https://orcid.org/0000-0002-6813-0288
Figures
Abstract
Autosomal dominant polycystic liver disease (ADPLD) commonly results from pathogenic variants in one of 7 genes (GANAB, ALG8, ALG9, PRKCSH, SEC63 and possibly LRP5 and SEC61B). These genes are also included in kidney panels because pathogenic variants cause kidney cysts but rarely kidney failure. This study determined the population frequency of predicted pathogenic variants in the ADPLD genes in the general population. Variants for each gene were downloaded from gnomADv.2.1.1 and annotated with ANNOVAR. The population frequencies were calculated from the number of people with ‘predicted pathogenic’ variants in gnomAD: loss-of-function structural and copy number; null; and rare, computationally-damaging missense changes that affected a conserved residue. Frequencies were also estimated from the number of gnomADv.4.1 variants assessed as Pathogenic or Likely pathogenic in ClinVar. Predicted pathogenic variants affected one in 91 people using our strategy and gnomAD v.2.1.1, and one in 130 with ClinVar assessments of gnomAD v.4.1 variants. LRP5 and ALG8 which are associated with a milder clinical phenotype, were the commonest affected genes with both strategies. Predicted pathogenic variants in ADPLD appear more frequent in admixed American (one in 91), Finnish (one in 110) and African/African American (one in 43) people (p all < 0.0001 compared with Europeans (one in 187). Predicted pathogenic variants for ADPLD may be even more common with the finding of additional causative genes. Assessments will become more accurate too. However not all ADPLD variants result in liver cysts, nor indeed cystic kidneys, because of incomplete penetrance and variable expressivity.
Citation: Varughese S, Huang M, Savige J (2026) The population frequency of predicted pathogenic variants in the genes associated with Autosomal Dominant Polycystic Liver Disease (ADPLD) and kidney cysts. PLoS One 21(9): e0348109. https://doi.org/10.1371/journal.pone.0348109
Editor: Klaus Brusgaard, Odense University Hospital, DENMARK
Received: April 10, 2026; Accepted: August 4, 2026; Published: September 28, 2026
Copyright: © 2026 Varughese et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: 1. The minimal data set is available in the manuscript or in the supplementary file.
Funding: JS, Polycystic Kidney Disease Australia, https://pkdaustralia.org/, The funders did not play any role in the study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Introduction
Autosomal Dominant Polycystic Liver Disease (ADPLD) overlaps genetically and clinically with Autosomal Dominant Polycystic Kidney Disease (ADPKD) [1]. Multiple liver cysts occur in 80% of people with ADPKD due to PKD1 and PKD2 variants [2], especially women [3], but the cysts are often smaller and less numerous than those in ADPLD [4]. Liver cysts are also associated with autosomal recessive (AR) PKD caused by PKHD1 [5] and DZIP1L [6] changes.
ADPLD results from pathogenic variants in PRKCSH, SEC63, ALG8, ALG9, GANAB and possibly LRP5 and SEC61B[7–9], which together account for at least 40% of people with ADPLD (Table 1) [8]. We have previously reported the population frequency of ALG9 [10]. These genes typically encode proteins involved in the transport of fluid or epithelial cell growth in the liver [11]. Some affect the glycosylation, folding or trafficking of ER proteins that process the polycystin 1 (PKD1) molecule that is affected in ADPKD [12,13].
PRKCSH variants are often associated with isolated PLD [14], and, in general, PRKCSH, SEC63, and GANAB result in a more severe phenotype than ALG8 and LRP5. All these genes are also associated with multiple kidney cysts, and ALG8 and GANAB variants have previously been reported to be the commonest ADPLD-related causes of kidney cysts [7]. However the ADPLD-associated genes rarely result in kidney failure [1].
These 7 genes are all associated with AD inheritance, and, while, for some genes, few variants have been reported, currently nonsense, splice site, and missense changes as well as insertions and deletions, have been described throughout most genes without obvious hotspots [11] (Fig 1, Simple ClinVar, https://simple-clinvar.broadinstitute.org/). Penetrance is more variable than for ADPKD, even within a family, and there appear to be no obvious genotype–phenotype correlations [15,16].
There were no Pathogenic or Likely pathogenic variants for SEC61B in Simple ClinVar (https://simple-clinvar.broadinstitute.org/).
The reported population frequency of ADPLD varies depending on whether the diagnosis has been based on clinical features, imaging, genetics or from the association with AD or AR polycystic kidney disease. The formal diagnosis of ADPLD depends on the demonstration of at least 4 liver cysts and another affected family member, or ten cysts without a family history [17]. Population studies usually underestimate the disease burden because most polycystic liver disease is asymptomatic. Isolated polycystic liver disease has been reported in one in 100,000 people, and in fewer than one in 10,000 people using medical records and imaging reports from the Mayo clinic [18]. In one study, ADPLD affected one in 400 or one in 20,000 people when associated with ADPKD or ARPKD [5]. However the population frequency was one in 496 when truncating variants were counted in three genes (PRKCSH, SEC63, GANAB) from two large public sequencing databases (gnomAD v.2.0.2, BRAVO) [19].
A more accurate population frequency for ADPLD will increase clinician awareness of the condition and its association with ADPKD, improve genotype-phenotype correlations and encourage pharmaceutical companies to develop targeted treatments or at least monitor the response of liver cysts to therapies for cystic kidney disease.
The present study determined the population frequency of predicted pathogenic variants in seven genes affected in ADPLD from two minimally-overlapping population datasets (gnomAD v.2.1.1 and v.4.1). Our strategy used computational tools based on the American College of Medical Genetics and Genomics (ACMG)/Association for Molecular Pathology (AMP) criteria for pathogenicity [20]. The population frequency was calculated from the number of people with structural or copy number changes that were assessed as loss of function; together with null variants; and missense changes that were rare, structurally or functionally damaging in computational tools and affected a conserved amino acid, in gnomADv.2.1.1. This population frequency was then compared with those derived from variants found in gnomAD that had also been assessed disease-causing in the ClinVar (https://www.ncbi.nlm.nih.gov/clinvar/), LOVD (https://www.lovd.nl/) or HGMD (https://www.hgmd.cf.ac.uk/ac/index.php) databases. These results were then confirmed using ClinVar analysis of the larger gnomAD v.4.1 database.
Our approach for calculating the population frequency for ADPLD has been used for many other monogenic diseases including Alport syndrome, AD Polycystic kidney disease, Wilson disease, Menkes disease, Gitelman syndrome, Mucopolysaccharidoses, Fabry disease, and Lysosomal acid lipase deficiency [21–27], and, in some cases, these results have been confirmed independently with epidemiological, histological or biochemical studies [22,24].
Methods
Population datasets
Two datasets were examined: gnomADv.2.1.1 (GRCh37/hg19) and gnomAD v.4.1 (GRCh38/hg38). gnomADv.2.1.1 comprises anonymised Whole Exome Sequencing (WES, n = 141,456 samples) and Whole Genome Sequencing (WGS, n = 10,847 samples) from multiple large-scale sequencing projects of adults with diabetes, neuropsychiatric or cardiac disease and has equal numbers of men and women, together with their ancestries but not their clinical data. gnomAD v.2.1.1. also includes a Control subset (n = 60,146) of people who were age- and gender-matched for those with these diseases but were themselves unaffected.
gnomAD v.4.1 includes a larger number of samples (n = 807,162) with some examined for structural (n = 63,046) and copy number (n = 464,297) variants and with people of more diverse ancestries. gnomADv.4.1 comprises data from gnomAD v.2.1.1 (n = 314,392), mainly community-based people recruited to the UK Biobank (n = 492,770), and other sources. GnomAD v.2.1.1 and gnomAD v.4.1 variants overlap by about 20% of variants.
Variants were downloaded between August 2024 and December 2024, and reviewed in May 2025.
IRB approval
This study did not use human participants but instead examined anonymised genomic data from gnomAD. All gnomAD participants had previously consented for the use of their anonymized demographic and genomic data so that secondary Institutional Review Board approval was not required.
Population frequency of predicted pathogenic variants in the ADPLD genes using gnomAD v.2.1.1
The population frequency of ADPLD was calculated from variants in 7 genes (GANAB, ALG8, ALG9, LRP5, PRKCSH, SEC61B, SEC63) associated with ADPLD (https://curation.clinicalgenome.org/) [28]. These genes were chosen on the basis of the KDIGO guidelines [9] and a previous study [25]. Pathogenic variants in ALG9 are associated with both AD polycystic kidney and liver disease and we have published its population frequency previously. The genes were downloaded, annotated with ANNOVAR (https://annovar.openbioinformatics.org/), and assessed for pathogenicity as we have described [26,29].
Variants in the intronic and 5’ or 3’ UTR or that were synonymous were excluded. The remaining variants were filtered using the following strategy (Fig 2).
Structural variants.
Structural variants that were classified as loss of function by gnomAD and affected an exon were assessed as predicted pathogenic. The number of people with a predicted pathogenic variant in the smaller structural variant sample was corrected to be equivalent to the whole cohort. All predicted pathogenic variants were included in the total regardless of their allele frequency for structural, copy number and null variants. Structural variants were available for gnomAD v.2.1.1 and v.4.1.
Copy number variants.
Copy number variants that resulted in the deletion of a least one exon were assessed as predicted pathogenic. Again, the number of people with each predicted pathogenic variant was corrected for the smaller copy number sample size. Copy number variants were only available for gnomAD v.4.1.
Null variants.
Null variants, including terminating codons, splicing and frameshift changes (except those in the last exon or the last 50 nucleotides of the penultimate exon, which do not undergo nonsense-mediated decay), were assessed as predicted pathogenic. The canonical mRNA transcripts and the locations of the final nucleotide beyond which variants do not result in nonsense mediated decay were obtained from Alamut (https://www.sophiagenetics.com).
Missense variants.
Missense variants were classified as predicted pathogenic if they were rare (≤ 5) and disease-causing in the following three bioinformatic tools: PP2 (Polymorphism Phenotyping v2, PolyPhen-2) score ≥ 0.95 (http://genetics.bwh.harvard.edu/pph2/); SIFT4G (Sorting Intolerant From Tolerant) score ≤ 0.05 (https://sift.bii.a-star.edu.sg/; and MT (MutationTaster) where they were “disease causing, or probably deleterious” (D) or “disease-causing automatic and known to be deleterious” (A) (https://www.mutationtaster.org/info/documentation.html). Predicted pathogenic variants also affected an amino acid that was conserved in vertebrates (the same (*) or similar (:) in chicken, mice, humans, using Clustal Omega (https://www.ebi.ac.uk/Tools/) and the Ensembl reference sequences (http://asia.ensembl.org/index.html).
We chose to examine missense variants using multiple computational tools (PP2, SIFT, Mutation Taster, GERP) rather than metapredictors such as REVEL and BAYESdel, which both include these three tools and others. Metapredictors have the weakness that many of their tools measure the same variant feature and the performance of the better ones is diminished by other under-performers. We did not exclude variants that were homozygous in gnomAD (which is a common strategy to exclude benign changes) because our studies have found that homozygous pathogenic variants were unlikely to be included with a minor allele frequency of ≤ 5.
The accuracy of our missense variant assessment was evaluated as follows. Twenty Benign and 20 Pathogenic missense variants were selected randomly in genes for ADPLD from ClinVar (Suppl Table S1 in S1 Data), assessed by our criteria for missense changes, and the sensitivity and specificity, positive and negative predictive values calculated. Benign variants included 4 GANAB, 4 ALG8, 1 LRP5, 7 PRKSH and 4 SEC63 missense changes; and Pathogenic variants included 4 GANAB, 3 ALG8, 8 LRP5, and 5 SEC63 variants.
Calculation of the population frequency of predicted pathogenic variants in ADPLD genes.
The population frequencies of predicted pathogenic variants for individual genes were calculated from the total number of people with pathogenic variants in each gene divided by the mean number of people who had undergone sequencing for that gene in gnomAD. This presumed that each affected person had only one pathogenic change in an ADPLD gene. Biallelic variants have been reported in ADPLD [30], and a ‘two hit hypothesis’ with 2 pathogenic variants is consistent with the variability in the number and size of cysts [31]. It is possible but laborious to identify compound heterozygous variants in a single gene in gnomAD and it is not possible currently to identify digenic variants in the same person.
The population frequencies were also calculated from the number of predicted pathogenic variants in the Control subset (n = 60,146) in gnomADv.2.1.1.
Population frequencies of predicted pathogenic variants in the ADPLD genes in gnomAD v.2.1.1 using ClinVar, LOVD and HGMD
The population frequencies of predicted pathogenic variants in ADPLD genes were also estimated from gnomAD variants assessed as disease-causing in ClinVar (Pathogenic, Likely pathogenic or Conflicting (Variant of Uncertain Significance (VUS) plus Pathogenic or Likely pathogenic, but not Benign or Likely benign)) (https://www.ncbi.nlm.nih.gov/clinvar/) from LOVD (https://databases.lovd.nl/shared/genes); or from HGMD (https://www.hgmd.cf.ac.uk/ac/index.php) even if variants were found in more than 5 people. This approach was taken because assessments were made on sequencing from people tested by accredited diagnostic genetic laboratories, who used the ACMG/AMP criteria [20] and recent population datasets for rarity. While errors occur in database assessments, these have become fewer with time [32], and this strategy enabled us to detect common founder variants.
Population frequency for predicted pathogenic variants in the ADPLD genes using gnomAD v.4.1
The results from gnomAD v.2.1.1 were then compared with the replication dataset from the ClinVar assessment of variants in gnomAD v.4.1, which included structural and copy number changes.
Variants in individual ADPLD genes were again downloaded from gnomAD v.4.1, and checked for assessments in ClinVar. Variants that were reported as Pathogenic, Likely pathogenic, or Conflicting (VUS plus Pathogenic or Likely pathogenic) were assessed as predicted pathogenic even if they were found in more than 5 people.
The total number of people with a predicted pathogenic variant in gnomAD v.4.1 included those with a loss of function structural or copy number change that affected an exon where numbers from the subsets were corrected to be equivalent to the whole cohort, in addition to the changes assessed as disease-causing in ClinVar.
The population frequency for ADPLD was calculated from the total number of people with a predicted pathogenic variant in our assessment of gnomAD v.4.1 divided by the mean number of people who had undergone WES and WGS for the gene in gnomAD, and, again, presumed that each person had only one pathogenic variant in an ADPLD gene.
Population frequencies of predicted pathogenic variants in ADPLD genes in people of different ancestries
Population frequencies of predicted pathogenic variants were then derived from gnomAD v.4.1 for people of African/African American, Admixed/American, Ashkenazim, East Asian, Finnish, Middle Eastern, European (Non-Finnish), South Asian or Remaining ancestries. These estimates included structural and copy number variants.
Statistical analysis
Results were compared with Chi-squared calculations with Yates’ correction using Graphpad (https://www.graphpad.com/).
Results
Population frequency of predicted pathogenic variants in the ADPLD genes using gnomAD v.2.1.1
Population frequency of predicted pathogenic variants in the ADPLD genes using our strategy.
Overall, 18 of the 20 Benign variants were benign in our assessment, and 15 of the Pathogenic variants from ClinVar were predicted pathogenic. This corresponded to a 75% sensitivity, 90% specificity, and an 88% positive and 78% negative predictive value (S1 Table).
Overall, there were 644 predicted pathogenic variants in 1,268 people in the 7 ADPLD genes (GANAB, ALG8, ALG9, LRP5, PRKCSH, SEC61B, and SEC63) in 115,143 people corresponding to a population frequency of one in 91 for ADPLD (Table 2, Suppl Tables S2-S8 in S1 Data). The commonest affected genes were LRP5 (one in 274), and ALG8 (one in 277), and the least common was SEC61B (one in 12,146). No pathogenic structural variants were found in ALG9, GANAB, LRP5 or SEC61B, and predicted pathogenic missense variants were about as common (n = 675) as null (n = 537) changes overall.
Population frequency of predicted pathogenic variants in the ADPLD genes in the control cohort.
There were 548 people with a predicted pathogenic variant in the control cohort of 50,404 people corresponding to a population frequency of ADPLD of one in 92 people, which was not different from our result for the whole gnomAD v.2.1.1 cohort (p = 0.99) (Table 3). However the control cohort did not include structural or copy number changes, which suggested a population frequency that was even more greater.
Population frequency of predicted pathogenic variants in the ADPLD genes using ClinVar, LOVD or HGMD databases.
Overall, the population frequencies of disease-causing variants in the ADPLD genes in gnomAD v.2.1.1 assessed by other variant databases were one in 318 using ClinVar, one in 40 using HGMD, and one in 5,041 using LOVD (Table 3). No pathogenic variants were found in SEC61B using ClinVar, LOVD or HGMD.
Predicted pathogenic variants in the ADPLD genes using ClinVar in gnomAD4.1
The overall population frequency for ADPLD was one in 130 people according to the ClinVar assessments of variants in these 7 genes (GANAB, ALG8, ALG9, LRP5, PRKCSH, SEC63, SEC61B) (Table 4). Only copy number variants were found in SEC61B in gnomAD v.4.1 using ClinVar. Again null variants (n = 2,736) were about as common as missense changes (n = 2,616).
Again the commonest predicted pathogenic variants in the ADPLD genes were LRP5 (one in 324) and ALG8 (one in 614). SEC63 variants were also common (one in 674).
Population frequencies of predicted pathogenic variants in ADPLD genes in gnomADv.4.1 in different ancestries.
ADPLD appeared to be commonest in African/African-Americans (one in 43) and Admixed Americans (one in 91), compared with Europeans (one in 187, p < 0.0001) (Table 5).
Discussion
Our assessment found that the population frequency of predicted pathogenic variants for the ADPLD genes (GANAB, ALG8, ALG9, LRP5, PRKCSH, SEC61B, SEC63) was one in 91 using our assessment for the gnomADv.2.1.1 cohort, and one in 130 using ClinVar assessments of gnomADv.4.1. The difference is explained by the inclusion of structural and copy number changes in gnomAD4.1 but also the lack of ClinVar assessments for all gnomAD v.4.1 changes.
However genetic causes of liver cysts are probably even more common. Up to 80% of people with typical ADPKD have liver cysts [33] and we have used a similar strategy to demonstrate that typical ADPKD (PKD1, PKD2 variants) affects one in 314 people, which corresponds to liver cysts from these genes occurring in one in 378 people. This means that, overall, liver cysts from pathogenic variants in the ADPKD and ADPLD genes occur in at least one in 73 of the population. There are even more genetic causes of liver cysts such as the atypical forms of ADPKD (ALG9, DNAJB11, ALG5, IFT140, NEK8) that were not considered here and further still-to-be-identified genes.
There are also other non-genetic reasons why our population frequencies may be underestimates. gnomAD v.2.1.1 does not include data for structural and copy number changes in everyone, and our strategy was relatively insensitive for pathogenic missense changes (75%), and will have excluded founder variants because of the requirement that the missense allele frequency was ≤ 5.
Nevertheless not everyone with a predicted pathogenic variant in an ADPLD-associated gene develops liver cysts nor indeed kidney cysts because of incomplete penetrance and variable expressivity [19]. Missense variants which are typically associated with less penetrant and milder disease were common in all these genes except for SEC61B.
Assessments of both gnomAD v.2.1.1 and v.4.1 demonstrated that the commonest genes affected in polycystic liver disease were LRP5 and ALG8. SEC1B variants were rare, and their role in liver cyst development ‘limited’ according to the ClinVar gene-disease association framework. Indeed, ClinVar also considers the evidence for LRP in ADPLD to be limited. These results differ from earlier reports that suggested PRKCSH and SEC63 variants were the commonest disease-causing genes, accounting for up to one third of cases of cystic liver disease [16,34], and that LRP5 variants accounted for fewer than 1% [35]. However the early studies were based on small cohorts of patients with severe disease, while our population frequencies were derived from all pathogenic variants in large cohorts of normal people including some likely without liver cysts.
Most of the ADPLD genes considered here are also associated with kidney cysts. Pathogenic variants in ALG8, ALG9 and GANAB are commonly associated with kidney cysts, but variants in SEC63 less so. Variants in LRP5 and SEC61B may not be associated with kidney cysts at all. The number of ADPLD-associated kidney cysts is very variable and kidney failure is uncommon [36]. We have previously demonstrated that predicted pathogenic variants in the atypical ADPKD genes affect about one in 283 people. Since most predicted pathogenic variants in ADPLD genes are also associated with kidney cysts this means that the population frequency of polycystic kidney disease is even more common than one in 283.
The strengths of this study were the use of two assessment strategies that included our approach, and confirmation using variant databases, especially ClinVar, in a minimally-overlapping gnomAD dataset. Our approach’s main advantage was that it examined each variant in gnomAD v.2.1.1 using the ACMG/AMP criteria based on rarity and computationally-assessed pathogenicity. However individual variant assessments were less accurate because of the limitations of the computational tools, the lack of confirmatory clinical data, and because common founder variants may have been excluded.
On the other hand, individual variant assessments of pathogenicity with ClinVar are mostly accurate with 84% concordance in the assessments between laboratories [37] and improving over time [32] because they have been submitted from accredited genetic testing laboratories, and use ACMG/AMP criteria and rarity based on large datasets. For these reasons, all variants considered pathogenic by ClinVar were used to calculate the population frequencies in gnomADv.4.1, no matter how often they were found. However, ClinVar included no pathogenic SEC61B variants at all [38]. Importantly, ClinVar does not assess each gnomAD variant meaning that the corresponding population frequencies represent underestimates. Population frequencies derived from variants characterised as pathogenic in LOVD or HGMD share some of the limitations of the ClinVar assessments. They may also include more errors because many of their variants were reported prior to the availability of the ACMG/AMP criteria and of the large variant datasets used to evaluate rarity [37].
The study’s major limitations were the lack of clinical confirmation of variant pathogenicity, underestimation of structural and copy number changes, and errors, especially in our interpretation of missense changes.
On the other hand, the alternative strategy included all disease-causing variants found in ClinVar, LOVD or HGMD even when they were very common which may have inflated the corresponding population frequencies. ClinVar variants can still be inaccurate. These inaccuracies arise from outdated classifications, improved ACMG guidelines and subjective interpretations from different clinical testing laboratories.
Our strategy also required missense variants to be positive in all three computational tools (PP2, SIFT, Mutation Taster) rather than using a meta score such as REVEL or BAYESdel. Both REVEL and BAYESdel incorporates the same three tools that we used but the advantage of the better tools may be diminished by the less well- performing ones. In addition, our population frequencies assumed that each person had only one pathogenic variant in one gene but some people possibly had two disease-causing variants, even in different genes [30]. The two-hit hypothesis of cyst formation might explain the variation in penetrance and variability in clinical presentation [31].
We have estimated the population frequencies of predicted pathogenic variants as a surrogate for ADPLD but are uncertain of how often cystic liver disease results, because the penetrance of individual genes and, indeed, of types of variants in the ADPLD genes is not known. In general, the number of liver cysts and the number of kidney cysts are highly variable.
Two major efforts will improve the accuracy of these estimates of the prevalence of ADPLD and of associated kidney cysts. One is the use of biobanks that include both genomic variants and imaging data. The other is a systematic classification of all the variants in the ADPLD genes in gnomAD using the ACMG/AMP criteria and all available evidence including previously reported cases. However this will still include errors. Our approach and the use of ClinVar variants was less laborious but not flawless.
In conclusion, this study demonstrated that predicted pathogenic variants in the ADPLD genes are more common than suspected previously. These population frequencies will become more accurate as computational tools improve, and more ADPLD genes and founder variants identified. Variants in ADPLD genes may be reported when genetic testing for kidney cysts is performed or a broad genetic kidney panel is examined, and it is useful for renal physicians to be familiar with these genes, and their contribution to the kidney cysts seen in their patients.
Acknowledgments
We would like to thank gnomAD, Simple ClinVar, ClinVar, LOVD and HGMD, for the use of their databases and the patients who contributed their data to gnomAD v.2.1.1 and v.4.1, as well as the laboratories that designed and maintain the PolyPhen2, SIFT and Mutation Taster computational tools, ANNOVAR, and the Ensembl group.
References
- 1. Cornec-Le Gall E, Torres VE, Harris PC. Genetic complexity of autosomal dominant polycystic kidney and liver diseases. J Am Soc Nephrol. 2018;29(1):13–23. pmid:29038287
- 2. Arjune S, Todorova P, Bartram MP, Grundmann F, Müller R-U. Liver manifestations in autosomal dominant polycystic kidney disease (ADPKD) and their impact on quality of life. Clin Kidney J. 2024;18(1):sfae363. pmid:40008356
- 3. Hogan MC, Abebe K, Torres VE, Chapman AB, Bae KT, Tao C, et al. Liver involvement in early autosomal-dominant polycystic kidney disease. Clin Gastroenterol Hepatol. 2015;13(1):155-64.e6. pmid:25111236
- 4. Hoevenaren IA, Wester R, Schrier RW, McFann K, Doctor RB, Drenth JPH, et al. Polycystic liver: clinical characteristics of patients with isolated polycystic liver disease compared with patients with polycystic liver and autosomal dominant polycystic kidney disease. Liver Int. 2008;28(2):264–70. pmid:17927714
- 5. Masyuk TV, Masyuk AI, LaRusso NF. Polycystic liver disease: advances in understanding and treatment. Annual Rev Pathol. 2022;17:251–69. pmid:34724412
- 6. Lu H, Galeano MCR, Ott E, Kaeslin G, Kausalya PJ, Kramer C, et al. Mutations in DZIP1L, which encodes a ciliary-transition-zone protein, cause autosomal recessive polycystic kidney disease. Nat Genet. 2017;49(7):1025–34. pmid:28530676
- 7. Besse W, Dong K, Choi J, Punia S, Fedeles SV, Choi M, et al. Isolated polycystic liver disease genes define effectors of polycystin-1 function. J Clin Invest. 2017;127(5):1772–85. pmid:28375157
- 8. Cornec-Le Gall E, Olson RJ, Besse W, Heyer CM, Gainullin VG, Smith JM, et al. Monoallelic mutations to DNAJB11 cause atypical autosomal-dominant polycystic kidney disease. Am J Hum Genet. 2018;102(5):832–44. pmid:29706351
- 9. Kidney Disease: Improving Global Outcomes AWG. KDIGO 2025 clinical practice guideline for the evaluation, management, and treatment of autosomal dominant polycystic kidney disease (ADPKD). Kidney Int. 2025;107(2S):S1–239. pmid:39848759
- 10. Varughese S, Huang M, Savige J. Typical and atypical ADPKD: predicted pathogenic genetic variants and population frequencies. Nephrol Dial Transplant. 2026;41(2):286–96. pmid:40810452
- 11. Cnossen WR, Drenth JPH. Polycystic liver disease: an overview of pathogenesis, clinical manifestations and management. Orphanet J Rare Dis. 2014;9:69. pmid:24886261
- 12. Besse W, Dong K, Choi J, Punia S, Fedeles SV, Choi M, et al. Isolated polycystic liver disease genes define effectors of polycystin-1 function. J Clin Invest. 2017;127(9):3558. pmid:28862642
- 13. Hu J, Harris PC. Regulation of polycystin expression, maturation and trafficking. Cell Signal. 2020;72:109630. pmid:32275942
- 14. Reynolds DM, Falk CT, Li A, King BF, Kamath PS, Huston J, et al. Identification of a locus for autosomal dominant polycystic liver disease, on chromosome 19p13.2-13.1. Am J Hum Genet. 2000;67(6):1598–604. pmid:11047756
- 15. Van Keimpema L, De Koning DB, Van Hoek B, Van Den Berg AP, Van Oijen MGH, De Man RA, et al. Patients with isolated polycystic liver disease referred to liver centres: clinical characterization of 137 cases. Liver Int. 2011;31(1):92–8. pmid:20408955
- 16. Waanders E, te Morsche RHM, de Man RA, Jansen JBMJ, Drenth JPH. Extensive mutational analysis of PRKCSH and SEC63 broadens the spectrum of polycystic liver disease. Hum Mutat. 2006;27(8):830. pmid:16835903
- 17. Qian Q. Isolated polycystic liver disease. Adv Chronic Kidney Dis. 2010;17(2):181–9. pmid:20219621
- 18. Suwabe T, Chamberlain AM, Killian JM, King BF, Gregory AV, Madsen CD, et al. Epidemiology of autosomal-dominant polycystic liver disease in Olmsted county. JHEP Rep. 2020;2(6):100166. pmid:33145487
- 19. Lanktree MB, Haghighi A, Guiard E, Iliuta I-A, Song X, Harris PC, et al. Prevalence estimates of polycystic kidney and liver disease by population sequencing. J Am Soc Nephrol. 2018;29(10):2593–600. pmid:30135240
- 20. Richards S, Aziz N, Bale S, Bick D, Das S, Gastier-Foster J, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405–24. pmid:25741868
- 21. Borges P, Pasqualim G, Giugliani R, Vairo F, Matte U. Estimated prevalence of mucopolysaccharidoses from population-based exomes and genomes. Orphanet J Rare Dis. 2020;15(1):324. pmid:33208168
- 22. 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. pmid:34400539
- 23. Kaler SG, Ferreira CR, Yam LS. Estimated birth prevalence of Menkes disease and ATP7A-related disorders based on the Genome Aggregation Database (gnomAD). Mol Genet Metab Rep. 2020;24:100602. pmid:32528851
- 24. Kondo A, Nagano C, Ishiko S, Omori T, Aoto Y, Rossanti R, et al. Examination of the predicted prevalence of Gitelman syndrome by ethnicity based on genome databases. Sci Rep. 2021;11(1):16099. pmid:34373523
- 25. Lanktree MB, Haghighi A, Guiard E, Iliuta I-A, Song X, Harris PC, et al. Prevalence estimates of polycystic kidney and liver disease by population sequencing. J Am Soc Nephrol. 2018;29(10):2593–600. pmid:30135240
- 26. Kermond-Marino A, Weng A, Zhang SK, Tran Z, Huang M, Savige J. Population frequency of undiagnosed Fabry disease in the general population. Kidney Int Rep. 2023;8(7):1373–9. pmid:37441486
- 27. Carter A, Brackley SM, Gao J, Mann JP. The global prevalence and genetic spectrum of lysosomal acid lipase deficiency: a rare condition that mimics NAFLD. J Hepatol. 2019;70(1):142–50. pmid:30315827
- 28. Strande NT, Riggs ER, Buchanan AH, Ceyhan-Birsoy O, DiStefano M, Dwight SS, et al. Evaluating the clinical validity of gene-disease associations: an evidence-based framework developed by the clinical genome resource. Am J Hum Genet. 2017;100(6):895–906. pmid:28552198
- 29. Choi K, Huang M, Savige J. Developing a more accurate population frequency of Marfan syndrome from predicted pathogenic FBN1 variants in the gnomAD cohorts. Sci Rep. 2025;15(1):9292. pmid:40102521
- 30. Thammathiwat T, Besse W, Yeung MY, Chandraker A, Onuchic-Whitford AC. Co-occurrence of heterozygous variants in PKD1 and ALG9 in a patient with early-onset ESKD: a case report. Kidney Int Case Rep. 2026;1(1):100013.
- 31. Janssen MJ, Waanders E, Te Morsche RHM, Xing R, Dijkman HBPM, Woudenberg J, et al. Secondary, somatic mutations might promote cyst formation in patients with autosomal dominant polycystic liver disease. Gastroenterology. 2011;141(6):2056-2063.e2. pmid:21856269
- 32. Sharo AG, Zou Y, Adhikari AN, Brenner SE. ClinVar and HGMD genomic variant classification accuracy has improved over time, as measured by implied disease burden. Genome Med. 2023;15(1):51. pmid:37443081
- 33.
Harris PC, Torres VE. Polycystic kidney disease, autosomal dominant. GeneRev. 2022.
- 34. Davila S, Furu L, Gharavi AG, Tian X, Onoe T, Qian Q, et al. Mutations in SEC63 cause autosomal dominant polycystic liver disease. Nat Genet. 2004;36(6):575–7. pmid:15133510
- 35. van de Laarschot LFM, Drenth JPH. Genetics and mechanisms of hepatic cystogenesis. Biochim Biophys Acta Mol Basis Dis. 2018;1864(4 Pt B):1491–7. pmid:28782656
- 36. Apple B, Sartori G, Moore B, Chintam K, Singh G, Anand PM, et al. Individuals heterozygous for ALG8 protein-truncating variants are at increased risk of a mild cystic kidney disease. Kidney Int. 2023;103(3):607–15. pmid:36574950
- 37. Amendola LM, Muenzen K, Biesecker LG, Bowling KM, Cooper GM, Dorschner MO, et al. Variant classification concordance using the ACMG-AMP variant interpretation guidelines across nine genomic implementation research studies. Am J Hum Genet. 2020;107(5):932–41. pmid:33108757
- 38. Boerrigter MM, Bongers EMHF, Lugtenberg D, Nevens F, Drenth JPH. Polycystic liver disease genes: practical considerations for genetic testing. Eur J Med Genet. 2021;64(3):104160. pmid:33556586