Juvenile Neuronal Ceroid Lipofuscinosis

1. Disease Information

1.1 Overview

The neuronal ceroid lipofuscinoses (NCLs) are a group of inherited, mostly autosomal recessive, lysosomal-storage neurodegenerative disorders unified by the intracellular accumulation of autofluorescent, PAS- and Sudan-black-positive "ceroid-lipofuscin" storage material in neurons and extraneural tissue. They have been subclassified since Santavuori's era along two orthogonal axes:

1. Age of onset — congenital, infantile, late-infantile, juvenile, adult (Kufs)
2. Causal gene — CLN1–CLN8, CLN10–CLN14 (CLN9 is withdrawn; see §4.6)

MONDO:0019262 is a class on the first axis. It denotes the cohort of NCL patients whose first symptom appears in the juvenile window — conventionally ~5–10 years of age (some authors use 4–10 or extend to early adolescence) — regardless of which gene is mutated.

The clinical gestalt of the juvenile window is distinctive and largely gene-independent:

• Vision is usually the herald symptom. A previously normal school-age child presents to an ophthalmologist with rapidly progressive central visual loss and a pigmentary retinopathy/bull's-eye maculopathy. Frequent initial misdiagnoses are Stargardt disease, retinitis pigmentosa, or cone–rod dystrophy.
• Cognitive and behavioural decline follows within 1–3 years — school failure, then frank dementia, often with a striking psychiatric prodrome (anxiety, psychosis, hallucinations) in the CLN3 subgroup.
• Seizures (generalised tonic-clonic and myoclonic) begin typically age 8–13.
• Motor decline — extrapyramidal (parkinsonism, dystonia, rigidity) plus cerebellar ataxia and pyramidal signs — leads to loss of ambulation in the second decade.
• Death in the late second to third decade for CLN3; earlier or later for other genotypes.

The critical exceptions to this gestalt are diagnostically load-bearing and are detailed gene-by-gene in §4: CLN8/EPMR ("Northern epilepsy") presents with seizures and no visual failure, and some CLN6 juvenile patients present with ataxia/spasticity and explicitly no visual loss.

1.2 Key identifiers

Contrast — MONDO:0008767 (do not conflate): label "neuronal ceroid lipofuscinosis 3"; definition "A condition associated with mutation(s) in the CLN3 gene, encoding battenin…"; xrefs OMIM:204200, Orphanet:228346, MEDGEN:155549, NCIT:C61258, UMLS:C0751383, DOID:0110731, GARD:0005897, NORD:843; material basis in germline mutation in CLN3 (HGNC:2074).

1.3 MONDO hierarchy and children (retrieved live)

Parents of MONDO:0019262: - MONDO:0016295 — neuronal ceroid lipofuscinosis - MONDO:0020143 — cerebral lipidosis with dementia

Direct children returned by OLS4 (/children and /descendants both returned the same four):

Plus, confirmed by direct class lookup (it is dual-parented and was missed by the descendants call):

| MONDO:0979346 | juvenile neuronal ceroid lipofuscinosis 3 | CLN3 | syn. "juvenile CLN3 disease"; xrefs Orphanet:699780, MEDGEN:1897244, UMLS:C6012317. Parents: MONDO:0019262 and MONDO:0008767. |

Two curation-relevant ontology observations:

1. MONDO:0979346 is the correct bridging term between the onset grouping and CLN3. It sits under both MONDO:0019262 and MONDO:0008767. This is exactly the right modelling: "juvenile CLN3 disease" is simultaneously a juvenile NCL and a form of CLN3 disease. If a KB entry needs to reference "the CLN3 member of the juvenile grouping", MONDO:0979346 — not MONDO:0008767 — is the term.
2. The grouping is under-populated relative to the literature. Orphanet has minted a juvenile CLNx disease series (ORPHA:699739 = juvenile CLN1, ORPHA:699769 = juvenile CLN2, ORPHA:699780 = juvenile CLN3), and MONDO has imported three of them. But juvenile-onset presentations are well documented for CLN5, CLN6, CLN7/MFSD8, CLN8, and CLN10/CTSD (§4), and no corresponding MONDO children exist. MONDO:0019262's asserted extension is therefore substantially narrower than its textual definition. This is a genuine ontology gap worth reporting upstream, and it means a KB entry for this grouping should enumerate members from the literature rather than from the MONDO child list.

1.4 Synonyms

Safe (onset-neutral): juvenile NCL; JNCL; juvenile neuronal ceroid lipofuscinosis; juvenile-onset neuronal ceroid lipofuscinosis; juvenile Batten disease (with caution).

Hazardous — carried by MONDO but in practice CLN3-specific in the literature: "Batten disease"; "Spielmeyer-Vogt disease"; "Spielmeyer-Sjögren disease"; "Vogt-Spielmeyer disease"; "Batten-Spielmeyer-Vogt disease". Note that "Vogt Spielmeyer disease" and "Spielmeyer Sjogren disease" are also synonyms on MONDO:0008767 — the synonym sets overlap, which is the mechanical root of the conflation.

1.5 Data provenance

Information in this report is aggregated disease-level (ontologies, OMIM/Orphanet, GeneReviews, cohort studies, case series), not individual-patient/EHR-derived. There is no EHR-scale phenotyping resource for this grouping; the DEM-CHILD and NCL Resource (UCL) patient/mutation databases are the closest thing to patient-level aggregation, and the Rochester UBDRS natural-history cohort is patient-level but CLN3-restricted.

2. Etiology

2.1 Causal factors

Monogenic, overwhelmingly autosomal recessive, biallelic loss-of-function or hypomorphic variants in a lysosomal/endolysosomal gene. There is no infectious, toxic, or environmental etiology. The one non-recessive member of the wider NCL family, autosomal-dominant DNAJC5/CLN4, is adult-onset and is not a member of this grouping (§4.7).

The unifying etiological statement for the grouping is a quantitative one rather than a gene-level one: juvenile onset arises when residual function of an NCL gene product is reduced enough to cause progressive storage, but not so severely as to produce infantile or late-infantile presentation. This "residual-activity gradient" model is the single most important mechanistic concept for MONDO:0019262 and is supported directly for CLN2:

"loss of function variants abolishing TPP1 enzyme activity lead to CLN2 disease, whereas variants that diminish TPP1 enzyme activity lead to SCAR7." — Sun Y et al., Hum Mutat 2013;34(5):706-13 (PMID:23418007)

and is the framing of the comprehensive mutation-spectrum review:

"Different mutations within the NCL spectrum can cause variable disease severity. The NCLs exemplify both phenotypic convergence or mimicry and phenotypic divergence. For example, mutations in CLN5, CLN6, MFSD8, or CLN8 can underlie the clinically similar late infantile variant NCL disease. Phenotypic divergence is exemplified by different CLN8 mutations giving rise to two very different diseases, the mild CLN8 disease, EPMR (progressive epilepsy with mental retardation), and the more severe CLN8 disease, late infantile variant." — Kousi M, Lehesjoki A-E, Mole SE. Hum Mutat 2012;33(1):42-63 (PMID:21990111)

Practical corollary for curation: the causal chain for this grouping should be modelled as [gene-specific hypomorphic lesion] → [partial residual protein function] → [slower storage accumulation] → [juvenile-window onset], with the gene as a substitutable slot — this is structurally analogous to a lysosomal_substrate_accumulation module conformer with the severity/timing dimension made explicit.

2.2 Genetic risk factors

• Causal variants: biallelic in PPT1, TPP1, CLN3, CLN5, CLN6, MFSD8, CLN8, CTSD, ATP13A2 (see §4 for the defensibility grading).
• Consanguinity is a major risk factor for all recessive members and is documented in most non-CLN3 juvenile reports (Cypriot CLN6 families, Chinese CLN6 family, Somali CTSD sibship, Turkish/Roma CLN7 cohorts).
• Founder effects are strong and population-specific (§9.3): the CLN3 1.02-kb deletion in Northern Europeans; CLN5 p.Tyr392 in Finns; CLN8 p.Arg24Gly in Finns (EPMR); MFSD8 p.Thr294Lys in Roma of the former Czechoslovakia; PPT1* p.Arg122Trp in Finns.
• No established common-variant susceptibility loci. There is no GWAS signal for NCL — this is a Mendelian entity and GWAS Catalog/PheGenI are not informative sources here.

2.3 Modifier genes

Formally identified genetic modifiers are not established for the juvenile NCLs. There is, however, strong indirect evidence of modification:

• Intrafamilial variability with an identical genotype is documented. In a consanguineous Chinese CLN6 family, "Both patients exhibited seizures and progressive psychomotor decline and mental deterioration without visual impairment. They had different ages of onset, although they carried the same missense mutation. The affected female showed a pronounced abnormal MRI signal in the bilateral hippocampus, while her younger brother only showed a very slight abnormal signal." (Neurodegener Dis 2021;21(5-6):126-131; PMID:35609511)
• Cross-CLN protein interdependence is a plausible modifier mechanism: "Loss of CLN3 has been shown to affect PPT1, TPP1, CLN5, and CTSD" (Zhang Y et al., CNS Neurosci Ther 2025;31(2):e70261; PMID:39925015). A hypomorphic allele in a second NCL gene could therefore plausibly modify severity — hypothesis, not established finding.

2.4 Protective factors

None identified. There are no reported protective alleles, dietary factors, or lifestyle exposures that modify onset or course. Claims to the contrary should be treated as unsupported. The only "protective" genetic phenomenon is intra-locus: a hypomorphic allele in trans to a null allele is protective relative to two nulls, shifting the phenotype from late-infantile toward juvenile/protracted. This is the mirror image of the residual-activity gradient in §2.1 and is best modelled as allelic severity, not as a protective factor.

2.5 Gene–environment interactions

Not applicable / none documented. No GxE interaction has been demonstrated for any NCL. Intercurrent febrile illness can lower seizure threshold and precipitate clinical deterioration, but this is a nonspecific epilepsy phenomenon and not a disease-modifying interaction.

3. Phenotypes

⚠️ Frequency caveat, stated once and applying throughout this section. Quantitative frequency data for "juvenile NCL" in the literature are overwhelmingly derived from CLN3 cohorts (the Rochester UBDRS cohort, the DEM-CHILD/NCL Resource registries, Scandinavian and Danish series). Applying those frequencies to the grouping over-weights CLN3 and imports precisely the conflation this entry exists to avoid. Where I give a frequency below, I state which population it came from. For a dismech entry, the defensible position is to curate the grouping's phenotypes with frequency: omitted for most terms, per the frequency-evidence SOP, and to attach quantitative frequencies only at the member (juvenile CLN3 disease, MONDO:0979346) level.

3.1 Ophthalmological (the cardinal presenting domain)

Quality-of-life impact: vision loss in this window is uniquely destructive because it lands at the start of formal literacy acquisition. It forces immediate transition to braille/assistive technology — which is then itself lost as dementia advances, producing a documented "double loss" and a well-described family-reported crisis point. Loss of independent mobility and reading are the two dominant QoL domains in the CLN3 literature.

3.2 Cognitive / behavioural / psychiatric

The psychiatric phenotype is a genuine differentiator within the grouping: florid psychosis with hallucinations in an adolescent with visual failure is characteristically CLN3, whereas the CLN6 and CLN8 juvenile forms are dominated by seizures and motor decline with less prominent psychosis.

3.3 Seizures / epilepsy

Seizure onset is the defining first symptom in the CLN8/EPMR member — Ranta et al. describe EPMR as "an autosomal recessive disorder characterized by onset of generalized seizures between 5 and 10 years, and subsequent progressive mental retardation" (Nat Genet 1999;23(2):233-6; PMID:10508524). That onset window is squarely juvenile, but the phenotype lacks retinopathy — which is why a purely vision-anchored definition of "juvenile NCL" would wrongly exclude it.

In a mixed pediatric NCL cohort (median onset 5.46 ± 1.95 y), "myoclonic seizures in 68%, and motor difficulty in 24%" were the presenting symptoms, with "visual impairment (80%), global developmental delay (56%), and disc pallor (56%)" as primary features (Pak J Med Sci 2024;40(8):1638-1643; PMID:39281238). Note this cohort was CLN6-dominant (42%), not CLN3-dominant, which explains the seizure-first skew relative to classic CLN3 descriptions.

3.4 Motor: extrapyramidal, cerebellar, pyramidal

3.5 Other systemic

3.6 Neuroimaging and laboratory

3.7 Phenotype characteristics summary

• Onset: juvenile, ~5–10 y (grouping-defining). Range across defensible members: ~4 y (some CLN6/CLN5) to ~15 y (CTSD family A; some CLN7 protracted).
• Severity: severe and uniformly fatal in the classic members; variable across the grouping — protracted CLN2, CLN5, and CLN7 forms can survive into the fourth decade.
• Progression: progressive, monotonic, without remission. Not episodic or relapsing.
• Frequency: see the caveat opening §3.

4. Genetic / Molecular Information — the gene-by-gene core of this report

This is the section the entity requires. Below, each NCL gene is graded for whether a juvenile-onset presentation is defensibly attributable to it.

4.0 Grading key

• Tier A — Established. Multiple independent reports, or an authoritative classification source (GeneReviews Table 1 / Kousi 2012 / Mole & Cotman 2015) lists juvenile onset as a recognised phenotype for the gene.
• Tier B — Reported, limited. Juvenile onset reported in one or few families; real but thinly evidenced.
• Tier C — Not defensible as juvenile. The gene's recognised onset windows are infantile, late-infantile, or adult. Do not list as a member.
• Tier X — Withdrawn / invalid.

The two anchor classification sources agree substantially. GeneReviews Neuronal Ceroid-Lipofuscinoses (NBK1428) Table 1, retrieved this session:

Mole & Cotman 2015 (Biochim Biophys Acta 1852(10 Pt B):2237-41; PMID:26026925) Table 2 concurs and additionally lists CLN12/ATP13A2 → juvenile.

4.1 Tier A — CLN3 (HGNC:2074) — the modal, but not the definitional, member

• MONDO: disease MONDO:0008767; juvenile member term MONDO:0979346. OMIM: 204200. Protein: CLN3/battenin, a polytopic lysosomal/endosomal membrane protein of incompletely defined function.
• Onset: 4–7 y with visual failure. Course: vision → cognition/behaviour → seizures (~8–13 y) → extrapyramidal motor decline → death typically in the third decade.
• Variant spectrum: dominated by a single founder allele, a ~1.02-kb genomic deletion removing exons 7 and 8 (c.461-280_677+382del966, historically "1 kb deletion"), reported in roughly 80–85% of disease alleles in Northern European ancestry, with ~70–75% of patients homozygous. ⚠️ These specific percentages are from the standard literature (International Batten Disease Consortium, Cell 1995) but were not re-verified against a fetched abstract in this session — verify before curating as evidence. Remaining alleles are missense, nonsense, frameshift, and splice.
• Functional consequence: loss of function. The common deletion produces a frameshifted truncated product.
• Allelic non-NCL phenotype: isolated/non-syndromic retinal degeneration from hypomorphic CLN3 genotypes — clinically important because such patients are juvenile-onset and present to retina clinics.
• Ultrastructure: fingerprint profiles, with curvilinear and rectilinear components (GeneReviews Table 2).
• Discriminating lab feature: vacuolated peripheral lymphocytes, essentially unique to CLN3 within the grouping.
• Carrier frequency: ~1/500 in the US when adjusted for ethnic diversity (Gene 2016; see §4.14).

4.2 Tier A — PPT1 / CLN1 (HGNC:9325) — juvenile CLN1 disease

• MONDO: MONDO:0979341 ("juvenile neuronal ceroid lipofuscinosis 1"; Orphanet:699739); gene-level MONDO:0009744. OMIM: 256730. Protein: palmitoyl-protein thioesterase 1, a soluble lysosomal enzyme removing thioester-linked palmitate from S-acylated proteins.
• Defensibility: strong and long-established. The defining paper is explicitly titled for this phenotype: "Mutations in the palmitoyl-protein thioesterase gene (PPT; CLN1) causing juvenile neuronal ceroid lipofuscinosis with granular osmiophilic deposits" (Mitchison HM et al., Hum Mol Genet 1998;7(2):291-7; PMID:9425237). It reported: "Five mutations in the PPT gene were identified: three missense mutations, Thr75Pro, Asp79Gly, Leu219Gln, and two nonsense mutations, Leu10STOP and Arg151STOP."
• The diagnostic trap this paper solved: juvenile CLN1 combines a juvenile clinical course with infantile-type ultrastructure (GROD, granular osmiophilic deposits). A pathologist who reads GROD and reports "infantile NCL" will contradict the clinician. Mitchison et al. concluded this demonstrates "the correlation which exists between genetic basis and ultrastructural changes in the NCLs" — i.e. ultrastructure tracks the gene, not the onset age. This is the single most useful ultrastructural rule for triaging a juvenile NCL: GROD in a juvenile patient means PPT1 or CTSD, not CLN3.
• Genotype–phenotype: juvenile onset associates with missense/hypomorphic alleles retaining partial PPT1 activity; null/null gives classic infantile CLN1 (Santavuori-Haltia).
• Biochemically confirmable: PPT1 enzyme assay in leukocytes/fibroblasts/dried blood spot. This makes juvenile CLN1 one of only three members with a cheap, definitive, non-sequencing first-line test.
• Carrier frequency: highest of any NCL gene — 1/75 in Finns; ~1/500 US-adjusted (Gene 2016).

4.3 Tier A — TPP1 / CLN2 (HGNC:2073) — juvenile and late-juvenile/protracted CLN2

• MONDO: MONDO:0979345 ("juvenile neuronal ceroid lipofuscinosis 2"; Orphanet:699769); gene-level MONDO:0009746. OMIM: 204500. Protein: tripeptidyl peptidase 1, a soluble lysosomal serine protease.
• Defensibility: strong. GeneReviews lists both "Juvenile" and "Late juvenile/protracted" as recognised atypical CLN2 phenotypes. The historically important report is Wisniewski KE et al., "Reevaluation of neuronal ceroid lipofuscinoses: atypical juvenile onset may be the result of CLN2 mutations" (Mol Genet Metab 1999) — ⚠️ the exact-title PubMed query failed this session and I could not confirm its PMID; verify before citing.
• The best-characterised juvenile TPP1 phenotype is SCAR7. Sun et al. showed that autosomal recessive spinocerebellar ataxia 7 is allelic to CLN2: SCAR7 patients "showed ataxia and low activity of tripeptidyl-peptidase 1, but no ophthalmologic abnormalities or epilepsy", and proposed that "loss of function variants abolishing TPP1 enzyme activity lead to CLN2 disease, whereas variants that diminish TPP1 enzyme activity lead to SCAR7." (Hum Mutat 2013;34(5):706-13; PMID:23418007). SCAR7 = MONDO:0012452 / OMIM:609270.
• Why this matters for MONDO:0019262: the TPP1 juvenile/protracted phenotype is a cerebellar-ataxia-first presentation without the retinopathy that anchors the classic juvenile gestalt. A juvenile NCL grouping defined only by "vision loss + seizures + dementia" will miss it.
• Biochemically confirmable: TPP1 enzyme assay (leukocyte / dried blood spot). This is the highest-yield single test in the entire grouping, because CLN2 is the only NCL with an approved disease-modifying therapy (§12.1).
• Ultrastructure: curvilinear profiles.
• Carrier frequency: ~1/500 US-adjusted (Gene 2016).

4.4 Tier A — CLN5 (HGNC:2076) — juvenile is the predominant onset outside Finland

• MONDO: MONDO:0008768 (gene-level) — no juvenile-specific MONDO child exists. OMIM: 256731. Protein: CLN5, a soluble lysosomal glycoprotein; recently characterised as a lysosomal bis(monoacylglycero)phosphate synthase.
• Defensibility: strong, and the evidence is directionally surprising. CLN5 is classically the "Finnish variant late-infantile" gene (vLINCL^Fin, onset 4.5–6 y — already at the late-infantile/juvenile boundary). But in non-Finnish populations juvenile onset predominates. Xin W et al. screened 47 clinically diagnosed, molecularly unsolved NCL patients and found 10 with pathogenic CLN5 variants (11 previously undescribed), concluding: "The age at disease onset in this cohort is predominantly juvenile rather than late infantile. Importantly, we have identified 2 adult-onset patients who share a common pathogenic allele." (Neurology 2010;74(7):565-71; PMID:20157158). The title itself is the claim: "CLN5 mutations are frequent in juvenile and late-onset non-Finnish patients with NCL."
• Clinical texture: the same paper notes most patients presented with motor and visual impairment rather than seizures.
• Founder allele: the Finnish major mutation p.Tyr392* (historically "2467A>T").
• Ultrastructure: rectilinear, curvilinear, fingerprint.
• Curation note: CLN5 is arguably the strongest single argument that MONDO:0019262 must not be modelled as CLN3 — a non-Finnish patient with a juvenile NCL and no CLN3 variant has CLN5 as a leading candidate.

4.5 Tier A — CLN6 (HGNC:2077) — juvenile onset is in the classic, not atypical, range

• MONDO: MONDO:0011503 (gene-level) — no juvenile-specific child. OMIM: 601780 (CLN6), 204300 (Kufs type A). Protein: CLN6, a non-glycosylated ER transmembrane protein implicated in lysosomal acidification and in the CLN6–CLN8 (EGRESS) complex trafficking lysosomal enzymes from ER to Golgi.
• Defensibility: strong. GeneReviews classifies the classic phenotype as "Late infantile to juvenile" — CLN6 straddles the boundary by default. Mole & Cotman list "juvenile cerebellar ataxia" and "teenage progressive myoclonic epilepsy" among CLN6 phenotypes.
• Two recent, well-characterised juvenile-onset families, both with a phenotype that breaks the vision-first rule:

Kyriakou K et al. reported two Greek-Cypriot families: "We report clinical and genetic findings of three patients from two Greek-Cypriot families (families 915 and 926) with JNCL. All patients were males, and the first symptoms appeared at the age of 6 years. The proband of family 926 presented with loss of motor abilities, ataxia, spasticity, seizure, and epilepsy. The proband of family 915 had ataxia, spasticity, dysarthria, dystonia, and intellectual disability. Both probands did not show initial signs of vision and/or hearing loss." Molecular findings: "family 926 revealed two CLN6 biallelic variants: the novel, de novo p.Tyr295Cys and the known p.Arg136His variants. In family 915, both patients were homozygous for the p.Arg136His CLN6 variant." — Front Genet 2021;12:746101 (PMID:34868216), titled "A Novel CLN6 Variant Associated With Juvenile Neuronal Ceroid Lipofuscinosis in Patients With Absence of Visual Loss as a Presenting Feature."

A consanguineous Chinese family with a novel homozygous CLN6 c.14G>T (p.Arg5Leu): "Both patients exhibited seizures and progressive psychomotor decline and mental deterioration without visual impairment." — Neurodegener Dis 2021;21(5-6):126-131 (PMID:35609511), titled "Juvenile-Onset Kufs Disease in a Chinese Consanguineous Family due to CLN6 Mutation."

• The absent-visual-loss signature is the key CLN6 discriminator within the juvenile grouping and should be curated explicitly. Together with CLN8/EPMR, it establishes that visual failure is typical of MONDO:0019262 but not necessary.
• Also note: CLN6 additionally causes adult Kufs type A (autosomal recessive progressive myoclonic epilepsy) — so one gene spans late-infantile, juvenile, teenage, and adult windows. CLN6 is the clearest example of why gene ≠ onset class.
• Epidemiological weight: CLN6 was the most common genotype (42%) in a 153-patient pediatric NCL cohort (PMID:39281238), well ahead of CLN2 (16%) and CLN7 (12%).

4.6 Tier A — MFSD8 / CLN7 (HGNC:28486) — juvenile / late-juvenile protracted

• MONDO: MONDO:0012588 (gene-level) — no juvenile-specific child. OMIM: 610951. Protein: MFSD8/CLN7, a lysosomal major-facilitator-superfamily transmembrane transporter.
• Defensibility: strong. GeneReviews lists "Juvenile/late juvenile" as the recognised atypical phenotype. Kousi et al.'s foundational MFSD8 paper, while framed around variant late-infantile disease — "With one exception, the CLN7/MFSD8 mutation positive patients present a phenotype indistinguishable from the other vLINCL forms" (Brain 2009;132(Pt 3):810-9; PMID:19201763) — describes that exception as a Dutch patient with a protracted course who presented at age 11 with visual failure, with motor impairment and seizures in his mid-twenties and mental/speech regression in his thirties. That is a juvenile-onset, decades-long MFSD8 NCL.
• Founder allele: p.Thr294Lys, homozygous in 14 Roma patients from 12 families of the former Czechoslovakia.
• Allelic non-syndromic juvenile eye disease: Roosing S et al. identified compound-heterozygous MFSD8 variants causing nonsyndromic autosomal recessive macular dystrophy with central cone involvement, normal/subnormal full-field ERG but reduced multifocal ERG. Both families carried the mild missense p.Glu336Gln in trans to a severe allele (protein-truncating in one family, splicing-defect in the other), supporting an explicit dose model: "proposing a genotype-phenotype model where variant combinations determine disease severity." (Ophthalmology 2015;122(1):170-9; PMID:25227500).
• Why this matters here: MFSD8 produces a graded juvenile-onset visual spectrum from isolated maculopathy (no neurodegeneration) through juvenile NCL — the residual-activity gradient of §2.1 made visible in one gene.
• Ultrastructure: rectilinear, fingerprint.

4.7 Tier A — CLN8 (HGNC:2079) — EPMR / Northern epilepsy: juvenile by age, atypical by phenotype

• MONDO: MONDO:0009746-adjacent; gene-level MONDO:0008776 (Northern epilepsy) / MONDO:0012531 (CLN8 vLINCL) — ⚠️ these two CURIEs were not individually verified this session. OMIM: 600143 (EPMR/Northern epilepsy), 610003 (CLN8 vLINCL). Protein: CLN8, an ER/ERGIC transmembrane protein; partner of CLN6 in the EGRESS complex.
• Defensibility: strong on age, with an explicit clinical caveat. GeneReviews classifies the classic CLN8 phenotype as "Late infantile to juvenile." Ranta et al.'s positional cloning paper defines EPMR as "an autosomal recessive disorder characterized by onset of generalized seizures between 5 and 10 years, and subsequent progressive mental retardation", caused by a homozygous missense mutation "(70C-->G, R24G) that was not found in homozygosity in 433 controls" (Nat Genet 1999;23(2):233-6; PMID:10508524).
• 5–10 years is the textbook juvenile window. EPMR therefore belongs in MONDO:0019262 on the grouping's own onset criterion — but it presents with epilepsy, not vision loss, and is comparatively mild (patients survive into middle age with intellectual disability).
• CLN8 is also the canonical illustration of intra-genic phenotypic divergence. Kousi et al.: "Phenotypic divergence is exemplified by different CLN8 mutations giving rise to two very different diseases, the mild CLN8 disease, EPMR (progressive epilepsy with mental retardation), and the more severe CLN8 disease, late infantile variant." (PMID:21990111)
• Founder allele: p.Arg24Gly, essentially restricted to a region of northern Finland (Kainuu), hence "Northern epilepsy". The Turkish vLINCL CLN8 alleles are distinct and produce the more severe late-infantile disease.
• Ultrastructure: curvilinear-like fingerprint, granular.

4.8 Tier B — CTSD / CLN10 (HGNC:2529) — juvenile-onset ataxia with retinopathy

• MONDO: MONDO:0012350 (gene-level) — ⚠️ CURIE not verified this session. OMIM: 610127. Protein: cathepsin D, a soluble lysosomal aspartyl protease.
• Defensibility: reported and credible, but few families. GeneReviews lists "Juvenile" among CTSD atypical phenotypes (classic = congenital). The primary evidence:
• Steinfeld R et al., "Cathepsin D deficiency is associated with a human neurodegenerative disorder" (Am J Hum Genet 2006; PMID:16685649) — established CTSD as an NCL gene, including a juvenile-onset sibship.
• Two consanguineous pedigrees "both with a juvenile onset of NCL" were characterised in Neurology 2014;83(20):1873-5 (PMID:25298308), titled "Cathepsin D deficiency causes juvenile-onset ataxia and distinctive muscle pathology." Family A carried a "homozygous missense mutation (p.G149V in exon 4 of CTSD)" with "juvenile onset of cerebellar ataxia and retinitis pigmentosa at around 15 years, which progressed to significant motor impairment and cognitive decline." Family B carried a "homozygous missense mutation… (p.Arg399His in exon 9 of CTSD)" with an "earlier age at onset of 8 years" and additionally "sensory axonal neuropathy." Fibroblast assay showed "a significant reduction in enzyme activity compared to controls." Muscle biopsy showed "granulovacuolar material in angular atrophic fibers in addition to the granular osmiophilic deposits that are diagnostic for neuronal ceroid lipofuscinosis."
• Notably, in a 153-patient pediatric NCL cohort, CLN10 was the only genotype presenting exclusively as juvenile (PMID:39281238) — a small-n but striking observation.
• Biochemically confirmable: cathepsin D enzyme activity assay in fibroblasts. Third of the three assayable members.
• Ultrastructure: GROD — the other GROD-in-a-juvenile gene alongside PPT1.

4.9 Tier B — ATP13A2 / CLN12 (HGNC:30213) — juvenile NCL / Kufor-Rakeb syndrome

• MONDO: MONDO:0017809 ("parkinsonism due to ATP13A2 deficiency") — this is an asserted direct child of MONDO:0019262. OMIM: 606693. Protein: ATP13A2/PARK9, a lysosomal P5B-type polyamine transporting ATPase.
• Defensibility: reported, single-family origin for the NCL designation, but ontologically endorsed. Bras J et al. described a family with typical NCL pathology in which exome sequencing found a homozygous ATP13A2 mutation segregating with disease, noting that "Mutations in ATP13A2 are a known cause of Kufor-Rakeb syndrome (KRS), a rare parkinsonian phenotype with juvenile onset", and concluding that NCL and KRS may share etiological mechanisms and "implicate the lysosomal pathway in Parkinson's disease." (Hum Mol Genet 2012;21(12):2646-50; PMID:22388936). Mole & Cotman Table 2 lists CLN12/ATP13A2 as juvenile. Zhang et al. 2025 give onset ~13 y.
• Phenotype: juvenile-onset parkinsonism with pyramidal signs, supranuclear gaze palsy, and cognitive decline (Kufor-Rakeb), plus NCL storage. The extrapyramidal dominance distinguishes it, though note that CLN3 adolescents also develop parkinsonism.
• Caveat for curation: the NCL designation for ATP13A2 rests on a small evidence base and the gene is far better known as a parkinsonism gene. Model as a member with explicit acknowledgement of the thin evidence, and preserve the KRS identity rather than flattening it into "juvenile NCL".

4.10 Tier B/borderline — GRN / CLN11 (HGNC:4601) — adolescent-to-young-adult, mostly outside the juvenile window

• MONDO: MONDO:0013839 (⚠️ not verified this session). OMIM: 614706. Protein: progranulin, a secreted glycoprotein processed to granulin peptides; lysosomal chaperone functions including interaction with prosaposin and cathepsin D.
• Defensibility: weak as a juvenile member; strong as an adult/teenage member. GeneReviews classifies CLN11 as "Teenage to adult." The defining paper reported two siblings homozygous for c.813_816del (p.Thr272Serfs*10), and its central point is the dosage dichotomy: "Heterozygous mutations in GRN are a major cause of frontotemporal lobar degeneration with TDP-43 inclusions (FTLD-TDP)… The age-at-onset and neuropathology of FTLD-TDP and NCL are markedly different. Our findings reveal an unanticipated link between a rare and a common neurological disorder and illustrate pleiotropic effects of a mutation in the heterozygous or homozygous states." (Smith KR et al., Am J Hum Genet 2012;90(6):1102-7; PMID:22608501). Reexamination of progranulin-deficient mice "revealed rectilinear profiles typical of NCL."
• Onset in homozygous GRN NCL is typically ~20–25 y (retinal dystrophy first, then ataxia, seizures, cognitive decline). Zhang et al. 2025 quote a wider "5–25 years" band; I could not verify a specific well-documented childhood-onset homozygous GRN case in this session.
• Recommendation: do NOT list GRN/CLN11 as a core member of MONDO:0019262. List it as an adjacent, adult-boundary entity to be excluded in differential reasoning, with a note that the youngest reported onsets brush the upper edge of adolescence.

4.11 Tier C — DNAJC5 / CLN4 (HGNC:24586) — adult only; exclude

• OMIM: 162350. Protein: cysteine-string protein alpha (CSPα), a synaptic-vesicle co-chaperone.
• Autosomal DOMINANT adult-onset Kufs disease, onset typically mid-20s to 40s, with progressive myoclonic epilepsy and dementia. GeneReviews: classic phenotype "Adult"; no atypical phenotypes listed.
• Explicitly not juvenile. The current gene list for adult NCL is stated cleanly in Jedličková I et al.: "Adult-onset neuronal ceroid lipofuscinoses (ANCL, Kufs disease) are rare hereditary neuropsychiatric disorders characterized by intralysosomal accumulation of ceroid in tissues… Although several causative genes have been identified (DNAJC5, CLN6, CTSF, GRN, CLN1, CLN5, ATP13A2), the genetic underpinnings of ANCL in some families remain unknown." (Eur J Hum Genet 2020;28(6):783-789; PMID:31919451). That paper also carries a diagnostic caution worth recording: a 30-bp in-frame DNAJC5 duplication "was not detected initially by standard Sanger sequencing due to a preferential PCR amplification of the shorter wild-type allele and allelic dropout of the mutated DNAJC5 allele. It was also missed by subsequent whole-exome sequencing (WES)."
• Note the overlap this creates: CLN6, CTSF, GRN, CLN1, CLN5, ATP13A2 appear in both the adult list above and (for CLN1, CLN5, CLN6, ATP13A2) the juvenile list. The same gene can be a member of the juvenile grouping and of the adult grouping via different alleles. This is the central structural fact about onset-defined NCL groupings and must not be modelled as an inconsistency.

4.12 Tier C — CTSF / CLN13 (HGNC:2531) — adult Kufs type B; exclude

• OMIM: 615362. Protein: cathepsin F, a lysosomal cysteine protease. GeneReviews: "Adult Kufs type B", no atypical phenotypes. Onset typically after age 20 (Zhang 2025: "After 20 years"). Ultrastructure: GROD and fingerprint.

4.13 Tier C — KCTD7 / CLN14 (HGNC:21957) — infantile/late-infantile PME; exclude

• OMIM: 611726 (progressive myoclonic epilepsy 3 with or without intracellular inclusions). Protein: potassium channel tetramerisation domain-containing 7, a cytoplasmic/peripherally membrane-associated protein. GeneReviews: classic "Late infantile"; Zhang 2025 gives onset ~14 months. No credible juvenile-onset NCL attribution.

4.14 Tier X — "CLN9" — withdrawn; the label survives in ontologies as an artefact

• MONDO:0012188 ("neuronal ceroid lipofuscinosis 9") is an asserted direct child of MONDO:0019262 — but there is no CLN9 gene. OMIM 609055 remains as a legacy entry.
• History: Schulz A et al. (2004; PMID:15349861, "Impaired cell adhesion and apoptosis in a novel CLN9 Batten disease variant") described a juvenile-onset NCL variant in two Serbian sisters and two German brothers, attributed to a putative new gene "CLN9". Kousi et al. still listed CLN4 and CLN9 as "provisionally named" in 2012 (PMID:21990111). El Haddad et al. (2012) subsequently identified a homozygous CLN5 nonsense mutation in affected siblings from one of the Schulz families, reclassifying that family as CLN5. ⚠️ The El Haddad PMID and the exact CLN5 variant nomenclature were reported to me in secondary search summaries and were not verified against a primary abstract this session — verify before citing.
• Curation guidance: CLN9 should be recorded as an invalid/withdrawn designation, retained only to explain legacy literature and legacy ontology terms. Do not curate it as a distinct genetic member. Its presence as a MONDO child of MONDO:0019262 is a defect worth reporting upstream, alongside the missing CLN5/CLN6/CLN7/CLN8/CLN10 juvenile children (§1.3).

4.15 Population allele frequencies and carrier burden

The best single quantitative source is an ExAC-based analysis of ~61,000 exomes across twelve NCL genes:

"Estimates of NCL incidence range from 0.6 to 14 per 100,000 live births but vary widely between populations and are influenced by whether patients are classified based upon clinical or genetic criteria. We investigated mutations in twelve NCL genes in ~61,000 individuals represented in the Exome Aggregation Consortium (ExAC) whole exome sequencing database… Carrier frequency was dependent on ethnicity, with the highest (1/75) observed for PPT1 in the Finnish. When data are adjusted for ethnic diversity within the USA, PPT1, TPP1 and CLN3 carrier frequencies were found to be the highest of the NCLs, each at ~1/500." — Gene 2016;593(2):284-91 (PMID:27553520)

That paper also carries a warning directly relevant to ClinVar-based curation:

"the analysis identified numerous variants that are annotated as pathogenic in public repositories but have a predicted frequency that is not consistent with patient studies. These variants appear to be neutral polymorphisms that are reported as pathogenic without validation."

Variant classification and origin: all variants are germline; ACMG/AMP classification applies; COSMIC/TCGA/ICGC are not applicable (no somatic component). ClinVar and the NCL Mutation and Patient Database (UCL, ucl.ac.uk/ncl-disease) are the two primary variant resources; the latter is NCL-specific and organises Patient Datasheets and Mutation Datasheets per gene. Kousi et al. catalogued 365 NCL-causing mutations across eight genes as of 2012 (PMID:21990111); Mole & Cotman put the figure at "more than a dozen genes containing over 430 mutations" by 2015 (PMID:26026925).

4.16 Epigenetics and chromosomal abnormalities

• Epigenetics: no established disease-causing epigenetic mechanism. No imprinting, no methylation-defined subtype. ENCODE/Roadmap/DiseaseMeth are not informative for this entity.
• Chromosomal abnormalities: the only recurrent structural variant of note is the CLN3 ~1.02-kb intragenic deletion — a small CNV detectable by targeted PCR or by exon-level dosage analysis, not by routine karyotype or standard-resolution chromosomal microarray. Larger multi-exon deletions occur in several NCL genes and are a recognised cause of "one variant found, one missing" cases. Karyotype and FISH have no role. CMA has a limited role only if it is exon-resolution over the relevant genes.

5. Environmental Information

• Environmental factors: none. No toxin, radiation, pollutant, or occupational exposure contributes to NCL causation. CTD/TOXNET/EPA are not informative sources for this entity.
• Lifestyle factors: none causal. Nutritional status and aspiration risk affect outcome, not etiology.
• Infectious agents: none. NCL is not infectious, not triggered by infection, and has no zoonotic dimension. Intercurrent febrile illness may transiently worsen seizure control — a nonspecific epilepsy effect.

This section is genuinely empty for MONDO:0019262 and should be curated as such, rather than padded.

6. Mechanism / Pathophysiology

6.1 The shared final common pathway

All members converge on lysosomal dysfunction with accumulation of autofluorescent ceroid-lipofuscin, then on neuronal death with regional selectivity. The proximal defects are heterogeneous — soluble lysosomal enzymes (PPT1, TPP1, CTSD, CTSF), a soluble lysosomal protein (CLN5), a secreted protein (GRN), cytosolic/membrane-peripheral proteins (DNAJC5, KCTD7), and multiple transmembrane proteins at different subcellular locations (CLN3, CLN6, MFSD8, CLN8, ATP13A2):

"These genes encode lysosomal enzymes (CLN1, CLN2, CLN10, CLN13), a soluble lysosomal protein (CLN5), a protein in the secretory pathway (CLN11), two cytoplasmic proteins that also peripherally associate with membranes (CLN4, CLN14), and many transmembrane proteins with different subcellular locations (CLN3, CLN6, CLN7, CLN8, CLN12). For most NCLs, the function of the causative gene has not been fully defined." — Mole & Cotman, PMID:26026925

Proposed causal chain for the grouping (upstream → downstream):

1. [MOLECULAR] Biallelic hypomorphic variant → partial loss of gene-product function (residual activity in the juvenile-permissive range).
2. [MOLECULAR] Failure of the specific lysosomal degradative/transport step → substrate accumulation.
3. [CELLULAR] Progressive intralysosomal accumulation of autofluorescent ceroid-lipofuscin storage material.
4. [CELLULAR] Autophagic-lysosomal pathway failure — accumulation of autophagic vacuoles, impaired autophagosome-lysosome fusion, impaired lysosomal acidification.
5. [CELLULAR/TISSUE] Reactive microgliosis and astrogliosis; neuroinflammation, which in NCL models precedes and predicts regional neuron loss.
6. [TISSUE] Selective neurodegeneration — photoreceptors and retinal ganglion cells early (accounting for the vision-first phenotype), then cortical layers II/III/V, cerebellar Purkinje and granule cells, thalamic relay nuclei.
7. [ORGANISM] Visual failure → cognitive/behavioural decline → epilepsy → motor decline → death.

6.2 The stored material — a genuine mechanistic discriminator

The storage body composition is not uniform, and this maps onto ultrastructure and onto gene:

• In most NCLs the major stored protein is subunit c of mitochondrial ATP synthase (SCMAS) — established by Palmer and colleagues: "Mitochondrial ATP synthase subunit c storage in the ceroid-lipofuscinoses (Batten disease)", Am J Med Genet 1992 (PMID:1535179), and the companion immunocytochemical study, Am J Med Genet 1995 (PMID:7668326). This corresponds to curvilinear/rectilinear/fingerprint ultrastructure.
• In CLN1/PPT1 and CLN10/CTSD, the predominant stored proteins are instead saposins A and D, and the ultrastructure is GROD. ⚠️ The saposin attribution is standard in the field (Tyynelä et al.) but I did not verify a specific PMID for it this session.

Curation consequence: a juvenile NCL entry should model two storage-composition branches (SCMAS-type vs saposin/GROD-type), not one.

6.3 Gene-specific proximal mechanisms

⚠️ GO IDs above are suggestions from domain knowledge and were not validated against OAK/OLS this session. Run just validate-terms before committing any of them.

Additional shared-process GO terms: GO:0007041 lysosomal transport; GO:0006914 autophagy; GO:0061919 process utilizing autophagic mechanism; GO:0006954 inflammatory response; GO:0050808 synapse organization; GO:0070997 neuron death. (Same validation caveat.)

6.4 Cross-gene interdependence

An important and under-modelled mechanism: NCL proteins regulate one another, so a single-gene lesion produces a multi-protein lysosomal deficit. "Loss of CLN3 has been shown to affect PPT1, TPP1, CLN5, and CTSD" (Zhang et al. 2025, PMID:39925015). CLN6 and CLN8 act as an obligate complex. Progranulin regulates prosaposin and cathepsin D. This explains both phenotypic convergence and why enzyme assays can be mildly abnormal in the "wrong" NCL.

6.5 Immune system involvement

Neuroinflammation, not autoimmunity or immunodeficiency. Microglial and astrocytic activation is early, regionally patterned, and in models precedes neuron loss: "Neuroimmune responses mediated by astrocytes and microglia are integral to the progression of neurodegenerative diseases" (PMID:39925015). Autoantibodies to GAD65 have been reported in CLN3 patients and provide the rationale for the immunosuppression trials in §12.4 — but CLN3 disease is not an autoimmune disease, and the entry should not be modelled as one.

Relevant CL terms: CL:0000129 microglial cell; CL:0000127 astrocyte; CL:0000540 neuron; CL:0000573 retinal cone cell; CL:0000604 retinal rod cell; CL:0000740 retinal ganglion cell; CL:0000121 Purkinje cell; CL:0000117 CNS neuron (sensu Vertebrata). (Not OAK-validated this session.)

6.6 Tissue damage mechanisms

Progressive neuronal death (apoptotic and non-apoptotic), synaptic loss preceding somatic loss, axonal/neuritic dystrophy, oxidative stress, and secondary mitochondrial dysfunction. Photoreceptor outer-segment degeneration precedes ganglion-cell loss in the retina. Reactive gliosis and progressive brain atrophy on MRI, most marked cerebellar and cortical.

6.7 Molecular profiling

• Transcriptomics/proteomics/lipidomics: substantial data exist for CLN3 mouse and ovine CLN5/CLN6 models; human data are sparse. GEO/PRIDE hold NCL model datasets. Lipidomics is mechanistically important given the CLN5-BMP-synthase finding and the lipid nature of the storage material.
• Single-cell / spatial: emerging in NCL mouse models (microglial state transitions); no established human single-cell atlas for any juvenile NCL member.
• Functional genomics screens: DepMap and CRISPR screens have been used for CLN3 interactor discovery; nothing definitive at grouping level.

Honest statement for the KB: molecular-profiling evidence for MONDO:0019262 as a grouping is thin and almost entirely model-derived; gene-specific human omics is largely absent.

7. Anatomical Structures Affected

7.1 Organ level

Primary: - Central nervous system — UBERON:0001017 central nervous system; UBERON:0000955 brain; UBERON:0000956 cerebral cortex; UBERON:0002037 cerebellum; UBERON:0001897 dorsal thalamus / UBERON:0001879? (verify); UBERON:0002420 basal ganglia (verify); UBERON:0002240 spinal cord. - Eye / retina — UBERON:0000970 eye; UBERON:0000966 retina; UBERON:0001782 macula lutea (verify); UBERON:0000941 optic nerve (verify).

Secondary / systemic (storage is ubiquitous even where dysfunction is not): - Skin/eccrine sweat glands (the classic EM biopsy site), rectal mucosa, conjunctiva, skeletal muscle (granulovacuolar change in CTSD deficiency), peripheral blood lymphocytes (CLN3), myocardium (CLN3). - Body systems: nervous (primary), visual/sensory (primary), musculoskeletal (secondary — contractures, scoliosis), cardiovascular (CLN3-specific, late), respiratory (aspiration), digestive (dysphagia, malnutrition).

7.2 Tissue and cell level

Neural tissue is the target. Affected cell populations: retinal photoreceptors (rods and cones), retinal ganglion cells, cortical pyramidal neurons (layers II/III and V), cerebellar Purkinje cells and granule cells, thalamic relay neurons, with prominent involvement of microglia and astrocytes as active participants rather than bystanders. Storage material is also present in non-neural cells — fibroblasts, lymphocytes, eccrine gland epithelium — which is what makes peripheral biopsy diagnostically possible.

7.3 Subcellular level

• GO:0005764 lysosome — the primary compartment.
• GO:0005765 lysosomal membrane — locus of the transmembrane members (CLN3, MFSD8, ATP13A2).
• GO:0005783 endoplasmic reticulum — locus of CLN6/CLN8.
• GO:0005794 Golgi apparatus — the EGRESS trafficking route.
• GO:0005776 autophagosome; GO:0005739 mitochondrion (SCMAS origin; secondary dysfunction); GO:0008021 synaptic vesicle (CSPα/CLN4, adult).

(GO CC IDs not OAK-validated this session.)

7.4 Localization and laterality

Bilateral and symmetric throughout. Retinopathy is bilateral; cerebral and cerebellar atrophy are symmetric. Asymmetry is not a feature and should prompt reconsideration of the diagnosis. The characteristic MRI pattern is early cerebellar atrophy with progressive generalised cerebral atrophy, periventricular white-matter T2 hyperintensity, and thalamic T2 hypointensity (best described in CLN2/late-infantile but seen across members).

8. Temporal Development

8.1 Onset

• Age: the grouping's defining criterion — ~5–10 years, extended by some authors to 4–15 y. Practically: first symptom after the child has been developmentally normal through early schooling, and before the adult (>18 y) Kufs window.
• Pattern: insidious and chronic-progressive. Not acute, not subacute. The apparent "sudden" presentation is usually delayed recognition of insidious visual loss.
• Onset symptom by member: visual failure (CLN3, CLN5, CLN7, CLN10, many CLN1-juvenile) | seizures (CLN8/EPMR, many CLN6) | ataxia (TPP1-SCAR7, CLN6, CLN10) | parkinsonism (ATP13A2/CLN12) | behavioural/school problems (CLN3, frequently the true first sign in retrospect).

8.2 Progression and staging

A four-stage frame applies across the grouping (mapped most precisely for CLN3):

Formal instruments (CLN3-specific, not grouping-general): - Unified Batten Disease Rating Scale (UBDRS) — four subscales: physical (28 items, 0–112), seizure (12 items, 0–54), behaviour (9 items, 0–55), capability (5 items, 0–14). Validated in an independent CLN3 sample (PMC9879304). - CLN3 Disease Staging System (PMID:32300063). - CLN2 Clinical Rating Scale (Hamburg/Weill Cornell motor-language domains) — used as the primary endpoint in the cerliponase alfa trial.

There is no validated rating instrument for MONDO:0019262 as a whole; this is a real gap for any grouping-level natural-history or trial work.

8.3 Progression rate and course

Progression rate is genotype- and allele-dependent, and is the main axis of within-grouping variation: - Rapid: classic CLN3 (death typically third decade); juvenile CLN1 (faster than CLN3). - Intermediate: CLN6 juvenile, CLN5 juvenile, CLN10 juvenile. - Slow/protracted: TPP1-SCAR7 (ataxia only for decades); protracted MFSD8 (visual failure at 11, motor/seizures in mid-20s, mental/speech regression in the 30s); CLN8-EPMR (survival into middle age with seizure attenuation after puberty).

Course pattern: progressive, chronic, lifelong. No remission, spontaneous or treatment-induced, has ever been reported for any member. Disease duration: from onset to death, ~10–25 years for the classic members; longer for protracted forms.

8.4 Critical periods

• Diagnostic window: the 1–3 years between visual failure and cognitive decline is the therapeutic opportunity. Diagnostic delay in this window is the norm and is the dominant modifiable failure in current care.
• Therapeutic window: for the only disease-modifying therapy available (cerliponase alfa in CLN2), benefit is preservation of remaining function, not recovery — so the window closes as function is lost. The same logic applies to every gene therapy in trial. Zhang et al. put this bluntly: "these therapies are unlikely to achieve partial disease reversal, and complete reversal remains improbable."
• Presymptomatic identification (sibling cascade testing) is currently the only route to treatment before loss.

9. Inheritance and Population

9.1 Epidemiology

Grouping-level, from the Scandinavian survey (Neuropediatrics 1997;28(1):6-8; PMID:9151309) — the best direct data on juvenile NCL as such:

"For juvenile NCL 40 Swedish living patients were identified. The corresponding number for Finland was 61, for Norway 28, for Denmark 16 and for Iceland three. The prevalence of juvenile NCL was thus 4.6, 12.2, 6.5, 3.1 and 11 per million inhabitants in Sweden, Finland, Norway, Denmark, and Iceland, respectively. For calculating incidence the years 1976-85 were used. The incidence was 2.2 per 100,000 live births in Sweden, 4.8 in Finland, 3.7 in Norway, 2.0 in Denmark, and 7.0 in Iceland."

Structured for curation:

Italy (Orphanet J Rare Dis 2013;8:19; PMID:23374165), a contrasting low-incidence population, and directly informative about the juvenile fraction:

"One hundred eighty-three NCL patients from 156 families were recruited between 1966 and 2010… Late infantile onset NCL (LINCL) accounted for 75.8% of molecularly confirmed cases, the most frequent form being secondary to mutations in CLN2 (23.5%). Juvenile onset NCL patients accounted for 17.7% of this cohort, a smaller proportion than found in other European countries. … An incidence rate of 0.98/100,000 live births was found in 69 NCL patients born between 1992 and 2004, predicting 5 new cases a year. Prevalence was 1.2/1,000,000."

All-NCL range (Gene 2016, PMID:27553520): "Estimates of NCL incidence range from 0.6 to 14 per 100,000 live births but vary widely between populations."

Synthesis: juvenile NCL is best characterised as ~2–5 per 100,000 live births in Northern Europe, ~0.2–1 per 100,000 in Southern Europe, with point prevalence of roughly 0.3–1.2 per 100,000 population. In Orphanet prevalence-class terms this is BAND_1_9_PER_1000000 for point prevalence in most European populations. The juvenile fraction of all NCL ranges from ~18% (Italy) to a majority in Northern Europe.

9.2 Inheritance genetics

• Pattern: autosomal recessive for every defensible member. HPO HP:0000007 Autosomal recessive inheritance. (The one AD NCL, DNAJC5/CLN4, is adult-onset and excluded.)
• Penetrance: complete for biallelic pathogenic genotypes. No reported non-penetrant biallelic carriers.
• Expressivity: variable, including within families sharing a genotype (documented in the Chinese CLN6 sibship, PMID:35609511). Age of onset and MRI severity varied between siblings with the identical homozygous missense variant.
• Anticipation: not applicable — no repeat-expansion mechanism in any NCL gene.
• Germline mosaicism: not documented as a recurrence mechanism; recurrence risk for AR members is the standard 25%. Note the de novo p.Tyr295Cys CLN6 allele in Cypriot family 926 (PMID:34868216) — a de novo event on one allele in an otherwise recessive disorder, which alters recurrence counselling for that family.
• Consanguinity: a major contributor. Homozygosity for private missense alleles in consanguineous pedigrees is the modal route to the non-CLN3 juvenile forms (Cypriot, Chinese, Somali, Turkish, Roma reports).
• Carrier frequency: PPT1 1/75 in Finns; PPT1, TPP1, CLN3 each ~1/500 US-adjusted (PMID:27553520).

9.3 Population demographics and geography

• Sex ratio: 1:1. Autosomal recessive; no sex bias in incidence. (The Cypriot CLN6 series happened to be all male — a chance finding in n=3, not a sex effect.)
• Age distribution of affected individuals: by definition onset 5–10 y; the prevalent population spans childhood through the third decade (longer for protracted members).

10. Diagnostics

10.1 The diagnostic algorithm for a suspected juvenile NCL

This is the practical heart of the entity, and it is genuinely different from the CLN3-only algorithm.

Step 1 — Recognise the syndrome. School-age child with progressive visual failure + retinal dystrophy, or new-onset epilepsy with cognitive regression, or progressive ataxia with cognitive decline.

Step 2 — Enzyme assays first (fast, cheap, and immediately actionable). In leukocytes, fibroblasts, or dried blood spot: - TPP1 (CLN2) — do this first: it is the only NCL with an approved therapy. - PPT1 (CLN1) — will catch juvenile CLN1, which EM would mislabel as infantile. - Cathepsin D (CLN10).

A normal result on all three excludes three of the nine defensible members in days.

Step 3 — Blood film for vacuolated lymphocytes. Positive → strongly suggests CLN3. Cheap, immediate, and one of the few within-grouping discriminators available at the bedside.

Step 4 — Molecular testing. Targeted CLN3 common-deletion PCR if the phenotype is classic; otherwise, and in all enzyme-negative cases, a multigene NCL panel covering at minimum PPT1, TPP1, CLN3, DNAJC5, CLN5, CLN6, MFSD8, CLN8, CTSD, GRN, ATP13A2, CTSF, KCTD7. Exome or genome sequencing where panel is negative or where the differential is broader (juvenile-onset ataxia, PME, or retinal dystrophy differentials). GTR lists dedicated NCL/Batten panels. - Copy-number analysis must be included — the CLN3 1.02-kb deletion and multi-exon deletions in other NCL genes are missed by SNV-only pipelines. - A known WES failure mode: in-frame duplications can be missed by both Sanger and WES through allelic dropout — the DNAJC5 case in PMID:31919451. Reanalysis of raw WES data with modified protocols recovered it.

Step 5 — Electron microscopy (skin/conjunctival/rectal biopsy) is now second-line but retains value in molecularly unsolved cases. Interpret ultrastructure as a pointer to the gene, not to the onset class (§4.2).

10.2 Ultrastructural patterns (GeneReviews Table 2)

10.3 Imaging, electrophysiology, and other tests

• MRI brain: cerebellar atrophy (early and often disproportionate), progressive generalised cerebral atrophy, periventricular T2 white-matter hyperintensity, thalamic T2 hypointensity. In one CLN6 sibship MRI severity diverged sharply between siblings with an identical genotype (PMID:35609511) — MRI is not a reliable genotype predictor.
• ERG: abnormal early, becomes extinguished; often the finding that first raises "retinal dystrophy" before the neurological diagnosis. HP:0000512 → HP:0000550.
• OCT / fundus autofluorescence: retinal thinning, outer-retinal loss, abnormal autofluorescence (HP:0030602).
• VEP: enlarged/giant responses early (as in other PMEs), attenuating later.
• EEG: progressive slowing; generalised epileptiform discharges; photoparoxysmal response at low flash frequencies is a classic PME/NCL clue.
• Nerve conduction: may show sensory axonal neuropathy (CTSD; PMID:25298308).
• Muscle biopsy: not routine, but in CTSD deficiency shows "granulovacuolar material in angular atrophic fibers in addition to the granular osmiophilic deposits" (PMID:25298308).
• CSF neurofilament light chain: an emerging progression biomarker in NCL — not validated for diagnosis; do not curate as a diagnostic test.

LOINC coding exists for the enzyme assays and for ERG; ⚠️ specific LOINC codes were not retrieved this session.

10.4 Omics-based diagnostics

• RNA sequencing has a real role: resolving splice-region VUS in NCL genes (functional splicing evidence for ACMG PS3/BS3). Worth curating as an adjunct, not a first-line test.
• Proteomics / metabolomics / liquid biopsy: no validated clinical diagnostic role for any NCL.
• Epigenomics: no role.

10.5 Clinical criteria and differential diagnosis

There are no formal consensus diagnostic criteria for "juvenile NCL" as a grouping. Diagnosis is by demonstration of a biallelic pathogenic genotype in an NCL gene, with a compatible juvenile-onset phenotype (and, historically, characteristic storage on EM).

Differential diagnosis, organised by presenting syndrome — this is where the grouping earns its keep:

Niemann-Pick type C and Wilson disease deserve specific mention as the two mimics where missing the diagnosis has the greatest therapeutic cost.

10.6 Screening

• Newborn screening: not implemented anywhere for any NCL. TPP1 enzyme activity in dried blood spot is technically NBS-compatible and has been piloted; the argument for it strengthened materially once cerliponase alfa was approved, but it targets CLN2 (predominantly late-infantile), not the juvenile grouping. Adding NCL to RUSP-type panels remains an open policy question.
• Carrier screening: available for known familial variants; expanded carrier-screening panels increasingly include CLN3, PPT1, TPP1. Population carrier screening is not recommended outside founder populations.
• Cascade screening of siblings is the highest-yield screening activity, because it can identify a presymptomatic sibling within the therapeutic window.

11. Outcome / Prognosis

11.1 Survival and mortality

• Uniformly fatal for the classic members. No cure exists for any form.
• CLN3: life expectancy typically second to third decade; deaths reported from late teens into the 30s and occasionally 40s.
• Juvenile CLN1: generally more rapid than CLN3.
• Protracted members (TPP1-SCAR7, protracted MFSD8, CLN8-EPMR): survival into the fourth to sixth decade, with EPMR patients reaching middle age.
• Mortality mechanisms: aspiration pneumonia (the leading cause), respiratory failure, status epilepticus, and — in CLN3 specifically — cardiac arrhythmia/conduction disease in the second-to-third decade.
• ⚠️ No formal 5-/10-year survival statistics exist for the grouping. There is no SEER-equivalent registry. Any percentage survival figure encountered in the literature is almost certainly CLN3-specific and cohort-specific — do not generalise it to MONDO:0019262.

11.2 Morbidity and function

Profound and cumulative: blindness → dementia → epilepsy → loss of ambulation → loss of speech → gastrostomy dependence → total care dependence. Effectively 100% disability by the end of the second decade in the classic members. Very high caregiver burden; the psychiatric phase in CLN3 adolescence is repeatedly reported as the hardest for families.

Quality-of-life instruments: no NCL-specific validated QoL instrument is in general use. Generic pediatric instruments (PedsQL, EQ-5D-Y) are poorly suited once vision and cognition are lost. The UBDRS capability subscale is the closest available functional measure and is CLN3-validated only. This is a documented measurement gap, and the honest curation statement is that grouping-level QoL data do not exist.

11.3 Complications

Status epilepticus; aspiration pneumonia; malnutrition and failure to thrive; contractures and neuromuscular scoliosis; osteopenia/fractures; pressure injury; sleep disorder; behavioural crisis and psychosis; in CLN3, cardiac conduction disease and arrhythmia; drug-refractory epilepsy.

11.4 Recovery potential

None. No spontaneous or treatment-induced remission has been described. Even the best-evidenced disease-modifying therapy (cerliponase alfa) slows decline rather than reversing it — a point the field states plainly (Zhang et al., PMID:39925015). Rehabilitation preserves function and comfort but does not alter trajectory.

11.5 Prognostic factors

• Genotype is the dominant prognostic factor, and is the strongest argument for pursuing molecular diagnosis even when it does not change treatment: it changes the prognosis conversation from "second-to-third decade" (CLN3) to "possibly middle age" (EPMR, SCAR7, protracted MFSD8).
• Allele severity within a gene: null/null → earlier onset and faster decline; hypomorph in trans → later onset, slower decline (the TPP1 CLN2-vs-SCAR7 dichotomy is the cleanest demonstration).
• Age at onset: earlier onset predicts faster progression, consistently across members.
• Seizure control and nutritional/respiratory management are the main modifiable prognostic factors.
• Prognostic biomarkers: none validated. CSF NfL and MRI volumetrics are under investigation.

12. Treatment

Overarching statement: there is no approved disease-modifying therapy for MONDO:0019262 as a grouping. There is exactly one approved disease-modifying therapy for one member gene (TPP1/CLN2), and a set of gene therapies in trial. Everything else is symptomatic and supportive.

12.1 Enzyme replacement therapy — TPP1/CLN2 only

Cerliponase alfa (Brineura) — recombinant human TPP1 delivered by intracerebroventricular infusion via an implanted reservoir, 300 mg every 2 weeks. Approved by FDA (2017) and EMA for CLN2 disease. Pivotal evidence:

"The mean (±SD) unadjusted rate of decline in the motor-language score per 48-week period was 0.27±0.35 points in treated patients and 2.12±0.98 points in 42 historical controls." — Schulz A et al. "Study of Intraventricular Cerliponase Alfa for CLN2 Disease." N Engl J Med 2018;378(20):1898-1907 (PMID:29688815)

The trial enrolled 24 children aged 3–16, all receiving 300 mg for at least 96 weeks; median time to a 2-point motor-language decline was not reached in treated patients versus 345 days in controls (P<0.001). Adverse events: convulsions, fever, vomiting, hypersensitivity reactions; two patients developed device-related infections requiring antibiotic therapy and device replacement.

Relevance to this entity: the age range 3–16 y means the trial population included juvenile-onset CLN2 patients. Any child in the juvenile window with a compatible phenotype should have a TPP1 assay early, because this is the one branch of the differential with an approved therapy. This single fact is the strongest clinical justification for modelling MONDO:0019262 as gene-heterogeneous rather than as CLN3.

• NCIT: NCIT:C15986 Pharmacotherapy; NCIT:C158784? (a specific cerliponase alfa NCIT code likely exists but was not verified this session). therapeutic_modality: PROTEIN_REPLACEMENT.
• ERT is not extensible to the transmembrane members (CLN3, CLN6, MFSD8, CLN8, ATP13A2) — there is no soluble enzyme to replace. Zhang et al.: ERT is "limited to soluble lysosomal enzyme deficiencies due to blood-brain barrier challenges."

12.2 Gene therapy (investigational)

AAV-vectored gene transfer, largely AAV9 by intrathecal or intracerebroventricular route. Trials have been run or are running for CLN2, CLN3, CLN5, CLN6, and CLN7:

Preclinical support is strongest where large-animal models exist: intracerebroventricular scAAV9.CB.CLN6 "significantly alleviates motor defects, delays learning and memory impairment, and extends lifespan" (reviewed in PMID:39925015), and the naturally occurring ovine CLN5/CLN6 models have carried much of the translational work (§15).

NCIT: NCIT:C15238 Gene Therapy; therapeutic_modality: GENE_THERAPY.

12.3 RNA-based therapy

Milasen — the landmark n-of-1 patient-customised splice-modulating antisense oligonucleotide, designed against a cryptic splice-acceptor site created by a MFSD8/CLN7 retrotransposon insertion, designed, manufactured, and dosed within about a year (Kim J et al., N Engl J Med 2019;381:1644-1652). ⚠️ The PMID for this paper could not be confirmed by the searches run this session — verify before citing. Referenced in the 2025 review as "Milasen, designed to target…cryptic splice-acceptor site" (PMID:39925015).

Significance for this entity: milasen is the proof of concept that a private allele in a rare member of this grouping can be drugged, and is a strong argument for exact molecular diagnosis rather than a syndromic "juvenile NCL" label.

NCIT: NCIT:C15986; therapeutic_modality: ANTISENSE_OLIGONUCLEOTIDE; aso_mechanism: SPLICE_MODULATION_EXON_INCLUSION (mechanism assignment should be confirmed against the primary paper).

12.4 Other investigational and repurposed approaches

• Miglustat — substrate-reduction agent; open-label safety/PK/efficacy study in CLN3 (NCT05174039).
• Mycophenolate mofetil — immunosuppression rationale from the autoimmune/neuroinflammatory arm of CLN3 pathogenesis (NCT01399047).
• Small molecules: NtBuHA (a cysteamine-derived thioesterase mimetic, CLN1), trehalose (autophagy inducer), gemfibrozil (PPARα agonist) — all preclinical/early (PMID:39925015).
• Hematopoietic stem-cell gene therapy: "Overexpressing PPT1 on hematopoietic stem cells… has been shown to extend the lifespan of CLN1-deficient mice" (PMID:39925015). Unmodified HSCT has not shown benefit in NCL and should not be offered.
• Microglial replacement therapies — an emerging concept given the centrality of neuroinflammation.
• Investigations of Juvenile Neuronal Ceroid Lipofuscinosis — NCT03307304; Natural History Study of Batten Disease — NCT04644549.

12.5 Symptomatic and supportive care (the mainstay)

12.6 Pharmacogenomics

No NCL-specific pharmacogenomic guidance exists. Standard CPIC guidance applies to the drugs used (e.g. HLA-B*15:02 and carbamazepine — moot here, since carbamazepine is relatively contraindicated; CYP2C9/CYP2C19 for valproate/clobazam metabolism). PharmGKB has no NCL-specific entries.

12.7 Treatment strategy

The algorithm is short and genotype-gated:

1. Establish the gene. Enzyme assays → panel/WES → CNV analysis.
2. If TPP1/CLN2 → refer for cerliponase alfa immediately. This is the only branch with an approved therapy, and benefit depends on remaining function.
3. If another member → assess trial eligibility (CLN3 NCT03770572, CLN6 NCT02725580/NCT07582484, CLN7 AAV9, CLN5 natural history) and enrol in natural-history registries.
4. In all cases → multidisciplinary symptomatic care (neurology, ophthalmology/low vision, epileptology, gastroenterology/nutrition, rehabilitation, palliative care, genetics).
5. In all cases → sibling cascade testing, to catch a presymptomatic sibling while the therapeutic window is open.

13. Prevention

• Primary prevention: not possible. These are germline monogenic disorders. No vaccination, no risk-factor modification, no behavioural intervention affects occurrence. Curate this section as explicitly not-applicable rather than inventing content.
• Reproductive prevention is the only route that reduces incidence:
• Genetic counselling (NCIT:C15240) — 25% recurrence risk for AR members; discussion of consanguinity where relevant.
• Carrier testing of at-risk relatives once the familial variants are known.
• Prenatal diagnosis (CVS/amniocentesis) and preimplantation genetic testing for monogenic disease (PGT-M) — both routine once the biallelic genotype is defined.
• Population carrier screening in founder populations (Finland; Roma communities for MFSD8 p.Thr294Lys) is technically justifiable; consanguineous-community screening programmes are the highest-yield setting.
• Secondary prevention (early detection):
• Cascade testing of siblings — the single highest-value preventive act, and the only one that can place a child in the therapeutic window.
• Newborn screening — not implemented; TPP1 dried-blood-spot assay is the leading candidate now that CLN2 is treatable. This is a live policy question, not current practice.
• Awareness-driven earlier diagnosis: an ophthalmologist encountering a school-age child with rapidly progressive maculopathy and an abnormal ERG should consider NCL, not stop at "Stargardt". Diagnostic-delay reduction is the most tractable secondary-prevention target for this entity.
• Tertiary prevention (complication avoidance): seizure-medication optimisation with avoidance of myoclonus-aggravating agents; dysphagia surveillance and timely gastrostomy to prevent aspiration; scoliosis and contracture surveillance; cardiac surveillance in CLN3; vaccination and respiratory care.
• Public health / environmental interventions: not applicable.

14. Other Species / Natural Disease

NCL is one of the best examples in medicine of a human rare disease with naturally occurring, breed-defined large-animal counterparts — which is why NCL gene therapy has an unusually strong translational pipeline.

14.1 Taxonomy and natural disease

Breeds (VBO): Tibetan Terrier (ATP13A2/CLN12; onset 4–6 y, i.e. adult-equivalent in dog terms), American Staffordshire Terrier (ARSG, an NCL-like disorder with no confirmed human juvenile NCL counterpart — ARSG in humans causes Usher syndrome type IV), Border Collie and Golden Retriever (CLN5), Australian Shepherd and Schapendoes (CLN6), English Setter (CLN8), Dachshund (TPP1, PPT1), Miniature Schnauzer, Chihuahua. ⚠️ Specific VBO identifiers were not retrieved this session.

Veterinary importance: canine NCL is a genuine clinical veterinary disease with commercial DNA tests offered by breed clubs for carrier avoidance — a real-world instance of the carrier-screening logic in §13. Border Collie NCL in Japan has been the subject of a dedicated molecular-epidemiological study (PMID:22919312).

14.2 Orthologous genes

All human NCL genes have well-conserved orthologues across mammals; PPT1, TPP1, CTSD, CLN3, CLN5, CLN6, CLN8, MFSD8, and ATP13A2 orthologues exist in mouse, rat, dog, sheep, and (for most) zebrafish and Drosophila. Alliance of Genome Resources and HomoloGene are the reference sources; ⚠️ specific NCBI Gene IDs were not retrieved this session.

14.3 Comparative biology

• Conservation of mechanism is high: lysosomal storage, SCMAS accumulation, autofluorescence, neuroinflammation, and retinal plus CNS neurodegeneration recur across species. Ranta et al. put the CLN8/mnd correspondence as "the first description of the molecular basis of a naturally occurring animal model for NCL" (PMID:10508524).
• Key comparative divergence: disease tempo and retinal involvement differ. Ovine CLN5/CLN6 recapitulate retinal degeneration well (PMC8901734 — natural history of retinal degeneration in ovine CLN5/CLN6) and brain atrophy is trackable by MRI (PMC9830986); rodent models often under-recapitulate the retinal phenotype that dominates the human juvenile presentation.
• Zoonotic potential / cross-species transmission: none. Genetic disease; not transmissible.

15. Model Organisms

15.1 Mouse (Mus musculus, NCBITaxon:10090) — MGI, IMPC, IMSR, JAX

Conditional and cell-type-specific alleles exist for several (notably Cln3 and Ppt1), enabling dissection of the neuron-vs-glia contribution to neuroinflammation.

Cross-cutting mouse limitation, stated honestly: the mouse models under-recapitulate the two features that define the human juvenile phenotype — early profound visual failure and death in the second-to-third decade. A dismech entry should record this as a HUMAN_MODEL_MISMATCH discussion rather than a generic knowledge gap: the evidence exists in the model, but its translational validity for the juvenile-onset human phenotype is the open question.

15.2 Large animals — the translational workhorses

• Sheep: the Borderdale CLN5 and South Hampshire / Merino CLN6 flocks (New Zealand) are naturally occurring, well-characterised, and gyrencephalic with a brain size and lifespan permitting realistic dosing, surgical delivery, and longitudinal imaging. Published resources include progressive MRI brain-volume studies (PMC9830986) and natural-history studies of retinal degeneration (PMC8901734). These models carry much of the credibility of the CLN5/CLN6 gene-therapy programmes.
• Dog: TPP1 Dachshund, CLN5 Border Collie/Golden Retriever, CLN6 Australian Shepherd/Schapendoes/mixed-breed, CLN8 English Setter, ATP13A2 Tibetan Terrier. The Dachshund TPP1 model contributed to ERT development.
• Cattle: Devon CLN5.

15.3 Non-mammalian and in vitro

• Zebrafish (Danio rerio, NCBITaxon:7955) — ZFIN; cln3, mfsd8, ppt1, tpp1 morphants/mutants. Value: rapid, optically transparent, well-suited to retinal phenotyping and small-molecule screening — arguably the best system for the visual arm of this grouping.
• Drosophila melanogaster (NCBITaxon:7227) — FlyBase; Cln3, Ppt1, Cln7 models for genetic-modifier screens.
• C. elegans, yeast — used for CLN3 and MFSD8 orthologue function.
• Patient-derived fibroblasts — the practical workhorse for enzyme assays and storage-material characterisation; the substrate for the CTSD activity measurements in PMID:25298308.
• iPSC-derived neurons, cerebral organoids, and retinal organoids — the most promising human-relevant systems, and the only ones that can model the human-specific retinal vulnerability. Retinal organoids are particularly apt here given the vision-first phenotype.

15.4 Applications

Mechanism dissection (lysosomal storage, autophagy, neuroinflammation), biomarker discovery, preclinical efficacy and safety for AAV gene therapy and ERT, dose-finding and route-of-administration studies (large animals), and high-throughput drug screening (zebrafish, iPSC).

15.5 Resources

MGI, IMPC/KOMP, IMSR, JAX, EMMA, MMRRC (mouse); RGD (rat); ZFIN (zebrafish); FlyBase; WormBase; OMIA (OMIA:000181 and gene-specific entries) for natural animal disease; Alliance of Genome Resources for orthology; Cellosaurus/ATCC and Coriell (NIGMS repository holds NCL patient fibroblast lines) for cell models; the UCL NCL Resource (ucl.ac.uk/ncl-disease) for the mutation and patient database.

16. Curation guidance and verification status

16.1 The three claims this entry must make that a CLN3-anchored entry would not

1. At least nine genes — CLN3, PPT1, TPP1, CLN5, CLN6, MFSD8, CLN8, CTSD, ATP13A2 — have defensible juvenile-onset presentations. CLN3 is the most prevalent, not the definition.
2. Vision loss is typical but not necessary. CLN8/EPMR and several CLN6 juvenile families present without visual failure. A definition requiring retinopathy would wrongly exclude real members.
3. Onset class and gene are orthogonal axes. The same gene can appear in the juvenile, late-infantile, and adult groupings via different alleles — CLN6 spans all three. This is not an inconsistency to be resolved; it is the structure of the domain.

16.2 Members to exclude, and why

16.3 Ontology defects observed (worth reporting upstream)

1. MONDO:0019262 asserts only five children (MONDO:0979341 CLN1, MONDO:0979345 CLN2, MONDO:0979346 CLN3, MONDO:0012188 "NCL 9", MONDO:0017809 ATP13A2), while the literature supports at least nine members. CLN5, CLN6, CLN7/MFSD8, CLN8, and CLN10 juvenile forms have no corresponding MONDO term.
2. MONDO:0012188 ("neuronal ceroid lipofuscinosis 9") is asserted as a child of the juvenile grouping despite CLN9 being a withdrawn designation whose index family was reassigned to CLN5.
3. The synonym overlap with MONDO:0008767 (Vogt Spielmeyer disease, Spielmeyer Sjogren disease, Batten disease) is the mechanical driver of the historical conflation and is worth flagging even though it accurately reflects historical usage.
4. MONDO:0979346 is correctly dual-parented (MONDO:0019262 + MONDO:0008767) but was not returned by the OLS4 /descendants endpoint — a retrieval inconsistency that could cause an automated member-enumeration script to silently miss the CLN3 member.

16.4 Verification status of citations in this report

Fully transcribed abstracts (single-PMID E-utilities fetch; quotes in this report are verbatim from those transcriptions): PMID:21990111 · PMID:10508524 · PMID:31919451 · PMID:34868216 · PMID:35609511 · PMID:27553520 · PMID:23374165 · PMID:9151309 · PMID:22608501 · PMID:25227500 · PMID:22388936 · PMID:29688815

Partial quotes only (fragments returned inside multi-record fetches or PMC full-text extraction; the quoted strings are reliable but the surrounding abstract was summarised): PMID:22778232 · PMID:9425237 · PMID:23418007 · PMID:20157158 · PMID:19201763 · PMID:39281238 · PMID:25298308 · PMID:26026925 · PMID:39925015

Cited but PMID or content NOT verified this session — verify before curating as evidence: - Wisniewski KE et al., "Reevaluation of neuronal ceroid lipofuscinoses: atypical juvenile onset may be the result of CLN2 mutations", Mol Genet Metab 1999 (exact-title query returned no results) - International Batten Disease Consortium, "Isolation of a novel gene underlying Batten disease, CLN3", Cell 1995 (author/title queries returned no results) - Kim J et al., "Patient-Customized Oligonucleotide Therapy for a Rare Genetic Disease" (milasen), NEJM 2019 - El Haddad et al. 2012, reassignment of the CLN9 family to CLN5 - Tyynelä et al., saposins A and D as the stored proteins in CLN1/CLN10 - The CLN3 1.02-kb deletion allele frequencies (~80–85% of alleles; ~70–75% homozygous) - ICD-10 E75.4 assignment - All GO, CL, UBERON, CHEBI, and NCIT identifiers suggested in this report, and HP:0001922 (vacuolated lymphocytes) and HP:0011675 (arrhythmia), which were not found in the local HP cache

Verified against the local cache/hp/terms.csv: every HP identifier in §3 other than HP:0001922 and HP:0011675.

For dismech curation specifically: every PMID cited here must go through just fetch-reference PMID:XXXXXXXX, and every snippet through just count-verified-snippets, before it enters a kb/disorders/ entry. Several of the quotes above are drawn from PMC full text rather than the abstract (notably PMID:25298308) and will therefore fail the --no-full-text check that just validate-disorders and CI run — replace those with abstract-resident quotes or move the claims to notes. Ontology terms need just validate-terms.

Sources

Ontology / database records (retrieved live 2026-08-08) - MONDO:0019262 — OLS4 · MONDO:0008767 — OLS4 · MONDO:0979346 — OLS4 - GeneReviews: Neuronal Ceroid-Lipofuscinoses (NBK1428) - UCL NCL Resource — Mutation and Patient Database - OMIA:000181-9615 — NCL, generic, in dog - GARD: Juvenile neuronal ceroid lipofuscinosis

Primary literature - PMID:21990111 — Kousi, Lehesjoki, Mole. Hum Mutat 2012;33(1):42-63 - PMID:26026925 — Mole & Cotman. Biochim Biophys Acta 2015;1852:2237-41 · PMC4567481 - PMID:22778232 — Williams & Mole. Neurology 2012;79(2):183-91 - PMID:9425237 — Mitchison et al. Hum Mol Genet 1998;7(2):291-7 (juvenile CLN1/GROD) - PMID:23418007 — Sun et al. Hum Mutat 2013;34(5):706-13 (TPP1/SCAR7) - PMID:20157158 — Xin et al. Neurology 2010;74(7):565-71 (CLN5 juvenile) - PMID:34868216 — Front Genet 2021;12:746101 (CLN6 juvenile, no visual loss) · PMC8640139 - PMID:35609511 — Neurodegener Dis 2021;21:126-131 (juvenile-onset Kufs, CLN6) - PMID:19201763 — Kousi et al. Brain 2009;132:810-9 (CLN7/MFSD8) - PMID:25227500 — Roosing et al. Ophthalmology 2015;122(1):170-9 (MFSD8 macular dystrophy) - PMID:10508524 — Ranta et al. Nat Genet 1999;23(2):233-6 (CLN8/EPMR, mnd mouse) - PMID:25298308 — Neurology 2014;83(20):1873-5 (CTSD juvenile ataxia) · PMC4240432 - PMID:16685649 — Steinfeld et al. Am J Hum Genet 2006 (cathepsin D deficiency) - PMID:22388936 — Bras et al. Hum Mol Genet 2012;21(12):2646-50 (ATP13A2/CLN12) - PMID:22608501 — Smith et al. Am J Hum Genet 2012;90(6):1102-7 (GRN dosage) - PMID:31919451 — Jedličková et al. Eur J Hum Genet 2020;28(6):783-9 (DNAJC5, adult NCL gene list) - PMID:15349861 — Schulz et al. 2004 (the "CLN9" variant) - PMID:1535179 — Palmer et al. Am J Med Genet 1992 (SCMAS storage) · PMID:7668326 - PMID:27553520 — Gene 2016;593(2):284-91 (ExAC carrier frequencies) - PMID:23374165 — Orphanet J Rare Dis 2013;8:19 (Italian molecular epidemiology) - PMID:9151309 — Neuropediatrics 1997;28(1):6-8 (Scandinavian epidemiology) - PMID:39281238 — Pak J Med Sci 2024;40(8):1638-43 (pediatric NCL cohort) - PMID:29688815 — Schulz et al. N Engl J Med 2018;378(20):1898-1907 (cerliponase alfa) - PMID:39925015 — Zhang et al. CNS Neurosci Ther 2025;31(2):e70261 - PMID:32300063 — CLN3 Disease Staging System · UBDRS validation, PMC9879304 - Nat Rev Neurol 2025 — NCL mechanisms and therapeutic targets

Trials - NCT03770572 — Gene Therapy for Children With CLN3 Batten Disease · NCT02725580 — CLN6 gene therapy · NCT07582484 — CLN6 scAAV9 · NCT05174039 — Miglustat in CLN3 · NCT03822650 — CLN5 natural history · NCT04644549 — Natural History Study of Batten Disease · NCT03307304 — Investigations of JNCL · CLN7 AAV9 phase 1, PMC12703863

Animal models - Ovine CLN5/CLN6 MRI brain volume, PMC9830986 · Ovine CLN5/CLN6 retinal natural history, PMC8901734 · CLN6 mixed-breed dog, PMC11203140