Diagnosis, management and treatment of the Alport syndrome - 2024 guideline on behalf of ERKNet, ERA and ESPN

INTRODUCTION

The focus of this guideline is to describe the current state-of-art in the diagnosis, management, and treatment of the disorders resulting from the genetic defects in COL4A3/4/5 genes including the classical Alport syndrome (AS) (ORPHA 63). The document is a joint initiative of ERKNet, ERA Genes&Kidney, and ESPN Inherited renal disorders working group along with patient advocacy groups. There is an ongoing discussion on better defining the name for this entity [1]. The key clinical manifestation is persistent haematuria defined as present in two or more samples over a 6- to 12-month period. The classical triad of symptoms comprises familial progressive kidney disease with sensorineural hearing loss and ocular defects, although the penetrance and expressivity of these phenotypes is variable from isolated haematuria to early kidney failure (KF). Kidney biopsies may show various light microscopic abnormalities including normal appearances, focal segmental glomerulosclerosis, tubulointerstitial atrophy, and even IgA deposits mimicking IgA nephropathy; however, electron microscopic analysis typically reveals abnormalities, splitting, or thickening of the glomerular basement membrane (GBM) [2].

MATERIALS AND METHODS

Eight working groups (WG) were established to examine different clinical aspects of AS (Definition/Epidemiology, Prognosis, Management, Management Ear, Management Eye, Treatment, Transplantation). The WGs defined a total of 21 PICO (Patient or Population covered, Intervention, Comparator, Outcome) research questions. A systematic literature search was conducted in PubMed based on relevant key words. Electronic records of the references retrieved by searches were stored using a reference management web application (Rayyan) [3]. For the initial screening process, at least two members of the writing committee scanned titles and abstracts to exclude irrelevant publications. Original articles for selected references were retrieved, and full-text screening was processed.

Initial statements were elaborated and discussed by each WG. Evidence grading was done according to the methodology of the American Academy of Pediatrics [4]. External voting groups (ERKNet WG for hereditary glomerulopathies, ESPN Inherited renal disorders working group, ERA Genes&Kidney, international experts, and Alport Alliance) were asked to provide the level of agreement via a Delphi survey. A minimum of 70% consensus was required for the final adoption of the recommendations.

Q1: XLAS AND ARAS: WHAT ARE THE DIFFERENCES AMONG THEM AND WHAT IS THEIR PREVALENCE?

1-Statement: XLAS is much more prevalent than ARAS and affects males and females differently. (Grade A, strong)

AS (ORPHA 63) can be subdivided according to either the observed inheritance pattern or the underlying molecular defect into X-linked Alport syndrome (XLAS; ORPHA 88917), which is linked with pathogenic variants affecting the COL4A5 gene on the X chromosome, and autosomal recessive Alport syndrome (ARAS; ORPHA 88919), linked with biallelic pathogenic variants affecting COL4A3 or COL4A4, which are situated next to each other on chromosome 2 [5]. In exceptional cases the defect in COL4A5 also affects part of the adjacent COL4A6 gene leading to additional development of smooth muscle tumours (ORPHA 1018) [6].

AS is the second most common Mendelian cause of kidney disease. Estimates of the combined phenotype-based prevalence of XLAS and ARAS in historical literature vary, ranging from 1:5000 in Utah [7] to 1:17 000 in Sweden [8] and 1:53 000 in Finland [9], with the recessive form accounting for around 15% of families. More recently, analysis of population-based genome sequencing data in GnomAD showed that 1:2320 individuals harbour a predicted pathogenic COL4A5 variant with the proportion being higher in those with European ancestry (1:1800) compared to those with East Asian (1:2310) or Ashkenazi Jewish (1:4961) ancestry [10]. The population frequency of a predicted pathogenic variant in COL4A3 or in COL4A4 was 0.41% and 0.42%, respectively. Homozygous pathogenic COL4A3 or COL4A4 are reported in GnomAD but it is not possible to infer compound heterozygosity in this dataset. However, calculation of the expected frequency of biallelic disease, inferred from the observed frequency of heterozygous pathogenic variants in either of these genes, was 1:88 866 under the assumption of independent occurrence of variants. This may underestimate the true frequency of autosomal recessive disease given the presence of founder pathogenic variants in certain populations [e.g. COL4A4: c.2906C>G; p.(Ser969Ter) in the UK] and assortative mating that is prevalent across human populations [11].

However, since not all individuals with qualifying variants in these population studies exhibit haematuria or other clinical evidence of AS, this may lead to an overestimation of the disease prevalence, which would need future studies addressing penetrance (see Q11).

Q2: WHAT ARE THE PENETRANCE AND PREVALENCE OF HETEROZYGOUS COL4A3/COL4A4 SINGLE PATHOGENIC VARIANTS?

2-Recommendation: Single pathogenic variants in COL4A3/COL4A4 are relatively frequent in the general population and therefore their presence should not be regarded as the only cause of the kidney disease and other/additional diagnoses (e.g. IgA nephropathy) should be considered. (Grade B, strong)

The presence of one pathogenic or likely pathogenic (P/LP) allele of COL4A3 or COL4A4 is relatively common in the general population, occurring in 0.94% (or 1:106) of individuals in GnomAD, with the frequency ranging from 0.24% in Finnish to 1.58% in Latino populations [10]. While the presence of such a variant in heterozygosity can be regarded as a risk factor for having offspring with ARAS (which requires inheritance of the P/LP allele along with a defect in the other allele), the term ‘ARAS carrier’ is generally avoided to describe such individuals. This is because the presence of a monoallelic P/LP COL4A3 or COL4A4 variant is itself sufficient to cause clinical manifestations, with haematuria and microalbuminuria strongly associated in population-based genome sequencing studies [10, 12, 13]. The lifetime risk of clinically significant kidney impairment or KF attributable to heterozygosity for such variants remains less clear. While several studies have identified multigenerational families with clinically significant phenotypes such as focal segmental glomerulosclerosis, chronic kidney disease (CKD), and KF segregating with monoallelic P/LP COL4A3 or COL4A4 variants, and sequencing studies in unselected kidney disease cohorts recognize these variants as important contributors to kidney disease in the population [14, 15], the absolute risk of KF due to these genetic variants remains poorly understood. It is likely much lower than the 15% risk previously estimated from older family-based studies, which were likely influenced by ascertainment and reporting biases [16, 17]. Given that the prevalence of monoallelic P/LP COL4A3 or COL4A4 variants in the general population is close to 1%, the absolute risk of KF attributable to the presence of such a variant must be far smaller, with more recent estimates putting it below 3% [18]. Where kidney biopsies are performed in individuals with a monoallelic P/LP COL4A3 or COL4A4 variant, various light microscopic abnormalities have been reported, including normal appearances, minimal change disease, mesangioproliferative nephritis, membranous nephritis, focal segmental glomerulosclerosis, tubulointerstitial nephritis and atrophy, and even IgA deposits [19, 20]. However, electron microscopy typically reveals thinning of the GBM in most individuals. Other abnormalities, such as irregularities, areas of thickening, lamination, basket-weaving, and wrinkling, may also be present in some cases [16].

Due to the various ways of identifying patients with heterozygous P/LP COL4A3 or COL4A4 variants, several diagnostic labels are (or have been) used, including ‘thin basement membrane nephropathy’, ‘autosomal dominant Alport syndrome’, ‘Type IV collagen-associated kidney disease’, ‘Alport spectrum nephropathy’ and, historically, ‘benign familial haematuria’. There is a strong consensus among medical professionals that the term ‘benign’ should no longer be used. This decision stems from the understanding that, although the absolute risk of KF is modest, it is significantly higher than that of the general population [2, 16, 21]. Despite the current uncertainty over nomenclature there is an emerging appreciation of the high frequency, and hence clinical and epidemiological importance of heterozygous P/LP COL4A3 or COL4A4 variant-associated disease [22].

Q3: IS THE TERM ‘DIGENIC’ JUSTIFIED IN AS?

3-Statement: The presence of pathogenic variants in more than one of COL4A3/4/5, sometimes called ‘digenic’, may lead to an age at onset of KF that is intermediate between that reported for XLAS or ARAS and monoallelic autosomal disease. (Grade D, expert opinon)

The term ‘digenic’ AS refers to the situation in which there are P/LP changes in two of the COL4A3, COL4A4, or COL4A5 genes in an individual. This can be subdivided into (i) a change in COL4A5 plus a change in one of either COL4A3 or COL4A4, and (ii) a change in one copy of COL4A3 and a change in one copy of COL4A4 (which may be in cis or in trans) in an individual. In the latter situation the phase of the variants affects the risk to the offspring of inheriting digenic or monoallelic disease [23, 24].

It is likely that the severity of AS (measured by the median age at onset of KF) can be increased by digenic disease if the proportion of normal type IV collagen trimer is reduced compared to the monoallelic setting. For instance, in males with a single defective COL4A5 allele all relevant trimers will be abnormal, so the presence of an additional COL4A3 or COL4A4 variant may not be clinically apparent. This is consistent with the lack of earlier onset of KF in 16 male individuals reported in the largest series to date [23]. In females with heterozygous pathogenic changes in both COL4A5 and one of COL4A3 or COL4A4, there will be a reduction in a fraction of normal trimers compared with females with a faulty copy of only COL4A5, so severity of disease is predicted to be greater. While current empirical evidence is weak, showing only a trend towards earlier onset proteinuria in 26 females with digenic disease in the largest series to date [23], future analyses with larger sample sizes may confirm this logical hypothesis.

Individuals of either sex with heterozygous pathogenic alleles of both COL4A3 and COL4A4 (autosomal digenic disease) are predicted to produce normal trimeric type IV collagen chains (i.e. that incorporate only wild-type alleles) in only 25% of the time, resulting in more severe disease than individuals with a single pathogenic variant in either COL4A3 or COL4A4, where 50% of the trimers are predicted to be normal. Determining whether the variants are in cis or in trans is important for genetic counselling as the risk to offspring is different for each situation. Analysis of a cohort of 32 individuals with autosomal digenic disease revealed a median age at onset of KF that was intermediate between that of ARAS [23, 25], and monoallelic disease [17]. Given that the frequency of heterozygous pathogenic variants in COL4A3 or COL4A4 is approximately 1%, this is the proportion of individuals with XLAS who are predicted to harbour an additional autosomal variant, assuming independent inheritance of variants in the relevant genes. This estimate is supported by analysis of 417 individuals with XLAS in one series, which revealed that six individuals (1.4%, 95% confidence interval 0.6–3%) carried an additional COL4A3 or COL4A4 likely pathogenic variant [26].

Q4: WHEN TO ORDER ALPORT GENE TESTING? WHAT IS THE LIST OF CLINICAL AND/OR HISTOLOGICAL FINDINGS THAT REQUIRE EXCLUDING ALPORT DIAGNOSIS IN DIFFERENTIAL DIAGNOSTICS AND WHAT SHOULD BE DONE IN MINORS?

4.1-Recommendation: Indications for genetic testing for AS should include any of these:

1. Children and young adults (especially females of childbearing age) with isolated persistent glomerular (dysmorphic) haematuria
2. Individuals with persistent haematuria and family history of either well-documented haematuria or unexplained KF (at least in one first- or second-degree relative)
3. Kidney biopsy with characteristic findings (Box 1)
4. Individuals with persistent haematuria and high tone sensorineural hearing loss
5. Individuals with persistent haematuria and certain ocular findings (fleck retinopathy and anterior lenticonus).

When genetic testing is done to investigate proteinuric kidney disease, cystic kidney disease, unexplained KF, or hearing loss, COL4A3/4/5 gene testing should be included (Box 2). (Grade B, strong)

Genetic testing for AS should comprise analysis of all three COL4A3/4/5 genes. Comprehensive analysis for single nucleotide variants and structural variants comprising entire coding sequences and adjacent intronic sequences (minimum ± 5) of these three genes should be performed. In addition, structural analysis should include the first four exons of COL4A6 [6]. Awareness should be made that complex rearrangements, small copy-number variants and changes within regulatory fragments (promotor, introns, enhancers, silencers, etc.) might be missed by standard protocols that only target exons. The interpretation of variant pathogenicity should be performed in line with current state-of-art guidelines (e.g. American College of Medical Genetics/Association of Molecular Pathologists [ACMG/AMP] criteria, ACGS Best Practice Guidelines for Variant Classification, local recommendations) taking into account several disease-particularities (e.g. variant frequency thresholds and the position in the Gly-X-Y motifs) [1, 27, 28]. For autosomal recessive and digenic disease it is crucial to test the parents of the index case, if possible, to establish phase and help in assessing the pathogenicity of a detected variant.

The experts acknowledge that genetic screening strategies might vary depending on country-specific peculiarities including availability and access to genetic testing and reimbursement policy of national healthcare systems.

4.2-Recommendation: The experts support the need for genetic testing in minors presenting signs of renal and/or extra-renal impairment belonging to Alport spectrum but refrain from proposing genetic testing in an asymptomatic child (including absence of haematuria), without the family having received genetic counselling. (Grade B, strong)

In the case of minors, genetic counselling advocates advising parents to defer elective genetic testing until adulthood to prevent potential harms such as stigma, discrimination, and the loss of the child's ability to make her/his own decisions as an adult [34]. However, consensus on this policy may differ in the context of childhood-onset diseases like AS, where early management protocols are available. Regular clinical evaluation of haematuria (urinalysis) should be offered in the first place to all children born to patients with AS, irrespective of parental subtype (XL, AD, AR), except for boys with fathers with XLAS, who are not at risk (see Q12 for details).

Q5: SHOULD GENETIC TESTING IN RELATIVES BE ENCOURAGED? SHOULD IT BE LIMITED TO (FAMILY) VARIANT TESTING OR RATHER COVER ALL THREE GENES? SHOULD PARTNERS BE TESTED WHEN WISHING TO CREATE A FAMILY?

5.1-Suggestion: Where the molecular diagnosis of AS is established in an index case, genetic testing for that variant could be offered to all adult relatives at risk of inheriting the disease. (Grade D, expert opinon)

A positive genetic testing should be accompanied by clinical evaluation of relatives at risk, including basic laboratory and image (ultrasound), kidney function assessment, and detailed pedigree analysis. Accordingly, the experts do not endorse consumer-initiated genetic testing services [34]. Testing of each second-degree relative should be preceded by an attempt to offer to test the intervening first-degree relative of the index case. Genetic testing in adult relatives should concern only the pathogenic variant(s) identified in the index case. It should be noted that approximately 10–15% of COL4A5 and an unknown rate of COL4A3/COL4A4 cases occur de novo [35].

In the case of a couple undergoing preconception counselling, a genetic test can be offered to the partner of the affected individual if detection of a high risk of ARAS would influence their decision-making or lead to reproductive intervention. In case of consanguinity, a test for the specific variant(s) present in the proband can be offered to the partner. In the absence of consanguinity, if the partner has persistent haematuria or a family history possibly fitting COL4A3/4/5-related kidney disease, full genetic testing of the three genes should be performed if this would influence reproductive decisions. The likelihood of having an offspring affected by ARAS for a patient with a monoallelic or biallelic P/LP variant and an unrelated, asymptomatic partner who does not have haematuria is less than 0.5%. As a result, genetic testing of the partner before starting a family is generally not recommended.

The timing of the genetic test, as well as decisions about preimplantation genetic diagnosis and prenatal genetic testing, should be discussed regarding local financial, social, and legal considerations.

Q6: SHOULD ALPORT GENE TESTING BE INCLUDED IN GENE PANELS DESIGNED FOR KIDNEY DISEASES?

6-Recommendation: Where genetic testing is done to investigate proteinuric kidney disease, unexplained KF or possibly cystic kidney disease, COL4A3/4/5 gene testing should be included. (Grade B, strong)

There are numerous reports of AS being diagnosed following genetic testing in patients with various kidney disease presentations, such as proteinuria, glomerulonephritis, focal segmental glomerulosclerosis, podocytopathies, IgA nephropathy, cysts, and unexplained KF. This suggests that it is logical to include AS genes in kidney disease gene panels. However, given the frequency of heterozygous P/LP variants in COL4A3 and COL4A4 in the general population, caution should be exercised before asserting that the identified variant fully explains the phenotype. This may require exclusion of an additional genetic or acquired kidney disease.

Q7: SHOULD ALPORT GENE TESTING BE INCLUDED ON THE LIST OF SECONDARY FINDINGS/ACTIONABLE FINDINGS?

7-Suggestion: Where genomic testing was performed for an indication other than to investigate kidney disease, a secondary finding of pathogenic variants in the COL4A5 gene and cases with two or more pathogenic variants in COL4A3/COL4A4 should be reported. (Grade C, moderate)

Pathogenic variants in the COL4A5 gene and cases where two or more P/LP variants in COL4A3/COL4A4 are observed should be reported to the patient so that clinical evaluation of kidney function and co-segregation studies (to verify their biallelic character) might be undertaken.

More data on disease risk attributable to heterozygous P/LP autosomal variants in COL4A3/COL4A4 is needed before these are reported to patients undergoing genomic testing for reasons unrelated to evidence of kidney disease.

Q8: GIVEN THE RELATIVELY HIGH FREQUENCY OF INDIVIDUALS WITH HETEROZYGOUS COL4A3/4 VARIANTS IN THE GENERAL POPULATION AND THEIR POTENTIALLY DETERIORATING EFFECT ON THE PHENOTYPE, SHOULD PATIENTS WITH PATHOGENIC COL4A3/4 VARIANTS BE SCREENED FOR OTHER KIDNEY DISEASES IN THE CLINICAL PRACTICE?

8-Suggestion: Caution should be exercised before asserting that a heterozygous P/LP COL4A3/COL4A4 variant fully explains the phenotype. This may require exclusion of an additional genetic or acquired kidney diseases. (Grade C, moderate)

The clinical impact of a heterozygous pathogenic variant in the COL4A3/4/5 genes should be interpreted in the context of the clinical presentation of the tested individual. Accordingly, basic laboratory, imaging (ultrasound), kidney biopsy assessment and/or large panels of genes involved in (glomerular) kidney diseases should be considered before making the final genetic diagnosis.

Q9: WHAT IS THE SENSITIVITY OF CURRENT GENETIC TESTS FOR AS? WHAT ARE THE POTENTIAL DIAGNOSTIC TOOLS IF NEXT-GENERATION SEQUENCING (NGS)-BASED GENETIC TESTING IS NEGATIVE?

9-Statement: No targeted gene panel is 100% sensitive for the different types of genetic changes that can cause AS. In well-designed gene panels the sensitivity of COL4A3/4/5 gene analysis to detect pathogenic variants is estimated to exceed 85%. (Grade D, expert opinon)

The sensitivity of genetic testing has significantly improved in recent years; however, limitations remain in regions of the genome that are inadequately covered or unanalysed. In a Japanese study with comprehensive targeted sequencing using NGS (including search for structural variants) in which strong clinical inclusion criteria were used, the mutation detection rate was 86% [36]. The lower ratios reported in other cohorts [37] are due to less stringent inclusion criteria. Exome sequencing has lower sensitivity/detection rates [38].

When COL4A3/4/5 gene testing and subsequent analysis of the larger panels of genes involved in (glomerular) kidney diseases are negative, the possibility of deep-intronic, regulatory, and structural variants such as inversions that might have been overlooked should be considered.

In such cases, having strong clinical suspicion of AS, targeted cDNA on skin RNA (for X-linked forms) or preferably urine RNA testing for the three genes should be considered; alternatively, or in parallel, genome sequencing might be attempted [39, 40].

Q10: WHAT IS THE ESTIMATED LIFELONG RELATIVE RISK OF STAGE 5 CKD IN INDIVIDUALS WITH HETEROZYGOUS COL4A3/COL4A4 VARIANTS VERSUS THAT IN THE GENERAL POPULATION?

10.1-Statement: The risk of reaching KF at advanced age is higher for individuals with heterozygous P/LP COL4A3/COL4A4 variants compared with the general population. (Grade B, strong)

10.2-Statement: Nephroprotective therapy can significantly reduce this risk. (Grade B, strong)

It is crucial to provide the most accurate prognosis possible for the risk of stage 5 CKD, as KF poses the highest morbidity risk among individuals with AS. Heterozygous P/LP COL4A3/COL4A4 variants, present in approximately 1% of the population [10] are associated with an expectedly low risk of CKD G5. The kidney prognosis of AS is significantly influenced by early treatment with renin-angiotensin system (RAS) inhibitors. However, studies based on hospital-based cohorts of patients and families with heterozygous P/LP COL4A3/COL4A4 variants may suffer from ascertainment bias, skewing towards more severely affected individuals. Additionally, calculating the risk in affected families without excluding the index patient may also lead to overestimation.

Recent studies examining the frequency of P/LP COL4A3/COL4A4 and COL4A5 variants in hospital-based cohorts with KF suggest that heterozygous P/LP COL4A3 and COL4A4 variants occur about as frequently as P/LP COL4A5 variants [14, 41–47]. Gibson et al. found that P/LP COL4A3 and COL4A4 variants were about 20 times more common than P/LP COL4A5 variants in population datasets of participants without known kidney genetic disease [10]. Consequently, individuals with heterozygous P/LP COL4A3/COL4A4 variants have a much lower risk of CKD progression than those with COL4A5 pathogenic variants. Yet it is important to acknowledge that the risk associated with heterozygosity for a COL4A3/COL4A4 variant in a family known to be affected by severe kidney disease is likely higher than in unselected populations, given the shared genetic and environmental background. Consequently, these are the families that often seek and receive genetic advice.

Based on population studies, the risk of CKD G5 by the age of 60 years in individuals with heterozygous P/LP COL4A3/COL4A4 variants is <3%, and just above 3% by the age of 80 [18]. However, this risk may be modified by nephroprotective therapies [48]. Risk factors for estimated glomerular filtration rate (eGFR) decline are thought to be similar to those for other kidney diseases, and include hypertension, cardiovascular disease, diabetes, and obesity. Polygenic risk scores may play a certain role to explain phenotypic heterogeneity in AS, but further studies are needed [48]. Albuminuria (urine albumin : creatinine ratio; UACR) remains the best marker to evaluate the risk of eGFR decline. A summary of prognostic and progression factors is presented in Table 1 (Q15).

Q11: WHAT IS THE ESTIMATED LIFELONG RELATIVE RISK AND AGE-RELATED RISK OF STAGE 5 CKD DEPENDING ON PATTERN OF INHERITANCE AND SEX?

11.1-Statement: The majority of untreated XLAS males will reach KF in the second to fifth decade of life. (Grade A, strong)

11.2-Statement: The vast majority of XLAS males progress to CKD G5 with a variable speed of progression influenced by genotype, timing of nephroprotective therapy, and other environmental factors. (Grade B, strong)

11.3-Statement: XLAS females have a variable kidney outcome. (Grade A, strong)

11.4-Statement: This is probably mostly influenced by variable chromosome X inactivation, early start of nephroprotective therapy, and other environmental factors. (Grade C, moderate)

11.5-Statement: There is currently no evidence that there are sex differences in clinical symptoms and incidence for individuals with heterozygous P/LP COL4A3 or COL4A4 variants and individuals with ARAS. (Grade B, strong)

11.6-Statement: ARAS patients all progress to KF with a variable speed of progression influenced by genotype, early start of nephroprotective therapy, and other environmental factors. The majority of untreated ARAS patients will require kidney replacement therapy (KRT) in the second to third decade of life. (Grade B, strong)

Q12: WHAT IS THE SUGGESTED FOLLOW-UP FOR INDIVIDUALS WITH HETEROZYGOUS P/LP COL4A3/COL4A4 VARIANTS?

12.1-Suggestion: In asymptomatic adults with heterozygous P/LP COL4A3/COL4A4 variants, we suggest performing regular screenings (1- to 2-year intervals) for microhaematuria and microalbuminuria (employing UACR), along with annual blood pressure monitoring. The interval of screening should be individualized based on age, familial history, and comorbidities. (Grade D, expert opinion)

12.2-Suggestion: In individuals with microalbuminuria (defined by UACR > 30 mg/g on two distinct urine samples) and after initiation of renin-angiotensin-aldosterone system (RAS) blockers, we suggest monitoring eGFR and microalbuminuria every 6 months to 1 year. In individuals with CKD G2–G5, the interval for eGFR and UACR/UPCR monitoring should be adjusted to CKD stage and treatments. (Grade D, expert opinion)

12.3-Suggestion: In at-risk children and in children diagnosed with heterozygous P/LP COL4A3/COL4A4 variants (for instance diagnosed in the setting of persistent haematuria), we suggest monitoring urinalysis every 1–4 years from 4–6 years old. (Grade D, expert opinion)

There is a large variability in terms of disease severity in individuals with heterozygous P/LP COL3A4/COL4A4 variants, ranging from asymptomatic to presentation with haematuria alone or with proteinuria, hypertension, and possible kidney function decline with subsequent KF in a subset of individuals [17]. Individuals with heterozygous P/LP COL4A3/COL4A4 variants very rarely have the hearing loss or ocular abnormalities, typical of XLAS or ARAS [16–18, 59].

In a large cohort of 252 individuals with heterozygous pathogenic variants of COL4A3 or COL4A4 from 82 families, tested due to a positive family history of microhaematuria and proteinuria, or abnormal GBM in the proband, microhaematuria, and proteinuria were present in 92% and 65% of cases, respectively. Additionally, 24.6% reached KF at the median age of 67 years [17]. A systematic review of 777 affected individuals from 258 families in 48 publications yielded similar results with, amongst those with available data, 94.8% microhaematuria, 46.4% proteinuria, and KF in 15% [16]. However, because the aforementioned studies include individuals evaluated in hospitals, they are likely to be biased towards more severe disease [18]. Indeed, publicly available databases indicate that predicted pathogenic heterozygous variants of COL4A3 and COL4A4 are present in 1 of 106 ‘ostensibly healthy’ individuals [10]. Overall, the risk of CKD G5 by the age of 60 years in individuals with heterozygous COL4A3/COL4A4 variants is estimated to be <3%, and slightly >3% by the age of 80 [18].

Q13: TO WHOM, WHEN, AND HOW SHOULD HEARING BE ASSESSED?

13.1-Suggestion: In patients with ARAS and males with XLAS, we suggest a first hearing evaluation at 4 years of age, followed by a hearing evaluation every year until adulthood. In adulthood, we suggest a hearing evaluation every 3 years. We recommend hearing screening until the age of 50 and testing based on symptoms afterwards. (Grade D, expert opinion)

13.2-Suggestion: We do not recommend regular hearing evaluation in individuals with a heterozygous P/LP COL4A3/COL4A4 variant and only suggest testing in patients with subjective hearing loss. (Grade D, expert opinion)

13.3-Suggestion: In female patients with XLAS we suggest hearing evaluation at diagnosis or when reaching adulthood and then every 5 years in the absence of hearing loss symptoms. (Grade D, expert opinion)

13.4-Suggestion: We suggest pure tone audiometry as a standard hearing test. (Grade D, expert opinion)

Every patient with AS exhibits an elevated risk to develop hearing loss, albeit with a different likelihood justifying a stratification strategy. Male patients with XLAS and individuals with ARAS are at high risk of develop hearing loss, usually in adolescence but sometimes in childhood [57, 60]. Especially in childhood, close monitoring is advised as untreated hearing loss can lead to delayed speech and language development. In individuals with a heterozygous P/LP COL4A3/COL4A4 variant and female patients with XLAS the occurrence of hearing loss is very rare [61].

We recommend pure tone audiometry in all patients. If hearing loss is present, additional speech audiometry quantifying speech intelligibility could be considered. In very young patients with unreliable pure tone audiometric thresholds, transient evoked otoacoustic emissions (TEOAE) can be considered if the middle ear status is normal, as well as auditory brainstem potentials (ABR).

Q14: TO WHOM, WHEN AND HOW SHOULD OPHTHALMOLOGIC EXAM BE PERFORMED?

14-Suggestion: A full ophthalmologic examination should be performed at the time of diagnosis, including dilated fundoscopy. Periodical eye examinations could be performed from the time of diagnosis for XLAS and ARAS. (Grade D, expert opinion)

Each examination should include clinical examination of all ocular segments with a specific focus on cornea, lens, and retinal alterations [62, 63].

Retinal alterations, including dot-and-fleck retinopathy, temporal retinal thinning, staircase foveopathy, and hypoplastic foveal pit, as well as further retinal alterations should be documented using optical coherence tomography (OCT) if the age of the patients allows [64–66]. For children aged ≤12 years: Ensure compliance with national screening for refractive errors. In case of high refractive errors detected, an ophthalmologic referral and screening for lenticonus should be performed.

Cataract surgery in patients with AS holds specific challenges, leading to the recommendation of performance in specialized centres with experience regarding the unique situation of intraocular lens (IOL) calculation in this situation (i.e. disregarding anterior chamber depth in case of lenticonus).

In individuals with a heterozygous P/LP COL4A3/COL4A4 variant and female patients with XLAS the occurrence of eye involvement is very rare [61].

Q15: WHAT ARE THE PROGRESSION FACTORS IN AS?

Progression factors are provided in the following table 1.

Q16: WHAT IS THE ROLE OF KIDNEY BIOPSY IN PATIENTS WHO ALREADY HAVE A GENETIC DIAGNOSIS OF (OR FAMILY HISTORY OF CONFIRMED) AS?

16.1-Suggestion: A kidney biopsy is not needed in individuals with a genetic diagnosis of XLAS or ARAS. (Grade D, expert opinion)

16.2-Suggestion: A kidney biopsy can sometimes be useful in individuals with heterozygous P/LP COL4A3/COL4A4 variants because other aetiologies of glomerular diseases may coexist. (Grade D, expert opinion)

16.3-Suggestion: In individuals with heterozygous P/LP COL4A3/COL4A4 variants and no microalbuminuria, over 40 years of age, who have been informed about the risk of subsequent development of CKD and want to donate a kidney, a kidney biopsy with electron microscopy should be discussed to evaluate the risk of developing severe CKD post-donation. (Grade D, expert opinion)

Due to the relatively common occurrence of P/LP variants in the COL4A3/COL4A4 genes in the general population, the experts emphasize the importance of considering kidney biopsy, especially in the absence of a family history [10]. This approach is critical for avoiding the oversight of potential diagnoses of kidney disease that may require different therapeutic interventions. For individuals with kidney disease and heterozygous P/LP variants in the COL4A3/A4 genes, it is advisable to pursue familial co-segregation analysis. This will help to confirm that the identified variant is indeed causally contributing to the observed kidney phenotype. In singletons with heterozygous P/LP COL4A3/A4 variants, the experts assume that genetic testing alone is not sufficient to obtain a diagnosis of precision, and that kidney biopsy might be needed. In addition, the experts emphasize that in case of unexpected disease course in individuals with heterozygous P/LP COL4A3/A4 variants (e.g. rapid degradation of kidney function, unexplained acute kidney injury, sudden onset of nephrotic-range proteinuria), a kidney biopsy should be discussed to exclude superimposed kidney disease.

Q17: WHEN (CONSIDERING BOTH AGE AND CLINICAL SITUATION) SHOULD RAS BLOCKERS BE STARTED IN A PATIENT WITH A COL4A3/4/5 VARIANT?

17.1-Recommendation: Early initiation of RAS blockade treatment is recommended in AS, with the most extensive evidence for ramipril due to the most abundant preclinical and clinical studies on safety and efficacy. (Grade B, strong)

17.2-Recommendation: RAS blockers should be initiated and titrated to maximum dose when microalbuminuria is detected irrespective of the genetic subgroup. (Grade B, strong)

17.3-Suggestion: In males with XLAS and in both sexes with ARAS, after 2 years of age, we suggest initiating RAS blockers as soon as microscopic haematuria is detected. (Grade C, moderate)

17.4-Suggestion: In heterozygous individuals with autosomal AS or females with heterozygous XLAS, we suggest not to start RAS blockers if they only show isolated microscopic haematuria. (Grade D, expert opinion)

Early RAS blockade is the standard of care therapy in AS and it is supported by consistent preclinical and clinical data [5, 68–70].

A recent meta-analysis suggested that [71]:

1. RAS blockers could be considered as a specific therapy for patients with AS to delay KF with any genetic type, especially at the early stage of the disease, and
2. Additional therapy should be given in combination with RAS blockade as standard of care.

The classification of RAS blockade as a specific therapy for AS is a consequence of this meta-analysis and is based on the pathogenesis of AS. In brief, the type IV collagen defect leads to abnormal GBM assembly and altered turnover of GBM components. This abnormal GBM is believed to be vulnerable to mechanical load, leading to podocyte loss once several nephrons are damaged, and to FSGS in the remaining nephrons. Later in the course of the disease hypertension, the release of profibrotic proinflammatory chemokines, tubulointerstitial inflammation, and fibrosis may occur [72–74]. RAS blockade may influence various points of this pathogenesis without all mechanisms being exactly known.

Q18: SHOULD SGLT2-INHIBITORS BE USED IN AS? IF SO, WHEN SHOULD THEY BE PRESCRIBED?

18.1-Statement: Existing evidence supports the use of SGLT2-inhibitors (SGLT2i) in adults with proteinuria and CKD and this benefit is likely to apply to individuals with AS. (Grade D, expert opinion)

18.2-Suggestion: In adults with AS and CKD G2A3 or higher (G3A1–3 and G4A1–3) with albuminuria, despite RAS blockers, the addition of a SGLT2i should be considered. (Grade D, expert opinion)

18.3-Statement: SGLT2i have not been trialled in children with CKD (no data available) and therefore a recommendation cannot be made on their use in children. (Grade D, expert opinion)

18.4-Suggestion: We suggest not to use SGLT2i as monotherapy without underlying RAS blockade. (Grade C, moderate)

Dapagliflozin and empagliflozin are approved in the EU for the treatment of CKD (which includes AS) in adults (and post-pubertal teenagers in the UK), but not for children with CKD, as neither drug has been tested in this population. Dapagliflozin and empagliflozin have a marketing authorization for the treatment of Type 2 diabetes mellitus (T2DM) in children 10 years and older [85–87].

SGLT2i are approved for adults with CKD, including glomerular diseases such as AS, without eGFR-limitations in most countries or with the limitation of eGFR < 75–90 and >30 ml/min/1.73 m² in some countries. Existing evidence supports the use of SGLT2i in adults with proteinuria and CKD, and this benefit is likely to apply to individuals with AS [88].

Clinical trials have demonstrated positive cardiovascular and kidney outcomes of SGLT2i in adult patients with diabetic and other forms of CKD. Whether benefits extend to children, teenagers, and young adults with early-stage CKD is unknown as participants in the DAPA-CKD and EMPA-Kidney trials had a mean age of 60–65 years and CKD stages 3–4. The trials included less than 0.3% patients younger than 40 years of age with an eGFR >60 ml/min/1.73 m² [89, 90].

The DOUBLE PRO-TECT Alport trial (NCT05944016) with dapagliflozin will close this knowledge gap. The RCT started recruiting and aims to demonstrate superiority of the SGLT2i dapagliflozin in delaying change in UACR as a surrogate marker for CKD-progression in children and young adults at early stages of CKD in AS (age range 10 to 39 years; eGFR > 60 ml/min/1.73 m² in adults). A similar trial in children with any kind of CKD is being planned with empagliflozin.

Q19: SHOULD FURTHER STUDIES BE PERFORMED IN A PATIENT PRESENTING AT DIAGNOSIS WITH NEPHROTIC SYNDROME IN WHOM A COL4A3/4/5 PATHOGENIC VARIANT IS IDENTIFIED?

19.1-Suggestion: A kidney biopsy is suggested in an individual with a P/LP COL4A3/4/5 variant if the onset of nephrotic syndrome is abrupt or the degree of proteinuria does not match the expected clinical presentation of the disease. (Grade D, expert opinion)

19.2-Suggestion: In the presence of an unexpected nephrotic syndrome and a non-specific kidney biopsy, we suggest expanding genetic testing by using a genetic panel test that includes all genes associated with inherited kidney disease or exome sequencing. (Grade D, expert opinion)

The typical histological feature of later stages of AS is FSGS, which is thought to be triggered by progressive podocyte loss and consequent focal segmental glomerular scarring [72]. Although the exact patho-mechanisms are not well understood, most male patients with XLAS and males and females with ARAS with progressive disease can develop nephrotic range proteinuria and some even develop nephrotic syndrome [57, 68]. In heterozygous autosomal patients with AS, however, nephrotic syndrome is very rare and typically associated with additional contributing factors [56].

The amount of proteinuria can be aggravated by infections, poor blood pressure control, high body mass index, high protein intake, additional kidney disease, and during pregnancy [72].

In AS, nephrotic syndrome develops gradually over months and years, rather than presenting with an abrupt onset.

Q20: COULD INDIVIDUALS WITH HETEROZYGOUS COL4A3/COL4A4/COL4A5 PATHOGENIC VARIANTS BE CONSIDERED AS LIVING DONORS?

20.1-Recommendation: We recommend determining the exact genotype in all possible donors. (Grade B, strong)

20.2-Suggestion: All relatives with heterozygous AS variants (males or females with heterozygous P/LP variants in COL4A3/COL4A4 or females with a variant in COL4A5) should only be considered as the last possible resource for living kidney donation. (Grade D, expert opinion)

20.3-Suggestion: Living kidney donation is not advisable in individuals heterozygous for a single P/LP variant in COL4A3, COL4A4, or COL4A5 aged under 40 years, or at any age if there is clinical or histological evidence of kidney damage (e.g. albuminuria, reduced eGFR or interstitial fibrosis/tubular atrophy greater than what would be normal for their age). (Grade D, expert opinion)

20.4-Suggestion: Where kidney donation is considered by an individual heterozygous for a single P/LP variant in COL4A3, COL4A4, or COL4A5 aged over 40 years, in the absence of albuminuria or reduced eGFR, a kidney biopsy could be performed to detect evidence of subclinical kidney damage (i.e. scarring greater than what would be normal for their age) that would preclude donation. A decision to proceed with donation should only be made after careful consideration of the risks and benefits for that individual/family. (Grade D, expert opinion)

20.5-Suggestion: Lifelong monitoring (with prompt nephroprotective RAS blockade therapy if microalbuminuria or hypertension develop) could be offered to individuals heterozygous for a single P/LP variant in COL4A3 or COL4A4 or COL4A5 who have donated a kidney, as for all living kidney donors. (Grade D, expert opinion)

The experts are aware that when a young patient with AS needs KRT, the pressure on families to find a suitable living donor can be immense [96]. Additionally, waiting times for kidney transplants other than living donations vary by country—in Europe from 2–3 years to 8–10 years—leading to differing approaches in evaluating potential living donors with heterozygous variants of COL4A3/COL4A4. The high prevalence of COL4A3 and COL4A4 variants in the population also increases the likelihood that a donor will have one of these variants.

Q21: WHAT IS THE PREVALENCE AND THE MANAGEMENT OF ALPORT POST-TRANSPLANT NEPHRITIS OR DE NOVO GBM DISEASE?

21-Statement: While the risk of anti-GBM disease following a kidney transplant for AS without immunosuppressive therapy may be significant, current immunosuppressive treatments reduce this risk to less than 3%. (Grade D, expert opinion)

AS is a genetic disease, which does not affect the transplanted kidney and does not re-occur in the transplanted kidney. Therefore, the long-term outcome after transplantation is better than for age-matched control patients with other kidney diseases [55].

The risk of anti-GBM disease in AS patients after transplantation is explained by the presence of alpha 3/4/5 type IV network in the transplanted ‘non-Alport’ kidney, which is unknown to the immune system. Risk of anti-GBM disease is higher in non-missense variants, which may lead to the absence of alpha 3/4/5 type IV collagen network in this individual, rather than in missense variants. A negative serological test does not exclude anti-GBM antibodies [97].

In some European countries, the occurrence of anti-GBM antibodies is monitored in patients with XLAS and ARAS, especially in truncating variants, in the first year (first 2 years) post-transplantation. The Goodpasture antigen is located in the NC1 domain of the alpha 3(IV) chain. Therefore, serological assays that specifically detect antibodies against this antigen, may yield negative results even in cases of anti-GBM disease. This can occur in situations where there is de novo exposure to the alpha 5(IV) chain, such as in XLAS recipients who develop an alloimmune response to the newly introduced protein [97, 98]. When anti-GBM antibodies or linear immune staining in the GBM/crescents are detected, augmenting/adding immunosuppression could be considered in the presence of kidney function deterioration.

PATIENT NEEDS FOR PEOPLE WITH AS

Patient and family well-being is at the heart of this guideline for AS, a life-long genetic condition that can impact generations in a family. Depending on the severity of the genetic variant, different individuals or families can have clinical signs and symptoms on a spectrum from mild to severe. Even within a family, these can be different, and the differences cause a range of different patient journeys and needs. Patients should first understand where their genetic variant falls on the spectrum of potential outcomes for hearing, kidneys, and eyes. Prognosis information may be unavailable for their particular genetic variant, but patients need to know what is available and what research is going on to understand it better. Being diagnosed with a rare disease, like AS, can feel isolating Box 3 and Box 4.

Further information on support available for people living with AS is provided in the Supplementary Material.

CLOSING REMARKS AND FUTURE PERSPECTIVES

This guideline provides clear directives on diagnosing and managing XLAS and ARAS. However, for individuals with heterozygous variants in COL4A4 or COL4A3, the evidence is currently limited, and the recommendations may evolve as new information emerges. There is still a lack of evidence regarding the true prevalence and impact of these variants. In the near future, polygenic risk scores, specifically genome-wide polygenic scores, may play a significant role in interpreting these gene variants [48]. Larger, well-phenotyped population studies are needed to accurately assess the contribution of COL4A4 or COL4A3 pathogenic variants to CKD.

Besides that, the name of the disease is also undergoing an international renaming process.

Despite limited evidence for certain aspects of this guideline, the experts believe that addressing all facets of AS—a leading genetic cause of CKD and the second most common genetic cause of KF after autosomal dominant polycystic kidney disease—is crucial. This guidance, developed by a multidisciplinary team of internationally renowned experts and evaluated through a Delphi survey involving most of the remaining global experts, represents a significant contribution to the field.

Supplementary Material