Stub-like entry. Let me fetch primary sources.

I'm pulling abstracts for the remaining COA7 papers from Europe PMC, because the PMC pages are blocked behind a verification check.

I have the core papers. One last check for animal models, then I'll write the report.

Spinocerebellar Ataxia, Autosomal Recessive, With Axonal Neuropathy 3 (SCAN3; COA7-related disorder): research report

Causal gene: COA7 (cytochrome c oxidase assembly factor 7; formerly C1orf163 / RESA1; HGNC:25716; OMIM gene 615623; UniProt Q96BR5; chromosome 1p32.3). All findings below are tied to MONDO:0020770 and OMIM #618387. SCAN1 (TDP1) and SCAN2 (SETX*/AOA2) are different diseases caused by different genes. They are not covered here.

Note on ontology IDs. The HP, GO, CL, UBERON, NCIT and CHEBI identifiers in this report are candidates to check with a lookup (runoak, OLS or cache/*/terms.csv) before they are written into the KB. Several were suggested from memory. The disease identifiers (MONDO xrefs, OMIM, PMIDs) and the quoted abstract text were checked against Europe PMC or OLS in this session.

1. Disease information

Overview. SCAN3 is an autosomal recessive mitochondrial disease. It usually starts in the first or second decade and has two main features: - an axonal sensorimotor polyneuropathy that looks like Charcot-Marie-Tooth disease; - cerebellar ataxia with cerebellar atrophy.

Some patients also have leukoencephalopathy, spinal cord atrophy, mild intellectual disability, subclinical mitochondrial myopathy, and later dystonia or parkinsonism. The cause is biallelic loss-of-function variants in COA7, which lead to cytochrome c oxidase (complex IV) deficiency.

The disease was named in 2018: "this new entity of disease would be referred to as spinocerebellar ataxia with axonal neuropathy type 3" (Higuchi et al., Brain 2018, PMID:29718187).

The field now treats it as a broader COA7-related disorder spectrum: "COA7-related disorders typically encompass a spectrum of conditions characterized by a variety of major (cerebellar ataxia and axonal polyneuropathy) and minor (leukoencephalopathy, dystonia, and parkinsonism) symptoms" (Higuchi et al., J Neurol 2024, PMID:37750949).

Identifiers (from the MONDO record via OLS):

Synonyms. SCAN3; spinocerebellar ataxia with axonal neuropathy type 3; COA7-related spinocerebellar ataxia with axonal neuropathy; COA7-related disorder; COA7-related mitochondrial leukoencephalopathy with COX deficiency (the first-reported presentation).

Nature of the data. All knowledge comes from individual case reports and small case series, roughly 11–13 published individuals. There are no registries, no natural-history cohorts and no EHR-derived data.

2. Etiology

• Cause. The disease is genetic and monogenic. It requires biallelic (homozygous or compound heterozygous) pathogenic COA7 variants.
• Genetic risk factors. The only established risk factor is the COA7 genotype. The recurrent missense c.17A>G p.(Asp6Gly) appears in most Japanese patients and in one Chinese patient. It was rare in ExAC (1/120,268 alleles, PMID:29718187). It may be an East Asian founder or recurrent allele, but no haplotype analysis has been published.
• Modifier genes. None reported.
• Environmental risk or protective factors. None reported, and none expected for a Mendelian loss-of-function disorder.
• One possible trigger: in the Ban et al. 2021 patient, regression worsened after an episode of diarrhea and fever (PMID:34322155). This is anecdotal. It fits the general pattern of metabolic decompensation in mitochondrial disease, but has not been shown specifically for COA7.
• Gene–environment interactions. None documented.

3. Phenotypes

The frequencies below are drawn from about 12 published individuals. Denominators are small and the reporting is uneven.

Absent features. Higuchi et al. state there was no nystagmus or oculomotor apraxia, which separates SCAN3 from AOA1/AOA2. Both Cicala siblings had normal brain MRI (PMID:42182620): "Brain MRI findings were unremarkable in both siblings."

Quality of life. No formal QoL instruments (EQ-5D, SF-36) have been reported. The functional burden is mostly mobility: - one patient used a wheelchair from age 9 (PMID:29718187); - one walked only with ankle-foot orthoses at age 19 (PMID:27683825).

Dysarthria, mild cognitive impairment and, in severe cases, feeding difficulty and respiratory failure add to the burden.

Key quotes - "all four patients had characteristic neurological features of peripheral neuropathy and ataxia with cerebellar atrophy" (PMID:29718187) - "Both presented with early-onset, slowly progressive axonal sensorimotor neuropathy" (PMID:42182620) - "infantile-onset generalized dystonia without cerebellar ataxia" (PMID:37750949) - "Levodopa administration improved parkinsonism in this patient." (PMID:37264311)

4. Genetic and molecular information

Gene. COA7, HGNC:25716 (repository form hgnc:25716). It encodes a small (~231 aa) cysteine-rich protein (13 Cys) with five Sel1-like repeats, a subclass of the TPR fold.

Published pathogenic variants (NM_023077.2):

• Variant classes. Mostly missense, plus a frameshift, a nonsense, a splice-site variant and an exon deletion. The exon deletion means MLPA/CNV analysis is needed when only one variant is found.
• ACMG classification. The ClinVar status of each variant was not retrieved here; check ClinVar before entering a clinical_significance. PanelApp (Genomics England) lists COA7 with biallelic inheritance.
• ClinGen / GenCC. A GenCC submission exists (SGC-107318.2, which lists PMID:37750949 among six publications). No ClinGen CGGV record for COA7 is cached locally, so do not assign a validity tier (CLAUDE.md: a tier is copied from a source, never assigned).
• Population frequency. D6G was 1/120,268 in ExAC (PMID:29718187). N189S has 3 alleles in gnomAD and A171T is absent (PMID:34322155).
• Origin. All variants are germline. There are no somatic or mosaic reports.
• Functional consequence. Loss of function / hypomorphic (functional_impact_category: LOSS_OF_FUNCTION; partial LOF is plausible for the missense alleles).
• Wild-type COA7 re-expression rescued COX assembly and activity in patient fibroblasts (PMID:27683825).
• Y137C and the exon-2-deleted protein are imported into mitochondria less efficiently (about 50% and 70% lower) and are degraded by the cytosolic proteasome (PMID:30885959).
• Modifier genes, epigenetics, chromosomal abnormalities. None reported. Not applicable.

5. Environmental information

No environmental, lifestyle or infectious causes are known. The intercurrent febrile or diarrheal illness before regression in one infant (PMID:34322155) should be recorded at most as a possible exacerbating factor (environmental_effect: EXACERBATES, with low confidence). Otherwise this section is not applicable.

6. Mechanism and pathophysiology

Ordered causal chain

1. Biallelic COA7 loss-of-function variants (missense, truncating, splice, exon deletion) lead to reduced levels of functional COA7 protein. COA7 was undetectable by western blot in Martinez Lyons patient cells (PMID:27683825).
2. Branch 1a (shown for missense and in-frame alleles): the mutant precursors are imported more slowly through the MIA40/CHCHD4 disulfide-relay (MIA) pathway. COA7 forms a disulfide intermediate with MIA40 through Cys55. The unimported protein is degraded by the cytosolic ubiquitin–proteasome system (PMID:30885959): "mislocalized proteins are degraded in the cytosol by the proteasome."
3. Branch 1b: truncating alleles (G144fs, W185*) probably cause loss of protein outright. This is inferred; NMD and protein loss were not tested.
4. Loss of COA7 in mitochondria results in failure of its proposed heme-dependent disulfide-reductase activity. COA7 binds heme with His/Met axial ligation (reduction potential −353 mV), interacts transiently with the copper metallochaperones SCO1 and SCO2, and reduces the disulfides in their copper-binding sites. This regenerates the copper relay to COX2 (Formosa et al., PNAS 2022, PMID:35210360): "COA7 interacts transiently with the copper metallochaperones SCO1 and SCO2."
5. Shown biochemically and in knockout cells. That patient variants act through this route is inferred.
6. Impaired Cu_A delivery stalls complex IV assembly after the COX1 module forms and before COX2/COX3 are added (PMID:35210360): "loss of COA7 blocks complex IV assembly after the initial step where the COX1 module is built." The result is lower levels of MT-CO2 and MT-CO3 and of assembled CIV and the III₂+IV supercomplex (PMID:27683825). Assembled CIV fell to 3.8–9.9% of control in Higuchi fibroblasts (PMID:29718187).
7. Complex IV deficiency results in reduced mitochondrial oxygen consumption and OXPHOS capacity. Maximal respiration was about 46% of control (PMID:34322155). Secondary complex I reduction was seen in one patient (PMID:29718187). Elevated lactate shows the shift to glycolysis.
8. Bioenergetic failure in high-demand, long-axon neurons (inferred) leads to length-dependent axonal degeneration of motor and sensory peripheral nerves. Sural nerve biopsy showed loss of large and medium myelinated fibres, with no onion bulbs or inflammation (PMID:29718187).
9. Drosophila neuron-specific dCOA7 knockdown shortens neuromuscular-junction synaptic branches by about 50% and impairs climbing (PMID:29718187). This supports a cell-autonomous neuronal requirement.
10. In parallel (branching), energy failure in CNS neurons and glia (inferred) produces:
11. 6a cerebellar atrophy, leading to cerebellar ataxia and dysarthria;
12. 6b degeneration of the spinal cord long tracts and dorsal columns, leading to spinal cord atrophy and sensory ataxia;
13. 6c white-matter injury and cavitating leukoencephalopathy (the oligodendrocyte or myelin vulnerability is inferred), leading to cognitive impairment and spasticity;
14. 6d nigrostriatal dopaminergic involvement, leading to dystonia or levodopa-responsive parkinsonism. Striatal hypo-uptake on DaT SPECT is shown (PMID:37264311); the mechanism is inferred.
15. Skeletal muscle (mild) shows ragged-red, COX-negative and SDH-increased fibres and mildly raised CK, i.e. a subclinical mitochondrial myopathy (PMID:29718187). Neurogenic atrophy adds to the distal weakness.
16. Clinical result: a progressive axonal sensorimotor neuropathy plus cerebellar ataxia (SCAN3), with variable CNS features.

Unresolved controversy (subcellular location). Martinez Lyons et al. found COA7 behaving "as the soluble matrix protein Citrate Synthase," which suggests a matrix location (PMID:27683825). Mohanraj et al. 2019 (PMID:30885959) and Formosa et al. 2022 (PMID:35210360) place it in the intermembrane space (IMS), imported by MIA. The IMS location also fits the SCO1/SCO2 function. Curate the IMS location as the better-supported current view and record the earlier matrix report in a note.

Genotype–biochemistry discordance. The Tang 2022 patient had normal respiratory-chain activity in fibroblasts despite pathogenic variants and abnormal muscle (PMID:35603692). Fibroblast testing can therefore give false-negative results, which is a useful diagnostic caveat.

Ontology suggestions (verify)

• GO biological process: GO:0033617 mitochondrial cytochrome c oxidase assembly; GO:0045333 cellular respiration; GO:0006123 mitochondrial electron transport, cytochrome c to oxygen; GO:0045041 protein import into mitochondrial intermembrane space; GO:0008090? (check for a "retrograde/axonal transport" term only if justified; not shown for COA7).
• GO molecular function: GO:0020037 heme binding; a disulfide oxidoreductase activity term (for example GO:0015036 disulfide oxidoreductase activity, verify).
• GO cellular component: GO:0005758 mitochondrial intermembrane space; GO:0005759 mitochondrial matrix (historical report); GO:0005751 mitochondrial respiratory chain complex IV.
• Cell types (CL): CL:0000100 motor neuron; CL:0000101 sensory neuron; CL:0000121 Purkinje cell (inferred from cerebellar atrophy; not directly shown); CL:0000128 oligodendrocyte (inferred); CL:0000700 dopaminergic neuron; CL:0000057 fibroblast (assay system); CL:0000188 cell of skeletal muscle.
• Molecular profiling. No transcriptomic, proteomic, metabolomic, single-cell or spatial datasets are reported for COA7 disease. A future datasets: search (just discover-datasets) is likely to return none.

7. Anatomical structures affected

• Primary. Peripheral nervous system: peripheral nerves, length-dependent and lower limbs first (UBERON:0001021 nerve; sural nerve, verify the UBERON ID). Cerebellum (UBERON:0002037).
• Secondary or variable. Spinal cord (UBERON:0002240); cerebral white matter (UBERON:0002437); striatum / basal ganglia (UBERON:0002435 striatum); skeletal muscle (UBERON:0001134 skeletal muscle tissue); heart in one atypical infant (UBERON:0000948).
• Body systems. Nervous (central and peripheral), neuromuscular, and possibly cardiovascular.
• Subcellular. Mitochondrial intermembrane space and inner membrane, specifically respiratory complex IV.
• Lateralization. Bilateral and symmetric. DaT hypo-uptake was bilateral (PMID:37264311).

8. Temporal development

• Onset. Usually <5 years: foot deformity, steppage gait or poor dexterity. The range runs from 3 months (encephalopathic form) through adolescence (15 y) to 30 years (adult-onset D6G homozygote, PMID:37264311). Cicala siblings: 5 y and 12 y.
• Pattern. Insidious onset, then a chronic, slowly progressive course. The infantile form progressed rapidly (PMID:34322155).
• Progression.
• Wheelchair from age 9 in one patient.
• Ankle-foot orthoses at 19 in another.
• The oldest reported patient was 63 years old and still described as alive.
• Parkinsonism can appear decades after onset (age 58).
• Stages, remission and critical periods. No staging system exists and no remission has been reported. Possible vulnerability to intercurrent illness in infancy.

9. Inheritance and population

• Inheritance. Autosomal recessive, HP:0000007. Consanguinity was present in Higuchi patient 1, who was homozygous.
• Penetrance and expressivity.
• Penetrance appears complete in biallelic individuals. Heterozygous parents are unaffected, with normal CIV in parents' fibroblasts (PMID:34322155).
• Expressivity is highly variable, from isolated axonal neuropathy with a normal MRI to infantile encephalopathy, generalized dystonia, or adult-onset ataxia with parkinsonism.
• Some genotypes recur: D6G homozygosity was seen with both childhood onset (Higuchi patient 1) and adult onset (Ouchi).
• Anticipation and mosaicism. None.
• Prevalence. Unknown, ultra-rare. Prevalence would be measure_type: CASES_IN_LITERATURE with prevalence_class: ULTRA_RARE or RARE.
• Published count: Martinez Lyons 1; Higuchi 2018 4; Ban 2021 1; Tang 2022 1; Ouchi 2023 1; Higuchi 2023 3 new (giving 7 Japanese patients in that series); Cicala 2026 2 siblings. That is about 11–13 individuals, plus an unknown number in a 2026 Turkish hereditary-polyneuropathy cohort (Avcı et al., PMID:42552261).
• Possible double count: the Ouchi 2023 patient (homozygous D6G, levodopa-responsive parkinsonism) and Higuchi 2023 patient 2 (ataxia, neuropathy and parkinsonism with a slight levodopa response) are both Japanese. Check whether they are the same person before counting.
• Sex discrepancy in the literature: Martinez Lyons describe their patient as a 19-year-old woman, but Ban et al.'s summary table lists that patient as M. Use the primary report.
• Diagnostic yield. 4 of 1,396 in a Japanese inherited-neuropathy / CMT series (~0.3%) (PMID:29718187): "identified four unrelated patients with recessive mutations in COA7 among a Japanese case series of 1396 patients."
• Populations. Japanese (most patients, recurrent D6G), Chinese (2), Italian (Martinez Lyons; Cicala siblings), Turkish (2026 cohort).
• Carrier frequency and sex ratio. Carrier frequency is unknown. The sex ratio is roughly balanced (M ≈ F).

10. Diagnostics

• Genetic testing (first line). Exome or genome sequencing, or ataxia/CMT gene panels that include COA7.
• Add CNV/MLPA when only one heterozygous variant is found, because the exon 2 deletion was missed by sequencing (PMID:29718187).
• Trio sequencing helps establish phase.
• Sequence mtDNA to exclude mtDNA COX defects (done in PMID:34322155).
• Electrophysiology. Nerve conduction studies show an axonal sensorimotor neuropathy. Lower-limb CMAPs and SNAPs are absent or severely reduced, and median motor conduction velocity is near normal (49.6–51.5 m/s). One patient had a mixed axonal-demyelinating picture (PMID:27683825).
• Imaging.
• Brain MRI: cerebellar atrophy; ± white-matter T2 hyperintensity or cavitation; ± cerebral atrophy.
• Spinal MRI: cord atrophy.
• DaT SPECT (¹²³I-ioflupane) when parkinsonism is present; MIBG was normal in the reported case.
• Biochemistry.
• Respiratory-chain enzyme assays in fibroblasts, muscle or skin show isolated complex IV deficiency.
• BN-PAGE shows reduced assembled CIV.
• Fibroblast results can be normal (PMID:35603692), so muscle is preferred.
• Serum and CSF lactate; CK.
• Relevant LOINC targets: lactate in serum/plasma and CSF, creatine kinase. Bind the LOINC Part or panel codes after a lookup.
• Pathology.
• Sural nerve: chronic axonal degeneration with loss of large and medium myelinated fibres.
• Muscle: ragged-red, COX-negative and SDH-increased fibres.
• Differential diagnosis.
• SCAN1 (TDP1); AOA1 (APTX); AOA2/SCAN2 (SETX), which have oculomotor apraxia and raised AFP, both absent in SCAN3.
• Friedreich ataxia (FXN); ataxia-telangiectasia.
• POLG-related ataxia-neuropathy; other COX-assembly disorders (SCO2, COA8/APOPT1 cavitating leukoencephalopathy, COX18-related neuropathy, PMID:40830826).
• Axonal CMT types; SACS (ARSACS); PNPLA6.
• Clinical criteria. None standardized.
• Screening. No newborn screening. Carrier and cascade testing are possible in families with known variants.

11. Outcome and prognosis

• Survival into the 7th decade is documented (Higuchi patient 1, aged 63).
• The infantile encephalopathic form (Ban 2021, last seen at 2 y) and the form with respiratory failure and feeding difficulty (Higuchi 2023 patient 3) suggest worse outcomes at the severe end. No mortality data exist.
• Morbidity comes from progressive loss of mobility (wheelchair or orthoses), falls, dysarthria and mild cognitive impairment.
• No prognostic biomarkers or validated QoL data exist.
• Possible negative prognostic signs, from anecdote only: infantile onset, leukoencephalopathy, cardiac involvement.

12. Treatment

There is no disease-modifying therapy. Management is supportive.

Experimental and preclinical. - Proteasome inhibitors (MG132 in HEK293; bortezomib, carfilzomib in patient fibroblasts) rescued COA7 mitochondrial localization and raised complex IV activity by about 35% (PMID:30885959): "proteasome inhibition rescued both the mitochondrial localization of COA7 and complex IV activity." - This is IN_VITRO only. Clinical use is limited by proteasome-inhibitor toxicity, including bortezomib-induced peripheral neuropathy, which is a direct concern in this disease. No clinical trials were found. - Gene therapy, ASO, cell therapy: none. - Pharmacogenomics: none.

13. Prevention

• Primary. Genetic counseling for carrier couples; preimplantation or prenatal diagnosis once familial variants are known.
• Secondary. Early genetic diagnosis in children with axonal neuropathy plus ataxia.
• Tertiary.
• Physical therapy and orthotics.
• Avoid neurotoxic drugs, especially those that cause neuropathy. This is general mitochondrial-disease advice and not COA7-specific evidence.
• Prompt management of intercurrent illness.
• Vaccines, prophylaxis and public-health measures: not applicable.

14. Other species and natural disease

• Natural animal disease. None reported. No OMIA entry was found.
• Orthologs. COA7 is conserved across metazoans but absent from yeast (PMID:27683825). Orthologs: mouse Coa7; Drosophila dCOA7 (CG?, check the FlyBase ID; FlyBase FBrf0238967 refers to the Higuchi 2018 paper).

15. Model organisms

Suggested knowledge gaps. No vertebrate model; no iPSC-derived motor-neuron or cerebellar model; the basis of the selective neuronal vulnerability; why fibroblast CIV is normal in some patients. A HUMAN_MODEL_MISMATCH discussion is not warranted yet because there is no model that diverges from the human picture.

Key references (verified in this session via Europe PMC/PMC)

Curation notes - The quoted strings above come from abstracts as returned by Europe PMC, but some were relayed through a summarizing fetch tool. Run just fetch-reference and just count-verified-snippets on each before using it as an exact snippet. - The existing kb/disorders/Spinocerebellar_Ataxia_Autosomal_Recessive_With_Axonal_Neuropathy_3.yaml is a 25-line stub with only MONDO and one HP term, so almost everything here is new content.

Sources: - Higuchi 2018, Brain (PMC5972596) - Martinez Lyons 2016, J Med Genet (PMC5264227) - Mohanraj 2019, EMBO Mol Med (PMC6505684) - Ban 2021, Front Genet (PMC8312223) - Formosa 2022, PNAS (PMC8892353) - Formosa bioRxiv preprint - Higuchi 2023, J Neurol (PMC10769979) - Ouchi 2023, BMC Neurol (PMC10234071) - Europe PMC search results (Cicala 2026, Tang 2022, Avcı 2026, COX18 2026) - MONDO:0020770 via OLS - MalaCards SCAN3 - KEGG H02573 - GenCC submission - PanelApp COA7 - Alliance COA7 - RCSB 7MQZ

Reference Validation

Checked with linkml-reference-validator 0.4.0.

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:

• MONDO:0020770 (5 mentions) - the report calls it "MONDO"; MONDO calls it spinocerebellar ataxia, autosomal recessive, with axonal neuropathy 3
• DOID:0070465 (1 mention) - the report calls it "DOID"; DOID calls it autosomal recessive spinocerebellar ataxia with axonal neuropathy 3

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:0005751 (obsolete mitochondrial respiratory chain complex IV) (1 mention) - replaced by GO:0045277

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:

• NCBITaxon:7227 (1 mention) - the report calls it "Drosophila melanogaster neuron- or eye-specific dCOA7 RNAi knockdown"; NCBITaxon calls it Drosophila melanogaster*

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: GARD.