Spinocerebellar Ataxia, Autosomal Recessive 26 (SCAR26; XRCC1-related ataxia): Research Report

Prepared: 2026-10-08 · Target: MONDO:0033116 · Category: Mendelian

About this report. The quotes marked "verbatim" were checked against the cache files I fetched in this session with just fetch-reference (references_cache/PMID_*.md), so they can be used as snippets. Ontology IDs marked (verify) came from memory. They are leads only and must be checked with runoak or the term caches before binding. Two 2026 case reports have no abstract in PubMed (PMIDs 41631675 and 41506093), and their publisher pages blocked retrieval. They are listed but not quoted.

1. Disease Information

Overview. SCAR26 is a very rare autosomal recessive cerebellar ataxia caused by biallelic variants in XRCC1. XRCC1 is the scaffold protein for DNA single-strand break repair (SSBR) and base excision repair (BER). The core phenotype is: - progressive cerebellar ataxia with cerebellar atrophy - sensorimotor, predominantly axonal, length-dependent peripheral neuropathy - oculomotor abnormalities (ocular motor apraxia in the index case; slow saccade initiation and nystagmus in later cases)

The disease belongs to the SSBR-defective ataxia family with AOA1 (APTX), SCAN1 (TDP1) and AOA4 (PNKP). One 2022 Brazilian cohort paper calls it "AOA5" (PMID:35426160). I could not confirm that synonym in OMIM or Orphanet.

• Verbatim (PMID:28002403, Hoch et al., Nature 2017): "Here we show that biallelic mutations in the human XRCC1 gene are associated with ocular motor apraxia, axonal neuropathy, and progressive cerebellar ataxia."

Identifiers

Synonyms: SCAR26; autosomal recessive spinocerebellar ataxia 26; XRCC1-related cerebellar ataxia; "AOA5" (single source, unconfirmed).

Source of information: Essentially all data come from individual patient reports. At most about 4–6 published probands are known (see §9). No registry or aggregated cohort exists.

2. Etiology

• Cause: Biallelic germline loss-of-function variants in XRCC1.
• Genetic risk factors:
• c.1293G>C (p.Lys431Asn) is recurrent in South Asian (Pakistani and East Indian) families. It sits at the last base of exon 11 and disrupts the intron 11 donor splice site.
• O'Connor et al. found it heterozygous in four South Asian ExAC individuals (AF 0.0002449). They wrote that it "likely exists at a higher frequency than previously described" (PMID:29472272).
• Modifier genes: None in humans. In mice, Parp1 dosage strongly modifies the phenotype: losing one Parp1 allele extended lifespan more than losing both (PMID:33932076).
• Environmental, protective and gene–environment factors: None documented. Mechanistically, endogenous oxidative single-strand breaks are the relevant "exposure". This is inferred; no human exposure data exist.

3. Phenotypes

The sources below are the index patient (PMID:28002403) and two homozygous c.1293G>C patients (PMID:29472272). With N≈3, every frequency is qualitative.

Onset - Adult onset in the index case: balance and gait difficulties "first noticed at twenty-eight years", with diagnosis at 41 (PMID:28002403). - Early childhood onset in O'Connor's patients: observable signs by about 3 years. - 2026 reports describe one childhood-onset homozygous c.1293G>C patient (PMID:41631675) and one late-onset case (PMID:41506093). I could not read either.

Course: Slowly progressive. Secondary sources (e.g., MalaCards/OMIM synopsis) mention eventual loss of independent ambulation.

Confounder: O'Connor P2 also carried homozygous CLCN1 myotonia congenita. Myotonia, calf hypertrophy and probably the cramps in that patient should not be attributed to SCAR26.

Quality of life: No formal QoL data exist. Expect impact on mobility and falls, fine motor function, speech, and possibly fertility.

4. Genetic / Molecular Information

Gene: XRCC1, X-ray repair cross-complementing 1. It is a non-enzymatic scaffold with: - an N-terminal domain that binds POLβ - a central BRCT1 domain that binds PAR and DNA - a C-terminal BRCT2 domain that binds LIG3α - a CK2-phosphorylated linker that binds PNKP, APTX and APLF

Reported pathogenic variants

• Verbatim (PMID:28002403): "1293G>C mutation is located at the end of exon 11 and is also part of the donor splice site for intron 11, most likely affecting splicing and inducing premature stop codons/nonsense-mediated decay and/or encoding XRCC1 with the missense mutation, K431N."
• Protein consequence (PMID:28002403): XRCC1 protein is reduced in patient cells. The partner protein is also destabilized: "Levels of DNA ligase IIIα (Lig3α) were also greatly reduced (by >80%) in patient cells".
• Functional class: Loss of function, hypomorphic. Complete XRCC1 loss is embryonic lethal in mice, so the human alleles must retain some residual function (inference).
• Other recorded SCAR26 variants: ClinVar lists about 13 XRCC1 variants under SCAR26 with mixed classifications. A 2026 Cell Death Differ study tested three "SCAR26-associated" missense constructs: c.1196A>G (BRCT1), c.1293G>C (linker) and c.1738C>T (BRCT2). It found that one variant disrupts TADA2B binding while LIG3 binding is preserved (PMID:42399641). Their individual clinical provenance is unverified, so do not curate them as patient variants without a primary report.
• Gene–disease validity: A ClinGen GDV record for XRCC1–SCAR26 was not found in the local cache; check before asserting a tier. PanelApp England and Australia rate XRCC1 Green (high evidence) for hereditary ataxia in current versions; earlier versions were Amber/Red.
• Origin: Germline. No somatic, epigenetic or chromosomal mechanisms are described.
• Common polymorphisms: p.Arg399Gln and similar variants are extensively studied cancer-association SNPs, unrelated to SCAR26. Avoid confusing them with disease variants. Hoch used R399Q only as a phasing marker.

5. Environmental Information

There are no environmental, lifestyle or infectious factors. The relevant "lesions" are endogenous oxidative and topoisomerase-1–induced single-strand breaks. Not applicable for environmental:.

6. Mechanism / Pathophysiology

Ordered causal chain

1. Biallelic hypomorphic XRCC1 variants (splice / missense K431N, nonsense Q465) lead to* reduced XRCC1 mRNA and protein (human patient cells, demonstrated; PMID:28002403).
2. Reduced XRCC1 leads to destabilization and loss of LIG3α (>80% reduced), so the XRCC1–LIG3α–POLβ–PNKP repair complexes fail to assemble (demonstrated in patient cells; PMID:28002403).
3. Loss of XRCC1 complexes results in slow repair of single-strand breaks and BER intermediates (demonstrated in patient cells and XRCC1⁻/⁻ RPE-1 cells; PMID:28002403, 34102106).
4. Verbatim (PMID:28002403): "Cells from a patient with mutations in XRCC1 exhibited not only reduced rates of single-strand break repair but also elevated levels of protein ADP-ribosylation."
5. Unrepaired breaks plus the missing "anti-trapper" lead to PARP1 trapping on BER intermediates and PARP1 hyperactivation (excess poly- and mono-ADP-ribosylation).
6. Verbatim (PMID:34102106): "As a result, PARP1 becomes "trapped" on BER intermediates in XRCC1-deficient cells in a manner similar to that induced by PARP inhibitors, including in patient fibroblasts from XRCC1-mutated disease."
7. Also shown in mouse cerebellum: "Strikingly, we detected elevated levels of ADP-ribose in the cerebellum of Xrcc1Nes-Cre mice…" (PMID:28002403).
8. XRCC1 loss also amplifies mono-ADPr and RNF114 recruitment (PMID:41922367).
9. Feedback: trapped PARP1 blocks POLβ access and further impedes repair, which sustains step 3: "This excessive PARP1 engagement and trapping renders BER intermediates inaccessible to enzymes such as DNA polymerase β and impedes their repair." (PMID:34102106)
10. PARP1 hyperactivity then branches into three effects:
11. 5a. Transcriptional failure. PARP1 hyperactivity leads to excessive USP3 recruitment, loss of H2A/H2B monoubiquitination, and failed recovery of RNA polymerase I and II transcription after oxidative damage. This was shown in XRCC1 patient fibroblasts and Xrcc1-deficient cerebellar neurons (PMID:34811483).
    • Verbatim: "We show that aberrant PARP1 activity suppresses transcriptional recovery during base excision repair by promoting excessive recruitment and activity of the ubiquitin protease USP3, which as a result reduces the level of monoubiquitinated histones important for normal transcriptional regulation."
12. 5b. NAD⁺ depletion / PAR toxicity. Proposed and supported in zebrafish discussion, but not directly measured in human SCAR26 (inferred; PMID:40379758).
13. 5c. Deregulated presynaptic Ca²⁺ signalling. Leads to hyperexcitability and seizures in mice (PMID:33932076). Seizures have not been reported in human SCAR26.
14. In parallel, persistent DNA damage in differentiating cerebellar interneuron progenitors leads to p53-dependent cell-cycle arrest. This produces loss of basket, stellate and Golgi interneurons, granule-neuron apoptosis and reduced Purkinje-cell spike activity (mouse; PMID:19633665, 28002403).
15. Verbatim (PMID:19633665): "This cell loss was linked to p53-dependent cell cycle arrest and occurred as interneuron progenitors commenced differentiation."
16. Upstream of this branch: PARP1 deletion restores interneuron density about 4-fold, so PARP1 acts upstream of much of the neuron loss (PMID:28002403).
17. Cerebellar neuron loss and dysfunction result in cerebellar atrophy and progressive ataxia. This is direct in mice, and inferred in humans from MRI and the clinical picture.
18. PARP1 is causal: "genetic deletion of Parp1 rescued normal cerebellar ADP-ribose levels and reduced the loss of cerebellar neurons and ataxia in Xrcc1-defective mice" (PMID:28002403).
19. Zebrafish give the same answer: "parp1 knockdown alone does not significantly affect neural development, and instead rescues the cerebellar defects observed in xrcc1 mutant larvae" (PMID:40379758).
20. Peripheral axonal neuropathy and oculomotor apraxia presumably arise from the same SSBR/PARP1 mechanism in long peripheral axons and brainstem/cerebellar oculomotor circuits. This is inferred: no model has studied the peripheral nerve.

Upstream vs downstream: steps 1–3 form the lesion; step 4 is the pivotal hub node (PARP1 hyperactivation); steps 5–6 are cellular effectors; steps 7–8 are the organ and clinical phenotypes.

Suggested terms (verify all)

Molecular profiling: No omics, single-cell or spatial datasets specific to SCAR26 were found.

7. Anatomical Structures

• Primary: cerebellum (UBERON:0002037), with vermis and hemispheres; peripheral nerves (UBERON:0001021).
• Secondary: pons and middle/superior cerebellar peduncles (O'Connor P1); oculomotor control circuits.
• Possible: testis/gonad (one patient, unconfirmed).
• Cells: cerebellar interneurons, granule and Purkinje cells; peripheral sensory and motor axons.
• Subcellular: nucleus, nucleolus, chromatin at single-strand break sites.
• Distribution: bilateral and symmetric; length-dependent neuropathy.

8. Temporal Development

• Onset: ranges from early childhood (~3 y; homozygous c.1293G>C) to adult (28 y; compound heterozygote) to late onset (PMID:41506093, unread).
• Onset pattern: insidious.
• Course: chronic, slowly progressive. Cerebellar atrophy can develop after a normal adolescent MRI (O'Connor P1).
• Remission and staging: no remission; no formal staging.
• Critical period (mouse, inferred): XRCC1 is most needed as cerebellar interneuron progenitors differentiate around birth (PMID:19633665).

9. Inheritance and Population

• Inheritance: autosomal recessive (HP:0000007). LOVD labels it "digenic". That looks erroneous because all cases are biallelic at one locus.
• Penetrance: apparently complete in reported biallelic individuals; an unaffected heterozygous sibling was reported.
• Expressivity: variable, with age of onset from ~3 to ~28 years or more. No anticipation.
• Consanguinity and founder effect: consanguinity was present in both Pakistani families. They share a chromosome 19 haplotype around c.1293G>C, which suggests a South Asian founder allele. The index patient was of East Indian descent with non-consanguineous parents.
• Carrier frequency: c.1293G>C AF about 2.4×10⁻⁴ overall in ExAC, enriched in South Asians. Re-check gnomAD v4 before quoting.
• Prevalence: unknown. Fewer than 10 published patients. Suggested prevalence_class: CASES_IN_LITERATURE + ULTRA_RARE.
• Negative cohort data: a Brazilian ataxia-with-oculomotor-apraxia cohort found no XRCC1 variants (PMID:35426160).
• Sex ratio: reported cases are 1 female (index) and 2 males (O'Connor).

10. Diagnostics

• Genetic testing: the mainstay. Use a hereditary ataxia multigene panel (XRCC1 is on PanelApp hereditary ataxia panels) or exome/genome sequencing. RNA studies can help confirm splice variants such as c.1293G>C. Single-variant testing is reasonable in South Asian probands with a consistent phenotype.
• Imaging: brain MRI showing cerebellar (± pontine) atrophy.
• Electrophysiology: NCS/EMG showing length-dependent sensorimotor axonal neuropathy.
• Biomarkers: No validated clinical biomarker. Research-level elevated ADP-ribose / PARP1 activity in patient fibroblasts was proposed as a "biomarker of PARP1 hyperactivity" (PMID:28002403). Labs such as AFP, albumin and cholesterol (useful in AOA1/2/4) are not reported for SCAR26.
• Differential diagnosis:
• AOA1 (APTX): hypoalbuminemia, hypercholesterolemia
• AOA2 (SETX): elevated AFP
• AOA4 (PNKP): plus microcephaly and seizures
• SCAN1 (TDP1)
• ataxia-telangiectasia (ATM)
• Friedreich ataxia
• ARCA with neuropathy (SYNE1, POLG, RFC1/CANVAS)
• Screening: no newborn screening. Cascade carrier testing is possible in affected families.

11. Outcome / Prognosis

• No survival or mortality data exist. Patients surveyed were alive at ages 22–47.
• Morbidity comes from progressive gait and limb ataxia, falls, dysarthria, distal weakness and sensory loss. Secondary sources mention possible loss of independent ambulation.
• No prognostic biomarkers. Earlier onset appears to go with the homozygous splice genotype (N too small to conclude).
• The mouse Xrcc1^Nes-Cre^ median lifespan of 3–4 weeks (seizures) is far more severe than the human disease. Do not extrapolate it to human prognosis.

12. Treatment

No disease-modifying therapy and no clinical trials. Management is supportive.

Experimental and preclinical directions - PARP1 suppression is the leading therapeutic hypothesis. Genetic deletion rescues the mouse ataxia and interneuron loss (PMID:28002403), the zebrafish cerebellar hypotrophy (PMID:40379758), and the mouse seizures and lifespan (PMID:33932076). - Caution: current clinical PARP inhibitors trap PARP1 and "do not mimic PARP1 genetic deletion" (PMID:28002403). Non-trapping PARP1 inhibitors are suggested instead. Note also that in mice, losing one Parp1 allele beat losing both, so dosage matters. - USP3 is a second candidate target (PMID:34811483). NAD⁺ repletion is hypothesized (an inference from the NAD⁺-depletion model; the ALS NAD⁺–PARP1–XRCC1 review PMID:42086408 is not SCAR26-specific). - Pharmacogenomics: none.

13. Prevention

• Primary: none.
• Secondary: cascade carrier testing and prenatal or preimplantation diagnosis once familial variants are known. Consider c.1293G>C in carrier screening for consanguineous South Asian families.
• Tertiary: fall prevention and neuropathy foot care.
• No vaccines, prophylaxis or public-health measures apply.

14. Other Species / Natural Disease

• No naturally occurring XRCC1 ataxia was found (OMIA not checked in detail).
• Orthologs: mouse Xrcc1 and zebrafish xrcc1. XRCC1 recruitment strategies vary by species (PMID:35026705).
• Not zoonotic.

15. Model Organisms

Possible model divergences to type when curating (per CLAUDE.md taxonomy): - SPECIES_MISMATCH - BOUNDARY_OMISSION: no peripheral nerve in any model - INCOMPLETE_PHENOTYPE: zebrafish has no ataxia readout - a severity mismatch: the mouse null is far more severe than the human disease

Evidence gaps and curation notes

1. Very small case base: about 3 well-described patients plus 2 unread 2026 case reports. Retry fetching PMID:41631675 and PMID:41506093 later. Both are currently cached with content_type: unavailable and must not be quoted.
2. No quotable gene–disease validity tier: no ClinGen GDV, Orphanet or GeneReviews chapter was found. Run just check-genereviews on the entry.
3. No human data on NAD⁺ depletion, seizures, or the peripheral nerve mechanism. These are candidate KNOWLEDGE_GAP / HUMAN_MODEL_MISMATCH discussions.
4. Repository housekeeping from this session:
5. I cached PMID_28002403, 29472272, 41631675, 41506093, 40379758, 19633665, 34811483, 34102106, 41922367 and 33932076 in references_cache/.
6. I also fetched PMID_33779029.md by mistake. It is an unrelated FXR2P seizure paper and should be deleted; my rm was blocked by permissions.
7. Prune any of the others the entry does not end up citing.

Key references

Web sources: - PubMed 28002403 - PMC6227798 (O'Connor 2018) - Komulainen 2021, PMC8097344 - OMIM 617633 - MedGen C4539948 - Disease Ontology DOID:0080260 - LOVD disease 06425 - PanelApp England XRCC1 - PanelApp Australia XRCC1 - MalaCards SCAR26 - Brazilian AOA cohort, PubMed 35426160 - Kim 2026, PubMed 42399641

Reference Validation

Checked with linkml-reference-validator 0.3.0.

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:0033116 (2 mentions) - the report calls it "MONDO"; MONDO calls it spinocerebellar ataxia, autosomal recessive 26
• DOID:0080260 (4 mentions) - the report calls it "Disease Ontology"; DOID calls it autosomal recessive spinocerebellar ataxia 26
• NCIT:C121351 (1 mention) - the report calls it "Fine motor"; NCIT calls it Occupational Therapy
• NCIT:C159273 (1 mention) - the report calls it "Dysarthria"; NCIT calls it Speech Language Therapy
• NCIT:C15240 (1 mention) - the report calls it "AR recurrence 25%"; NCIT calls it Genetic Counseling

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:0006977 (obsolete DNA damage response, signal transduction by p53 class mediator resulting in cell cycle arrest) (1 mention) - replaced by GO:0030330

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:

• CHEBI:57540 (1 mention) - the report calls it "NAD⁺"; CHEBI calls it NAD(1-), and lists "NAD(+)" among its other names
• UBERON:0002037 (1 mention) - the report calls it "Primary: cerebellum"; UBERON calls it cerebellum**