the microbiology standard, with boric acid tubes and flow cytometry, conditions that do not allow analysis of fragile elements; the nephrologist should request quantitative urine cytology on fresh urine with manual search for casts and red cell dysmorphism, analysed within 2 hours of voiding (collection at the laboratory or the clinic), midstream, after washing with water or saline and more than 72 hours after intense exertion or menstruation [1].

5.4. Imaging

Renal and bladder ultrasound is the first-line test, with an explicit request for the elements to be described [1]. Small kidneys with a thin cortex reflect chronicity; normal or enlarged size points to amyloidosis or diabetes; hyperechogenicity is non-specific but marked in CTIN and NPH; medullary nephrocalcinosis suggests Dent disease or primary hyperoxaluria [1]. Ultrasound is the key test for CAKUT (hypoplasia, dysplasia, single kidney, duplex collecting system, cortical reflux scars, seminal vesicle cysts, bicornuate uterus in the context of HNF1B) and should confirm the absence of obstruction or ectopic ureteric insertion [1].

The location of cysts has aetiological value: diffuse bilateral cortical cysts with nephromegaly suggest ADPKD; cysts at the corticomedullary junction in normal or small kidneys suggest NPH (young adults) or ADTKD; multiple cortical and medullary microcysts may result from prolonged lithium treatment [1]. Medullary sponge kidney (Cacchi-Ricci), with dilatation of the collecting ducts and papillary calcifications, is diagnosed by CT urography with a delayed excretory phase or by flexible ureteroscopy [1]. In the liver, simple polycystic liver disease (PKD1/PKD2, ALG8, GANAB) is distinguished from biliary anomalies and congenital hepatic fibrosis (PKHD1, Caroli disease); in the pancreas, cysts or atrophy of the body and tail suggest HNF1B or VHL [1]. T2-weighted renal and abdominal MRI is the reference test for phenotyping, superior to ultrasound and CT for microcysts, corticomedullary differentiation, liver and pancreas, and is recommended in atypical or discordant cystic phenotypes and in internal genital anomalies [1]. The ERA adds non-contrast CT for lithiasis and nephrocalcinosis in adults [2].

When ADPKD is suspected, Pei's unified ultrasound criteria for at-risk individuals are: age 15–29, at least 3 unilateral or bilateral cysts (positive predictive value 100%; sensitivity 94% for PKD1 and 69% for PKD2); age 30–39, at least 3 cysts; age 40–59, at least 2 cysts in each kidney; age 60 or over, at least 4 cysts in each kidney; fewer than 2 cysts at age 40 or over excludes the diagnosis [54, 76]. On MRI, more than 10 cysts above age 16 have 100% sensitivity and specificity, which is useful in evaluating a related living donor [10, 54].

5.5. Kidney biopsy

Biopsy remains indispensable in acquired diseases, in particular immunological glomerulonephritides, but in CKDx genetic analysis is a first-line alternative when hereditary disease is suspected, especially with a family history or a contraindication to biopsy [1]. Biopsy is justified when an acquired and treatable disease is suspected (rapidly progressive glomerulonephritis, overt adult nephrotic syndrome) or to assess activity and chronicity and guide immunosuppression; the risks increase with small kidneys, haemostatic abnormalities and uncontrolled hypertension [1]. In categories G4–G5, the bleeding risk increases and findings become non-specific [2]. The sample should have sufficient glomeruli and cortex for light microscopy, immunofluorescence and, when necessary, EM [1].

The histological result should be read as a deep phenotype [1]. A "non-specific" biopsy (nephroangiosclerosis, moderate interstitial fibrosis) or a "non-contributory" one is an important argument in favour of genetics: in the series of 52 adults of Robert et al., the exome yield was 72.7% in patients with non-specific lesions [1, 44]. FSGS without immune deposits is a common phenotype of many genetic podocytopathies, including collagen IV disease; interstitial foam cells may point to collagen IV; EM looks for thinning, irregularity or lamellation of the GBM; and interstitial fibrosis with an infiltrate without tubulitis has been observed in hereditary tubulopathies and ciliopathies [1]. GBM thickness cut-offs are 250–330 nm in adults and 200–250 nm in children; "basket-weave" lamellation is pathognomonic; α3/α4/α5 immunostaining is absent in males with the X-linked form (with positive staining in 30–40%, especially with missense variants), shows a mosaic pattern in females, lacks α3/α4 in the GBM with α5 in Bowman's capsule in the recessive form and is normal in the dominant form [9]. In ADTKD, biopsy should not be performed for diagnosis, because the findings are non-specific and molecular testing is safer and cheaper [11, 12, 53].

Mesangial IgA deposits do not exclude hereditary disease. Variants in COL4A3, COL4A4 or COL4A5 have been identified in patients labelled with IgA nephropathy and, conversely, a variant in these genes in a patient with IgA does not necessarily explain the whole phenotype, because collagen IV disease may coexist with true IgA nephropathy or with another acquired glomerulopathy [1]. Testing of COL4A3/4/5 is justified in IgA nephropathy with atypical features (family history of kidney disease, familial haematuria, deafness, GBM abnormalities, unusual course or resistance to treatment), and the histological diagnosis of IgA nephropathy should be re-evaluated when it rests on incomplete criteria (minimal deposits without proliferation or discordance with clinical severity) [1]. In typical Alport syndrome, genetics is the first-line test and may make biopsy unnecessary; in atypical CKDx, biopsy helps interpret variants, and a variant or VUS may prompt targeted re-analysis of the histology, retrophenotyping [1]. In the 2024 series of Robert et al., the exome-first strategy gave a diagnosis in 10 of 22 patients and avoided 11 biopsies [5].

5.6. CKDx subcategories with orienting value

Source: HAS PNDS, section 3.6 [1].

5.7. Indications for genetic testing

The PNDS recommends discussing genetic investigation in every adult with CKD without an evident aetiology after the first assessment, and makes the indication formal in the presence of at least one of the following criteria: familial context (at least one first- or second-degree relative with undetermined CKD or confirmed hereditary kidney disease); CKD diagnosed before age 60 without an evident cause, the indication being maintained after 60 when there is an expected benefit for the patient or relatives; at least one extrarenal manifestation suggestive of a genetic syndrome; a picture or course discordant with common acquired nephropathies, with a formal indication in adult SRNS and in rapid progression (for example, IgA nephropathy with discordant histological lesions); a "non-specific", "non-contributory" biopsy or one with only chronic lesions; and organ donation, in which the CKDx recipient of a related living donor should be tested, and possibly the donor, to exclude a hereditary disease that would contraindicate donation or threaten the donor [1, 77]. The ERA consensus considers genetic testing increasingly the standard in CKDx before about age 60, requires explicit justification for omitting it and lists other reasons to test: reproductive decisions, surveillance and donation among relatives, avoiding invasive procedures, starting or stopping therapies (for example, immunosuppression in FSGS) and relieving anxiety [2]. In a multi-step pathway, early exome sequencing reduced actual costs by 41% [2]. The KDIGO conference recommends that all nephrologists know how to select patients, take the personal and family history and collaborate with the geneticist or genetic counsellor [3].

The PNDS considers that testing is generally not indicated in a palliative situation, at the end of life or in the multimorbid elderly patient without impact on management or young relatives at risk; when there is no impact for the patient or for first- and second-degree relatives; when the request arises from curiosity, anxiety or third-party pressure without a clinical sign or family history, in which case the patient should be directed to psychological support; and when the patient refuses it, in respect of the right not to know [1]. ERKNet advises against testing SRNS with secondary resistance after an initial response to corticosteroids [43].

5.8. Choice of test

The PNDS describes the evolution from targeted panels to a first-line genome-wide strategy (exome or genome) in adults, justified by genetic and phenotypic heterogeneity and by the frequency of phenocopies; when access is limited, an extended panel with all kidney disease genes (more than 600, as in the Groopman cohorts) is an acceptable alternative [1]. The ERA distinguishes the sequencing method (capture panel, exome, genome) from the analysis applied (virtual panel on exome or genome); the "clinical exome" covers the coding part of disease-associated genes, fewer than 30% of genes; the PanelApp kidney panel has more than 600 genes; and the minimum required in CKDx is a comprehensive kidney panel with genes of moderate to definitive evidence, curated in ClinGen or PanelApp [2]. The 2022 ERA and ERKNet recommendations propose a tiered approach, in which the extended panel of all nephropathy genes is analysed first and, if negative, the full exome is opened [43]. The exome represents 1–2% of the genome [43]. In undiagnosed syndromic involvement, the genome is preferable because of its ability to detect rearrangements [1]. CNV detection from exome data, assessed in 2418 cases, produced 30 additional diagnoses (about 1.2%), an argument used by the PNDS for first-line exome without a prior comparative genomic hybridisation array [50, 78]. Describing the phenotype in HPO (Human Phenotype Ontology) terms improves bioinformatic prioritisation of variants [2, 43]. The recommended reporting times are 8–12 weeks, and 2 weeks in urgent prenatal or neonatal situations [43].

The requesting clinician should have specific training in the indications, patient information, the limits of the test and interpretation, and prepares a summary clinical and paraclinical file, an indispensable condition for a relevant interpretation; in case of doubt, the PNDS recommends seeking the opinion of a competence or reference centre, by tele-expertise or at a multidisciplinary team meeting (MDT) on hereditary kidney diseases [1].

5.9. Technical limits: what a negative test may have missed

The ERA requires the report of a negative test to state the genes analysed, the technology, the year, whether CNVs were analysed and, when relevant, whether the MUC1 VNTR and mitochondrial DNA were assessed [2]. Long-read sequencing and RNA sequencing are reserved for selected cases [2]. The proportion of variants detectable by a method (for example, UMOD sequencing detects more than 95% of known variants) is not equivalent to the clinical sensitivity of the test in CKDx, which depends on the prior probability of each disease [2, 11].

5.10. Pre-test information and consent

In France, the pre-test information consultation is mandatory and ensures free, informed and specific consent, written and signed in duplicate (articles L1131-1 and R1131-1-1 of the Public Health Code), with a possible reflection period on request [1]. The information explains the indication, the methodology (panel, exome or genome) and the possible results: positive; negative, which does not exclude a genetic cause; VUS, which is not a diagnosis and does not in itself change management; and risk factor or susceptibility variant [1]. It also covers incomplete penetrance and variable expressivity, family implications and the legal obligation to inform relatives, confidentiality and revocability of consent (with possible destruction of the sample), protection against insurers, sharing of anonymised data and emotional impact [1]. ERKNet recommends specific consent for large panels, exome and genome, a clear local policy on incidental findings and pre- and post-test counselling [43].

Incidental findings, discovered unintentionally during analysis directed at the nephropathy, are distinguished from secondary findings, which would result from an active search of a predefined list of actionable genes, such as the ACMG SF v3.2 list, and which are not authorised in France [1, 80]. The patient decides before the test whether they wish to be informed of actionable incidental findings (genes such as BRCA1/BRCA2, TP53, LDLR, RYR1 or KCNQ1, the list being consulted only after the discovery), may revoke this decision at any time and, if a serious and actionable finding is communicated, is subject to the same obligation to inform relatives [1]. In the Groopman et al. cohort, 1.6% of patients assessed had actionable findings unrelated to the nephropathy [8]. To explain the value of the test, the PNDS offers the clinician three reasons to convey to the patient: guiding treatment (avoiding ineffective drugs such as corticosteroids, monitoring the heart and eyes, refining the renal prognosis), guiding transplantation (risk of recurrence and choice of donor) and enabling family screening with early nephroprotection of carriers [1].

5.11. Interpretation of results

Interpretation is a co-interpretation between the nephrologist, who provides a precise, longitudinal and familial phenotype, and the laboratory, which assesses the variant at population, bioinformatic and functional levels [1]. A variant is pathogenic (class 5) or likely pathogenic (class 4) according to the ACMG/AMP (American College of Medical Genetics and Genomics/Association for Molecular Pathology) criteria, combining the null nature of the variant, functional data, segregation, gnomAD frequency and computational predictions, in line with the French ANPGM good practice guidelines [1, 81, 82]. Adapted ACMG criteria exist for collagen IV [9, 47]. Returning a diagnosis requires strong concordance between the variant and the renal or extrarenal phenotype, exclusion of competing variants and confirmation at an MDT or by the reference centre's opinion in case of doubt [1]. The ERA points out that a P/LP variant does not necessarily imply causality in the patient, especially in fragmented healthcare systems, and that retrophenotyping is indispensable for the result to be interpreted in the clinical context [2]. The report should identify the gene, the versioned reference transcript, the variant in HGVS nomenclature, zygosity, ACMG/AMP class and the analyses performed, including CNVs [2, 43].

VUS. VUS (class 3) are frequent, are not diagnostic and should not, by themselves, change patient management [1]. The laboratory reports them sparingly, after analysis of phenotypic consistency and mode of transmission, low gnomAD frequency, in silico predictions (SIFT, REVEL, SpliceAI) and consistency with the known mechanism and hotspots of the gene [1]. The PNDS describes three situations: lack of clinical data, in which retrophenotyping is performed (for example, looking for early gout or familial hyperuricaemia in the presence of a UMOD VUS); lack of functional or family data, in which a segregation or functional study is proposed (for example, minigene); and a poorly described candidate gene, in which a specialised centre is consulted [1]. A VUS should not label the patient or lead to irreversible decisions (selection of a related living donor, targeted screening of relatives), cannot form the basis of prenatal or preimplantation diagnosis and should be re-evaluated every 3–5 years or according to laboratory procedures, or when new clinical data arise [1]. The ERA consensus suggests a review perhaps every 2–3 years and reports only VUS with a high probability of causality and a realistic prospect of reclassification [2]. The KDIGO conference did not reach consensus on which VUS to report and recommends interdisciplinary contact before communicating them [3].

Retrophenotyping. This consists of actively seeking, once a variant has been identified, the expected clinical, laboratory, radiological or histological signs, through targeted review of the record and multidisciplinary assessment (nephrologist, clinical geneticist, laboratory scientist, radiologist, pathologist, ENT surgeon, ophthalmologist) [1]. Useful tests include slit lamp and OCT (collagen IV, Fabry), audiogram (collagen IV, MYH9), uric acid and FEUA (ADTKD, HNF1B), renal and abdominal MRI and brain MRI (mitochondrial disease, COL4A1/COL4A2), biopsy EM (GBM) and functional tests (homocysteine and methylmalonic acid in cobalamin defects) [1]. The risk is over-interpretation of non-specific signs in the light of the result, and retrophenotyping should begin by checking phenotypic consistency [1]. In the PNDS example, a 40-year-old man with a GFR of 50 mL/min/1.73 m², hypertension and thrombotic microangiopathy on biopsy had on exome a pathogenic MMACHC variant, NM_015506.3:c.566G>A p.(Arg189His), and a PRDX1 variant, c.515-1G>T (NM_181697.3, MANE; the PNDS uses NM_002574.3, with the same position), initially a VUS and capable of inducing a secondary epimutation of MMACHC; homocysteine of 346 µmol/L (normal <15) and raised plasma and urinary methylmalonic acid allowed the PRDX1 variant to be reclassified as likely pathogenic and the diagnosis of methylmalonic aciduria with homocystinuria due to cobalamin C deficiency to be established [1, 51].

Segregation. The segregation study looks for the variant in affected and unaffected relatives, ideally across three generations; it strengthens pathogenicity (criterion PP1, whose strength depends on the number of informative meioses, or PS2 if the variant is de novo) or weakens it (BS4 if it does not segregate) [1]. First-degree relatives (parents, siblings, adult children) are prioritised, each gives their own written consent, and their renal phenotype should be documented with creatinine and eGFR, PCR or ACR, quantitative urine cytology (microscopic haematuria above 5 red cells/mm³) and renal ultrasound [1]. In an asymptomatic relative, the analysis is a presymptomatic test, performed in a specialised clinic (nephrologist with genetics training, geneticist or genetic counsellor under medical responsibility); in a symptomatic relative, it may be requested by the nephrologist; and it should preferably be performed in the index case's laboratory, with local sampling and dispatch when the relative lives far away [1].

Inconclusive result. When segregation and retrophenotyping do not allow a conclusion, a clinical diagnosis without molecular confirmation (for example, Gitelman syndrome with a characteristic phenotype) is distinguished from a diagnostic impasse, in which no clinical or molecular diagnosis can be retained; in France, the latter can be recorded with a specific code in the National Rare Disease Database (BNDMR) [1]. The VUS remains in the report without change in family management and with the possibility of recontact [1].

5.12. Negative result and escalation

A result without a pathogenic, likely pathogenic or relevant VUS variant is "inconclusive" at that point in time, because knowledge is incomplete and some regions are not explored by standard techniques; the patient should be informed that the genetic hypothesis has not been ruled out [1, 2]. When suspicion is strong, the PNDS proposes genetic re-evaluation in 2–3 years or upon phenotypic change, and assessment of first-degree relatives with creatinine and eGFR, ACR, sediment and, depending on the context, ultrasound, repeated periodically in young people with a normal assessment [1]. The strategy includes longitudinal phenotyping (new auditory, ophthalmological, neurological or metabolic manifestations), updating the pedigree, periodic bioinformatic re-analysis of the raw data (new genes, variant reclassification, better tools), escalation at an MDT to short-read genome or to long-read sequencing (Nanopore, PacBio) for repetitive regions (MUC1), rearrangements with pseudogenes (PKD1) and fusion genes (CFH/CFHR), systematic discussion at a nephrogenetics MDT and sharing in databases such as GeneMatcher [1]. Meanwhile, control of progression factors remains strict, with honest information about the uncertainty [1].

5.13. APOL1 genotyping

The ERA recommends routinely including APOL1 in broad CKDx panels, because attributed ancestry is imprecise and there is population admixture, especially in patients with West African ancestry with proteinuria or FSGS [2]. The KDIGO consensus on APOL1 considers that the genotype by itself is not actionable and that there is no evidence for population screening or routine testing, but that genotyping may inform risk stratification, frequency of follow-up, living donation and eligibility for trials; its principles require informed consent, kidney disease, candidacy for donation or a relative with a high-risk genotype, an acceptable risk of harm and qualified counselling, and call for race, ethnicity, ancestry and genotype to be distinguished [64]. The term "APOL1-associated kidney disease" does not obviate diagnostic evaluation, including biopsy when indicated [64]. The PNDS adopts the AMKD (APOL1-mediated kidney disease) classification into four categories, which incorporates the M1 modifier [1].

Source: HAS PNDS, Table 6 [1]. M1 denotes p.(Asn264Lys) in cis with G2; M0, its absence; G0-M1 is equivalent to G0.

5.14. Algorithm

5.15. Prenatal and preimplantation diagnosis

Prenatal diagnosis (chorionic villus sampling at 11–13 weeks of gestation or amniocentesis at 15–17 weeks) and preimplantation genetic diagnosis require an identified familial pathogenic or likely pathogenic variant, are considered only in diseases of particular severity and cannot be based on a VUS [1].

6. Differential Diagnosis

6.1. Acquired and developmental causes to exclude

The differential diagnosis of CKDx has two sides: the non-genetic causes that the assessment must exclude before labelling the disease as undetermined, and the genetic diseases that mimic one another. Among the former, the PNDS highlights the exposome, that is, the set of non-genetic exposures throughout life: occupational (solvents, heavy metals such as lead, cadmium and mercury, silica), environmental (drinking water, proximity to industrial or agricultural sites, air pollution, hair straighteners containing glyoxylic acid), toxic (pesticides and herbicides, prolonged use of NSAIDs or proton pump inhibitors, uncontrolled herbal medicine containing aristolochic acid, cocaine and heroin associated with FSGS, amphetamines and MDMA associated with rhabdomyolysis and malignant hypertension, cannabis, ketamine associated with obstructive uropathy) and lifestyle-related (chronic heat exposure, diet) [1]. Congenital oligonephronia due to low birth weight, prematurity or neonatal kidney injury, reflux nephropathy, obstructive uropathy, lithium and autoimmune causes belong to the same differential [1, 58]. Endemic CKDu is a distinct entity, defined by epidemiological and environmental context [2]. These causes may coexist with a genetic predisposition, and the PNDS describes them as a cause or cofactor [1].

6.2. Differential table of the main genetic diseases

6.3. Differentials with the greatest practical impact

The first pair of differentials is IgA nephropathy and type IV collagen nephropathies. IgA deposits may be secondary or coexist with a COL4A3/4/5 variant, and patients with a family history, familial haematuria, deafness, GBM abnormalities, an unusual course or resistance to treatment should be tested; a heterozygous carrier with disproportionate disease should be investigated for an associated glomerulopathy [1]. The second is "primary" FSGS versus genetic FSGS: abrupt onset, response to corticosteroids or immunosuppressants and rapid recurrence after transplantation suggest the immunological form, and their absence requires a search for a monogenic cause, including collagen IV, because three of the collagen IV diagnoses in the Berlin cohort had been classified as FSGS [1, 50]. The third is "hypertensive nephropathy" in young adults: in the French cohort of patients younger than 45, 15% had a monogenic diagnosis, more frequent in those without African ancestry, in whom Mendelian forms compete less with APOL1 [16].

Among CTINs, ADTKD is distinguished from NPH by its dominant mode of transmission and later ESKD (third to seventh decade, versus adolescence or the 20s in NPH), from Fabry disease by the more marked proteinuria and ESKD in the third to fifth decade of the classic form, and from HNF1B-related disease by the absence of MODY, CAKUT and liver abnormalities [11]. In cystic disease, the distinction between typical ADPKD and atypical forms (DNAJB11, IFT140, ALG5, ALG9, GANAB) has consequences for prognosis and donor evaluation [1, 10]. A kidney biopsy with "zebra bodies" on EM points to Fabry, but may be mimicked by hydroxychloroquine-induced renal phospholipidosis [17].

7. Treatments and Risk Management Strategies

7.1. Principle: the diagnosis guides treatment

The PNDS is limited to the diagnostic pathway and refers treatment of each identified disease to the specific PNDS and recommendations: Alport syndrome and familial haematurias, idiopathic nephrotic syndrome (including genetic FSGS and SRNS), HUS (including aHUS), C3 glomerulopathy, ADPKD, tuberous sclerosis, nephronophthisis, Bardet-Biedl syndrome, Gitelman, Bartter, distal renal tubular acidosis, X-linked hypophosphataemic rickets, nephrogenic diabetes insipidus, Fabry, primary hyperoxaluria types 1 to 3, cystinuria, cystinosis, nail-patella syndrome and nephrocalcinosis [1]. The PNDS identifies three direct therapeutic consequences of the genetic diagnosis: starting a targeted treatment, stopping an inappropriate treatment and giving access to trials whose inclusion criterion is the genotype [1].

7.2. Common nephroprotection

After diagnosis, the attending physician ensures shared nephroprotection: BP control below 130/80 mmHg, avoidance of NSAIDs and iodinated contrast media, smoking avoidance, physical activity and an adapted diet [1]. The KDIGO 2024 CKD guideline recommends, in most patients, three first-line classes: a renin-angiotensin-aldosterone system inhibitor (RAASi: ACE inhibitor or ARB), a sodium-glucose cotransporter 2 inhibitor (SGLT2i) and a statin [85]. An ACE inhibitor or ARB is recommended in CKD without diabetes with A3 albuminuria (grade 1B) and suggested with A2 albuminuria (grade 2C); an SGLT2i is recommended in adults with type 2 diabetes and CKD with eGFR ≥20, in CKD with heart failure with reduced ejection fraction and with eGFR ≥20 and ACR ≥200 mg/g, should be continued until dialysis or transplantation and is associated with an expected initial dip in eGFR [85]. KDIGO proposes a systolic BP target below 120 mmHg with standardised office measurement (below 130 if measurement is not standardised), adjusted for frailty, which differs from the 130/80 mmHg target of the PNDS summary [1, 85]. The STOP-ACEi trial showed no benefit from stopping the RAASi in advanced CKD, and the sick day rules for temporary discontinuation in acute illness ("SADMANs" rules) apply to these drugs [85]. Early nephroprotection of carriers identified through family screening is one of the pillars for delaying severe kidney failure [1].

7.3. Stopping inappropriate treatments

Many patients with genetic glomerulopathy (Alport syndrome, monogenic FSGS) are initially considered to have "immunological" disease and receive immunosuppressants; the diagnosis allows these to be stopped [1]. The 2020 IPNA (International Pediatric Nephrology Association) recommendations, summarised by GeneReviews, advise avoiding ineffective prednisolone once SRNS is established, using RAASi to reduce proteinuria, stopping calcineurin inhibitors and other immunosuppressants when there is evidence of genetic SRNS (individual decision), treating hypertension, using diuretics, vitamin D and thyroid hormone with caution, and considering transplantation, including pre-emptive transplantation in certain diseases such as those related to WT1 [14]. In the 2023 cohort of Robert et al., the diagnosis avoided immunosuppression in about 10% of patients diagnosed [44, 50]. In INF2-related disease, there is no response to corticosteroids or immunosuppression [52].

7.4. Treatments conditioned by the molecular diagnosis

The PNDS lists the diseases in which the molecular diagnosis conditions access to targeted therapy: Fabry, primary hyperoxaluria type 1, atypical thrombotic microangiopathies and C3 glomerulopathies due to alternative pathway dysregulation, cblC defect, cystinosis and nephrotic syndrome due to COQ8B deficiency [1]. The table summarises the doses; full dosing, start criteria and monitoring follow the labels and guidelines for each disease.

7.5. Disease-modifying treatment in the most frequent diseases

In ADPKD, tolvaptan slowed the growth of kidney volume and the decline in function over 3 years, with a high discontinuation rate [54]. The FDA label for tolvaptan in ADPKD provides for initiation at 60 mg/day (45 mg on waking and 15 mg 8 hours later), titration to 60+30 and 90+30 mg/day at intervals of at least one week, and measurement of ALT, AST and bilirubin before and periodically because of the risk of hepatotoxicity, with adjustment for moderate CYP3A inhibitors [100]. Sodium restriction was beneficial [54]. In type IV collagen nephropathies, RAAS blockade is the disease-modifying treatment and its value depends on early initiation; in males with the X-linked form, ACE inhibitors extended the age at ESKD by a mean of 13 years, and male and female carriers should start ACE inhibitors early [48, 52]. In ADTKD-MUC1, a small molecule, BRD4780, cleared MUC1fs protein deposits in cells, organoids and a mouse model, with no clinical data [12].

7.6. Transplantation and living donation

The genetic diagnosis informs the risk of recurrence in the graft and the choice of donor [1]. Genetic forms of SRNS are corrected by transplantation, whereas about 70% of non-genetic SRNS recur rapidly [14]. ADTKD-MUC1 and ADTKD-UMOD do not recur in the graft, which makes these patients good candidates [12, 52]. In Alport syndrome, transplantation is the treatment of choice and 1–5% of males with the X-linked form develop post-transplant anti-GBM nephritis, with a higher risk when the variant eliminates the NC1 domain [48]. In primary hyperoxaluria and APRT deficiency, the disease recurs in the graft if untreated, and the xanthine oxidase inhibitor should be continued after transplantation in APRT deficiency [1, 18, 60]. Extended genetic analysis of transplant recipients may reveal genetic comorbidities that influence follow-up [1, 101]. The PNDS also mentions that pharmacogenetic variants such as those of CYP3A5 influence tacrolimus metabolism, with variable use across centres [1].

For the related living donor, the PNDS indicates testing of the CKDx recipient and, if necessary, of the donor, according to the Agence de la biomédecine living kidney donation recommendations of December 2023 [1, 77]. An asymptomatic relative may carry the familial variant (incomplete penetrance or late onset) and should be informed that the reduction in nephron mass may accelerate their own disease; absence of the variant makes donation safe with regard to that disease [1]. In dominant or X-linked forms of SRNS and in the WT1, COL4A3/4/5 and NPHS2 genes, molecular testing of the donor is mandatory, and carriers of a dominant or X-linked variant cannot donate, regardless of clinical status; heterozygotes for recessive diseases are asymptomatic and may donate [14]. In aHUS, counselling before related donation considers recurrence in the recipient and de novo disease in the donor [43]. A clinical practice document published in the American Journal of Transplantation in 2025 advises testing transplant candidates when the result changes management [102].

The donor's APOL1 genotype influences graft survival: in the USA, living donors with a high-risk genotype develop more advanced CKD and kidneys from high-risk deceased donors are associated with faster graft failure [1, 64]. In a cohort of 653 living donors (445 Black and 208 White; median follow-up 18.5 years), 15.3% of Black donors had a high-risk genotype, with a relative risk of eGFR below 45 of 2.31 (95% CI 1.16–4.61) [103]. The KDIGO living donor guideline suggests offering genotyping to candidates of sub-Saharan ancestry, the British Transplantation Society recommends offering it, funded by the public system, to all potential donors of African ancestry younger than 60, and a Delphi panel recommended asking about African ancestry rather than "race" and using the result in shared decision-making, never to refuse donation unilaterally [52, 64, 104].

7.7. APOL1-associated kidney disease: therapy in development

There is no approved specific therapy for APOL1-associated kidney disease [52, 64]. Inaxaplin (VX-147) is an oral small-molecule APOL1 channel inhibitor; AZD2373 is a subcutaneous antisense oligonucleotide that reduces APOL1 expression; and JAK/STAT pathway inhibition is another strategy under study [64]. These approaches rest on the toxic gain-of-function mechanism: reducing the amount of protein or blocking the channel should attenuate podocyte injury, without the need to correct the gene [52, 64].

The phase 2a trial had no control group and the reduction in proteinuria is an intermediate endpoint; efficacy on GFR and long-term safety depend on the results of AMPLITUDE [105, 106]. Inclusion in these trials requires the genotype, which illustrates access to trials conditioned by the diagnosis [1].

7.8. Nephroprotection trials and limits of the evidence

Nephroprotection trials included few patients with a genetic diagnosis, and extrapolation to each monogenic disease is an inference; relative reductions should be read with the absolute event rate of each population, and indirect comparison between trials with different populations is not legitimate [85]. The inaxaplin trial, without a control group, does not allow the effect on GFR to be estimated [105].

7.9. Regulatory status

INFARMED SmPCs and ANVISA package inserts were not among the sources consulted for this article.

7.10. Developments in the last 12 months

The HAS PNDS on the diagnostic approach to undetermined kidney disease in adults was published on 6 October 2026 (national guideline) [1]. The ERA consensus on CKDx was published in the December 2025 issue of Nephrology Dialysis Transplantation (expert consensus) [2]. The conclusions of the KDIGO conference on APOL1-associated kidney disease were published in November 2025 (consensus conference) [64]. In 2026, a living donor cohort study quantified the risk associated with the high-risk APOL1 genotype (observational study) [103], and a precision diagnosis study integrated the p.(Asn264Lys) modifier into the risk classification, known only by its title in the sources consulted [107]. The current FDA labels for iptacopan (July 2026 revision) and pegcetacoplan (August 2026 revision) include C3 glomerulopathy (regulatory labels) [95, 96]. The AMPLITUDE trial became active, not recruiting, with 522 participants and no published results (trial registry) [106]. The search performed did not identify other therapeutic trials targeting CKDx as a category, which does not prove that none exist.

7.11. Agents to avoid, pregnancy, perioperative care and transition

NSAIDs and iodinated contrast media should be avoided in CKD [1]; in ADTKD-REN, dehydration and NSAIDs precipitate acute kidney injury and hyperkalaemia [53], and in ADTKD-UMOD nephrotoxins and volume depletion should be avoided [11]. In HNF1B disease and in tubulopathies, introduction of ACE inhibitors/ARBs or sodium restriction may decompensate an underlying salt wasting [1]. SGLT2i are withheld at least 3 days before surgery or prolonged fasting [87]. In ADPKD, pregnancy is generally well tolerated, with a risk of pre-eclampsia [54]; pre-eclampsia may also reveal a collagen IV or complement disease [1]. In ADTKD-REN, high doses of fludrocortisone in pregnancy may cause hyperaldosteronism in the newborn [53]. In the cblC defect, nitrous oxide is contraindicated, which has anaesthetic implications [73]. The EKFC is valid from age 2 to 100 and ensures continuity during the transition from paediatric to adult medicine [1].

7.12. Organisation of care in France

Patients with a diagnosed rare disease are directed to the ORKiD network, formed by the reference centres (CRMR) MARHEA (Necker-Enfants Malades, Paris), MaReGE (Bron, Lyon) and SORARE (Toulouse) and by the competence centres (CCMR), which provide expertise, a multidisciplinary approach (ENT, ophthalmology, cardiology), inclusion in registries and cohorts, access to trials, therapeutic education programmes and social and psychological support [1]. Chronic kidney failure stages 3 to 5 (eGFR <60 mL/min) is recognised as a long-term condition (ALD 19), with 100% coverage [1]. The attending physician ensures nephroprotection and follows up the information given to relatives [1].

7.13. Divergences between guidelines

8. Reproductive Options and Family Counselling

8.1. Genetic counselling after diagnosis

ERKNet recommends genetic counselling for all patients with a diagnosis, with information on family planning, reproduction and kidney donation by a relative [43]. Return of results takes place in a dedicated consultation by the requesting physician, who has sole responsibility for communicating the result (article L1131-1-3 of the French Public Health Code), and includes the meaning of the result, the extrarenal manifestations to look for through retrophenotyping and identification of at-risk relatives in order to offer them nephrological screening and early nephroprotection [1]. The table summarises the distribution of roles provided for by the PNDS.

Source: HAS PNDS, Table 3 [1]. Prescription may be made by an authorised genetic counsellor, under Decree no. 2023-1426, and a non-geneticist prescriber works in liaison with a clinical genetics team [1].

8.2. Recurrence risks by mode of inheritance

In the family liaison form proposed by the PNDS, the relative's physician receives the gene, the mode of transmission and the a priori risk (50% or 25%) that the relative is a carrier, with the recommendation of a baseline nephrological assessment (BP, creatinine, PCR, urinary sediment) and a genetics or nephrology consultation to discuss testing targeted at the familial variant [1].

8.3. Cascade testing and informing relatives

Testing targeted at the familial variant is offered to parents, siblings and children, and identifying carriers before symptoms or at a very early stage allows nephroprotection to be started [1]. In France, the patient has a legal obligation to inform relatives of a confirmed molecular diagnosis when preventive or care measures exist (article L1131-1 of the Public Health Code), may do so directly or ask the physician to do so, and the PNDS provides template letters: a named letter, handed over by the patient with their agreement, and an anonymous one, sent by the physician, which reveals neither the identity of the index case nor the name of the disease and invites the relative to a counselling consultation [1]. Indirect information is less favourable, because relatives are less prepared to receive it [1]. At the end of the results consultation, the patient signs a certificate and receives the result with a form for relatives, and the physician checks whether the patient has been a gamete donor [1]. In genetic SRNS, GeneReviews recommends screening first-degree relatives with urinalysis for proteinuria without waiting for molecular confirmation [14]. In Alport syndrome, at-risk relatives and carriers should have annual surveillance by a nephrologist, and testing for the familial variant is the preferred method [48].

When the result is negative in CKDx, first-degree relatives should have a nephrological assessment with creatinine and eGFR, ACR, sediment and, depending on the context, renal ultrasound, repeated periodically in young people with a normal assessment [1]. The PNDS summary adds, for first-degree relatives, blood creatinine, bicarbonate and full electrolytes, urine protein/creatinine and testing for haematuria, and renal imaging, and referral of minors to a paediatric nephrologist [1].

8.4. Minors and asymptomatic relatives

A predictive test should not be performed in an asymptomatic minor without direct medical benefit during childhood; it is recommended when the result may modify surveillance, guide management or allow treatment from paediatric age, including early nephroprotection, with consent from those holding parental responsibility and a prior genetics consultation [1]. In an asymptomatic adult relative, the test is presymptomatic and is performed in a specialised clinic [1]. For APOL1, the KDIGO consensus considers that the genotype informs prognosis in children with FSGS or CKD, but testing of asymptomatic relatives follows the same principles of consent and qualified counselling [64].

8.5. Prenatal and preimplantation diagnosis

The PNDS recommends, for at-risk couples, a preconception consultation with a geneticist (ideally in liaison with a paediatric geneticist when the disease may manifest in childhood) or genetic counsellor, and referral to a multidisciplinary prenatal diagnosis centre (CPDPN) [1]. Prenatal diagnosis is performed by chorionic villus sampling (11–13 weeks of gestation) or amniocentesis (15–17 weeks), requires prior identification of the familial pathogenic or likely pathogenic variant and is considered only when the disease is of particular severity (article L2131-1) [1]. Preimplantation genetic diagnosis, a form of prenatal diagnosis within in vitro fertilisation (article L2131-4), is performed in France in five authorised centres, with waiting times that may exceed 12–18 months, and is subject to the conditions of access to assisted reproduction, including the woman's age [1]. Non-invasive prenatal diagnosis of monogenic diseases is limited to selected indications in specialised centres, and each request is assessed case by case by the CPDPN [1]. A VUS cannot serve as the basis for prenatal or preimplantation diagnosis [1]. The ERA consensus includes reproductive decisions among the reasons for testing CKDx [2].

8.6. Insurance, data and psychological impact

In France, insurers may not request genetic testing or the results of tests performed, even if the patient wishes to disclose them, and the patient does not have to declare predictive genetic results (pathogenic variant, predisposition, asymptomatic carrier status) on health questionnaires, although clinically documented diseases must still be declared (articles L.1141-1 and L.1141-2); the AERAS convention, revised in January 2022, facilitates access to credit insurance for people with an aggravated health risk [1]. Minimum data on patients followed in reference centres are recorded in the BaMaRa application and transmitted in pseudonymised form to the BNDMR, with a right to object, and anonymised data may be shared to improve diagnosis, in accordance with the GDPR and without transfer to commercial entities [1]. The result may bring relief, by explaining a disease attributed to poor adherence or poor control of risk factors, but also guilt towards children or relatives, and an inconclusive result may cause frustration; the PNDS recommends systematically offering psychological support [1].

9. Special Considerations: Population Diversity and Epidemiology

9.1. Frequency of CKDx

In France, nephropathy of unknown cause is the third cause of stage 5 CKD in the REIN registry, and the PNDS states that 15–25% of adults in kidney replacement therapy registries and CKD cohorts have disease of undetermined origin; the summary for the attending physician refers to 20–25% of patients with end-stage chronic kidney failure [1]. The ERA consensus estimates at least 16–20% of CKD, from registries and cohorts, with underestimation due to diagnoses of convenience [2]. The Dutch cohort of de Haan et al. starts from the assumption that at least 20% of CKD remains unexplained [6]. These values depend on the definition used by each registry, biopsy coverage and the period, and cannot be transposed directly between countries.

9.2. Genetic yield by country and population

The yield of adult cohorts varies with selection criteria: 9.3% in an unselected US CKD cohort, with 35.6% of patients of self-reported non-European ancestry [8]; 37% in Irish adults selected for genetic suspicion [42, 50]; 11.1% pathogenic and likely pathogenic variants in transplant candidates in Berlin [66]; 24–40% in the French CKDx cohorts with or without biopsy [4, 5, 44]; 17% in Dutch adults with onset before age 50 [6]; and 11.1% in Japanese haemodialysis patients who started dialysis between ages 20 and 49 [67]. The relative weight of each gene changes with the population: NPHP1 was the most frequent gene in the Dutch cohort (13 of 59 diagnoses), collagen IV dominated the French series of Robert et al. (46.7% of diagnoses) and, in Japan, ADPKD was the most frequent diagnosis [5, 6, 67]. In early "hypertensive nephropathy", the Mendelian yield was 30% in patients without African ancestry and 7% in those of African ancestry [16].

9.3. Frequencies of the main entities

The two estimates for ADTKD-UMOD in ESKD (about 0.2% and about 1%) come from different sources cited by the UMOD and MUC1 GeneReviews, and the difference reflects the uncertainty of the estimate and underdiagnosis [11, 12].

9.4. Founder populations and geographical variation

Primary hyperoxaluria type 1 accounts for 1–2% of children with ESKD in Western Europe and North America and for 10–17% in the Middle East, with a founder effect of the p.Ile244Thr variant in the Canary Islands and high frequency in Tunisia, Syria and the Druze population, associated with consanguinity [18]. In Fabry disease, newborn screening programmes found 1 in 7879 in northern Italy (mainly late-onset or unclassified forms), 1 in 6000 to 1 in 9000 males in Washington and Illinois, 1 in 2913 to 1 in 3277 in Missouri, 1 in 3000 to 1 in 4000 in Hungary, Austria and Spain and about 1 in 12,000 in Japan; the cardiac variant c.640-801G>A reaches about 1 in 1600 males in Taiwan, and targeted analysis is indicated in patients from Nova Scotia or of Chinese ancestry with an atypical presentation [17]. The NPHS2 variant p.(Arg229Gln) has a carrier frequency of 3% in the multi-ethnic population, 7% in Finland and 0.01% in East Asia [14]. The UMOD variant designated C744G may have a founder effect [11].

The APOL1 risk alleles follow the gradient of migrations from West Africa: they are frequent in West and Central African populations, in African Americans, in Afro-Caribbeans (Antilles, Guiana) and in people of African descent in South America, including Brazil and Colombia, and almost absent in populations of European, Asian and Native American origin [1]. The p.(Asn264Lys) modifier exists only on the G2 haplotype and in less than 1% of people of African ancestry [64]. The KDIGO consensus calls for race, ethnicity, ancestry and genotype to be distinguished, to avoid racialising the test [64].

9.5. Portugal and Brazil

The sources consulted do not include estimates of the frequency of CKDx or genetic diagnosis cohorts of adults with undetermined CKD in Portugal or Brazil. The only explicit reference is the PNDS's mention of people of African descent in Brazil among the populations carrying APOL1 risk alleles [1]. The absence of these data in the sources consulted does not mean that no published studies exist.

9.6. Special populations

In the elderly, eGFR should be interpreted in the light of age-specific reference values: at age 80, the 5th percentile lies at 45–50 mL/min/1.73 m², and a stable, isolated eGFR between 45 and 60 after age 65–70 reflects senescence and does not justify complex genetic investigation [1, 72]. At the same time, there is no upper age limit for monogenic disease, and testing remains indicated after age 60 when there is benefit for the patient or relatives [1, 3]. In patients on dialysis or transplant candidates, laboratory and urinary markers lose value, the history and old test results become the main source of phenotype, and the yield is high [1, 2]. In women, the obstetric history is a source of diagnostic information, and in the 2024 cohort of Robert et al. the yield was higher in women [1, 5]. In consanguineous or endogamous families, the probability of recessive disease increases [1].

9.7. European networks and patient organisations

The European reference network is ERKNet (European Rare Kidney Disease Reference Network), which maintains a European registry to which the PNDS refers [1, 14]. In France, the ORKiD network brings together the MARHEA (Paris), MaReGE (Bron) and SORARE (Toulouse) centres, the competence centres and the professional societies (Société Francophone de Néphrologie, Dialyse et Transplantation; Société de Néphrologie Pédiatrique), and publishes emergency cards, a directory and a list of trials [1]. The patient organisations listed by the PNDS are AIRG-France (all genetic kidney diseases), Association Polykystose France, APMF (Fabry), VML (lysosomal diseases), AMSN (nephrotic syndrome), VHL France, ASTB (tuberous sclerosis), ASL (Lowe syndrome), Association Bardet-Biedl France and France REIN, as well as the Maladies Rares Info Services helpline [1]. The Oxford Desk Reference cites Alport UK and the Alport Syndrome Foundation, and the SRNS GeneReviews cites NephCure and the NEPTUNE and PodoNet networks [14, 48]. Orphanet maintains disease-specific pages and the list of specialised centres [1].

10. Illustrative Clinical Case

10.1. ADTKD-MUC1 with negative panel and exome

The ERA consensus on CKDx describes, in its Table 2, six real patients initially classified as CKDx and later diagnosed through genetics [2]. One of them was a 41-year-old man with hypertension, eGFR of 50 mL/min/1.73 m², without proteinuria or haematuria, classified as CGA G3a A1 [2]. Ultrasound showed normal-sized kidneys with a thinner cortex and loss of corticomedullary differentiation, and the kidney biopsy had 40% globally sclerosed glomeruli, chronic interstitial changes with a mononuclear infiltrate and a normal GBM [2]. His father had reached kidney failure at age 51 and a paternal aunt at 58; the only extrarenal finding was mild hyperuricaemia (407 µmol/L) [2]. Using the notation proposed by the ERA, the patient moved from "CKDx h+g−" to "CKDx h+g+" as the investigation progressed [2]. The kidney panel and exome were negative; immunohistochemistry for the MUC1fs protein on the biopsy was positive, and targeted analysis of the MUC1 VNTR identified the variant NM_001204286.1:c.428dupC, ACMG class 5, establishing the diagnosis of ADTKD-MUC1 (OMIM 174000) [2, 35]. Retrophenotyping found neither gout nor other abnormalities [2]. The consensus does not describe the subsequent course or corresponding figures, and the case is presented in the source as a table, without an image [2].

The case brings together several elements of the CKDx diagnostic pathway. Father-to-son transmission, slowly progressive CKD with a bland sediment and mild hyperuricaemia make up the picture of CTIN without an evident cause that the PNDS associates with UMOD, MUC1, HNF1B and REN [1]. The biopsy was non-specific, as GeneReviews predicts, advising against diagnostic biopsy in ADTKD [12]. The negative panel and exome result did not exclude the disease, because MUC1 VNTR variants are not detected by these techniques, and the ERA recommends that the report of CKDx with an ADTKD-like presentation state whether the VNTR was analysed [2, 12]. The absence of gout is compatible with ADTKD-MUC1, in which gout is not primary [12]. The strategy followed, with targeted analysis after a negative exome, corresponds to the escalation recommended by the PNDS for repetitive regions [1]; detection from the exome data themselves with tools such as VNtyper is an alternative described in France [2, 79]. The practical consequences, not described in the report, would be informing relatives, targeted testing of at-risk relatives and candidacy for transplantation without risk of recurrence in the graft [1, 12].

10.2. Two complementary cases from the same series

In the same table, a 43-year-old man with hypertension, eGFR of 34 mL/min/1.73 m² and proteinuria of 1 g/day had normal-sized, hyperechoic kidneys, FSGS with a thinner GBM (203 ± 38 nm) on biopsy and a normal audiogram; his father had had kidney failure at age 50 and a brother at 7, with transplantation [2]. The FSGS panel identified MYH9 NM_002473.6:c.1270C>T p.(Arg424Trp), classified as class 4 by the authors, and retrophenotyping showed occasional mild thrombocytopenia (143–153 × 10⁹/L; the article writes 10⁶/L) and raised mean platelet volume (11.4 fL; normal <10.4), with a normal ophthalmological examination [2]. In ClinVar, this variant has an aggregate classification of uncertain significance, which illustrates the difference between a laboratory's classification in a family context and the aggregate classification of public databases [51, 61].

A 68-year-old woman being assessed for a second transplant, with CKD since age 37 and kidney failure at 49, had CTIN on biopsy (without EM), short stature (147 cm), deafness and insulin-dependent diabetes of sudden onset after the first transplant; her mother had hearing loss and a sister had kidney failure, insulin-dependent diabetes, neuropathy and myoclonus [2]. A panel including mitochondrial genes identified m.3271T>C in MT-TL1, with 5% heteroplasmy in blood, and retrophenotyping showed mildly raised lactate, microvascular leucopathy and a lactate peak on brain MR spectroscopy and mild cognitive impairment, with a diagnosis of MELAS [2]. The case shows the need to state in the report whether mitochondrial DNA was analysed [2].

11. Prognosis and Long-Term Follow-up

11.1. Prognostic value of the genetic diagnosis

Genetic diseases follow a more predictable natural history than many acquired nephropathies, and knowing the gene, and sometimes the variant type, allows kidney replacement therapy to be anticipated [1]. In ADPKD, the mean age at ESKD is 58.0 years with PKD1 and 74.8 with PKD2, and 55.6 years with truncating PKD1 variants versus 67.9 with non-truncating ones; the PNDS cites the PROPKD score (Predicting Renal Outcomes in ADPKD), which combines the genotype with clinical data [1, 10]. In males with X-linked Alport, the variant type predicts the age at ESKD (50% before age 20 with rearrangements, nonsense or frameshift variants; 50% by age 40 with missense variants) [9]. In ADTKD-MUC1, the median age at ESKD is 46 years; in ADTKD-UMOD, ESKD occurs between the third and seventh decades; in HNF1B disease, ESKD during follow-up was 51% with the 17q12 deletion and 78% with an intragenic variant; and in DNAJB11 disease, ESKD occurred between ages 55 and 89 (median 75) in 32 of 77 individuals [10–13]. In heterozygous COL4A3/COL4A4 carriers, the absolute risk of ESKD is low (well below 3%) in the population, but 28.7% of patients in a French clinical cohort with the dominant form reached ESKD [9, 50]; the difference between the two figures reflects the difference between unselected carriers and referred patients. In the high-risk APOL1 genotype, progression to ESKD is accelerated (HR 2.3) [1].

The diagnosis had a clinical consequence in 73% of diagnosed patients in the Dutch cohort and an implication considered important in 98% in the Tenon cohort [4, 6]. In the 2023 series of Robert et al., the diagnosis reclassified the disease in 27.3% of those diagnosed and led to family screening in 57% [44, 50].

11.2. Follow-up of the patient with a diagnosis

After diagnosis, follow-up follows the PNDS or specific recommendation for the disease, with a targeted search for the expected extrarenal manifestations, even if asymptomatic, which may avoid invasive tests [1]. Referral to a reference or competence centre ensures a multidisciplinary approach, access to trials, inclusion in registries and therapeutic education [1]. Nephroprotection shared with the attending physician (BP below 130/80 mmHg according to the PNDS, avoidance of NSAIDs and iodinated contrast, lifestyle) is maintained in all patients [1].

11.3. Follow-up without a diagnosis or with a VUS

A patient with a negative test and strong suspicion remains under regular nephrological follow-up with longitudinal phenotyping and updating of the pedigree, periodic re-analysis of the data and genetic re-evaluation in 2–3 years or upon phenotypic change, discussed at a nephrogenetics MDT [1]. A VUS is re-evaluated every 3–5 years or according to the laboratory in the PNDS, and perhaps every 2–3 years according to the ERA, or earlier when new clinical data arise [1, 2]. Re-evaluation is a responsibility shared between clinicians and laboratories [2].

11.4. Surveillance table

11.5. Patient on dialysis or transplant candidate

In the patient on dialysis, the genetic diagnosis retains value for transplantation, choice of donor, prediction of recurrence in the graft and family information [1]. In primary hyperoxaluria, about 10% of adults are diagnosed only after recurrence in the graft, often with graft loss, which justifies testing before transplantation [18]. In genetic forms of SRNS, transplantation corrects the disease [14].

12. Summary and Final Considerations

Undetermined kidney disease in adults, or CKDx, is a category of exclusion that encompasses 15–25% of adults in kidney replacement therapy registries and CKD cohorts and that, in France, is the third cause of stage 5 CKD [1]. Diabetes and hypertension may be comorbidities rather than the cause, and histological labels such as FSGS, CTIN or nephroangiosclerosis describe lesions, not aetiologies; the ERA proposes recording the depth of investigation with the "h/g" notation [1, 2]. Between 10 and 30% of these patients have a genetic cause, with a higher yield before age 40, with a family history, with tubulointerstitial involvement, atypical cysts or extrarenal signs, and with a non-specific biopsy [1, 44].

Recognition rests on a history that seeks old clues (birth weight, prematurity, routine examinations, pre-eclampsia), on a three-generation pedigree that unpicks the pitfalls of masked heritability, on an examination focused on subtle extrarenal signs and on laboratory tests that give weight to persistent haematuria, early gout with low FEUA, hypomagnesaemia, atypical diabetes, thrombotic microangiopathy and crystalluria [1]. eGFR should be read with age-specific reference values, so as not to confuse senescence with disease [72]. The most frequent causes in adult cohorts are type IV collagen nephropathies, cystic diseases, ADTKDs, HNF1B-related disease, nephronophthisis, podocytopathies and treatable phenocopies such as Fabry disease, aHUS and C3 glomerulopathy, the cblC defect and primary hyperoxaluria [1, 5, 6].

The PNDS recommends exome or genome as the first-line test, with a panel of more than 600 genes as an acceptable alternative, and the ERA sets a comprehensive kidney panel as the minimum; both insist on the technical limits (MUC1 VNTR, PKD1 pseudogenes, NPHP1 deletion, CFHR hybrids, mitochondrial DNA) and on the fact that a negative result does not exclude a genetic cause [1, 2]. Interpretation is shared between clinician and laboratory, requires concordance between variant and phenotype, uses retrophenotyping and segregation for VUS and prohibits irreversible decisions based on a VUS [1]. Testing is preceded by an information consultation and written consent, with an explicit choice regarding incidental findings, which in France are distinguished from secondary findings, which are not authorised [1].

The diagnosis changes management even without curative therapy: it allows targeted treatments to be started (enzyme replacement or migalastat in Fabry disease, lumasiran in PH1, complement inhibitors in aHUS and C3G, hydroxocobalamin and betaine in cblC, cysteamine in cystinosis, coenzyme Q10 in COQ8B deficiency), futile immunosuppressants to be stopped, transplantation and living donation to be planned, cascade screening with early nephroprotection to be organised and reproductive decisions to be informed [1]. APOL1-associated kidney disease, with G1, G2 and the protective modifier p.(Asn264Lys), requires a risk-based reading rather than a Mendelian diagnosis, in a context of ongoing trials with inaxaplin and AZD2373 [1, 64, 106]. Common nephroprotection, longitudinal follow-up and periodic re-evaluation of negative tests and VUS complete the pathway [1, 85].

13. References

1. Haute Autorité de Santé; Centres de référence MARHEA et MaReGE, filière ORKiD. Protocole National de Diagnostic et de Soins (PNDS): Approche diagnostique des néphropathies indéterminées de l'adulte (PNDS text and summary for the attending physician). Saint-Denis La Plaine: HAS; text of September 2026, published 6 Oct 2026. https://www.has-sante.fr/jcms/p_4543542
2. Halbritter J, Figueres L, Van Eerde AM, Capasso G, Hoorn EJ, Nijenhuis T, et al. Chronic Kidney Disease of unexplained cause (CKDx): a consensus statement by the Genes & Kidney Working Group of the ERA. Nephrol Dial Transplant. 2025;40(12):2390–2400. doi:10.1093/ndt/gfaf092
3. KDIGO Conference Participants (Köttgen A, Cornec-Le Gall E, Halbritter J, et al.). Genetics in chronic kidney disease: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference. Kidney Int. 2022;101(6):1126–1141. doi:10.1016/j.kint.2022.03.019
4. Doreille A, Lombardi Y, Dancer M, Lamri R, Testard Q, Vanhoye X, et al. Exome-first strategy in adult patients with CKD: a cohort study. Kidney Int Rep. 2023;8(3):596–605. doi:10.1016/j.ekir.2022.12.007
5. Robert T, Raymond L, Dancer M, Torrents J, Jourde-Chiche N, Burtey S, et al. Beyond the kidney biopsy: genomic approach to undetermined kidney diseases. Clin Kidney J. 2024;17(1):sfad099. doi:10.1093/ckj/sfad099
6. de Haan A, Eijgelsheim M, Vogt L, Hoorn EJ, Rotmans JI, Fortrie G, et al. Genetic testing in a national cohort of adults with chronic kidney disease of unknown origin. Nephrol Dial Transplant. 2025;40(6):1225–1233. doi:10.1093/ndt/gfae270
7. Gbadegesin RA, Ulasi I, Ajayi S, Raji Y, Olanrewaju T, Osafo C, et al. APOL1 bi- and monoallelic variants and chronic kidney disease in West Africans. N Engl J Med. 2025;392(3):228–238. doi:10.1056/NEJMoa2404211
8. Groopman EE, Marasa M, Cameron-Christie S, Petrovski S, Aggarwal VS, Milo-Rasouly H, et al. Diagnostic utility of exome sequencing for kidney disease. N Engl J Med. 2019;380(2):142–151. doi:10.1056/NEJMoa1806891
9. Nozu K, Yamamura T, Horinouchi T. Alport syndrome. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Updated 27 Feb 2025. https://www.ncbi.nlm.nih.gov/books/NBK1207/
10. Harris PC, Torres VE. Polycystic kidney disease, autosomal dominant. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Updated 29 Sep 2022. https://www.ncbi.nlm.nih.gov/books/NBK1246/
11. Bleyer AJ, Kidd K, Živná M, Kmoch S. Autosomal dominant tubulointerstitial kidney disease – UMOD. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Updated 8 Apr 2021. https://www.ncbi.nlm.nih.gov/books/NBK1356/
12. Bleyer AJ, Živná M, Kidd K, Kmoch S. Autosomal dominant tubulointerstitial kidney disease – MUC1. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Updated 21 Oct 2021. https://www.ncbi.nlm.nih.gov/books/NBK153723/
13. Mitchel MW, Moreno-De-Luca D, Myers SM, Levy RV, Turner S, Ledbetter DH, Martin CL. 17q12 recurrent deletion syndrome. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Updated 6 Feb 2025. https://www.ncbi.nlm.nih.gov/books/NBK401562/
14. Lipska-Ziętkiewicz BS. Genetic steroid-resistant nephrotic syndrome overview. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Published 26 ago 2021. https://www.ncbi.nlm.nih.gov/books/NBK573219/
15. Stokman M, Lilien M, Knoers N. Nephronophthisis-related ciliopathies. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Updated 2 Mar 2023. https://www.ncbi.nlm.nih.gov/books/NBK368475/
16. Serre J, Doreille A, Raymond L, Suc G, Bobot M, Dancer M, et al. Diagnostic yield of exome sequencing in early-onset hypertensive nephropathy in adults. Am J Nephrol. 2024;55(4):468–471. doi:10.1159/000538173
17. Mehta A, Hughes DA. Fabry disease. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Updated 27 Jan 2022. https://www.ncbi.nlm.nih.gov/books/NBK1292/
18. Milliner DS, Harris PC, Sas DJ, Cogal AG, Lieske JC. Primary hyperoxaluria type 1. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Updated 15 Aug 2024. https://www.ncbi.nlm.nih.gov/books/NBK1283/
19. Noris M, Bresin E, Mele C, Remuzzi G. Genetic atypical hemolytic-uremic syndrome. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Updated 23 Sep 2021. https://www.ncbi.nlm.nih.gov/books/NBK1367/
20. Orphanet. Alport syndrome. ORPHA:63 [Internet]. https://www.orpha.net/en/disease/detail/63
21. Orphanet. Autosomal dominant polycystic kidney disease. ORPHA:730 [Internet]. https://www.orpha.net/en/disease/detail/730
22. Orphanet. Autosomal recessive polycystic kidney disease. ORPHA:731 [Internet]. https://www.orpha.net/en/disease/detail/731
23. Orphanet. Autosomal dominant tubulointerstitial kidney disease. ORPHA:34149 [Internet]. https://www.orpha.net/en/disease/detail/34149
24. Orphanet. Hereditary steroid-resistant nephrotic syndrome. ORPHA:656 [Internet]. https://www.orpha.net/en/disease/detail/656
25. Orphanet. Nephronophthisis. ORPHA:655 [Internet]. https://www.orpha.net/en/disease/detail/655
26. Orphanet. Fabry disease. ORPHA:324 [Internet]. https://www.orpha.net/en/disease/detail/324
27. Orphanet. C3 glomerulopathy. ORPHA:329918 [Internet]. https://www.orpha.net/en/disease/detail/329918
28. Orphanet. Primary hyperoxaluria. ORPHA:416 [Internet]. https://www.orpha.net/en/disease/detail/416
29. Orphanet. Cystinosis. ORPHA:213 [Internet]. https://www.orpha.net/en/disease/detail/213
30. Orphanet. Dent disease. ORPHA:1652 [Internet]. https://www.orpha.net/en/disease/detail/1652
31. Online Mendelian Inheritance in Man. #301050 Alport syndrome 1, X-linked (ATS1) [Internet]. https://www.omim.org/entry/301050
32. Online Mendelian Inheritance in Man. #173900 Polycystic kidney disease 1 with or without polycystic liver disease [Internet]. Baltimore: Johns Hopkins University. https://www.omim.org/entry/173900
33. Online Mendelian Inheritance in Man. #162000 Tubulointerstitial kidney disease, autosomal dominant, 1 (ADTKD1) [Internet]. https://www.omim.org/entry/162000
34. Online Mendelian Inheritance in Man. #613092 Tubulointerstitial kidney disease, autosomal dominant, 4 (ADTKD4) [Internet]. https://www.omim.org/entry/613092
35. Online Mendelian Inheritance in Man. #174000 Tubulointerstitial kidney disease, autosomal dominant, 2 (ADTKD2) [Internet]. https://www.omim.org/entry/174000
36. Online Mendelian Inheritance in Man. #617056 Tubulointerstitial kidney disease, autosomal dominant, 5 (ADTKD5) [Internet]. https://www.omim.org/entry/617056
37. Online Mendelian Inheritance in Man. #137920 Renal cysts and diabetes syndrome (RCAD) [Internet]. https://www.omim.org/entry/137920
38. Online Mendelian Inheritance in Man. #600995 Nephrotic syndrome, type 2 (NPHS2) [Internet]. https://www.omim.org/entry/600995
39. Genovese G, Friedman DJ, Ross MD, Lecordier L, Uzureau P, Freedman BI, et al. Association of trypanolytic ApoL1 variants with kidney disease in African Americans. Science. 2010;329(5993):841–845. doi:10.1126/science.1193032
40. Online Mendelian Inheritance in Man. #612551 Focal segmental glomerulosclerosis 4, susceptibility to (FSGS4) [Internet]. https://www.omim.org/entry/612551
41. Online Mendelian Inheritance in Man. #301500 Fabry disease [Internet]. https://www.omim.org/entry/301500
42. Connaughton DM, Kennedy C, Shril S, Mann N, Murray SL, Williams PA, et al. Monogenic causes of chronic kidney disease in adults. Kidney Int. 2019;95(4):914–928. doi:10.1016/j.kint.2018.10.031
43. Knoers N, Antignac C, Bergmann C, Dahan K, Giglio S, Heidet L, et al. Genetic testing in the diagnosis of chronic kidney disease: recommendations for clinical practice. Nephrol Dial Transplant. 2022;37(2):239–254. doi:10.1093/ndt/gfab218
44. Robert T, Greillier S, Torrents J, Raymond L, Dancer M, Jourde-Chiche N, et al. Diagnosis of kidney diseases of unknown etiology through biopsy-genetic analysis. Kidney Int Rep. 2023;8(10):2077–2087. doi:10.1016/j.ekir.2023.07.003
45. Eckardt KU, Alper SL, Antignac C, Bleyer AJ, Chauveau D, Dahan K, et al. Autosomal dominant tubulointerstitial kidney disease: diagnosis, classification, and management — a KDIGO consensus report. Kidney Int. 2015;88(4):676–683. doi:10.1038/ki.2015.28
46. Kashtan CE, Ding J, Garosi G, Heidet L, Massella L, Nakanishi K, et al. Alport syndrome: a unified classification of genetic disorders of collagen IV α345: a position paper of the Alport Syndrome Classification Working Group. Kidney Int. 2018;93(5):1045–1051. doi:10.1016/j.kint.2017.12.018
47. Savige J, Lipska-Zietkiewicz BS, Watson E, Hertz JM, Deltas C, Mari F, et al. Guidelines for genetic testing and management of Alport syndrome. Clin J Am Soc Nephrol. 2022;17(1):143–154. doi:10.2215/CJN.04230321
48. Firth HV, Hurst JA. Alport syndrome. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 340–343.
49. HUGO Gene Nomenclature Committee. MANE Select transcripts of the genes cited [Internet]. https://www.genenames.org/
50. Centres de référence MARHEA et MaReGE, filière ORKiD. Argumentaire scientifique — PNDS Approche diagnostique des néphropathies indéterminées de l'adulte. Saint-Denis La Plaine: HAS; September 2026. https://www.has-sante.fr/jcms/p_4543542
51. VariantValidator. Validation of NM_015506.3:c.566G>A, NM_181697.3:c.515-1G>T, NM_002473.6:c.1270C>T and NM_003661.4:c.792C>A (GRCh38) [Internet]. https://variantvalidator.org/
52. GeneGuidance. Records for the genes PKD1, PKD2, UMOD, MUC1, REN, HNF1B, SEC61A1, COL4A3, COL4A4, COL4A5, NPHS1, NPHS2, APOL1, GLA, INF2, TRPC6 and DNAJB11. Local version of 2026-09-26.
53. Živná M, Kidd K, Kmoch S, Bleyer AJ. Autosomal dominant tubulointerstitial kidney disease – REN. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews® [Internet]. Seattle (WA): University of Washington, Seattle; 1993–2026. Updated 10 Dec 2020. https://www.ncbi.nlm.nih.gov/books/NBK53700/
54. Firth HV, Hurst JA. Autosomal dominant polycystic kidney disease (ADPKD). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 354–357.
55. Senum SR, Li YSM, Benson KA, Joli G, Olinger E, Lavu S, et al. Monoallelic IFT140 pathogenic variants are an important cause of the autosomal dominant polycystic kidney-spectrum phenotype. Am J Hum Genet. 2022;109(1):136–156. doi:10.1016/j.ajhg.2021.11.016
56. Huynh VT, Audrézet MP, Sayer JA, Ong AC, Lefevre S, Le Brun V, et al. Clinical spectrum, prognosis and estimated prevalence of DNAJB11-kidney disease. Kidney Int. 2020;98(2):476–487. doi:10.1016/j.kint.2020.02.022
57. Snoek R, van Setten J, Keating BJ, Israni AK, Jacobson PA, Oetting WS, et al. NPHP1 (nephrocystin-1) gene deletions cause adult-onset ESRD. J Am Soc Nephrol. 2018;29(6):1772–1779. doi:10.1681/ASN.2017111200
58. Firth HV, Hurst JA. Urinary tract and renal anomalies (CAKUT). In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 796–799.
59. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 1, Clinical approach to inborn errors of metabolism in paediatrics, section 1.6.9 (Nephrology).
60. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 32, section 32.2 (Diseases with predominant kidney stones or kidney involvement).
61. National Center for Biotechnology Information. ClinVar: NM_002473.6(MYH9):c.1270C>T (VCV000871830); NM_003661.4(APOL1):c.792C>A (VCV001170737); APOL1 haplotypes G1/G2 (VCV001120175) [Internet]. https://www.ncbi.nlm.nih.gov/clinvar/
62. Kopp JB, Nelson GW, Sampath K, Johnson RC, Genovese G, An P, et al. APOL1 genetic variants in focal segmental glomerulosclerosis and HIV-associated nephropathy. J Am Soc Nephrol. 2011;22(11):2129–2137. doi:10.1681/ASN.2011040388
63. Gupta Y, Friedman DJ, McNulty MT, Khan A, Lane B, Wang C, et al. Strong protective effect of the APOL1 p.N264K variant against G2-associated focal segmental glomerulosclerosis and kidney disease. Nat Commun. 2023;14(1):7836. doi:10.1038/s41467-023-43020-9
64. Ojo AO, Adu D, Bramham K, Freedman BI, Gbadegesin RA, Ilori TO, et al. APOL1 kidney disease: conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO) Controversies Conference. Kidney Int. 2025;108(5):763–779. doi:10.1016/j.kint.2025.05.017
65. Schmidt RJ, Steeves M, Bayrak-Toydemir P, Benson KA, Coe BP, Conlin LK, et al. Recommendations for risk allele evidence curation, classification, and reporting from the ClinGen Low Penetrance/Risk Allele Working Group. Genet Med. 2024;26(3):101036. doi:10.1016/j.gim.2023.101036
66. Schrezenmeier E, Kremerskothen E, Halleck F, Staeck O, Liefeldt L, Choi M, et al. The underestimated burden of monogenic kidney disease in adults waitlisted for kidney transplantation. Genet Med. 2021;23(7):1219–1224. doi:10.1038/s41436-021-01127-8
67. Fujimaru T, Mori T, Chiga M, Mandai S, Kikuchi H, Ando F, et al. Genetic diagnosis of adult hemodialysis patients with unknown etiology. Kidney Int Rep. 2024;9(4):994–1004. doi:10.1016/j.ekir.2024.01.027
68. Bensouna I, Doreille A, Dancer M, Lebre AS, Robert T, Mesnard L. Nephrogenomics, precision medicine and the role of genetic testing in adult kidney disease management. Nat Rev Nephrol. 2025;21(9):597–612. doi:10.1038/s41581-025-00970-1
69. Olinger E, Hofmann P, Kidd K, Dufour I, Belge H, Schaeffer C, et al. Clinical and genetic spectra of autosomal dominant tubulointerstitial kidney disease due to mutations in UMOD and MUC1. Kidney Int. 2020;98(3):717–731. doi:10.1016/j.kint.2020.04.038
70. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Fabry disease. In: Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. p. 655–661.
71. Cornec-Le Gall E, van Eerde AM, Figueres L, Simons M, Capasso G, Perez Gomez MV, et al. 10 tips on how to take a proper family history in CKD patient care. Clin Kidney J. 2025;18(9):sfaf253. doi:10.1093/ckj/sfaf253
72. Astley ME, Chesnaye NC, Hallan S, Gambaro G, Ortiz A, Carrero JJ, et al. Age- and sex-specific reference values of estimated glomerular filtration rate for European adults. Kidney Int. 2025;107(6):1076–1087. doi:10.1016/j.kint.2025.02.025
73. Huemer M, Diodato D, Schwahn B, Schiff M, Bandeira A, Benoist JF, et al. Guidelines for diagnosis and management of the cobalamin-related remethylation disorders cblC, cblD, cblE, cblF, cblG, cblJ and MTHFR deficiency. J Inherit Metab Dis. 2017;40(1):21–48. doi:10.1007/s10545-016-9991-4
74. Pottel H, Björk J, Courbebaisse M, Couzi L, Ebert N, Eriksen BO, et al. Development and validation of a modified full age spectrum creatinine-based equation to estimate glomerular filtration rate: a cross-sectional analysis of pooled data. Ann Intern Med. 2021;174(2):183–191. doi:10.7326/M20-4366
75. Inker LA, Eneanya ND, Coresh J, Tighiouart H, Wang D, Sang Y, et al. New creatinine- and cystatin C-based equations to estimate GFR without race. N Engl J Med. 2021;385(19):1737–1749. doi:10.1056/NEJMoa2102953
76. Pei Y, Obaji J, Dupuis A, Paterson AD, Magistroni R, Dicks E, et al. Unified criteria for ultrasonographic diagnosis of ADPKD. J Am Soc Nephrol. 2009;20(1):205–212. doi:10.1681/ASN.2008050507
77. Agence de la biomédecine. Recommandations d'aide à la pratique clinique pour le don de rein du vivant. Saint-Denis La Plaine: ABM; December 2023. https://www.agence-biomedecine.fr/IMG/pdf/recommandations_dv_rein_argumentaire_vfinale_14-12-2023.pdf
78. Testard Q, Vanhoye X, Yauy K, Naud ME, Vieville G, Rousseau F, et al. Exome sequencing as a first-tier test for copy number variant detection: retrospective evaluation and prospective screening in 2418 cases. J Med Genet. 2022;59(12):1234–1240. doi:10.1136/jmg-2022-108439
79. Bensouna I, Robert T, Vanhoye X, Dancer M, Raymond L, Delaugère P, et al. Systematic screening of autosomal dominant tubulointerstitial kidney disease-MUC1 27dupC pathogenic variant through exome sequencing. J Am Soc Nephrol. 2025;36(2):256–263. doi:10.1681/ASN.0000000503
80. Miller DT, Lee K, Abul-Husn NS, Amendola LM, Brothers K, Chung WK, et al. ACMG SF v3.2 list for reporting of secondary findings in clinical exome and genome sequencing: a policy statement of the American College of Medical Genetics and Genomics (ACMG). Genet Med. 2023;25(8):100866. doi:10.1016/j.gim.2023.100866
81. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405–424. doi:10.1038/gim.2015.30
82. Réseau NGS-Diag; Association Nationale des Praticiens de Génétique Moléculaire. Recommandations de bonnes pratiques: interprétation des variants issus du séquençage NGS. BP-NGSDiag 001 v2. https://anpgm.fr/media/documents/BP-NGSDiag_001_Interpretation_Variants_v2.pdf
83. Firth HV, Hurst JA. Ciliopathies. In: Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 366–369.
84. Ribeiro EM, coord. Doenças de adultos. In: Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 163–167.
85. Awdishu L, Maxson R, Gratt C, Rubenzik T, Battistella M. KDIGO 2024 clinical practice guideline on evaluation and management of chronic kidney disease: a primer on what pharmacists need to know. Am J Health Syst Pharm. 2025;82(12):660–671. doi:10.1093/ajhp/zxaf044
86. U.S. Food and Drug Administration. COZAAR (losartan potassium) tablets — prescribing information. NDA 020386. Revised 20 Jan 2026. https://www.accessdata.fda.gov/scripts/cder/daf/
87. U.S. Food and Drug Administration. FARXIGA (dapagliflozin) tablets — prescribing information. NDA 202293. Revised 10 Jul 2026. https://www.accessdata.fda.gov/scripts/cder/daf/
88. U.S. Food and Drug Administration. JARDIANCE (empagliflozin) tablets — prescribing information. NDA 204629. openFDA record of 2023. https://www.accessdata.fda.gov/scripts/cder/daf/
89. Ribeiro EM, coord. Doença de Fabry. In: Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 113–115.
90. U.S. Food and Drug Administration. FABRAZYME (agalsidase beta) for injection — prescribing information. BLA 103979. Revised 22 Aug 2025. https://www.accessdata.fda.gov/scripts/cder/daf/
91. U.S. Food and Drug Administration. GALAFOLD (migalastat) capsules — prescribing information. NDA 208623. Revised 28 Aug 2025. https://www.accessdata.fda.gov/scripts/cder/daf/
92. U.S. Food and Drug Administration. OXLUMO (lumasiran) injection — prescribing information. NDA 214103. Revised 1 Dec 2025. https://www.accessdata.fda.gov/scripts/cder/daf/
93. U.S. Food and Drug Administration. SOLIRIS (eculizumab) injection — prescribing information. BLA 125166. Revised 8 Jul 2026. https://www.accessdata.fda.gov/scripts/cder/daf/
94. U.S. Food and Drug Administration. ULTOMIRIS (ravulizumab-cwvz) injection — prescribing information. BLA 761108. Revised 9 Oct 2025. https://www.accessdata.fda.gov/scripts/cder/daf/
95. U.S. Food and Drug Administration. FABHALTA (iptacopan) capsules — prescribing information. NDA 218276. Revised 16 Jul 2026. https://www.accessdata.fda.gov/scripts/cder/daf/
96. U.S. Food and Drug Administration. EMPAVELI (pegcetacoplan) injection — prescribing information. NDA 215014. Revised 21 Aug 2026. https://www.accessdata.fda.gov/scripts/cder/daf/
97. U.S. Food and Drug Administration. PROCYSBI (cysteamine bitartrate) delayed-release — prescribing information. NDA 203389/213491. Revised 25 Jun 2026. https://www.accessdata.fda.gov/scripts/cder/daf/
98. Drovandi S, Lipska-Ziętkiewicz BS, Ozaltin F, Emma F, Gulhan B, Boyer O, et al. Oral coenzyme Q10 supplementation leads to better preservation of kidney function in steroid-resistant nephrotic syndrome due to primary coenzyme Q10 deficiency. Kidney Int. 2022;102(3):604–612. doi:10.1016/j.kint.2022.04.029
99. U.S. Food and Drug Administration. Allopurinol tablets — prescribing information. ANDA 075798. Revised 8 Jul 2024. https://www.accessdata.fda.gov/scripts/cder/daf/
100. U.S. Food and Drug Administration. Tolvaptan tablets (ADPKD indication) — prescribing information. ANDA 216933. Revised 9 Mar 2025. https://www.accessdata.fda.gov/scripts/cder/daf/
101. Mann N, Braun DA, Amann K, Tan W, Shril S, Connaughton DM, et al. Whole-exome sequencing enables a precision medicine approach for kidney transplant recipients. J Am Soc Nephrol. 2019;30(2):201–215. doi:10.1681/ASN.2018060575
102. Ames EG, Anand PM, Bekheirnia MR, Doshi MD, El Ters M, Freese ME, et al. Evaluation for genetic disease in kidney transplant candidates: a practice resource. Am J Transplant. 2025;25(2):237–249. doi:10.1016/j.ajt.2024.10.019
103. Hsu CY, Gao Y, Freedman BI, Lunn MR, Muiru AN, Schnitzler MA, et al. Apolipoprotein L1 gene genotype and kidney outcomes after living kidney donation. JAMA Intern Med. 2026;186(8):943–950. doi:10.1001/jamainternmed.2026.0996
104. McIntosh T, Walsh H, Baldwin K, Iltis A, Mohan S, Sawinski D, et al. Evaluating ApoL1 genetic testing policy options for transplant centers: a Delphi consensus panel project with stakeholders. Clin J Am Soc Nephrol. 2024;19(4):494–502. doi:10.2215/CJN.0000000000000397
105. Egbuna O, Zimmerman B, Manos G, Fortier A, Chirieac MC, Dakin LA, et al. Inaxaplin for proteinuric kidney disease in persons with two APOL1 variants. N Engl J Med. 2023;388(11):969–979. doi:10.1056/NEJMoa2202396
106. ClinicalTrials.gov. NCT05312879 — Study of inaxaplin (VX-147) in participants with APOL1-mediated proteinuric kidney disease (AMPLITUDE) [Internet]. Updated 26 Oct 2026. https://clinicaltrials.gov/study/NCT05312879
107. Martinelli E, Ke J, Khan A, Wongboonsin J, Vanderwall DR, Lim TY, et al. Precision diagnosis in APOL1 kidney disease with the p.N264K M1 protective variant. JAMA Netw Open. 2026;9(3):e261452. doi:10.1001/jamanetworkopen.2026.1452
108. Devuyst O, Olinger E, Weber S, Eckardt KU, Kmoch S, Rampoldi L, et al. Autosomal dominant tubulointerstitial kidney disease. Nat Rev Dis Primers. 2019;5(1):60. doi:10.1038/s41572-019-0109-9