Alport syndrome (AS) is a hereditary kidney disorder of type IV collagen caused by pathogenic variants in the COL4A3, COL4A4, and COL4A5 genes. Previously several cases of digenic AS, caused by two pathogenic variants in two of the three COL4A genes, have been reported. Patients with digenic AS may present with a more severe phenotype compared to patients with single variants, depending on the percentage affected type IV trimeric collagen chain. We report a newly discovered case of trigenic AS. A 52-year-old female presented with hematuria at the age of 24 years and developed hypertension by the age of 30. Over the years, she developed chronic kidney disease; the most recent eGFR was 44 mL/min/1.73 m2. She has symmetric high-tone sensorineural hearing loss. Full genetic analysis revealed a heterozygous pathogenic variant c.2691del in COL4A3, a heterozygous pathogenic variant c.1663dup in COL4A4, and a complete heterozygous deletion of COL4A5. We describe the first patient with AS caused by pathogenic variants in all three COL4A genes, designated trigenic AS. This case report emphasizes the importance of examining all three COL4A genes, even in patients with a mild Alport phenotype, for optimal follow-up of the patient and adequate genetic counseling of family members.

Alport syndrome (AS) is an inherited disorder of type IV collagen expressed in the basement membranes of kidney glomeruli, eyes and cochleae. AS is caused by (likely) pathogenic variants in the COL4A genes: COL4A3, COL4A4, and COL4A5. Variants in the COL4A5 gene, located on the X-chromosome, cause X-linked AS (XLAS). Variants in COL4A3 and COL4A4, located head-to-head on chromosome 2, cause autosomal recessive or dominant AS [1, 2].

Conventionally, genetic analysis in patients with AS was performed using Sanger sequencing, which was time-consuming given the extensive size of the genes (COL4A3 52 exons, COL4A4 48 exons, and COL4A5 53 exons, respectively). Furthermore, a sequential gene analysis was terminated when a pathogenic variant was detected. Nowadays, next-generation sequencing (NGS) is the most efficient method for analyzing the COL4A genes collectively, detecting all variants present. Since NGS may fail to detect exon deletions and duplications, additional multiplex ligation-dependent probe amplification is performed [3]. Implementation of NGS resulted in the discovery of digenic AS: pathogenic variants in two of the COL4A genes.

In digenic AS, all possible combinations have been described and the phenotype is more severe compared to patients with one heterozygous variant in COL4A3 or COL4A4 [4‒6]. When variants are configured in cis (i.e., located on the same allele), this mimics autosomal dominant inheritance, however, with a more severe phenotype. Two variants in trans (i.e., located on different alleles) mimic autosomal recessive inheritance, though with a less severe phenotype for the offspring [4‒7]. Based on the population frequency of COL4A pathogenic variants, we expect more patients to be identified with digenic AS in the future using the current genetic analyses [8].

We present a case in which genetic analyses revealed pathogenic variants in all three COL4A genes. This case emphasizes the current guideline advice to analyze all three COL4A genes in patients suspected with AS [7].

A 52-year-old female (shown in Fig. 1; III:2 and referred to as index case) was referred for evaluation of her kidney disease. She was diagnosed with hematuria at age 24 and developed hypertension by the age of 30. During pregnancy (at 35 years of age) she was diagnosed with preeclampsia and nephrotic syndrome with severe proteinuria (up to 17 grams/day) for over 6 weeks. Post pregnancy the proteinuria decreased to levels between 0.5 and 1.5 g/day with ACE inhibitor treatment.

Fig. 1.

Family pedigree – family members with identified variants and their clinical phenotype del, deletion; dup, duplication; n, normal/no pathogenic variant detected; ?, individuals not genetically tested. The arrow indicates the index case.

Fig. 1.

Family pedigree – family members with identified variants and their clinical phenotype del, deletion; dup, duplication; n, normal/no pathogenic variant detected; ?, individuals not genetically tested. The arrow indicates the index case.

Close modal

The index case developed postpartum cardiomyopathy (with left ventricle dilatation), which recovered over the years. She underwent a gastric bypass at the age of 43 to treat obesity. She suffers from high-tone sensorineural hearing loss and has been wearing hearing aids since the age of 36.

She experienced rapid loss of kidney function in the period between 2006 and 2013 as shown in online supplementary File 1 (for all online suppl. material, see https://doi.org/10.1159/000538587). However, the rate of kidney function decline improved after recovery from other medical conditions (obesity, preeclampsia, and cardiomyopathy). Currently, chronic kidney disease (CKD) is classified as stage G3bA2 with an eGFR of 44 mL/min/1.73 m2 and an albumin/creatinine ratio of 46 mg/g with ACE inhibitor therapy. X-linked AS was suspected based on the disease symptoms (Table 1), and genetic analysis was performed.

Table 1.

Clinical characteristics of patients

PatientAge, yearsSexFirst presentationMost recent follow-upExtrarenal featuresGenetics
age, yearspresenting symptomscreatinine, mg/dLeGFRa, ml/min/1.73 m2hematuriaproteinuriamedicationhearing lossocular lesionsgeneexonnucleotide changeamino acid changemutation type
III:2 (index case) 52 24 Hematuria 1.40 44 Yes ACR 46 mg/g Lisinopril 1dd10 mg Yes (36 yrs) No COL4A3 17 c.2691del p. (Gly898Glufs*26) Frameshift 
COL4A4 23 c.1663dup p. (Ala555Glyfs*4) Frameshift 
COL4A5 1–53b c.(?_-1)_(*1_?)del p.0 Deletion 
IV:1 16 10 Hypertension 0.85 >90 No PCR 71 mg/g Enalapril 2dd5mg No NA COL4A3 17 c.2691del p. (Gly898Glufs*26) Frameshift 
II:2 75 63 Hypertensionhematuria, proteinuria 0.71 84 Yes ACR 25.7 mg/g Losartan 1dd 100 mg No NA COL4A4 23 c.1663dup p. (Ala555Glyfs*4) Frameshift 
II:1 Deceased  No renal phenotype, no hearing loss. Deceased at 48 years (cardiac event)         
III:4 50 No renal phenotype, no hearing loss. Denied (genetic) evaluation         
PatientAge, yearsSexFirst presentationMost recent follow-upExtrarenal featuresGenetics
age, yearspresenting symptomscreatinine, mg/dLeGFRa, ml/min/1.73 m2hematuriaproteinuriamedicationhearing lossocular lesionsgeneexonnucleotide changeamino acid changemutation type
III:2 (index case) 52 24 Hematuria 1.40 44 Yes ACR 46 mg/g Lisinopril 1dd10 mg Yes (36 yrs) No COL4A3 17 c.2691del p. (Gly898Glufs*26) Frameshift 
COL4A4 23 c.1663dup p. (Ala555Glyfs*4) Frameshift 
COL4A5 1–53b c.(?_-1)_(*1_?)del p.0 Deletion 
IV:1 16 10 Hypertension 0.85 >90 No PCR 71 mg/g Enalapril 2dd5mg No NA COL4A3 17 c.2691del p. (Gly898Glufs*26) Frameshift 
II:2 75 63 Hypertensionhematuria, proteinuria 0.71 84 Yes ACR 25.7 mg/g Losartan 1dd 100 mg No NA COL4A4 23 c.1663dup p. (Ala555Glyfs*4) Frameshift 
II:1 Deceased  No renal phenotype, no hearing loss. Deceased at 48 years (cardiac event)         
III:4 50 No renal phenotype, no hearing loss. Denied (genetic) evaluation         

Conversion factors for units: creatinine mg/dL to μmol/L, ×88.4; albumin/creatinine ratio mg/g to mg/mmol, ×0.113; protein/creatinine ratio in mg/g to mg/mmol, ×0.113.

F, female; M, male; ACR, albumin/creatinine ratio; PCR, protein/creatinine ratio; NA, not available.

aeGFR was estimated from CKD-EPI for age >18 years old and Bedside Schwartz formula for age <18 years old.

bIncluding at least exon 1–4 of COL4A6.

Genetic Analysis

Analysis of index case III:2’s DNA with NGS enhanced with single molecule molecular inversion probes, revealed two heterozygous pathogenic variants: c.2691del in COL4A3 and c.1663dup in COL4A4. These variants both predict a frameshift, possibly resulting in truncated COL4A3 (p.(Gly898Glufs*26)) and COL4A4 proteins (p.(Ala555Glyfs*4)), respectively. Both variants have not been described in the literature or gnomAD database. In the Dutch Leiden Open Variation Database (LOVD), the COL4A3 variant was submitted once (by our center) and the COL4A4 variant was not submitted before [9]. Furthermore, CNV analysis by single molecule molecular inversion probes suggested a heterozygous deletion of the entire COL4A5 gene expanding to at least exons 1–4 of COL4A6, which was confirmed by multiplex ligation-dependent probe amplification. A further description of the methods is given in online supplementary File 2.

Family History and Segregation Analysis

The father (II:1) died suddenly at the age of 48 years, probably due to a heart attack. The father was not known with hypertension or a kidney disorder and did not suffer from hearing loss Fig 1.

The 75-year-old mother (II:2) has hypertension, hematuria, and microalbuminuria. She has been treated with RAAS blockade since the age of 65. Her kidney function has remained stable for the past 10 years (CKD stage G2A1), the most recent eGFR is 84 mL/min/1.73 m2. Genetic analysis only revealed the heterozygous pathogenic variant c.1663dup in the COL4A4 gene.

The sister (III:4) does not have any signs of a kidney disorder or hearing disorder. She denied any (genetic) evaluation.

The son of the index case (IV:1) was referred to the pediatrician for evaluation of hypertension at the age of 10. His BMI was 15.8 kg per m2 and the hypertension was attributed to the use of methylphenidate [10]. Remarkably, there was variable proteinuria (varying from absent up to 2,067 mg/g creatinine) with normal serum albumin and minimal microscopic hematuria (maximum of 3–10 erythrocytes per field of view). His kidney function was normal (eGFR 120 mL/min/1.73 m2). After initiation of ACE inhibitor treatment, there is no persistent hematuria or proteinuria currently at the age of 16. This suggests a mild phenotype, with temporary pressure-dependent proteinuria during intercurrent events such as infections. Genetic analysis revealed the heterozygous pathogenic variant c.2691del in the COL4A3 gene.

We describe a female patient with AS and pathogenic variants in all three COL4A3, COL4A4, and COL4A5 genes. We designated this to be trigenic AS. Based on the population frequency of a digenic predicted pathogenic COL4A3 plus a COL4A4 variant of 1:44,793 and 1:245,920 for combinations of COL4A5 plus COL4A3 or COL4A4, trigenic AS must be extremely rare [8].

We were unable to investigate the inheritance of the three respective variants in the index case (III:2) in full detail. Because we only detected the COL4A3 variant in the son (IV:1), the COL4A3 and COL4A4 variants are inherited in trans. The COL4A4 variant is maternal (II:2); the COL4A3 variant is likely paternal (II:1). However, since the father passed away at an early age this could not be genetically confirmed. He was not reported as hypertensive or having hematuria, proteinuria or kidney failure, or deafness at the time of death, suggesting the COL4A5 deletion in the index patient is likely de novo. Yet, there is also a slight possibility of paternal (or maternal) postzygotic gonadal mosaicism of the COL4A5 deletion.

Deletions of COL4A5 extending to COL4A6 are associated with diffuse leiomyomatosis-AS (DL-AS). The deletion detected in the index case extends beyond exon 4 of COL4A6 and she does not have symptoms of DL-AS. This is in line with previously reported cases showing deletions of exon 1 and 2 of the COL4A6 gene cause DL-AS, while further expanding deletions do not lead to DL-AS. The exact pathogenesis of this phenomenon is unclear [11, 12].

In our index patient, the kidney function and proteinuria recovered over time and the decline in kidney function has slowed down ever since (now CKD stage G3bA2 at 52 years of age). In our view, this illustrates the contribution of lifestyle factors to CKD. The renal phenotype of AS appears relatively mild. Longer follow-up of this index case is needed to determine if kidney function will deteriorate to end-stage kidney disease in the future.

We can speculate on the possible genotype-phenotype correlation for all three pathogenic variants. Heterogenous phenotype has been described in females with heterozygous COL4A5 variants, heterozygous COL4A3 or COL4A4 variants, and in digenic AS. Disease severity spans the entire spectrum between normal phenotype and progression to end-stage kidney disease [13, 14]. Asymptomatic cases are reported in 2.5% of females with XLAS [13] and 5.2% of individuals with heterozygous COL4A3 or COL4A4 variants [14]. In the population-based Geisinger MyCode/DiscovEHR study the percentage of asymptomatic individuals was even higher; 36% for pathogenic variants in COL4A3 [15]. In digenic AS including a COL4A5 variant, although all women had hematuria, proteinuria was absent in 38% of them. Similarly, proteinuria was absent in 34% of cases of digenic AS with COL4A3 and COL4A4 variants [6]. The reported cases of digenic AS indicate that the phenotype is, in general, more severe compared to individuals with a single variant [6, 14]. Thus, we would expect the phenotype in trigenic AS to be even more severe than in digenic AS. However, the phenotype of our case matches the spectrum of phenotypes described for digenic AS (due to variants in COL4A3 and COL4A4), except our case presented with proteinuria at an earlier age than described in the literature for digenic AS due to variants in COL4A3 and COL4A4 and later than the median age described for digenic AS including a variant in COL4A5 [6].

We suggest several explanations for this mild phenotype. First, we hypothesize that heterozygosity of respective COL4A3, COL4A4, and COL4A5 variants explains the relatively mild phenotype [6]. Based on the calculated prediction and considering a 50:50 X-inactivation, we would expect 87.5% of triple helix collagen molecules to be defective. However, the contribution of the deletion of COL4A5 may be limited due to possible skewing of X-inactivation [16]. The percentage of residual functional triple helix collagen may therefore be >12.5%. The contribution of the COL4A5 variant is also potentially weak for the extrarenal phenotype: hearing loss is also described in patients with digenic AS (due to variants in COL4A3 and COL4A4) and also, but less frequent, in patients with heterozygous COL4A3 or COL4A4 variants [6, 14].

Also, the type of mutations could contribute to a more favorable phenotype. The heterozygous complete deletion of the COL4A5 gene results either in normal expression of the α5(IV) chain or in absent expression in cells. Jais et al. [13] described a large group of women with XLAS, including 12 families with large deletions of COL4A5 involving the COL4A6 gene, and 10 patients exhibiting DL-AS. Three of these individuals had normal urinalysis. The COL4A3 and COL4A4 pathogenic variants in our index patient were both probable protein-truncating variants. In the Geisinger MyCode/DiscovEHR study, only 52% of the patients with a protein-truncating variant in COL4A3 had any phenotypic feature, showing that protein-truncating variants are associated with a milder phenotype compared to glycine substitutions. The absence of disease symptoms in patients with protein-truncating mutations may be explained by production of functional collagen heteromers by the non-truncated allele [15].

In the literature, genetic modifiers in AS have been described [7]. Our patient also illustrates the role of lifestyle factors since proteinuria was reduced with treatment of blood pressure using ACE inhibitors, and weight reduction after bariatric surgery for obesity. The history of the son illustrates that hypertension and infections could have a modifying effect. This suggests that patients with collagenopathies could have a mild phenotype but show increased vulnerability of the glomerular basement membrane to intercurrent events, dependent on the percentage of residual functional triple helix collagen.

Although trigenic pathogenic variants have been described before in a family with hypertrophic cardiomyopathy, we describe the first case of AS in the literature with trigenic pathogenic variants [17]. There is an earlier report by Zhang et al. [18] describing an individual with three variants. However, the described variant in COL4A3 should be interpreted as a variant of unknown significance based on amino acid conservation between species and paralogous type IV collagen NC1 domains [19], allele frequency in the population (dbSNP, gnomAD) and the LOVD, so this case does not meet the criteria for trigenic AS.

Our findings are relevant for nephrologists, clinical geneticists, and patients with AS to be aware of the possibility of trigenic variants. It emphasizes the importance of examining all three COL4A genes in patients with suspected AS, even when a familial pathogenic variant is identified before. It is also important to identify the nature of the variants, e.g., inheritance in cis or trans, as this helps determine the risk of inheritance and prognosis for family members [5, 6].

In conclusion, this case report expands knowledge of the genetic diagnosis of AS, revealing the possibility of trigenic pathogenic variants. It is crucial to examine all three COL4A genes, even in patients with a relatively mild Alport phenotype, for optimal follow-up of the patient and counseling of family members.

In accordance with Dutch law, this Case Report did not require formal approval by the Medical Ethics Committee. This was confirmed by our hospital’s human research committee. The authors declare that they have obtained written consent from the patients reported in this Case Report. Consent was obtained from the participants for publication of the details of their medical case that appears within this Case Report and associated supplementary material.

The authors have no conflicts of interest to declare.

This work was supported by the Dutch Kidney Foundation (PhD student Grant 19OP020).

Data collection: Dipti Rao, Marlies Cornelissen, and Michel van Geel. Genetic analysis and interpretation: Michel van Geel. Clinical data interpretation: Dipti Rao, Rutger Maas, Marlies Cornelissen, and Jack Wetzels. Drafting the article: Dipti Rao. Review and revisions: Dipti Rao, Rutger Maas, Marlies Cornelissen, Jack Wetzels, and Michel van Geel. Reading and approving the final manuscript: Dipti Rao, Rutger Maas, Marlies Cornelissen, Jack Wetzels, and Michel van Geel.

The data used to support the findings of this case report are available from the corresponding author upon reasonable request.

1.
Mochizuki
T
,
Lemmink
HH
,
Mariyama
M
,
Antignac
C
,
Gubler
M-C
,
Pirson
Y
, et al
.
Identification of mutations in the α3(IV) and α4(IV) collagen genes in autosomal recessive Alport syndrome
.
Nat Genet
.
1994
;
8
(
1
):
77
82
.
2.
Nozu
K
,
Nakanishi
K
,
Abe
Y
,
Udagawa
T
,
Okada
S
,
Okamoto
T
, et al
.
A review of clinical characteristics and genetic backgrounds in Alport syndrome
.
Clin Exp Nephrol
.
2019
;
23
(
2
):
158
68
.
3.
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
(
7
):
1175
89
.
4.
Fallerini
C
,
Baldassarri
M
,
Trevisson
E
,
Morbidoni
V
,
La Manna
A
,
Lazzarin
R
, et al
.
Alport syndrome: impact of digenic inheritance in patients management
.
Clin Genet
.
2017
;
92
(
1
):
34
44
.
5.
Mencarelli
MA
,
Heidet
L
,
Storey
H
,
van Geel
M
,
Knebelmann
B
,
Fallerini
C
, et al
.
Evidence of digenic inheritance in Alport syndrome
.
J Med Genet
.
2015
;
52
(
3
):
163
74
.
6.
Savige
J
,
Renieri
A
,
Ars
E
,
Daga
S
,
Pinto
AM
,
Rothe
H
, et al
.
Digenic Alport syndrome
.
Clin J Am Soc Nephrol
.
2022
;
17
(
11
):
1697
706
.
7.
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
54
.
8.
Gibson
J
,
Fieldhouse
R
,
Chan
MMY
,
Sadeghi-Alavijeh
O
,
Burnett
L
,
Izzi
V
, et al
.
Prevalence estimates of predicted pathogenic col4a3-col4a5 variants in a population sequencing Database and their implications for Alport syndrome
.
J Am Soc Nephrol
.
2021
;
32
(
9
):
2273
90
.
9.
Fokkema
IFAC
,
Kroon
M
,
López Hernández
JA
,
Asscheman
D
,
Lugtenburg
I
,
Hoogenboom
J
, et al
.
The LOVD3 platform: efficient genome-wide sharing of genetic variants
.
Eur J Hum Genet
.
2021
;
29
(
12
):
1796
803
.
10.
Buitelaar
JK
,
van de Loo-Neus
GHH
,
Hennissen
L
,
Greven
CU
,
Hoekstra
PJ
,
Nagy
P
, et al
.
Long-term methylphenidate exposure and 24-hours blood pressure and left ventricular mass in adolescents and young adults with attention deficit hyperactivity disorder
.
Eur Neuropsychopharmacol
.
2022
;
64
:
63
71
.
11.
Nozu
K
,
Minamikawa
S
,
Yamada
S
,
Oka
M
,
Yanagita
M
,
Morisada
N
, et al
.
Characterization of contiguous gene deletions in COL4A6 and COL4A5 in Alport syndrome-diffuse leiomyomatosis
.
J Hum Genet
.
2017
;
62
(
7
):
733
5
.
12.
Zhou
X
,
Wang
J
,
Mao
J
,
Ye
Q
.
Clinical manifestations of Alport syndrome-diffuse leiomyomatosis patients with contiguous gene deletions in COL4A6 and COL4A5
.
Front Med
.
2021
;
8
:
766224
.
13.
Jais
JP
,
Knebelmann
B
,
Giatras
I
,
De Marchi
M
,
Rizzoni
G
,
Renieri
A
, et al
.
X-linked Alport syndrome: natural history and genotype-phenotype correlations in girls and women belonging to 195 families: a “European Community Alport Syndrome Concerted Action” study
.
J Am Soc Nephrol
.
2003
;
14
(
10
):
2603
10
.
14.
Matthaiou
A
,
Poulli
T
,
Deltas
C
.
Prevalence of clinical, pathological and molecular features of glomerular basement membrane nephropathy caused by COL4A3 or COL4A4 mutations: a systematic review
.
Clin Kidney J
.
2020
;
13
(
6
):
1025
36
.
15.
Solanki
KV
,
Hu
Y
,
Moore
BS
,
Abedi
V
,
Avula
V
,
Mirshahi
T
, et al
.
The phenotypic spectrum of COL4A3 heterozygotes
.
medRxiv
.
2023
;
8
(
10
):
2088
99
.
16.
Rheault
MN
.
Women and Alport syndrome
.
Pediatr Nephrol
.
2012
;
27
(
1
):
41
6
.
17.
Chen
Y
,
Barajas-Martinez
H
,
Zhu
D
,
Wang
X
,
Chen
C
,
Zhuang
R
, et al
.
Novel trigenic CACNA1C/DES/MYPN mutations in a family of hypertrophic cardiomyopathy with early repolarization and short QT syndrome
.
J Transl Med
.
2017
;
15
(
1
):
78
.
18.
Zhang
Y
,
Ding
J
,
Zhang
H
,
Yao
Y
,
Xiao
H
,
Wang
S
, et al
.
Effect of heterozygous pathogenic COL4A3 or COL4A4 variants on patients with X-linked Alport syndrome
.
Mol Genet Genomic Med
.
2019
;
7
(
5
):
e647
.
19.
Netzer
KO
,
Suzuki
K
,
Itoh
Y
,
Hudson
BG
,
Khalifah
RG
.
Comparative analysis of the noncollagenous NC1 domain of type IV collagen: identification of structural features important for assembly, function, and pathogenesis
.
Protein Sci
.
1998
;
7
(
6
):
1340
51
.