Polycystic Kidney Disease 3

1. Disease Information

Overview

PKD3 is an autosomal dominant cystic disease of the kidney and liver caused by heterozygous loss-of-function or hypomorphic variants in GANAB, which encodes the α subunit of glucosidase II (GIIα), a heterodimeric endoplasmic reticulum enzyme that trims glucose residues from N-linked glycans during glycoprotein quality control. GIIα is required for the maturation, surface delivery, and ciliary localization of polycystin-1 (PC1). PKD3 is therefore not a polycystin gene disease but a polycystin-biogenesis disease: the cystogenic endpoint is the same as classical ADPKD, reached one step upstream.

The clinical presentation is bimodal and, within families, highly variable:

• a mild polycystic kidney phenotype — few large cysts, preserved kidney function into late life, kidney failure rare or absent — often with liver cysts; or
• a severe isolated polycystic liver disease (ADPLD) phenotype with few or no kidney cysts, occasionally requiring liver transplantation.

The same allele can produce either. Delbarba et al. reported a family in which the p.Arg839Trp variant caused mild ADPKD, while the same variant had previously been reported in a patient with ADPLD severe enough to require liver transplant (PMID:34357571).

Key identifiers

Synonyms and alternative names

• ADPKD-GANAB (the preferred contemporary designation; Cornec-Le Gall et al. 2018)
• GANAB-related autosomal dominant polycystic kidney disease
• Polycystic kidney disease 3 with or without polycystic liver disease
• PKD3
• GANAB-related polycystic liver disease
• Glucosidase IIα-deficiency polycystic disease (descriptive, not standard)

⚠️ Do not list "polycystic kidney disease, type III (unmapped locus)" as a synonym without the historical caveat above.

Nature of the evidence base

Entirely aggregated disease-level and case-level literature. There is no EHR-derived cohort, no registry, and no natural-history study specific to GANAB. The total published experience is on the order of ~30–40 families worldwide. The evidence base is:

• one gene-discovery study with functional validation (Porath 2016, n = 20 affected individuals / 9 families)
• one ADPLD gene-discovery study (Besse 2017)
• one targeted cohort screen (van de Laarschot 2020, 625 patients)
• a scattering of single-family and single-case reports
• one population-sequencing prevalence estimate (Lanktree 2018)

Every quantitative claim below should be read against that denominator.

2. Etiology

Primary cause

Heterozygous germline variants in GANAB. The mechanism is haploinsufficiency / partial loss of glucosidase IIα function, with a probable requirement for a somatic or stochastic second hit at the tissue level (the standard ADPKD two-hit / dosage-threshold framework).

"Whole-exome sequencing of six GUR ADPKD-affected families identified one with a missense mutation in GANAB, encoding glucosidase II subunit α (GIIα). Because PRKCSH encodes GIIβ, GANAB is a strong ADPKD and ADPLD candidate gene." — Porath et al., Am J Hum Genet 2016;98:1193–1207. PMID:27259053, doi:10.1016/j.ajhg.2016.05.004

The candidate-gene logic is worth capturing in the entry: PRKCSH, the long-established isolated-PCLD gene (PCLD1, OMIM 174050), encodes the β subunit of the very same glucosidase II heterodimer. GANAB was pursued because its partner was already a cystic-disease gene.

Genetic risk factors

• Causal variants: heterozygous GANAB truncating (nonsense, frameshift, canonical splice) and hypomorphic missense variants. See §4.
• Trans-heterozygous / oligogenic modification: the strongest recurring theme in the ADPKD-spectrum literature. Cornec-Le Gall et al. state:

  "Recent data have shown that biallelic disease including at least one weak ADPKD allele is a significant cause of symptomatic, very early onset ADPKD." — J Am Soc Nephrol 2018;29:13–23. PMID:29038287, doi:10.1681/ASN.2017050483

A worked instance: a 12-year-old girl with bilateral renal cysts and nephrolithiasis carrying a GANAB nonsense variant (c.181C>T, p.Arg61) and a PKD1 VUS (c.182C>T, p.Pro61Leu) — reported as the first pediatric case combining PKD1 and GANAB variants (PMC6375066). - Modifier loci: unidentified for GANAB specifically. Lanktree et al. found substantial rare variation in candidate cyst-modifier genes at population scale (see §9), but no GANAB-specific modifier has been mapped. - Sex: female sex is a well-established risk factor for severe liver cystic disease* across all PLD genes (see §9); whether it modifies the kidney phenotype in GANAB carriers is unknown.

Environmental risk factors

No GANAB-specific environmental risk factor has been identified. Extrapolating from the ADPKD literature (label these clearly as ADPKD-general if used):

• Estrogen exposure (endogenous, pregnancy, exogenous HRT/oral contraceptives) — accelerates hepatic cystogenesis; the standard clinical advice is estrogen avoidance in symptomatic PLD.
• High dietary sodium, low water intake / high circulating vasopressin — associated with faster kidney disease progression in ADPKD.
• Smoking, caffeine, obesity — variably reported ADPKD progression associations, none established for GANAB.

Protective factors

• Genetic: none identified. There is no reported protective allele or modifier.
• Environmental: none GANAB-specific. High water intake (vasopressin suppression) and salt restriction are plausible-by-mechanism but unproven in ADPKD generally and untested in GANAB.

Gene–environment interaction

No data. ⚠️ Not available for this disease. The mechanistically obvious hypothesis — that vasopressin/cAMP tone modulates the penetrance of a partial polycystin-1 deficit — is untested in GANAB carriers and should be recorded as a KNOWLEDGE_GAP discussion rather than asserted.

3. Phenotypes

Kidney phenotypes

The kidney phenotype is the single most consistent descriptive claim in the literature:

"The phenotype was mild PKD and variable, including severe, PLD." — Porath et al. 2016, PMID:27259053

GANAB pathogenic variants cause "mild cystic kidney disease, usually without a decline in kidney function, with the majority of affected individuals having liver cysts." — GeneReviews, Polycystic Kidney Disease, Autosomal Dominant (NBK1246)

⚠️ The mildness claim is now formally contested. A 2025 case report describes an 18-year-old with a GANAB variant and Mayo Imaging Classification 1E — the highest-risk imaging class:

"The GANAB gene mutation found in this patient is typically associated with mild kidney disease; however, according to the Mayo Clinic Imaging Classification (MIC) for ADPKD, our patient falls under Classification 1E, which is predictive of rapid progression to end-stage renal disease (ESRD)... This case questions the assumption that GANAB-associated ADPKD progresses in a mild manner." — Agrawal G, Agarwal B, Chandrasekhara Pillai A, Kuriakose K. Cureus 2025;17(2):e79498. PMID:40134995, doi:10.7759/cureus.79498

This is a good candidate for a discussions entry with kind: KNOWLEDGE_GAP attached to the progression/prognosis section, since n = 1 does not overturn a cohort but does bound the confidence.

Liver phenotypes

Severity spans the full range: some carriers have no cysts, others require liver resection or transplantation (Porath 2016; Delbarba 2022 cites a prior p.Arg839Trp carrier who was transplanted).

Extrarenal / vascular phenotypes

⚠️ Weakly supported for GANAB specifically. Intracranial aneurysm is a canonical ADPKD extrarenal feature but was observed in only one GANAB family (M641) in the discovery cohort; the association is not established for the gene. Aortic root dilatation was reported in the Delbarba proband.

Laboratory abnormalities

No GANAB-specific biochemical signature. Standard ADPKD labs apply: serum creatinine/eGFR (LOINC 33914-3 eGFR; 2160-0 creatinine), urinalysis for hematuria, and liver enzymes (typically normal or mildly cholestatic in PLD — GGT and ALP may rise with large cyst burden). There is no clinically deployed glucosidase II activity assay for diagnosis.

Quality-of-life impact

⚠️ No GANAB-specific QoL data exist. No EQ-5D, SF-36, PROMIS, or disease-specific (e.g. ADPKD-IS, PLD-Q) instrument has been applied to a GANAB cohort. From the PLD literature generally:

"Liver cysts arise in patients with autosomal dominant PLD (ADPLD) or in co-occurrence with renal cysts... which constitute the main cause of morbidity and markedly affect the quality of life." — Olaizola P, Rodrigues PM, Caballero-Camino FJ, et al. Nat Rev Gastroenterol Hepatol 2022;19(9):585–604. PMID:35562534, doi:10.1038/s41575-022-00617-7

Because kidney function is typically preserved, the QoL burden in GANAB disease is predominantly hepatic mass effect — early satiety, abdominal distension, pain, dyspnea, malnutrition — rather than the dialysis-trajectory burden that dominates PKD1. This is a genuine, curatable difference from Polycystic_Kidney_Disease_2 and from Autosomal_Dominant_Polycystic_Kidney_Disease.

4. Genetic / Molecular Information

Causal gene

GANAB (hgnc:4138), glucosidase II alpha subunit, 11q12.3, OMIM 104160, UniProt Q14697, NM_198334. Aliases: GluII, G2AN, KIAA0088, GIIA, GIIalpha. Previous symbol: glucosidase, alpha; neutral AB.

Pathogenic variants

Variants reported in Porath et al. 2016 (⚠️ extracted from full text via automated retrieval — re-verify each variant against Table 1 of the paper before entering into the KB):

Nine variants, six truncating — consistent with loss of function as the mechanism.

Variants reported in van de Laarschot et al. 2020 (Orphanet J Rare Dis 2020;15:302, doi:10.1186/s13023-020-01585-4, PMC7585303):

In-silico interpretation from that paper: p.Arg612Pro "was predicted to disrupt the structure of the active site of the protein"; the truncating variants are "predicted to cause abnormal binding of α- and β-subunits of glucosidase II, thus affecting its enzymatic activity."

Structural variants. Large GANAB deletions are a real and under-ascertained class — sequence-only panels will miss them:

Wilson EM, Choi J, Torres VE, Somlo S, Besse W. Large Deletions in GANAB and SEC63 Explain 2 Cases of Polycystic Kidney and Liver Disease. Kidney Int Rep 2020;5(5):727–731. PMID:32405593, doi:10.1016/j.ekir.2020.01.009

Non-coding variants. At least one deep-intronic/non-coding GANAB variant has been reported to explain isolated PCLD in a large family (PMC5805583) — relevant to the diagnostic-yield discussion in §10.

⚠️ Note the discrepancy for curators: GeneReviews states no GANAB deletions/duplications had been documented, but Wilson et al. 2020 reported exactly that. Cite the primary paper, not the summary table.

Variant classification, origin, and functional consequence

• Classification (ACMG/AMP): truncating variants are classified pathogenic on PVS1 grounds — ClinVar records e.g. NM_198334.3(GANAB):c.490C>T (p.Arg164*) as pathogenic, "loss-of-function is an established mechanism of disease for this gene." Missense variants require functional support (see the rescue assay in §6) and many remain VUS.
• Origin: germline, heterozygous, autosomal dominant. No somatic-origin disease is described. Somatic second hits in cyst epithelium are presumed by analogy to PKD1/PKD2 but ⚠️ have not been demonstrated for GANAB.
• Functional consequence: loss of function / partial loss of function. Use functional_impact_category: LOSS_OF_FUNCTION on the GeneticContext, and modifier: DECREASED (not LOSS_OF_FUNCTION) on downstream GO-bound activity nodes unless a qualitative regulatory-escape claim is being made — the CLAUDE.md INCREASED/GAIN_OF_FUNCTION discipline applies here.
• Allele frequency: individual pathogenic variants are absent or ultra-rare in gnomAD. ⚠️ Specific gnomAD constraint metrics (pLI, LOEUF, o/e) for GANAB were not retrievable in this session — look them up directly at gnomad.broadinstitute.org before citing any constraint number. Note the mouse data (§15) showing homozygous lethality, which predicts strong constraint.

Modifier genes

None mapped for GANAB. The generic ADPKD-spectrum modifier framework applies: Lanktree et al. found "truncating mutations in ADPLD genes and genes of potential relevance as cyst modifiers were found in 20.2 cases and 103.9 cases per 10,000 sequenced, respectively" (PMID:30135240).

Epigenetics

⚠️ No GANAB-specific epigenetic data. No methylation, histone, or chromatin study has been performed on GANAB-related disease. (Epigenetic dysregulation — HDAC, bromodomain, miRNA — is an active theme in ADPKD generally, but nothing is GANAB-specific.) Record as a knowledge gap; do not import ADPKD-general epigenetics as PKD3 content.

Chromosomal abnormalities

None associated. GANAB disease is a single-gene, small-variant + intragenic-deletion disorder. No aneuploidy, translocation, or contiguous-gene syndrome involving 11q12.3 has been linked to PKD3.

5. Environmental Information

• Environmental factors: ⚠️ None identified. No toxin, radiation, pollutant, or occupational exposure is associated with PKD3. CTD/TOXNET searching yields no GANAB-disease exposure link.
• Lifestyle factors: No GANAB-specific data. If any environmental: block is curated, the estrogen-avoidance and vasopressin/water-intake items must be labelled as ADPKD/PLD-general extrapolation, with environmental_effect chosen conservatively (MODULATES, not TRIGGERS).
• Infectious agents: ⚠️ Not applicable. No infectious etiology or trigger.

Given the repo's check-environmental-evidence gate, the honest curation here is either an empty environmental: section or an entry carrying the review_notes: "Left deliberately uncited." waiver with the searches recorded — not a manufactured citation.

6. Mechanism / Pathophysiology

This is the section with the strongest, best-cited content, and the reason PKD3 deserves its own entry.

The causal chain

GANAB heterozygous LoF variant
  → reduced glucosidase IIα (GIIα) protein/activity in the ER
  → failure to trim glucose from N-glycans on nascent PC1
  → PC1 fails to complete GPS autoproteolytic cleavage / maturation; retained in ER
  → loss of mature PC1 glycoform; PC1 does not reach plasma membrane or cilium
  → PC2 ciliary localization lost (PC1-dependent trafficking)
  → loss of ciliary polycystin complex signalling
  → ↑ intracellular cAMP, ↑ epithelial proliferation, ↑ transepithelial Cl⁻/fluid secretion
  → focal tubular and biliary cyst initiation, cyst detachment, expansion
  → cystic kidney and liver disease

Upstream — ER glycoprotein quality control (the GANAB-specific step)

Glucosidase II is an ER heterodimer: GIIα (GANAB) is catalytic, GIIβ (PRKCSH/hepatocystin) is the regulatory/ER-retention subunit. Together they perform the second glucose-trimming step of N-glycan processing, gating entry into and exit from the calnexin/calreticulin folding cycle. Both subunits are cystic-disease genes — which is the structural argument that hepatic and renal cystogenesis run through a common protein-biogenesis bottleneck.

Besse et al. generalized this to a pathway-level model:

"Similarly to PRKCSH and SEC63, these genes encode proteins that are integral to the protein biogenesis pathway in the endoplasmic reticulum. We inactivated these candidate genes in cell line models to show that loss of function of each results in defective maturation and trafficking of polycystin-1, the central determinant of cyst pathogenesis. Despite acting in a common pathway, each PCLD gene product demonstrated distinct effects on polycystin-1 biogenesis." — Besse W, Dong K, Choi J, et al. Isolated polycystic liver disease genes define effectors of polycystin-1 function. J Clin Invest 2017;127(5):1772–1785. PMID:28375157, doi:10.1172/JCI90129

That last clause matters for curation: ALG8, GANAB, SEC61B, PRKCSH, and SEC63 are not interchangeable in their effect on PC1 — they converge on the same node by different routes. This is a strong candidate for a shared mechanism module in kb/modules/ (see the curation note at the end).

Core functional evidence (Porath 2016, in vitro)

"Analysis of GANAB-null cells showed an absolute requirement of GIIα for maturation and surface and ciliary localization of the ADPKD proteins (PC1 and PC2), and reduced mature PC1 was seen in GANAB(+/-) cells. PC1 surface localization in GANAB(-/-) cells was rescued by wild-type, but not mutant, GIIα. Overall, we show that GANAB mutations cause ADPKD and ADPLD and that the cystogenesis is most likely driven by defects in PC1 maturation." — Porath et al. 2016, PMID:27259053

Additional detail from the full text (⚠️ automated extraction — verify before quoting as snippet):

• GANAB-null cells: complete loss of the mature PC1 glycoform (PC1-NTR); full-length and immature PC1 accumulate; ciliary PC2 completely absent despite normal cilium formation; minimal effect on control glycoproteins (EGFR, E-cadherin) — i.e. the defect is selective for PC1, not a global glycosylation collapse.
• GANAB^+/− cells: ~50% reduction in PC1-NTR — gene-dosage-proportional, which is the cell-biological basis for haploinsufficiency as the human mechanism.
• Rescue: wild-type FLAG-GIIα restored PC1 surface localization; the disease missense variants p.Thr405Arg, p.Arg422Leu, p.Arg839Trp failed to rescue, while presumed-neutral variants did. This is a functional assay usable for ACMG PS3-level evidence.

Curation note on evidence_source: all of the above is IN_VITRO, not HUMAN_CLINICAL. The Porath abstract mixes human genetic and cell-biology claims in one paragraph — per CLAUDE.md, split the evidence items so each carries a single evidence_source.

Downstream — shared ADPKD cystogenic machinery

Once mature PC1 is lost, the mechanism is the canonical polycystin pathway already curated in Autosomal_Dominant_Polycystic_Kidney_Disease and Polycystic_Kidney_Disease_2: loss of the ciliary polycystin-1/polycystin-2 receptor-channel complex → derepression of adenylyl cyclase → ↑cAMP → PKA-driven proliferation (B-Raf/MEK/ERK) and CFTR-mediated transepithelial chloride and fluid secretion → mTOR activation, Wnt/planar-cell-polarity disturbance, and a metabolic shift toward aerobic glycolysis in cyst epithelium. This is conforms_to territory — reuse the existing nodes rather than re-deriving them, and cite the shared literature at the ADPKD entry.

Cellular processes, cell types, protein dysfunction

• Protein dysfunction: loss of enzymatic function (GIIα) → trans-acting maturation failure of a client glycoprotein (PC1). Note this is not misfolding or aggregation of the mutant protein itself; the disease protein is an enzyme whose absence strands a client. That distinction is worth stating explicitly in the entry's description, because it is what separates PKD3's molecular node from PKD1/PKD2.
• Cellular processes: ER protein quality control; N-linked glycan processing; ER-to-Golgi trafficking; ciliogenesis-independent ciliary cargo delivery; epithelial proliferation; transepithelial anion secretion.
• Metabolic changes: ⚠️ No GANAB-specific metabolomic data. The Warburg-like shift described in ADPKD cyst epithelium has not been demonstrated in GANAB disease.
• Immune involvement: ⚠️ None specific. Macrophage-driven interstitial inflammation is described in advanced ADPKD; no GANAB data.
• Tissue damage: cyst mass effect, compression of adjacent parenchyma, secondary interstitial fibrosis. Given preserved kidney function in most carriers, the fibrotic burden appears low — but this has not been formally studied.

Ontology term leads for the mechanism section

⚠️ All CURIEs below are leads. Run just validate-terms before binding any of them; per the dismech-terms rule, "no term beats a bad one."

GO — biological process | Concept | Lead CURIE | |---|---| | protein N-linked glycosylation | GO:0006487 | | protein folding in endoplasmic reticulum | GO:0034975 ⚠️ | | protein folding | GO:0006457 | | response to endoplasmic reticulum stress | GO:0034976 | | ERAD pathway | GO:0036503 | | protein localization to plasma membrane | GO:0072659 | | protein localization to cilium | GO:0061512 ⚠️ | | cilium assembly | GO:0060271 | | transepithelial chloride transport | GO:0030321 ⚠️ | | cAMP biosynthetic process | GO:0006171 ⚠️ | | positive regulation of cell population proliferation | GO:0008284 | | kidney development | GO:0001822 |

GO — molecular function / cellular component | Concept | Lead CURIE | |---|---| | glucosidase II complex | GO:0017177 ⚠️ | | endoplasmic reticulum lumen | GO:0005788 | | endoplasmic reticulum membrane | GO:0005789 | | ciliary membrane | GO:0060170 | | cilium | GO:0005929 | | α-glucosidase activity | ⚠️ unresolved — GANAB's EC is 3.2.1.207; confirm the exact GO MF term via OAK rather than guessing |

CL — cell types | Concept | Lead CURIE | |---|---| | kidney epithelial cell | CL:0002518 | | epithelial cell of proximal tubule | CL:0002306 | | kidney collecting duct principal cell | CL:1001431 ⚠️ | | cholangiocyte | CL:1000488 ⚠️ |

Molecular profiling and advanced technologies

⚠️ Essentially all absent for PKD3. Recording this honestly is more useful than importing ADPKD-general data:

• Transcriptomics: no GANAB-specific GEO/ArrayExpress dataset identified.
• Proteomics: none GANAB-specific.
• Metabolomics / lipidomics: none.
• Single-cell / spatial transcriptomics: none. (Human kidney and liver single-cell atlases exist and would establish GANAB expression by cell type, but no GANAB-disease scRNA-seq has been published.)
• Functional genomics screens: GANAB appears in DepMap and genome-wide CRISPR screens as a general-essentiality/glycosylation gene, but not in a PKD-disease-model screen.

Datasets caution: per CLAUDE.md's Named Entity Confusion warning (§2b), searching GEO for "GANAB" will surface cancer and glycosylation datasets that have nothing to do with polycystic disease. Any datasets: block here needs manual DIRECT/GENE_ONLY triage and just verify-datasets.

7. Anatomical Structures Affected

Organ level

Body systems: renal/urinary and hepatobiliary primarily; cardiovascular possibly.

Tissue and cell level

• Tissue type: simple/cuboidal epithelium — renal tubular epithelium (distal nephron and collecting duct predominate in ADPKD) and intrahepatic biliary epithelium.
• Cell populations: kidney tubular epithelial cells (CL:0002518), collecting-duct principal cells (CL:1001431 ⚠️), cholangiocytes (CL:1000488 ⚠️).
• ⚠️ No histopathology series exists for GANAB-related disease specifically. Cyst lining morphology is presumed identical to classical ADPKD/PLD but has not been separately characterized.

Subcellular level

This is the diagnostic subcellular signature of PKD3 and should be curated explicitly:

• Endoplasmic reticulum (GO:0005783), specifically ER lumen (GO:0005788) and ER membrane (GO:0005789) — the site of GIIα action and of the PC1 maturation block.
• Glucosidase II complex (GO:0017177 ⚠️).
• Primary cilium (GO:0005929) and ciliary membrane (GO:0060170) — the compartment PC1/PC2 fail to reach.
• Plasma membrane (GO:0005886) — reduced mature PC1 at the surface.

Localization and lateralization

Bilateral kidney involvement, but characteristically asymmetric and with a small number of large cysts rather than the diffuse symmetric enlargement of PKD1 — the "atypical" / Mayo class 2 pattern that the minor-ADPKD-gene literature emphasizes. Liver cysts are diffusely distributed through both lobes.

8. Temporal Development

Onset

• Typical onset: adult, and often late-onset. Delbarba et al. explicitly title their report "late-onset ADPKD" and describe diagnosis at 45 years incidentally, during screening for hernia repair, with elderly parents who had bilateral cystic kidneys and normal kidney function (PMID:34357571).
• Pattern: insidious, frequently incidental discovery on imaging performed for another indication.
• Pediatric/adolescent onset is reported but exceptional — the 18-year-old with MIC 1E disease (Cureus 2025) and the 12-year-old with a GANAB + PKD1 pair (PMC6375066). In the latter case a second locus is the likely explanation.
• Liver cysts in PLD generally are "undetectable early in life and usually appear after the age of 40 years" (Orphanet, isolated PCLD, ORPHA:2924) — consistent with the GANAB series.

Progression

• Kidney: slow or non-progressive. Kidney function is typically preserved; no ESKD was reported in the Porath discovery cohort. For the ADPKD-spectrum minor genes generally, the pattern is "a smaller number of cysts and less kidney enlargement than with PKD1 and PKD2 mutations, asymmetric distribution of cysts, and a slower decline in kidney function."
• Liver: progressive and the main driver of morbidity in the liver-dominant subset; can reach transplant-requiring severity.
• Course: chronic, lifelong, progressive-but-slow. Not episodic or relapsing-remitting. Symptom episodes (pain, hematuria, cyst infection/hemorrhage) punctuate an otherwise indolent course.
• Duration: lifelong.

Staging

⚠️ No PKD3-specific staging system. Two ADPKD-general instruments are applied and are directly relevant:

• Mayo Imaging Classification (MIC) — htTKV-and-age-based classification into class 1A–1E (typical) and class 2 (atypical). GANAB carriers are expected to fall in class 2 (atypical morphology); the 2025 Cureus case is notable precisely because the patient was 1E.
• CKD stage (KDIGO G1–G5) for kidney function.
• Gigot / Schnelldorfer classification for polycystic liver severity.

Critical periods and remission

• No spontaneous or treatment-induced remission. Cysts do not regress; surgical/interventional treatment debulks rather than cures.
• Critical intervention window (ADPKD-general, unproven in GANAB): disease-modifying therapy is most valuable early, while eGFR and kidney volume are still preserved — which is exactly why tolvaptan eligibility is tied to MIC class and eGFR ≥25 (§12). Whether that window concept applies to a genotype that rarely progresses is an open question and a good KNOWLEDGE_GAP.

9. Inheritance and Population

Inheritance

• Pattern: autosomal dominant (HP:0000006). All reported disease is heterozygous.
• Penetrance: incomplete and age-dependent — strongly so. Lanktree et al.'s population-sequencing finding is the key evidence:

  "Loss-of-function mutations in ADPLD genes are also more common than expected, suggesting the possibility of unrecognized cases and incomplete penetrance." — Lanktree MB, Haghighi A, Guiard E, Iliuta IA, Song X, Harris PC, Paterson AD, Pei Y. J Am Soc Nephrol 2018;29(10):2593–2600. PMID:30135240, doi:10.1681/ASN.2018050493

• Expressivity: highly variable, both between and within families — the single most reproducible clinical statement about GANAB.

  "The evidence that the GANAB variant may cause both ADPKD and ADPLD of variable severity supports that renal and hepatic cystogenesis are the result of a common defective polycystin-1 pathway." — Delbarba E, Econimo L, Dordoni C, et al. J Nephrol 2022;35(2):645–652. PMID:34357571, doi:10.1007/s40620-021-01131-w

• Genetic anticipation: not applicable — not a repeat-expansion disorder; no anticipation reported.
• Germline mosaicism: ⚠️ not reported. De novo variants are presumably possible but undocumented.
• Founder effects: ⚠️ none identified. The recurrent c.1914_1915delAG in two families (M641, 290100) may represent a shared haplotype or a mutational hotspot — unresolved in the source.
• Consanguinity: not relevant (dominant). Note that homozygous Ganab loss is embryonic lethal in mouse (§15), so a human biallelic PKD3 phenotype is unlikely to exist.
• Carrier frequency: the concept does not apply to a dominant disorder; the relevant quantity is population allele frequency of pathogenic variants (see below).

Epidemiology

⚠️ No direct prevalence estimate for PKD3/ADPKD-GANAB exists. What can be said:

Anchor figures for context (label as ADPKD-general, not PKD3): - ADPKD point prevalence 3–5 per 10,000 in recent epidemiologic studies; classic lifetime risk ~1 per 1,000 (Lanktree 2018). - Population-sequencing lower bound for lifetime ADPKD prevalence: 9.3 cases per 10,000 sequenced (Lanktree 2018) — higher than clinical ascertainment, the direct evidence for unrecognized mild disease. - Isolated ADPLD prevalence ~1/100,000 (Orphanet ORPHA:2924).

Prevalence curation guidance (per CLAUDE.md §8): use measure_type: POINT_PREVALENCE with prevalence_class: BELOW_1_IN_1000000 for the derived PKD3 figure, put the derivation in notes, and do not put a rate_per_100000 that implies measurement precision the literature does not support. CASES_IN_LITERATURE is arguably the more honest measure_type here — on the order of 30–40 reported families.

Population demographics

• Ethnic/geographic distribution: reported families are of European ancestry (US Mayo/HALT/CRISP cohorts, French Genkyst, Dutch, Italian) plus scattered single cases. ⚠️ This is ascertainment, not biology — GANAB screening has been performed almost exclusively in European-ancestry ADPKD cohorts, so the apparent distribution is an artifact.
• Sex ratio: ~1:1 for inheritance. But severe liver disease is female-predominant across PLD generally — "Women are predominantly affected and have a larger number of cysts than affected males" (Orphanet ORPHA:2924) — and van de Laarschot's 8 carriers were 6 female, mean age 56 (range 31–79). ⚠️ Small n; do not overstate.
• Age distribution of affected individuals: diagnosis clusters in the 4th–7th decades; the Porath carriers spanned ages 9–78 years (⚠️ full-text extraction, verify).

10. Diagnostics

The diagnostic problem, stated plainly

PKD3 is not diagnosable by phenotype alone. Its imaging appearance is a mild/atypical cystic kidney with liver cysts — which overlaps with simple cysts, early PKD2, IFT140-related disease, ALG5/ALG8/ALG9 disease, ADTKD, and localized cystic disease. The diagnosis is molecular, made by a multigene panel.

Imaging

• Kidney ultrasound — first line. The Pei unified criteria (Pei et al., J Am Soc Nephrol 2009, PMID:19118147) apply to at-risk individuals with a positive family history: ≥3 cysts (uni- or bilateral) at ages 15–39; ≥2 cysts in each kidney at 40–59; ≥4 cysts in each kidney at ≥60. Exclusion in an at-risk individual ≥40 requires fewer than 2 cysts. ⚠️ These criteria were derived and validated in PKD1/PKD2 families and are of uncertain sensitivity in GANAB carriers, whose cyst counts are low by definition — a real and citable limitation.
• MRI / CT — required for height-adjusted total kidney volume (htTKV) and Mayo Imaging Classification, and for characterizing liver cyst burden and planning intervention. GANAB carriers frequently land in MIC class 2 (atypical), in which the class-1 progression model does not apply.
• RadLex/imaging: abdominal MRI without and with contrast; T2-weighted sequences for cyst enumeration.

Laboratory tests

Non-specific: serum creatinine and eGFR (LOINC 33914-3 / 2160-0), urinalysis, urine albumin-to-creatinine ratio, liver panel (ALP and GGT may be elevated with high cyst burden; synthetic function usually preserved), CBC. There is no biomarker and no clinical enzyme assay for glucosidase II activity.

Genetic testing — the decisive modality

• Recommended approach: a PKD/cystic-kidney multigene NGS panel including PKD1, PKD2, GANAB, DNAJB11, IFT140, ALG5, ALG8, ALG9, PRKCSH, SEC63, SEC61B, LRP5, PKHD1, HNF1B, NEK8. GeneReviews recommends "a multigene panel that includes GANAB along with PKD1, PKD2, and other associated genes."
• Sequence analysis detects >95% of GANAB pathogenic variants (GeneReviews).
• Deletion/duplication analysis (CNV calling) must be included — Wilson et al. 2020 reported a large GANAB deletion explaining PKD/PLD (PMID:32405593). ⚠️ GeneReviews' table stating no GANAB del/dups is out of date on this point.
• WES/WGS: WES was the discovery modality (Porath 2016; Besse 2017) and is appropriate for panel-negative cases. WGS additionally captures non-coding variants — at least one non-coding GANAB variant explains PCLD in a large family (PMC5805583).
• Single-gene GANAB testing: appropriate only for cascade testing of a known familial variant.
• Not indicated: chromosomal microarray, karyotype, FISH, mtDNA testing, repeat-expansion testing.
• ⚠️ PKD1 pseudogene caveat: any panel must handle the six PKD1 pseudogenes correctly, since excluding PKD1 is a prerequisite for calling a GANAB case.

Omics-based diagnostics

⚠️ None validated. RNA-seq could in principle resolve the splice variants (c.2002+1G>C, c.2690+2_+7del, c.39−1G>C) but no RNA-based diagnostic workflow has been published for GANAB.

Clinical criteria and differential diagnosis

No PKD3-specific criteria exist. Diagnosis = ADPKD/ADPLD clinical-radiologic picture + a pathogenic GANAB variant. Cornec-Le Gall et al. propose the composite phenotype-plus-genotype designation:

"We therefore propose categorization of patients with a phenotypic and genotypic descriptor that will clarify etiology, provide prognostic information, and better describe atypical cases. In genetically defined cases, the designation would include the disease and gene names... Including a genic (and allelic) descriptor with the disease name will provide outcome clues, guide treatment, and aid prevalence estimates." — PMID:29038287

Differential diagnosis, with distinguishing features:

Screening

• Cascade screening of at-risk first-degree relatives — targeted testing for the known familial variant is preferred over imaging, given the low cyst counts.
• No newborn screening. Not appropriate: adult onset, mild course, no preventive intervention available in childhood.
• No population carrier screening. Not indicated for a dominant, low-penetrance, mild disorder.
• ⚠️ Ethical note worth capturing: predictive testing of asymptomatic minors is generally not recommended for adult-onset ADPKD; the mildness of the GANAB phenotype strengthens that position.

11. Outcome / Prognosis

Survival and mortality

⚠️ No survival, life-expectancy, or mortality data exist for PKD3. No registry, no cohort, no actuarial estimate. Any number entered here would be fabricated.

What is defensible: because kidney failure is uncommon and the kidney phenotype is mild, the ADPKD mortality model (dominated by ESKD and cardiovascular disease) is not obviously transferable, and life expectancy in GANAB carriers with kidney-limited disease is plausibly near-normal — but this is inference, not evidence, and should be flagged as such.

Morbidity and function

• Kidney: low morbidity. Kidney function preserved in most carriers; ESKD not reported in the discovery cohort.
• Liver: the dominant morbidity, and it can be severe — liver resection and liver transplantation are both documented in GANAB carriers. Hepatic morbidity in PLD is mass-effect morbidity: pain, early satiety, malnutrition, dyspnea, portal hypertension in advanced cases.
• Disability outcomes / ICF: no data.
• Quality of life: no GANAB-specific instrument data (see §3).

Complications

Kidney: hypertension, hematuria, nephrolithiasis, cyst hemorrhage, cyst infection, chronic pain. Liver: cyst hemorrhage, cyst infection, rupture, compression of the inferior vena cava or portal vein, cholestasis. Vascular: intracranial aneurysm — ⚠️ one family only; not established for the gene.

Prognostic factors

• Mayo Imaging Classification (class 1C–1E predicts rapid progression) — but note that MIC was derived in typical class-1 morphology, and GANAB disease is usually class 2, where MIC does not apply. The 2025 Cureus case is the exception that makes this worth curating explicitly.
• Historical eGFR slope (≥3 mL/min/1.73 m²/year decline).
• Genotype itself: carrying GANAB rather than PKD1 is prognostically favorable — that is the substance of the "genic descriptor" proposal in Cornec-Le Gall 2018.
• Sex for the hepatic phenotype (female predominance in severe PLD).
• Prognostic biomarkers: none. The PROPKD score was derived for PKD1/PKD2 and is not validated for GANAB.

12. Treatment

⚠️ There is no GANAB-specific therapy and no GANAB-specific trial. Every item below is ADPKD- or PLD-general and must be curated as such.

Kidney-directed pharmacotherapy

Tolvaptan — selective vasopressin V2 receptor antagonist; suppresses cAMP-driven cyst-cell proliferation and fluid secretion. The only disease-modifying drug approved for ADPKD.

KDIGO 2025 (Kidney Int 2025, KDIGO ADPKD guideline):

Tolvaptan is recommended in adults with ADPKD and eGFR ≥25 mL/min/1.73 m² at risk of rapidly progressive disease, with initiation criteria of Mayo Imaging Classification 1C–1E or historical eGFR decline ≥3 mL/min/1.73 m²/year.

⚠️ The direct implication for PKD3: most GANAB carriers would not meet these criteria, because they are neither MIC 1C–1E nor declining at ≥3 mL/min/yr. The 2025 Cureus case is precisely a report of a GANAB patient who did. This tension — a genotype that usually excludes itself from the only approved therapy — is the most clinically actionable thing in this entry and deserves a mechanistic_hypotheses or discussions node.

Key evidence (ADPKD-general): TEMPO 3:4 (NCT00428948, Torres et al. N Engl J Med 2012, PMID:23121377) and REPRISE (NCT02160145, Torres et al. N Engl J Med 2017, PMID:29105594). ⚠️ Fetch and verify both abstracts before quoting — I did not retrieve them in this session.

Adverse events: aquaresis (polyuria, nocturia, thirst) is dose-limiting; idiosyncratic hepatotoxicity requires monthly then quarterly LFT monitoring. ⚠️ A specific concern in this genotype: KDIGO lists "significant liver disease other than polycystic liver disease" as a contraindication, so a GANAB patient with severe PLD sits at an awkward intersection — worth noting.

Supportive/adjunct: salt restriction and thiazides reduce tolvaptan polyuria (KDIGO 2025); ACE inhibitors/ARBs for blood pressure with a target of ≤110/75 in younger patients with preserved eGFR (HALT-PKD); high water intake.

Liver-directed therapy

• Somatostatin analogues — octreotide LAR, lanreotide, pasireotide. Reduce liver volume modestly (~3–6%). Olaizola et al.: "Current therapeutic strategies, mainly based on surgical procedures and/or chronic administration of somatostatin analogues, show modest benefits, with liver transplantation as the only potentially curative option" (PMID:35562534).
• Estrogen avoidance; H2 blockers or PPIs for symptomatic liver cysts (GeneReviews).
• Surgical/interventional: cyst aspiration with sclerotherapy; laparoscopic fenestration; segmental hepatic resection; liver transplantation — the only curative option, and documented in GANAB carriers.

Kidney replacement

Dialysis and kidney transplantation — standard, but rarely needed in this genotype.

Advanced therapeutics

⚠️ None exist. No gene therapy, gene editing, ASO, siRNA, mRNA, cell therapy, targeted therapy, or immunotherapy for GANAB-related disease. Conceptually, an ER-proteostasis or chemical-chaperone approach aimed at rescuing PC1 maturation is the mechanistically indicated strategy — the Besse and Porath rescue experiments are its proof of concept — but nothing has entered development. Record as a KNOWLEDGE_GAP / mechanistic_hypotheses entry with status: EMERGING, not as a treatment.

Experimental / trials

⚠️ No trial has ever enrolled by GANAB genotype. ADPKD-general trials that would in principle include GANAB carriers: venglustat (NCT04705051, glucosylceramide synthase inhibitor — note its own glycosphingolipid mechanism), and pasireotide LAR in severe PLD (NCT01670110). ⚠️ Verify current status via just fetch-reference NCT… before curating any trial.

NCIT treatment term leads ⚠️

Verify each with uv run runoak -i sqlite:obo:ncit info "l^…" per the repo's term contract.

Pharmacogenomics

⚠️ No GANAB pharmacogenomic data. Tolvaptan is a CYP3A4 substrate — strong CYP3A inhibitors are an absolute contraindication (KDIGO 2025). That is drug-level, not genotype-level, PGx. No CPIC guideline applies.

Personalized medicine

The genotype-guided principle here is precisely Cornec-Le Gall's: knowing a patient is ADPKD-GANAB rather than ADPKD-PKD1-truncating changes surveillance interval, tolvaptan candidacy, reproductive counselling, and living-donor evaluation of relatives. That is the clinical payoff of making this diagnosis at all.

13. Prevention

There is no primary prevention — this is a germline dominant disorder. Prevention is reproductive and secondary/tertiary.

• Primary prevention: ⚠️ Not applicable. No vaccination, no modifiable-exposure prevention.
• Reproductive options: genetic counselling (50% transmission risk per pregnancy), preimplantation genetic testing for monogenic disorders (PGT-M), prenatal diagnosis. ⚠️ A real counselling difficulty specific to GANAB: the phenotype is mild and penetrance incomplete, so the proportionality of PGT-M/prenatal testing is genuinely debatable in a way it is not for PKD1-truncating disease. Worth curating explicitly.
• Secondary prevention: cascade genetic testing of at-risk relatives; abdominal imaging every 1–5 years and blood-pressure monitoring (GeneReviews surveillance).
• Tertiary prevention (preventing complications): blood-pressure control; adequate hydration; avoidance of nephrotoxins and NSAIDs; treatment of UTIs; estrogen avoidance in symptomatic PLD (GeneReviews); intracranial aneurysm screening — ⚠️ indications are ADPKD-general (family history of aneurysm or SAH, high-risk occupation); there is no evidence base for screening GANAB carriers specifically.
• Risk stratification: MIC + eGFR slope, with the caveats above.
• Behavioral interventions: low-sodium diet, adequate water intake, smoking cessation, weight management — ADPKD-general, unproven in GANAB.
• Public health / environmental interventions: ⚠️ Not applicable.
• Prophylaxis: none.

14. Other Species / Natural Disease

⚠️ No naturally occurring GANAB-related polycystic disease has been reported in any non-human species. OMIA has no GANAB entry for polycystic kidney disease. This section is close to empty and should be curated that way.

• Taxonomy of the human disease: Homo sapiens, NCBITaxon:9606.
• Orthologues: Ganab is broadly conserved across metazoa — mouse Ganab (NCBI Gene 14376 ⚠️ verify), rat, zebrafish ganab, Drosophila, S. cerevisiae GLS2/ROT2. Glucosidase II is one of the most deeply conserved components of ER glycoprotein quality control, and that conservation is itself a mechanistic point: the enzyme is ancient, while its cystic-disease relevance is a vertebrate-specific consequence of having a polycystin client.
• Breed (VBO): ⚠️ Not applicable — no breed-associated GANAB disease.
• Naturally occurring PKD in other species — relevant but NOT GANAB: feline ADPKD in Persian and Persian-derived cats is caused by PKD1 c.10063C>A (p.Cys3284Ter) and is the best-known animal ADPKD (OMIA 000807-9685); bull terrier hereditary nephritis and West Highland White Terrier PKD are also PKD1-related. These are Autosomal_Dominant_Polycystic_Kidney_Disease content, not PKD3 content — importing them here would be a Named Entity Confusion error.
• Comparative pathology: the mouse data (§15) show a species divergence that matters: mouse Ganab haploinsufficiency does not phenocopy human disease. This is a textbook HUMAN_MODEL_MISMATCH rather than a KNOWLEDGE_GAP.
• Zoonotic potential / cross-species transmission: ⚠️ Not applicable — non-communicable genetic disorder.

15. Model Organisms

The headline finding — and it is a negative one

"Homozygous mutation of the Ganab gene in C57BL/6 mice resulted in early embryonic lethality, and there were no cysts in the kidneys or livers of Ganab +/- mice." … "Homozygous Ganab mutations are lethal in the fetal stage, and Ganab haploinsufficiency does not cause kidney or liver cysts in mice, suggesting that it may not be the causative gene in polycystic kidney disease." — Geng G, Xiao Y, Zhang Y, et al. Ganab Haploinsufficiency Does Not Cause Polycystic Kidney Disease or Polycystic Liver Disease in Mice. Biomed Res Int 2020;2020:7469428. PMID:32550232, doi:10.1155/2020/7469428

The same paper reports that despite ~50% reduction in Ganab protein, "the expression of ADPKD proteins (PC1 and PC2) and acetylated tubulin was not affected" in the heterozygous mice — i.e. the mouse does not even reproduce the cellular intermediate that Porath demonstrated in human cells.

How to curate this. This is exactly the case CLAUDE.md describes for discussions with kind: HUMAN_MODEL_MISMATCH rather than KNOWLEDGE_GAP: evidence exists in the model, and its translational validity to human disease is the open question. Two readings are on the table and the entry should hold both:

1. Species divergence in dosage threshold — mouse retains enough GIIα activity at 50% to mature PC1, while human PC1 sits closer to a threshold; supported by the human GANAB^+/− cell data showing ~50% PC1-NTR reduction (a measurable deficit that mouse tissue apparently tolerates). Note also that mouse ADPKD models generally require conditional/inducible inactivation and a "third hit" to cyst; a plain heterozygote is a low bar.
2. The authors' own reading — that GANAB "may not be the causative gene." ⚠️ This is a minority position contradicted by the human genetic and rescue data across four independent cohorts (Porath, Besse, van de Laarschot, plus case reports), but it is in the literature and should be represented rather than suppressed.

For the model link itself: relationship: FAILS_TO_RECAPITULATE, which per CLAUDE.md requires both limitations and evidence — both are available here.

Available and needed models

Applications and limitations

• What the cell models establish well: the molecular mechanism (PC1 maturation dependence on GIIα), gene-dosage proportionality, and a functional assay for variant classification — the rescue assay separates disease missense variants from neutral ones, which is directly usable as ACMG PS3 evidence.
• What no model currently supports: cystogenesis in vivo, natural history, organ-specificity (why liver > kidney in many carriers), modifier discovery, and any preclinical therapeutic testing.
• Resources: MGI (Ganab), IMPC, Alliance of Genome Resources. ⚠️ Check IMPC for a Ganab allele and its viability call — this session did not query it, and IMPC viability data would independently corroborate the Geng lethality finding.

Suggested modeled_mechanisms skeleton

animal_models:
- name: Ganab heterozygous knockout mouse (C57BL/6, CRISPR/Cas9)
  species: Mouse
  genotype: Ganab +/- (CRISPR/Cas9-targeted)
  publication: PMID:32550232
  modeled_mechanisms:
  - target: <the GANAB haploinsufficiency node>
    relationship: FAILS_TO_RECAPITULATE
    fidelity: LOW
    limitations: >-
      Ganab +/- mice develop no kidney or liver cysts and show unaltered PC1/PC2
      expression, so the model reproduces neither the human cystic phenotype nor
      the polycystin-maturation defect demonstrated in human GANAB+/- cells.
      Homozygous loss is embryonically lethal, so a constitutive null cannot be
      assessed postnatally.
    evidence:
    - reference: PMID:32550232
      supports: REFUTE
      evidence_source: MODEL_ORGANISM
      snippet: "<exact quote — fetch via just fetch-reference PMID:32550232>"
      explanation: >-
        Reports the absence of cysts in Ganab+/- mice and embryonic lethality of
        the homozygote, establishing the human/model mismatch.

Consolidated reference list

Curation notes and caveats

Three things must be verified before any of this reaches a YAML file:

1. Every CURIE above is an unvalidated lead. Nothing was run through just validate-terms. The dismech-terms rule applies in full — no term beats a bad one. The GO molecular-function term for glucosidase IIα activity and the "intracranial aneurysm" HPO term are the two I would flag as most likely wrong.
2. The Porath variant table and clinical-range figures in §4 and §3 came from automated full-text extraction, not from my own reading of the source. Re-verify each variant, each eGFR/htTKV range, and the "~0.3% of total ADPKD" denominator against the paper before entering them. Similarly, no evidence snippet in this report should be pasted into YAML without first running just fetch-reference PMID:… and confirming the exact substring — several quotes above are abstract text relayed through a fetch tool.
3. OMIM was not reachable (HTTP 403). The OMIM identifiers, the gene-phenotype assignment, and the clinical synopsis should be confirmed directly.

Two structural recommendations for the entry:

• The Autosomal_Dominant_Polycystic_Liver_Disease.yaml entry already exists in this repo. GANAB straddles it and Autosomal_Dominant_Polycystic_Kidney_Disease.yaml. Decide deliberately whether PKD3's liver-dominant presentation is curated here, cross-referenced there, or split — and record the reasoning, because a curator six months from now will ask.
• The ER-protein-biogenesis bottleneck is a module, not a disease-specific mechanism. PRKCSH, SEC63, GANAB, ALG8, and SEC61B all converge on defective PC1 maturation, and Besse 2017 is the paper that says so at pathway level. If no module in kb/modules/ covers it, this is a strong candidate for create-module — and PKD3 would then conforms_to it rather than duplicating the chain. Check first: rg -il "polycystin\|glucosidase\|protein biogenesis" kb/modules.

Next step: confirm the entry-type decision (separate DISEASE vs SUBTYPE of ADPKD), then I can draft the full Polycystic_Kidney_Disease_3.yaml against this report — the current file is a 22-line placeholder with two stub nodes.

Reference Validation

Checked with linkml-reference-validator 0.2.1.

All extracted references resolved successfully.

Term Validation

Checked with linkml-term-validator 0.4.5, through the ols: adapter.

Terms whose name is worth a second look

The report's name for these is recognisably related to the term's own name without being one of them. A loose paraphrase reads the same way as a citation of the wrong sibling term - and so does a related synonym, which the ontology records precisely because it names something adjacent rather than the same thing - so these are listed rather than judged:

• HP:0003774 (1 mention) - the report calls it "Stage 5 CKD / kidney failure"; HP calls it Stage 5 chronic kidney disease
• HP:0000787 (1 mention) - the report calls it "Nephrolithiasis"; HP calls it Kidney stone, and lists "Nephrolithiasis" among its other names

Prefixes with no resolver

Terms carrying these prefixes were not checked either way, because no configured ontology covers them. An unrecognised prefix may name an ontology this run could not reach as easily as one that does not exist, so nothing here is evidence of fabrication: ORPHA.

56 of 58 terms resolved to a current term; the rest could not be looked up either way.