Pulling the NEK8 literature from PubMed to check identifiers before I write the report.

Checking one more source, a review of NEK8 cases reported 2013–2020, then I'll write the report.

Research report: Renal-Hepatic-Pancreatic Dysplasia 2 (RHPD2), caused by NEK8

Provenance. This report was put together by an AI agent (Claude Code) on 2026-11-03. PubMed E-utilities needed approval in this session and I didn't have it, so citations come from web search plus memory, cross-checked against two committed repository sources (cache/clingen/gene_validity.csv and the dismech Nephronophthisis deep-research report). Each PMID is tagged by how it was checked: - [confirmed]: a PubMed URL, or a PMC/MGI/ZFIN record, matched title and PMID. - [lead]: the PMID was read off a secondary site and not opened on PubMed. - [memory]: I recalled it; it is unverified.

None of the "quotes" below are verified exact abstract strings. They are paraphrases returned by the search tool. Before they become YAML evidence, fetch each with just fetch-reference PMID:… and take the snippet from references_cache/.

1. Disease information

RHPD2 is an autosomal recessive ciliopathy caused by biallelic variants in NEK8 (NIMA-related kinase 8; HGNC:13387; also called NPHP9). Its core features are: - enlarged cystic or dysplastic kidneys; - ductal plate malformation and cysts in the liver; - pancreatic cysts and dysplasia; - frequent congenital heart disease and laterality defects.

Most reported cases are lethal before or around birth. The original report described three fetuses from one consanguineous family with "enlarged cystic kidneys, liver and pancreas" and developmental heart disease, carrying a homozygous NEK8 nonsense variant (Frank et al. 2013, Hum Mol Genet 22:2177–85, doi:10.1093/hmg/ddt070; PMID 23418306 [lead], from the GO annotation page for NEK8).

Synonyms: RHPD type 2; NEK8-related RHPD; renal-hepatic-pancreatic dysplasia caused by mutation in NEK8. The umbrella term also goes by Ivemark II syndrome and renohepatopancreatic dysplasia.

Source class: case reports and small families only (aggregated disease level). There are no registries or EHR cohorts.

Lump/split decisions to surface (not quietly resolved here): 1. Within the RHPD umbrella, RHPD1 is the NPHP3 form (Bergmann et al. 2008, Am J Hum Genet; PMID 18371931 [memory]). RHPD2 is the NEK8 form. The two are separate gene-defined MONDO terms, so separate entries are consistent with design-decisions §3a/§18 Mendelian splitting. An umbrella grouping would be optional. 2. Within the NEK8 allelic series: - RHPD2 (AR, severe, multi-organ); - NPHP9 (AR, juvenile nephronophthisis; weak evidence: one homozygous case among 588 screened in Otto 2008); - NEK8-related autosomal dominant PKD (heterozygous kinase-domain missense, dominant negative; Claus 2023).

Gunther et al. 2023 argue that RHPD2 is "a multisystemic disorder with variable expressivity" rather than always lethal, which leans toward lumping AR NEK8 disease into one spectrum. MONDO and OMIM still keep RHPD2 and NPHP9 apart. My recommendation: curate RHPD2 as the disease, with NPHP9 noted as an allelic, milder AR phenotype rather than a has_subtypes row, because NPHP9 has its own MONDO/OMIM term. AD-PKD should be its own entry or stay inside the existing ADPKD entry. This is a curator's call.

2. Etiology

• Cause: biallelic germline loss-of-function or hypomorphic NEK8 variants. Monogenic, autosomal recessive.
• Genetic risk factors: consanguinity in several families (Frank 2013, plus Saudi cohorts in Al-Hamed 2016). Nothing else is known.
• Environmental, protective, and gene-environment factors: none reported. This is an evidence gap, not a negative finding.

3. Phenotypes

Sources are case series only, with about 17 reported patients by 2022 (Gunther et al. 2023, "Renal-hepatic-pancreatic dysplasia type 2: Perinatal lethal condition or a multisystemic disorder with variable expressivity", PMC10094071; PMID not retrieved). The HPO IDs below are unverified leads. Look each one up before binding.

• Quality of life: there are no QoL data. In survivors, the burden comes from dialysis, transplant, and liver failure.
• Pediatric framing: this is a prenatal or neonatal entity. Variable expressivity separates fetal lethality from childhood survival with progressive kidney and liver failure. NPHP9 (juvenile nephronophthisis) and AD-PKD (neonatal-onset cystic kidneys without consistent extra-renal disease, per Claus 2023) are different clinical entities from the same gene.

4. Genetics and molecular information

• Gene: NEK8, 17q11.2, HGNC:13387 (hgnc:13387 in repo form). The protein is a serine/threonine kinase: an N-terminal NIMA-like kinase domain followed by C-terminal RCC1 repeats.
• Variant classes:
• Nonsense and truncating variants, often homozygous, give the enlarged-cystic-organ RHPD2 picture.
• Missense variants give hypodysplasia (Grampa 2016: "missense and nonsense mutations were associated with hypodysplasia and enlarged cystic organs, respectively", paraphrase).
• Compound-heterozygous variants were found in siblings who survived to ESRD with hepatic and cardiac anomalies (Rajagopalan et al. 2016, Am J Med Genet A; PMID not retrieved, [memory] 27530628).
• NPHP9 variants sit in the RCC1 repeats, one in the same repeat as the mouse jck allele (Otto et al. 2008, JASN 19:587–92, PMID 18199800 [confirmed]). The search summary garbled the variant names, so take them from the abstract.
• Dominant: heterozygous kinase-domain missense p.Arg45Trp (recurrent, often de novo, including mosaic cases), p.Ile150Met and p.Lys157Gln cause AD-PKD in 12 families through a proposed dominant-negative effect (Claus et al. 2023, Kidney Int 104:995–1007, doi:10.1016/j.kint.2023.07.021, PMC10592035; PMID [memory] 37598857).
• Codon 45 is also, from memory, the site of the Frank 2013 nonsense variant (p.Arg45*). This needs checking. If true, the same codon gives AR truncation and AD missense disease.
• Functional effect in RHPD2: loss of function. Missense versus truncating alleles affect YAP differently (Grampa 2016).
• ACMG, gnomAD, ClinVar: not checked in this session.
• Somatic vs germline: RHPD2 is germline. Mosaicism has been reported only for the dominant R45W allele (Claus 2023).
• Modifier genes, epigenetic changes, chromosomal abnormalities: none reported.
• Second cohort source: an "inferred novel variant in NEK8" (deduced from parental DNA only) in 44 Saudi families with antenatal cystic kidney disease (Al-Hamed et al. 2016, J Med Genet 53:338–47, PMC4853542; PMID [memory] 26862157). This is weak, indirect support.

5. Environmental information

Not applicable. No environmental, lifestyle, or infectious contributors are reported.

6. Mechanism and pathophysiology

Causal chain. Steps marked (inferred) are not demonstrated in human tissue.

1. Biallelic NEK8 loss-of-function variants lead to absent or defective NEK8 kinase in the ciliary inversin compartment.
2. NEK8 is a core member of the INV complex with INVS (NPHP2), NPHP3 and ANKS6.
3. ANKS6 binds the NEK8 kinase domain to activate it, and needs NEK8 to reach that compartment (Czarnecki et al. 2015, Nat Commun 6:6023, PMID 25599650 [lead]; model organism and in vitro).
4. Loss of INV-complex kinase signaling results in Hippo pathway dysregulation.
5. NEK8 binds TAZ and drives it into the nucleus (Habbig et al. 2012, Hum Mol Genet 21:5528–38, PMID 23026745 [confirmed]; in vitro).
6. NEK8 variants alter YAP regulation in patient fibroblasts, Nek8-null spheroids and jck kidneys (Grampa et al. 2016, PLoS Genet 12:e1005894, PMC4788435; PMID [memory] 26967905).
7. Patient fibroblasts showed lower PKD1/PKD2 and higher c-MYC (Frank 2013).
8. Branch A (kidney, liver, pancreas): altered YAP/TAZ output and polycystin signaling lead to too much epithelial proliferation and failed tubule and duct differentiation, giving cystic dysplasia of kidney, biliary ductal plate and pancreatic ducts.
9. The polycystin link is supported: Nek8-null cells respond poorly to fluid shear, which suggests NEK8 mediates polycystin-2-dependent signaling (Manning et al. 2013).
10. Dominant R45W and K157Q reduce ciliary polycystin-2 (Claus 2023).
11. Verteporfin, a YAP inhibitor, partly rescued spheroid and zebrafish defects (Grampa 2016). That is pharmacological support for YAP as the mediator, so directness: INDIRECT.
12. Branch B (laterality and heart): NEK8 loss in node cilia leads to misexpression of left-sided markers, which causes heterotaxy and cardiac malformations. In Nek8-null mice cilia form normally but nodal signaling fails (Manning et al. 2013, JASN 24:100–12, PMID 23274954 [confirmed]; model organism).
13. Branch C (in parallel; contribution to the human disease not yet shown): NEK8 loss leads to replication stress, because without NEK8 the ATR pathway no longer holds back S-phase CDK activity. This causes DNA damage and disordered renal cell architecture, and ciliopathy alleles impair this function (Choi et al. 2013, Mol Cell 51:423–39, PMID 23973373 [lead]; model organism and in vitro).
14. Dysplasia and cysts result in clinical failure of the affected organs:
15. oligohydramnios and pulmonary hypoplasia lead to perinatal death (inferred);
16. in survivors, ESKD, cholestatic liver failure, and heart disease.

Ontology leads (all unverified): - GO: - GO:0097543 ciliary inversin compartment - GO:0004674 protein serine/threonine kinase activity - GO:0035329 hippo signaling - GO:0007368 determination of left/right symmetry - GO:0060271 cilium assembly - GO:0072001 renal system development - GO:0006974 DNA damage response - CL: - kidney tubule epithelial cell - cholangiocyte (CL:1000488?) - pancreatic ductal cell - node cilia-bearing cells - Subcellular: primary cilium (GO:0072372), centrosome (GO:0005813), nucleus (for TAZ/YAP).

Omics: none found for human RHPD2 (no GEO, single-cell or proteomics data).

7. Anatomical structures

• Primary: kidney (UBERON:0002113), liver and intrahepatic bile ducts, pancreas (UBERON:0001264).
• Secondary: heart (septa), body axis (situs).
• Organ involvement is bilateral and symmetric.
• The UBERON IDs are leads; look them up before binding.

8. Temporal development

• Onset: congenital. Most cases are seen on prenatal ultrasound.
• Course: usually perinatal death. Gunther 2023 reports, in the UTHealth abstract, that "13 [of 17] died within the perinatal period" (pregnancy to age 1).
• Survivors follow a progressive course:
• death at 5 years from renal failure;
• death at 30 months from liver and kidney failure;
• two alive at 1 and 13 years.
• Critical period: the second trimester and earlier, when organ morphogenesis happens. No staging scheme exists.

9. Inheritance and population

• Inheritance: autosomal recessive (ClinGen: AR, Definitive).
• Penetrance: appears complete for biallelic null alleles, but expressivity varies widely.
• Other genetic features: no anticipation reported; consanguinity contributes. A founder effect has not been established (Saudi and other Middle Eastern families appear among the reports).
• Prevalence: fewer than 20 patients in the literature by 2022. Use CASES_IN_LITERATURE (about 17, Gunther 2023), class ULTRA_RARE. Incidence, carrier frequency and sex ratio are not reported.

10. Diagnostics

• Prenatal ultrasound: enlarged hyperechoic or cystic kidneys, oligohydramnios, cardiac defects, situs anomalies.
• Pathology: autopsy shows cystic dysplasia, ductal plate malformation, and pancreatic cysts and fibrosis.
• Genetic testing: exome or genome sequencing, or a ciliopathy / cystic-kidney gene panel. PanelApp lists NEK8 on its Cystic kidney disease and Renal ciliopathies panels. Claus 2023 recommends including NEK8 on PKD panels for children and adults.
• Differential diagnosis:
• RHPD1 (NPHP3);
• Meckel syndrome (looks for encephalocele and polydactyly);
• ARPKD (PKHD1);
• ANKS6 and INVS ciliopathies;
• Jeune / short-rib syndromes;
• glomerulocystic disease.
• Screening: carrier testing and prenatal or preimplantation testing for known familial variants.
• No GeneReviews chapter was found. Confirm with just check-genereviews.

11. Outcome and prognosis

• Mortality: high perinatal mortality (13 of 17).
• Survivors: progress to ESKD and end-stage liver disease in childhood.
• Prognostic factors: a truncating versus missense genotype correlation is suggested (Grampa 2016) but not validated.
• Biomarkers: no prognostic biomarkers.

12. Treatment

There is no disease-specific therapy. Care is supportive. NCIT leads: - dialysis; - kidney transplantation and combined liver–kidney transplantation (NCIT:C15289 Organ Transplantation); - cardiac surgery for structural heart defects (NCIT:C15329 Surgical Procedure); - supportive or palliative perinatal care (NCIT:C15747 Supportive Care); - genetic counseling (NCIT:C15240).

Verteporfin is preclinical only (spheroid and zebrafish; Grampa 2016). No clinical trials were found, though no registry search was run.

13. Prevention

• Genetic counseling for families: 25% recurrence risk.
• Cascade carrier testing.
• Prenatal or preimplantation genetic testing.
• Tertiary prevention: manage kidney and liver failure and refer for transplant.

14. Other species

No naturally occurring animal disease was found (OMIA not searched). NEK8 is conserved in mouse (Nek8) and zebrafish (nek8).

15. Model organisms

Databases: MGI (Nek8), ZFIN.

Gaps and to-dos before curation

• Confirm the PMIDs for Frank 2013, Grampa 2016, Claus 2023, Rajagopalan 2016, Al-Hamed 2016, Gunther 2023 and the 2020 cholestasis report. Confirm the variant nomenclature against the abstracts.
• No exact snippets have been verified yet.
• ORPHA and OMIM 613824 still need checking.
• An untracked placeholder kb/disorders/Renal-Hepatic-Pancreatic_Dysplasia_2.yaml (name, date and MONDO term only) is already in this worktree. I left it unchanged; no branch, commit or claim was made.

Sources: - Frank 2013 (CORE); GO wiki, HUMAN:NEK8 - Grampa 2016 (PMC4788435) - Otto 2008 (PMC2391043) - Claus 2023 (PMC10592035) - Rajagopalan 2016 (Sapienza IRIS) - Atlas of Genetics, NEK8 - Liu 2002 (ZFIN) - Manning 2013 (PMC3537214) - Choi 2013 (PMC3790667) - Czarnecki 2015 (PMC4361001) - Habbig 2012 (PubMed 23026745) - Al-Hamed 2016 (PMC4853542) - Gunther 2023 (PMC10094071); UTHealth repository record - RHPD aggregator (DrugMAP) - PanelApp, NEK8 - LOVD RHPD2

Reference Validation

Checked with linkml-reference-validator 0.5.1.

All extracted references resolved successfully.

Term Validation

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

Terms the report names something else

These identifiers resolve, so nothing about them looks wrong, and the ontology calls them something unrelated to what the report calls them. That usually means the identifier is not the one the sentence needs:

• GO:0072372 (1 mention) - the report calls it "Subcellular: primary cilium"; GO calls it GO_0072372**

Obsolete terms

These terms are real but deprecated. Citing one is not a fabrication; it does mean the report is naming something the ontology has retired:

• GO:0072372 (GO_0072372) (1 mention) - replaced by GO:0005929

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:

• MONDO:0004691 (1 mention) - the report calls it "NEK8 / ADPKD"; MONDO calls it autosomal dominant polycystic kidney disease, and lists "ADPKD" among its other names
• UBERON:0002113 (1 mention) - the report calls it "Primary: kidney"; UBERON calls it kidney**, and lists "reniculate kidney" 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.