The recent Chandos House meeting of the Alport Variant Collaborative extended the indications for screening for pathogenic variants in the COL4A5, COL4A3 and COL4A4 genes beyond the classical Alport phenotype (haematuria, renal failure; family history of haematuria or renal failure) to include persistent proteinuria, steroid-resistant nephrotic syndrome, focal and segmental glomerulosclerosis (FSGS), familial IgA glomerulonephritis and end-stage kidney failure without an obvious cause. The meeting refined the ACMG criteria for variant assessment for the Alport genes (COL4A3–5). It identified ‘mutational hotspots’ (PM1) in the collagen IV α5, α3 and α4 chains including position 1 Glycine residues in the Gly-X-Y repeats in the intermediate collagenous domains; and Cysteine residues in the carboxy non-collagenous domain (PP3). It considered that ‘well-established’ functional assays (PS3, BS3) were still mainly research tools but sequencing and minigene assays were commonly used to confirm splicing variants. It was not possible to define the Minor Allele Frequency (MAF) threshold above which variants were considered Benign (BA1, BS1), because of the different modes of inheritances of Alport syndrome, and the occurrence of hypomorphic variants (often Glycine adjacent to a non-collagenous interruption) and local founder effects. Heterozygous COL4A3 and COL4A4 variants were common ‘incidental’ findings also present in normal reference databases. The recognition and interpretation of hypomorphic variants in the COL4A3–COL4A5 genes remains a challenge.
Estimates of the prevalence of Alport syndrome vary from one in 5,000 to one in 53,000 of the population [1, 2], but the frequent finding of likely pathogenic COL4A5 variants in normal reference datasets and cohorts with renal failure suggests that X-linked disease affects closer to closer to one in 5,000 [3, 4]. This is consistent with Alport syndrome being the second commonest cause of inherited kidney failure after autosomal dominant Polycystic Kidney Disease. It also means that many variants are undetected and that even apparently normal individuals have pathogenic variants in the COL4A3–COL4A5 genes.
Inheritance of Alport syndrome is X-linked (OMIM301050), autosomal recessive (OMIM 203780), digenic (typically with a pathogenic variant in each of COL4A3 and COL4A4 ) or autosomal dominant (OMIM 104200, or thin basement membrane nephropathy, due to heterozygous COL4A3 or COL4A4 variants ). The typical clinical features of Alport syndrome are persistent microscopic haematuria, end-stage kidney failure, and often a family history of haematuria or renal failure . Hearing loss and non-nephrotic range proteinuria are common. Ocular abnormalities such as lenticonus with abnormal lens protrusion, or fleck retinopathy may be present . Clinical features overlap between boys and men with X-linked disease, and in boys and girls, men and women with recessive or digenic inheritance [5, 8]. Many women with X-linked Alport syndrome, and males and females with heterozygous COL4A3 or COL4A4 variants, or sometimes digenic COL4A3 and COL4A4 variants, have only microscopic haematuria [5, 8, 9].
In February 2020, a group of 47 medical and scientific specialists from three continents, all with an interest in the molecular diagnostics Alport syndrome (the ‘Alport Variant Collaborative’), met in London to review the current recommendations for genetic testing for Alport syndrome and to further refine the American College of Medical Genetics/Association of Molecular Pathologists (ACMG/AMP) criteria for evaluating variants in the COL4A3–COL4A5 genes that have been published previously [8, 10]. The group comprised adult (n = 12) and paediatrics (n = 6) nephrologists, histopathologists (n = 2), geneticists (n = 17), laboratory scientists (n = 4), researchers (n = 2), industry scientists (n = 3) and a representative of the Alport UK Patient Support group (n = 1). Attendees had submitted difficult-to-evaluate COL4A3–COL4A5 variants, which were precirculated, and they spent two sessions at the meeting examining these variants, and discussing the evidence for pathogenicity. Their conclusions have been summarised into the following guidelines for genetic testing for Alport syndrome for use by molecular testing laboratories. A draft of these guidelines was then sent to all attendees, modified based on their comments, and the final document approved by the group prior to submission. Decisions about individual variants are ongoing with further clinical data being sought in order to reach at least two-thirds consensus on the assertions.
Broadening the phenotype
Increasingly, pathogenic variants in COL4A3–COL4A5 are found in individuals with proteinuria (‘nephrotic syndrome’, ‘nephrotic range proteinuria’, ‘steroid-resistant nephrotic syndrome’) or with the renal biopsy finding of focal and segmental glomerulosclerosis (FSGS)[11, 12]. FSGS is one of the commonest causes of glomerular disease resulting in progressive renal failure, and pathogenic COL4A3–COL4A5 variants are the commonest cause of adult-onset FSGS ranging from 5% to 20% of cases with sporadic or familial disease respectively [11, 12]. In such cases the glomerular basement membrane does not necessarily have the lamellation typical of Alport syndrome and COL4A3–COL4A5 variants may be associated with an older age at end-stage kidney failure . Pathogenic variants in any of the COL4A3–COL4A5 genes are also found in up to 10% of individuals with renal failure where the cause was not known . In addition families with IgA glomerulonephritis not uncommonly have mutations in the COL4A3–COL4A5 genes . Finally there are reports of COL4A3–COL4A5 variants being associated with renal cysts where autosomal dominant polycystic kidney disease has been excluded [15, 16].
Thus, not only individuals with the typical features of Alport syndrome should undergo genetic testing, but also those with proteinuria or FSGS, with familial IgA glomerulonephritis or with end-stage kidney failure where there is no obvious cause (Table 1). For individuals with proteinuria or FSGS, or end-stage kidney failure, the presence of haematuria or a family history of haematuria, and in the case of IgA disease, a family history of haematuria, all make the detection of a pathogenic COL4A3–COL4A5 variant more likely. However the meeting considered there was currently insufficient evidence to recommend that individuals with cystic kidney disease should also be tested for COL4A3–COL4A5 variants.
Genetic testing of the COL4A3–COL4A5 genes
Finding a COL4A3–COL4A5 variant that affects the structure or function of a collagen IV α chain confirms the diagnosis of Alport syndrome, and indicates that other family members should be investigated. Genetic testing also identifies the mode of inheritance, and which family members who are at risk of being affected. End-stage kidney failure is very likely in hemizygous COL4A5 males (90% by the age of 40 years) and individuals with biallelic defects in COL4A3 or COL4A4, but much less common in females with a heterozygous COL4A5 variant (15–30% by 60 years) and a person with a heterozygous pathogenic variant in COL4A3 or COL4A4 . The nature of individual variants also indicates the likelihood of early-onset kidney failure and extra-renal features such as ocular abnormalities [18, 19]. Hearing loss is common with X-linked and recessive disease . Targeted diagnostics can be used to identify other affected family members, to ensure that an affected person does not act as a kidney donor, and in preimplantation genetic diagnosis . Where a variant is identified in a person referred with FSGS for testing, the finding of a causative pathogenic variant in one of the COL4A3–COL4A5 genes means that treatment with corticosteroids or immunosuppressants is not usually useful [4, 20, 21], and that this type of FSGS does not typically recur after renal transplantation.
MPS usually identifies at least 80% of disease-causing variants in COL4A3–COL4A5 where Alport syndrome is suspected clinically  because of microscopic haematuria or progressive kidney failure, together with a positive family history of haematuria or renal failure. Where a causative variant or variants are not found, it is important to ensure that the genes are sufficiently ‘covered’ by the sequencing technique. Other explanations for the inability to detect pathogenic variants are the large genomic imbalances encompassing one or more exons (such as deletions, duplications or inversions) that MPS detects less well; a location deep within an intron or regulatory element or affecting a non-canonical splice site (for example, 5 nucleotides from the intron-exon boundary); there being insufficient evidence for pathogenicity and the variant being assessed as a Variant of Uncertain Significance (VUS); the variant affecting a processing enzyme or chaperone rather than a COL4 gene, as occurs in about 10% of cases of osteogenesis imperfecta;  and rarely because of mosaicism. Sometimes the disease is simply not Alport syndrome. Phenocopies of Alport syndrome where there is predominantly haematuria include Fechtner syndrome (OMIM155100), fibronectin glomerulopathy (OMIM 601894), Nail-Patella syndrome (OMIM 161200), Hereditary Angiopathy, Nephropathy and muscle Cramps (OMIM 611773), Dense deposit disease (OMIM 134370) and CFHR5 deficiency (OMIM 614809). However increasingly pathogenic variants are found in genes affected in proteinuric diseases.
More pathogenic variants are known for COL4A5 than for COL4A3 and COL4A4, and these are missense in 60%, nonsense in 10%, canonical splice sites in 10% and frameshifts in 20% (databases.lovd.nl/shared/genes/COL4A5). About half the pathogenic COL4A3–COL4A5 variants found in diagnostic laboartories are novel. There appear to be some changes in the types of pathogenic variants found in the last years. Fewer large deletions have been reported possibly because MPS detects these less well, or because the more severe disease that these people have means that they have already been studied. In addition, more non-canonical splice site variants, and non-Gly substitutions in the collagenous and carboxy NC1 domains have been reported.
Definition of pathogenicity
The definition of pathogenicity has become less clear for sequence variants in COL4A3–COL4A5. Previous meetings have concluded that the demonstration of a pathogenic variant was the best evidence for the diagnosis of Alport syndrome [8, 10]. Despite the introduction of the standards and guidelines for the interpretation of sequence variants (ACMG/AMP criteria ) there is sometimes insufficient data for a conclusion of pathogenicity. The association with haematuria or kidney failure, and a family history of haematuria or renal failure may be the strongest evidence for the diagnosis of Alport syndrome. Microscopic haematuria is found in 95% of those with a pathogenic COL4A5 variant , and at least 67% with a pathogenic heterozygous COL4A3 or COL4A4 variant . End-stage kidney failure occurs in all males, but many fewer females with a pathogenic COL4A5 variant . The situation is more complicated in recessive Alport syndrome by the requirement for two COL4A3 or COL4A4 variants in trans . Renal failure is less common, but recognised increasingly, with pathogenic heterozygous COL4A3 or COL4A4 variants .
The manual application of the ACMG/AMP guidelines is time-consuming, subject to error and yields inconsistent results . The tools themselves may be flawed . Several web-based tools, such as Alamut (www.interactive-biosoftware.com/alamut-visual/), Varsome (www.varsome.com)  and the ClinGen pathogenicity calculator (calculator.clinicalgenome.org) , may help in classifying variants by summarising the results for ACMG/AMP criteria such as the computational criteria (Polyphen2, SIFT, Mutation taster, conservation scores, and occurrence in gnomAD and other reference databases) in one site . The ACMG/AMP recognise that “those working in specific disease groups should continue to develop more focused guidance regarding the classification of variants in specific genes“, and thus the Chandos House meeting assessed the ACMG criteria in the context of Alport syndrome and the COL4A3- COL4A5 genes and made the following recommendations (Table 2) including the identification of significant functional domains and mutational hotspots (PM1); the degree of conservation of the protein sequences in different species (PP3); the ‘well-established’ functional assays that indicate pathogenicity (PS3, BS3); whether pathogenic variants are absent from databases with healthy control individuals (PM2)  and the threshold in these databases above which variants are likely to be benign (BA1, BS1).
The Chandos House meeting did not revise the weighting system for the ACMG/AMP criteria for Alport variants . Nor did it did it reach consensus on who had the responsibility for requesting a review of Variants of Uncertain Significance for a reclassification to Likely Pathogenic/Pathogenic or Likely Benign/Benign and how often this assessment should be undertaken. Many laboratories considered that it was the clinician’s responsibility to request a VUS review based on the ongoing need for a diagnosis or the availability of new clinical information. On the other hand, clinicians considered that laboratories were more aware of advances in algorithms for defining pathogenicity. Finally, the ACMG/AMP recommendations for reporting incidental findings in Clinical Exome and Genome Sequencing do not include pathogenic variants in the COL4A3–COL4A5 genes  despite these being common and clinically significant.
Modification of the ACMG/AMP criteria for the COL4A3–COL4A5 genes and rationale
Types of pathogenic variants
Missense variants are the commonest changes found in COL4A3–COL4A5. For the collagen IV α5 chain, more missense variants are found in the large intermediate collagenous domain (84–88% depending on the database) which is more than expected based on length (p ≤ 0.006, LOVD, Table 3) and many of which are Glycine substitutions (95% (278/294) in ARUP, arup.utah.edu/database/). Missense variants in the carboxy non-collagenous domain are also common (11–14%). Missense variants that are not Glycine substitutions are probably the most difficult to classify, but even some Glycine substitutions, especially those adjacent to non-collagenous interruptions, are more difficult. In-silico tools used to assess the effect of a particular amino acid substitution on the protein structure may not take into account the character of the Gly-Xaa-Yaa triplets (a Glycine followed by two other amino acids) typical of the collagen chains and as a result their predictions of the effect on structure may be inaccurate.
Most nonsense variants in the COL4A3–COL4A5 genes result in nonsense-mediated mRNA decay and loss of the corresponding collagen IV α chains from affected membranes. Because the α5 chain is part of the α3α4α5 heterotrimer, the α3 and α4 chains with which it normally forms the heterotrimer are imperfectly incorporated or not incorporated at all , and there is a compensatory increase in the expression of the α1α1α2 network . This means that males with a COL4A5 nonsense variant typically have no α3α4α5 (or α5α5α6) network in affected membranes. In females with COL4A5 heterozygous nonsense variants, however, this loss is segmental depending on lyonisation since clusters of cells produce normal amounts of the α5 chain . In males and females with COL4A3 or COL4A4 heterozygous nonsense variants, affected cells result in less of the α3α4α5 network and membrane thinning.
Variants in COL4A5 have been collated into a number of databases (Table 4), from the literature and submissions from individual laboratories. Currently all databases record variant assessments made by the submitting laboratories. The role of such databases in variant assertions is controversial [33, 34] but they still represent useful repositories. ClinVar is the only database that independently evaluates variants but it has no expert panel assessments for the COL4A3–COL4A5 variants. Most laboratories also consult their own databases of variants that they have assessed previously.
The in vitro or in vivo functional studies used in pathogenicity assessments must be specific for the individual gene variant and not for the gene as a whole. One of the commonest in vitro tests for Alport syndrome is immunohistochemistry of a renal or skin biopsy to demonstrate that the collagen IV α5 chain is absent in X-linked and autosomal recessive Alport syndrome [35, 36]. However this does not confirm that a particular variant is pathogenic but only that the individual has a collagen IV defect consistent with the diagnosis of Alport syndrome.
Among the functional studies that support a variant having a damaging effect on the gene or gene product, in vitro trimerization assays that examine secretion into the extracellular space  correlate well with clinical variant severity, especially in distinguishing pathogenic from non-pathogenic Glycine variants in COL4A5. In addition, for non-canonical splice sites, minigene assays or sequencing of urinary podocyte or fibroblast mRNA can be used, but are currently only available in specialised laboratories . Increasingly, COL4A5 variants affecting non-canonical splice sites, up to 11 nucleotides from the intron-exon boundary, are identified in individuals with X-linked Alport syndrome. Sometimes synonymous variants within the coding region result in splicing changes that are also disease-causing. Potential splicing changes can be examined with on-line prediction tools such as MaxEntScan or SpliceSiteFinder, but eventually require confirmation with in vitro assays.
Functional assays used in research include evaluation of abnormalities in electrophoresis and thermal stability of collagen IV (previously common in collagen type I); [39, 40] and mRNA quantitation . Again these assays are not specific for individual gene variants unless a zebrafish, mouse or normal cell line has been engineered to incorporate the variant being assessed.
Critical and well-established domains and mutational hotspots
Collagen IV differs from most other collagens in forming a chicken-wire network rather than fibrillar structure. The COL4A1–COL4A6 genes code for the collagen IV α1–α6 chains. Collagen IV is a heterotrimer with three of the 6 α chains producing a triple helix (α1α1α2, α3α4α5 or α5α5α6) .The α3α4α5 network is the main component of the glomerular basement membrane, encoded by the COL4A3, COL4A4 and COL4A5 genes respectively. The amino acid sequences of these chains are each highly conserved in different species, and also between the individual α5, α3 and α4 chains (Fig. 1, Supplementary Figs. 1 and 2).
Each collagen IV α chain has an amino non-collagenous (NC) domain; an intermediate collagenous domain; and a carboxy NC domain (Fig. 1). The intermediate collagenous domain has the sequence Gly-Xaa-Yaa where X and Y are often Proline or Hydroxyproline. The presence of a Glycine in every third position in the Gly-Xaa-Yaa repeats is critical to triple helix formation because Glycine is the smallest amino acid and fits within the helix interior. The Proline residues contribute flexibility. The three α3, α4 and α5 chains bind together through disulphide bonds at the Cysteine residues and by a sulfilimine bond between Lysine and Methionine residues all within the carboxy NC domain . There are also multiple non-collagenous interruptions in the collagenous domains that confer flexibility and facilitate network formation (Tables 5–7) . Each interruption has adjacent Glycine residues, and some incorporate a Glycine. However the biochemistry is even more complicated because while the transcript for individual α chains is read off forwards 5’ to 3’, from the amino to the carboxy terminus, the three chains bind through disulphide and sulphilimide bonds in the carboxy terminus from which the triple helix winds up ‘backwards’ [45, 46]. Substitution of these critical position 1 Glycine residues within the Gly-Xaa-Yaa sequence distorts and disrupts helix formation. Glycine substitutions in other inherited collagen diseases are typically pathogenic, for example, in COL1A1 in osteogenesis imperfecta . The effect of Glycine substitutions in positions 2 or 3 or within the interruption is not known. The local environment is also important and Glycine substitutions adjacent to interruptions, where there is more chain flexibility, appear to be less pathogenic [48, 49].
Proline substitutions in the collagen IV α5 chain are reported pathogenic much less often than Glycine substitutions (databases.lovd.nl/shared/genes/COL4A5). Proline confers flexibility and its replacement with another amino acid is less problematic than for Glycine. However, Proline residues are also found in both the integrin binding sites and in the GlyProProGlyProPro binding site for glycoprotein VI which is needed for platelet activation , but no Proline substitutions have been described in these residues in the COL4A3–5 genes.
Interestingly, pathogenic variants have been reported for 10 of the 12 Cysteine residues that are critical for cross-linking in the carboxy NC domain (databases.lovd.nl/shared/genes/COL4A5). No substitutions have been described for the Lysine and Methionine residues in the NC domain that form the sulfilimine bond.
Thus, the Chandos House meeting recommended that most Glycine residues in the collagenous domain of the collagen IV α5, α3 and α4 chains should be recognised as critical residues equivalent to a functional domain. However some Glycine substitutions adjacent to NC interruptions represent hypomorphic variants with a milder clinical phenotype [49, 51] such as p.(Gly624Asp) in COL4A5. Clinical information is critical for the complete evaluation of variants, and where there are discrepancies, the evidence from many laboratories, such as segregation of haematuria with the variant in multiple families or within many members of the same family, may confirm a pathogenic nature. The 12 Cysteines in the carboxy NC domain should also be considered critical domains, and any substitutions are likely to be pathogenic.
Hypomorphic variants are increasingly described in the COL4A5 gene [51,52,–53] although there are, as yet, no definitions or criteria for their identification. They may result in haematuria only, later-onset kidney failure, or GBM thinning rather than lamellation in a male. Variants in the COL4A3–COL4A5 genes may also modify the pathogenic effects of genes encoding other podocyte or glomerular membrane proteins.
The COL4A5 hypomorphic variant, NM_000495.5:c.1871G>A substitution (rs104886142); p.(Gly624Asp) results in late-onset kidney failure and is the commonest pathogenic variant causing X-linked Alport syndrome in Central and Eastern Europe . It has previously been considered benign  or a VUS. This variant is located adjacent to a non-collagenous interruption in the Gly-X-Y sequence which reduces the effect of the Gly substitution (Table 5). The Chandos House meeting considered that hypomorphic variants, such as p.(Gly624Asp) may still be actionable, requiring renin-angiotensin- aldosterone blockade consistent with published guidelines for other COL4A5 variants [8, 54].
The nature of other variants such as NM_000495.5:c.2858G>T (rs78972735; p.(Gly953Val)) in COL4A5 , and NM_000092.5:c.1634G>C (rs1800516; p.(Gly545Ala)) and NM_000092.4(COL4A4):c.2996G>A (rs13027659 p.(Gly999Glu)) in COL4A4 is less clear. All are very abundant in certain ancestries. p.(Gly953Val) is also located immediately adjacent to an interruption in the collagenous domain of the collagen IV α5 chain, but it is not on its own associated with haematuria, kidney failure or GBM lamellation , and has conflicting interpretations of pathogenicity in Clin Var (www.ncbi.nlm.nih.gov/clinvar/variation/24573/). Both p.(Gly545Ala) and p.(Gly999Glu) affect Gly residues within the collagenous domain of the collagen IV α4 chain itself. These are considered Benign** (www.ncbi.nlm.nih.gov/clinvar/variation/255015/) and with Conflicting interpretation of pathogenicity* (www.ncbi.nlm.nih.gov/clinvar/variation/191312/) respectively. Thus interpretation of pathogenicity is complicated, and variants while themselves not causing haematuria may still worsen renal failure progression.
The COL4A3–COL4A5 genes have particular characteristics that must be considered in assessing variants for a pathogenic or benign nature. Most notable are the importance of the position 1 Glycine substitutions in the collagenous domains, Cysteine substitutions in the carboxy NC domains and the recognition of hypomorphic variants associated with a milder clinical phenotype with microscopic haematuria only, late-onset renal failure or GBM thinning rather than lamellation. Evidence is emerging for the importance of other non-Glycine substitutions especially in the carboxy NC domains but these await further evaluation.
Hasstedt SJ, Atkin CL. X-linked inheritance of Alport syndrome: family P revisited. Am J Hum Genet. 1983;35:1241–51.
Pajari H, Kaariainen H, Muhonen T, Koskimies O. Alport’s syndrome in 78 patients: epidemiological and clinical study. Acta Paediatr. 1996;85:1300–6.
Groopman EE, Marasa M, Cameron-Christie S, Petrovski S, Kamalakaran S, Povysil G, et al. Diagnostic utility of exome sequencing for kidney disease. N Engl J Med. 2019;380:142–51.
Connaughton DM, Hildebrandt F. Personalized medicine in chronic kidney disease by detection of monogenic mutations. Nephrol Dial Transpl. 2020;35:390–7.
Mencarelli MA, Heidet L, Storey H, van Geel M, Knebelman B, Fallerini C, et al. Evidence of digenic inheritance in Alport syndrome. J Med Genet. 2015;52:163–74.
Savige J, Rana K, Tonna S, Buzza M, Dagher H, Wang YY. Thin basement membrane nephropathy. Kidney Int. 2003;64:1169–78.
Gubler M, Levy M, Broyer M, Naizot C, Gonzales G, Perrin D, et al. Alport’s syndrome. A report of 58 cases and a review of the literature. Am J Med. 1981;70:493–505.
Savige J, Gregory M, Gross O, Kashtan, Ding J, Flinter F. Expert guidelines for the management of Alport syndrome and thin basement membrane nephropathy. J Am Soc Nephrol. 2013;24:364–75.
Savige J, Colville D, Rheault M, Gear S, Lennon R, Lagas S, et al. Alport syndrome in women and girls. Clin J Am Soc Nephrol. 2016;11:1713–20.
Savige J, Ariani F, Mari F, Bruttini M, Renieri A, Gross O, et al. Expert consensus guidelines for the genetic diagnosis of Alport syndrome. Pediatr Nephrol. 2019;34:1175–89.
Gast C, Pengelly RJ, Lyon M, Bunyan D, Seaby E, Graham N, et al. Collagen (COL4A) mutations are the most frequent mutations underlying adult focal segmental glomerulosclerosis. Nephrol Dial Transpl. 2016;31:961–70.
Malone AF, Phelan PJ, Hall G, Cetincelik U, Homstad A, Alonso A, et al. Rare hereditary COL4A3/COL4A4 variants may be mistaken for familial focal segmental glomerulosclerosis. Kidney Int. 2014;86:1253–9.
Yao T, Udwan K, John R, Rana A, Haghighi A, Xu L, et al. Integration of genetic testing and pathology for the diagnosis of adults with FSGS. Clin J Am Soc Nephrol. 2019;14:213–23.
Li Y, Groopman E, D’Agati V, Prakash S, Zhang J, Mizerska Waskiak MM, et al. Type IV collagen mutations in IgA nephropathy. Kid Int Rep. 2020;5:1075–8.
Sevillano AM, Gutierrez E, Morales E, Hernandez E, Molina M, Gonzales E, et al. Multiple kidney cysts in thin basement membrane disease with proteinuria and kidney function impairment. Clin Kidney J. 2014;7:251–6.
Gulati A, Sevillano AM, Praga M, Gutierrez E, Alba I, Dahl N, et al. Collagen IV gene mutations in adults with bilateral renal cysts and CKD. Kidney Int Rep. 2020;5:103–8.
Pierides A, Voskarides K, Athanasiou Y, Ioannou K, Damianou L, Arsali M, et al. Clinico-pathological correlations in 127 patients in 11 large pedigrees, segregating one of three heterozygous mutations in the COL4A3/COL4A4 genes associated with familial haematuria and significant late progression to proteinuria and chronic kidney disease from focal segmental glomerulosclerosis. Nephrol Dial Transpl. 2009;24:2721–9.
Savige J, Storey H, Il Cheong H, Gyung Kang H, Park E, Hilbert P, et al. X-linked and autosomal recessive Alport syndrome: pathogenic variant features and further genotype-phenotype correlations. PLoS ONE. 2016;11:e0161802.
Bekheirnia MR, Reed B, Gregory MC, McFann K, Shamshirsaz A, Masoumi A, et al. Genotype-phenotype correlation in X-linked Alport syndrome. J Am Soc Nephrol. 2010;21:876–83.
Brown EJ, Pollak MR, Barua M. Genetic testing for nephrotic syndrome and FSGS in the era of next-generation sequencing. Kidney Int. 2014;85:1030–8.
Trautmann A, Lipska-Zietkiewicz BS, Schaefer F. Exploring the clinical and genetic spectrum of steroid resistant nephrotic syndrome: the PodoNet Registry. Front Pediatr. 2018;6:200.
Moriniere V, Dahan K, Hilbert P, Lison M, Lebbah S, Topa A, et al. Improving mutation screening in familial hematuric nephropathies through next generation sequencing. J Am Soc Nephrol. 2014;25:2740–51.
Marini JC, Reich A, Smith SM. Osteogenesis imperfecta due to mutations in non-collagenous genes: lessons in the biology of bone formation. Curr Opin Pediatr. 2014;26:500–7.
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;17:405–24.
Patel RY, Shah N, Jackson AR, Ghosh R, Pawliczek P, Paithankar S, et al. ClinGen pathogenicity calculator: a configurable system for assessing pathogenicity of genetic variants. Genome Med. 2017;9:3.
Vihinen M. Problems in variation interpretation guidelines and in their implementation in computational tools. Mol Genet Genom Med. 2020;8:e1206.
Kopanos C, Tsiolkas V, Kouris A, Chapple C, Albarca Aguilera M, Meyer R, et al. VarSome: the human genomic variant search engine. Bioinformatics. 2019;35:1978–80.
Rivera-Munoz EA, Milko LV, Harrison SM, Azzaritis D, Kurtz C, Lee K, et al. ClinGen Variant Curation Expert Panel experiences and standardized processes for disease and gene-level specification of the ACMG/AMP guidelines for sequence variant interpretation. Hum Mutat. 2018;39:1614–22.
Kalia SS, Adelman K, Bale SJ, Chung W, Eng C, Evans J, et al. Recommendations for reporting of secondary findings in clinical exome and genome sequencing, 2016 update (ACMG SF v2.0): a policy statement of the American College of Medical Genetics and Genomics. Genet Med. 2017;19:249–55.
Antignac C, Knebelmann B, Drouot L, Gros F, Deschenes G, Hors-Cayla M, et al. Deletions in the COL4A5 collagen gene in X-linked Alport syndrome. Characterization of the pathological transcripts in nonrenal cells and correlation with disease expression. J Clin Invest. 1994;93:1195–207.
Kalluri R, Shield CF, Todd P, Hudston B, Neilson E. Isoform switching of type IV collagen is developmentally arrested in X-linked Alport syndrome leading to increased susceptibility of renal basement membranes to endoproteolysis. J Clin Invest. 1997;99:2470–8.
Yoshioka K, Hino S, Takemura T, Maki S, Wieslander J, Takekoshi Y, et al. Type IV collagen alpha 5 chain. Normal distribution and abnormalities in X-linked Alport syndrome revealed by monoclonal antibody. Am J Pathol. 1994;144:986–96.
Biesecker LG, Harrison SM.ClinGen Sequence Variant Interpretation Working Group The ACMG/AMP reputable source criteria for the interpretation of sequence variants. Genet Med. 2018;20:1687–8.
Richards CS, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, et al. Response to Biesecker and Harrison. Genet Med. 2018;20:1689–90.
Kashtan CE, Kleppel MM, Gubler MC. Immunohistologic findings in Alport syndrome. Contrib Nephrol. 1996;117:142–53.
Gubler MC, Knebelmann B, Beziau A, Broyer M, Pirson Y, Haddoum F, et al. Autosomal recessive Alport syndrome: immunohistochemical study of type IV collagen chain distribution. Kidney Int. 1995;47:1142–7.
Kamura M, Yamamura T, Omachi K, Suico M, Nozu K, Kaseda S, et al. Trimerization and genotype-phenotype correlation of COL4A5 mutants in Alport syndrome. Kidney Int Rep. 2020;5:718–26.
Horinouchi T, Nozu K, Yamamura T, Minamakawa S, Omori T, Nakanishi K, et al. Detection of splicing abnormalities and genotype-phenotype correlation in X-linked Alport syndrome. J Am Soc Nephrol. 2018;29:2244–54.
Wang YF, Ding J, Wang F, Bu D. Effect of glycine substitutions on alpha5(IV) chain structure and structure-phenotype correlations in Alport syndrome. Biochem Biophys Res Commun. 2004;316:1143–9.
Yeo J, Qiu Y, Jung GS, et al. Adverse effects of Alport syndrome-related Gly missense mutations on collagen type IV: Insights from molecular simulations and experiments. Biomaterials. 2020;240:119857.
Wang Y, Zhang H, Ding J. Correlation between mRNA expression level of the mutant COL4A5 gene and phenotypes of XLAS females. Exp Biol Med. 2007;232:638–42.
Hudson BG, Reeders ST, Tryggvason K. Type IV collagen: structure, gene organization, and role in human diseases. Molecular basis of Goodpasture and Alport syndromes and diffuse leiomyomatosis. J Biol Chem. 1993;268:26033–6.
Vanacore R, Pedchenko V, Bhave G, Hudson B. Sulphilimine cross-links in Goodpasture’s disease. Clin Exp Immunol. 2011;164:4–6.
Zhou J, Hertz JM, Leinonen A, Tryggvason K. Complete amino acid sequence of the human alpha 5 (IV) collagen chain and identification of a single-base mutation in exon 23 converting glycine 521 in the collagenous domain to cysteine in an Alport syndrome patient. J Biol Chem. 1992;267:12475–81.
Vanacore R, Ham AJ, Voehler M, Sanders C, Conrads T, Veenstra T, et al. A sulfilimine bond identified in collagen IV. Science. 2009;325:1230–4.
Gunwar S, Ballester F, Noelken ME, Sado Y, Ninomiya Y, Hudson B. Glomerular basement membrane. Identification of a novel disulfide-cross-linked network of alpha3, alpha4, and alpha5 chains of type IV collagen and its implications for the pathogenesis of Alport syndrome. J Biol Chem. 1998;273:8767–75.
Beck K, Chan VC, Shenoy N, Kirkpatrick A, Ramshaw J, Brodsky B. Destabilization of osteogenesis imperfecta collagen-like model peptides correlates with the identity of the residue replacing glycine. Proc Natl Acad Sci USA. 2000;97:4273–8.
Yang W, Battineni ML, Brodsky B. Amino acid sequence environment modulates the disruption by osteogenesis imperfecta glycine substitutions in collagen-like peptides. Biochemistry. 1997;36:6930–5.
Demosthenous P, Voskarides K, Stylianou K, Hadjigavriel M, Arsali M, Patsias C, et al. X-linked Alport syndrome in Hellenic families: phenotypic heterogeneity and mutations near interruptions of the collagen domain in COL4A5. Clin Genet. 2012;81:240–8.
Parkin JD, San Antonio JD, Pedchenko V, Hudson B, Jensen S, Savige J. Mapping structural landmarks, ligand binding sites, and missense mutations to the collagen IV heterotrimers predicts major functional domains, novel interactions, and variation in phenotypes in inherited diseases affecting basement membranes. Hum Mutat. 2011;32:127–43.
Zurowska AM, Bielska O, Daca-Roszak P, Jankowski M, Szczepanska M, Roszkowska-Bjanid D, et al. Mild X-linked Alport syndrome due to the COL4A5 G624D variant originating in the Middle Ages is predominant in Central/East Europe and causes kidney failure in midlife. Kidney Int. 2020. https://doi.org/10.1016/j.kint.2020.10.040
Pierides A, Voskarides K, Kkolou M, Hadjigavriel M. Deltas C. X-linked, COL4A5 hypomorphic Alport mutations such as G624D and P628L may only exhibit thin basement membrane nephropathy with microhematuria and late onset kidney failure. Hippokratia. 2013;17:207–13.
Macheroux EP, Braunisch MC, Pucci Pegler S, Satanovskij R, Riedhammer K, Gunthner R, et al. The Hypomorphic Variant p.(Gly624Asp) in COL4A5 as a possible cause for an unexpected severe phenotype in a family with X-linked Alport syndrome. Front Pediatr. 2019;7:485.
Kashtan CE, Ding J, Gregory M, Gross O, Heidet L, Knebelman P, et al. Clinical practice recommendations for the treatment of Alport syndrome: a statement of the Alport Syndrome Research Collaborative. Pediatr Nephrol. 2013;28:5–11.
Zhang Y, Ding J, Wang S, Zhang H, Zhong H, Liu X, et al. Reassessing the pathogenicity of c.2858G>T(p.(G953V)) in COL4A5 gene: report of 19 Chinese families. Eur J Hum Genet. 2020;28:244–52.
Mariyama M, Leinonen A, Mochizuki T, Tryggvason K, Reeders S. Complete primary structure of the human alpha 3(IV) collagen chain. Coexpression of the alpha 3(IV) and alpha 4(IV) collagen chains in human tissues. J Biol Chem. 1994;269:23013–7.
Leinonen A, Mariyama M, Mochizuki T, Tryggvason K, Reeders S. Complete primary structure of the human type IV collagen alpha 4(IV) chain. Comparison with structure and expression of the other alpha (IV) chains. J Biol Chem. 1994;269:26172–7.
The following authors of this publication (BSLZ, RC, LM, AvE, KC, EA) are members of the European Reference Network for Rare Kidney Diseases (ERKNet)– Project ID No 739532. This research has been supported not financially by “European Reference Network for Rare Kidney Disease, ERKNet”. This ERN is partly co-funded by the European Union within the framework of the Third Health Programme “ERN- 2016 - Framework Partnership Agreement 2017-2021”.
Conflict of interest
The authors declare no competing interests.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The original online version of this article was revised: Update in the Acknowledgment section.
Rights and permissions
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this license, visit http://creativecommons.org/licenses/by/4.0/.
About this article
Cite this article
Savige, J., Storey, H., Watson, E. et al. Consensus statement on standards and guidelines for the molecular diagnostics of Alport syndrome: refining the ACMG criteria. Eur J Hum Genet 29, 1186–1197 (2021). https://doi.org/10.1038/s41431-021-00858-1
This article is cited by
Allelic prevalence and geographic distribution of cerebrotendinous xanthomatosis
Orphanet Journal of Rare Diseases (2023)
Pathogenic variant detection rate by whole exome sequencing in Thai patients with biopsy-proven focal segmental glomerulosclerosis
Scientific Reports (2023)
Variants in genes coding for collagen type IV α-chains are frequent causes of persistent, isolated hematuria during childhood
Pediatric Nephrology (2023)
Approach to genetic testing to optimize the safety of living donor transplantation in Alport syndrome spectrum
Pediatric Nephrology (2022)
The 2019 and 2021 International Workshops on Alport Syndrome
European Journal of Human Genetics (2022)