name: ADPRS-Related Stress-Induced Neurodegeneration
creation_date: "2026-08-01T00:00:00Z"
category: Mendelian
disease_term:
  preferred_term: neurodegeneration, childhood-onset, stress-induced, with variable ataxia and seizures
  term:
    id: MONDO:0100095
    label: neurodegeneration, childhood-onset, stress-induced, with variable ataxia and seizures
description: >-
  Stress-induced childhood-onset neurodegeneration with variable ataxia and
  seizures (CONDSIAS) is an autosomal recessive neurodegenerative disorder
  caused by bi-allelic inactivating variants in ADPRS, the gene encoding
  ADP-ribosyl-acceptor hydrolase 3 (ARH3). ARH3 is the only human enzyme able
  to remove the terminal serine-linked mono(ADP-ribose) that PARP1/HPF1 deposits
  on chromatin at sites of DNA damage, and it acts in tandem with PARG to clear
  poly(ADP-ribose) from the nucleus, cytoplasm and mitochondrial matrix. Losing
  that eraser leaves the ADP-ribosylation arm of the DNA-damage response unable
  to switch itself off. The clinical signature that names the disorder follows
  directly from that: after normal or near-normal early development, affected
  children deteriorate in discrete steps precipitated by physiological stress -
  most characteristically febrile illness or infection, but also surgery,
  trauma, diarrhoea, exercise and cold-water immersion - with episodes of
  ataxia, dystonic posturing, seizures and developmental regression from which
  recovery is incomplete, so that the disorder ratchets downward rather than
  progressing smoothly. In the largest systematic review, covering 47 published
  patients from 12 articles, a triggering factor was documented in 59.5%
  (28/47), developmental regression in 60.4%, axonal polyneuropathy in 39.1%
  (9/23 tested), and cerebral and cerebellar atrophy with white matter change in
  77.7% (28/36 imaged); the fatality rate was 44.6% (21/47) at a median age of
  7 years. Death is usually from respiratory failure, frequently central rather
  than obstructive. The disorder is genuinely a spectrum: the same gene produces
  a lethal infantile-onset encephalopathy at one pole and, at the other,
  adults presenting in their twenties with episodic psychosis, ataxia and a
  motor neuropathy, or with an almost pure hereditary motor neuropathy. No
  disease-modifying therapy exists. PARP inhibition rescues the cellular and
  organismal phenotype in every model tested, which is the strongest
  mechanistic lead in the field, but it has never been tested in a patient in a
  controlled way, and the one PARP-inhibitor-flavoured agent actually
  administered off-label to patients - minocycline - was subsequently shown to
  be a poor PARP1 inhibitor with no effect on ADP-ribosylation in
  ARH3-deficient cells.
parents:
- Inherited Neurodegenerative Disorder
synonyms:
- CONDSIAS
- childhood-onset stress-induced neurodegenerative ataxia-seizure syndrome
- ADPRHL2-related neurodegeneration
- ARH3 deficiency
- stress-induced epileptic ataxia syndrome
- PAMP syndrome
classifications:
  harrisons_chapter:
  - classification_value: NEUROLOGIC
    notes: >-
      Essentially the entire reported morbidity is neurological: ataxia,
      seizures, developmental regression, dystonia, axonal sensorimotor
      neuropathy, spasticity, cerebral and cerebellar atrophy, and terminally
      central hypoventilation. Non-neurological findings are reported but are
      either downstream of neurological failure (respiratory insufficiency from
      brainstem involvement, cardiac arrhythmia and hypertension from autonomic
      dysfunction) or rare add-ons in single reports (ichthyosis, cataract,
      endocrine deficits). A single NEUROLOGIC chapter is therefore the correct
      assignment, and no second chapter is asserted.
    evidence:
    - reference: PMID:30100084
      reference_title: "Biallelic Mutations in ADPRHL2, Encoding ADP-Ribosylhydrolase 3, Lead to a Degenerative Pediatric Stress-Induced Epileptic Ataxia Syndrome."
      supports: SUPPORT
      evidence_source: HUMAN_CLINICAL
      snippet: "Affected individuals exhibited a pediatric-onset neurodegenerative disorder with progressive brain atrophy, developmental regression, and seizures in association with periods of stress, such as infections."
      explanation: The founding cohort description is entirely neurological, which is the basis for the single NEUROLOGIC chapter assignment.
notes: >-
  Entity verification and gene-symbol trap. The MONDO term was verified with OAK
  before any content was written: `runoak -i sqlite:obo:mondo info MONDO:0100095
  -O obo` gives `relationship: RO:0004003 HGNC:21304 ! ADPRS`, `xref:
  OMIM:618170`, `xref: Orphanet:694922`, `xref: DOID:0070352`, `xref:
  MEDGEN:1648391`, and the synonym CONDSIAS; `runoak -i sqlite:obo:mondo
  relationships -p RO:0004003 MONDO:0100095` returns the same single gene.
  `runoak -i sqlite:obo:hgnc info hgnc:21304` returns ADPRS. The MONDO
  definition text still names the gene by its FORMER symbol ADPRHL2, and the
  literature uses three names for one gene - ADPRS (current HGNC symbol),
  ADPRHL2 (former symbol, used by both 2018 founding papers and most case
  reports) and ARH3 (the protein, used by essentially all of the biochemistry).
  A PubMed search restricted to any one of those three names misses most of the
  evidence base; every search underlying this entry covered all three. HGNC has
  no record under the string ADPRHL2 (`runoak -i sqlite:obo:hgnc search
  "ADPRHL2"` returns nothing), so the alias mapping cannot be recovered from the
  HGNC adapter alone.

  NEC preflight, and the specific trap this disease carries. The dominant gene
  named across the clinical corpus is unambiguously ADPRHL2/ADPRS/ARH3, matching
  the MONDO definition, and every OMIM number asserted in the sources
  (OMIM:618170 for the disease, OMIM:610624 for the gene) matches the MONDO
  xref. One concrete confusion was found and discarded rather than curated: a
  PubMed query for `ADPRS[TIAB]` on 2026-08-01 returned roughly 100 records, and
  the large majority were Alzheimer disease neuroimaging and epidemiology papers
  in which "ADPRS" is an author-coined abbreviation for "Alzheimer's Disease
  Polygenic Risk Score", not the gene. Those records (for example
  PMID:42528531, PMID:42470526, PMID:42221673, PMID:41947717, PMID:41121378)
  were rejected as a block and none of their content appears here. This is a
  textbook acronym-collision NEC risk and it is recorded so the next curator
  does not have to rediscover it.

  Nomenclature caution on one differential. The disorder caused by the
  structurally unrelated hydrolase TARG1/OARD1/C6orf130 (a macrodomain
  ADP-ribosylglutamate hydrolase) is also a recessive neurodegenerative disease
  of ADP-ribosylation reversal, and its gene's aliases overlap confusingly with
  ADPRS in casual writing. It is a different gene and a different MONDO entity
  and is deliberately not curated here; no claim in this entry is sourced from
  TARG1 literature.

  Evidence base and its shape. This entry rests on two 2018 founding cohorts
  (PMID:30100084, 16 individuals from 6 families; PMID:30401461, 12 individuals
  from 8 families), one 2024 PRISMA systematic review aggregating 47 published
  patients from 12 articles up to 3 February 2023 (PMID:38365196), a set of
  individual case reports extending the phenotype, and a substantial
  biochemical/cell-biological literature on ARH3 that is largely independent of
  the clinical reports. The clinical and the mechanistic literatures are
  therefore of very different maturity: the enzymology is well worked out, the
  natural history is a pooled case-report series with no prospective cohort, no
  registry, and no natural-history study.

  Quantitative discipline. Every number in this entry's prose traces to a
  cached source that was read directly. The 47-patient aggregate figures
  (triggering factors 59.5%, regression 60.4%, axonal polyneuropathy 39.1%,
  cerebral/cerebellar atrophy 77.7%, fatality 44.6% at median 7 years, median
  onset 2 years) are from the abstract of PMID:38365196 and are quoted as
  snippets. Two internal inconsistencies in that source are recorded rather than
  smoothed over. First, its abstract gives regression as "28/43; 60.4%" while
  its Table 1 gives 26/43 (60.4%); 60.4% of 43 is 26.0, so the table is
  self-consistent and the abstract numerator appears to be a typographical
  error - the percentage, not the numerator, is used here. Second, its Table 1
  reports seizure as a presenting symptom in 16/47 (34%) but as a clinical
  symptom in only 11/47 (23.4%), which cannot both be complete; no frequency
  band is asserted for seizures as a result. Third, the same review writes the
  recurrent Turkish variant as both "c.235A > C" and "c.235C > A" in different
  sentences for the same p.(Thr79Pro) protein change; ACG->CCG requires c.235A>C,
  so c.235A>C is used here and the discrepancy is flagged in `genetic:`.

  Frequency bands are asserted only where a percentage is quotable from a
  cached abstract. HPO annotation fractions for OMIM:618170 (retrieved
  2026-08-01 from https://ontology.jax.org/api/network/annotation/OMIM:618170)
  are recorded in individual phenotype `notes:` because HPOA is not a
  snippet-validatable reference in this repository. Where the review's pooled
  denominator and the HPOA per-cohort denominator disagree - ataxia is 11/11 in
  the Danhauser cohort but 17/47 (36.1%) in the pooled review, because the
  review's denominator is all 47 patients including those for whom the datum was
  not extractable - no band is asserted at all.

  Deliberately absent sections and why. `histopathology:` is absent: the only
  neuropathological observation in the entire literature is the statement in
  PMID:30830864 that parthanatos was observed in neurons of a deceased sibling,
  with no systematic autopsy series, no Purkinje-cell counts and no white-matter
  pathology described anywhere. Asserting a histopathology section from that
  single sentence would over-read it; it is instead cited on the parthanatos
  pathophysiology node and flagged as a knowledge gap. `biochemical:` is absent:
  no diagnostic biomarker exists, and the obvious candidate assay
  (poly/mono-ADP-ribose immunoblot on patient fibroblasts) is a research tool
  with no reference interval. `clinical_trials:` is absent: no source read for
  this entry reports a registered interventional trial in this disorder, and no
  trial registry was queried directly, so the section is left empty rather than
  asserted to be empty.

  Structured-source evidence was unavailable in this checkout.
  `data/orphadata/` and `data/clingen/` contain only `MANIFEST.yaml` with no
  downloaded payload, so no Orphanet prevalence class (ORPHA:694922 exists as a
  MONDO xref) and no ClinGen gene-disease-validity classification could be
  consulted or quoted. Per scope rules the manifests were not touched. This is
  recorded as a gap, not as an assertion that no such records exist.

  Two further reports were retrieved and cached during the literature sweep but
  yielded no quotable text, so they support no claim in this entry and appear in
  no `references:` block. PMID:33426173 (Mishra et al., Mov Disord Clin Pract
  2021, "Dystonia and Myelopathy in a Case of Stress-Induced Childhood-Onset
  Neurodegeneration with Ataxia and Seizures") is indexed without an abstract;
  its subject matter, dystonia with myelopathy, is covered here from the
  systematic review's spinal-cord figures instead. PMID:39417910 (Eslamiyeh et
  al., Acta Neurol Belg 2025, "Repurposing doxycycline for a case of CONDSIAS
  Syndrome with a novel ADPRHL2 missense mutation") is a letter whose cached
  record contains only bibliographic metadata; it is referred to in the
  minocycline treatment's notes without a snippet. Both cache files are
  committed so that the negative result of the sweep is reproducible.

  No GeneReviews chapter for this disorder was found on PubMed on 2026-08-01,
  so the usual GeneReviews phenotype baseline is not available and the
  phenotype list here is built from the primary reports plus the systematic
  review.
inheritance:
- name: Autosomal recessive inheritance
  description: >-
    Disease requires two damaging ADPRS alleles. Both founding cohorts were
    bi-allelic, and heterozygous parents are consistently reported as
    unaffected. Homozygosity strongly predominates over compound
    heterozygosity, which is a consequence of ascertainment through
    consanguineous families rather than a property of the locus: in the pooled
    review a history of consanguinity was present in 30 of 42 patients for whom
    it was recorded. Compound heterozygous genotypes are nonetheless documented
    (only 2 of the 47 patients in the pooled review, with further examples since:
    PMID:35664652, PMID:35222245, PMID:42015965).

    Penetrance is deliberately left unset rather than recorded as COMPLETE. No
    unaffected bi-allelic individual has been reported, which superficially
    argues for complete penetrance, but the entire literature is
    clinically ascertained case reports with no population screening, so
    non-penetrance could not have been detected even if it existed. The
    intra-familial variability that is reported - one adult sibling described as
    only mildly affected alongside a severely affected 8-year-old, and two of
    three affected sisters in the PAMP-syndrome family described as having
    "much milder phenotypes" than the index case - shows that expressivity is
    variable at fixed genotype, which is a separate question from penetrance and
    is the more clinically consequential one here. The consanguinity fraction
    (30/42) is from Table 1 of PMID:38365196, which is PDF-extracted full text
    rather than the cached abstract, so it is recorded here and not as an
    evidence snippet.
  inheritance_term:
    preferred_term: Autosomal recessive inheritance
    term:
      id: HP:0000007
      label: Autosomal recessive inheritance
  evidence:
  - reference: PMID:30100084
    reference_title: "Biallelic Mutations in ADPRHL2, Encoding ADP-Ribosylhydrolase 3, Lead to a Degenerative Pediatric Stress-Induced Epileptic Ataxia Syndrome."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Using linkage analysis and exome or genome sequencing, we identified recessive inactivating mutations in ADPRHL2 in six families."
    explanation: Establishes recessive inheritance of inactivating alleles in the founding six-family cohort.
  - reference: PMID:30401461
    reference_title: "Bi-allelic ADPRHL2 Mutations Cause Neurodegeneration with Developmental Delay, Ataxia, and Axonal Neuropathy."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "We used exome sequencing to identify five different bi-allelic pathogenic ADPRHL2 variants in 12 individuals from 8 families affected by a neurodegenerative disorder manifesting in childhood or adolescence"
    explanation: Independent replication of bi-allelic inheritance in a second, larger multi-family cohort.
prevalence:
- population: Worldwide
  measure_type: CASES_IN_LITERATURE
  prevalence_class: ULTRA_RARE
  notes: >-
    No incidence or prevalence estimate of any kind has been published, so no
    numeric rate is recorded and `rate_per_100000` is deliberately left empty.
    The qualitative ULTRA_RARE band is taken from the explicit characterisation
    in the source literature rather than derived from a count. The published
    denominator as of this curation is 47 patients from 12 articles up to
    3 February 2023 (PMID:38365196), plus subsequent individual reports curated
    here (PMID:38365196's own two cases, PMID:39580621, PMID:40493129,
    PMID:42015965), so the true published total is now somewhat above 50. No
    founder effect is asserted at population level, although the authors of
    PMID:38365196 raise the possibility of one for the recurrent p.Thr79Pro
    allele in Turkey; ascertainment is heavily weighted towards consanguineous
    populations and towards centres running research exome sequencing. Table 2
    of PMID:32746785, which tabulates the nationality of each reported family,
    lists Turkey (three families), Iran (two), the United Arab Emirates,
    Pakistan and Lebanon alongside Italy, Germany, Kosovo, Poland and China;
    the pooled review records consanguinity in 30 of 42 patients for whom it was
    documented. That distribution reflects reporting and consanguinity
    patterns, not the true geographic range of the disorder.
  evidence:
  - reference: PMID:35664652
    reference_title: "Child-Onset Cerebellar Ataxia Caused by Two Compound Heterozygous Variants in ADPRS Gene: A Case Report."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "an ultra-rare neurodegenerative autosomal recessive disorder"
    explanation: Explicit qualitative rarity characterisation, which is the basis for the ULTRA_RARE band in the absence of any quantitative estimate.
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "In total, 12 articles describing 47 patients were included in the final analysis."
    explanation: Establishes the published case denominator at the time of the most recent systematic review.
progression:
- phase: Normal or near-normal early development
  age_range: Birth to approximately 1-2 years
  notes: >-
    Most affected children are described as developmentally normal, or as having
    only mild isolated delay (commonly expressive language), before the first
    stress-related episode. In the pooled review 25.5% of patients for whom the
    datum was available had entirely normal milestones and a further 16.2% had
    pre-existing delay, so a normal early history neither supports nor excludes
    the diagnosis. Median age at symptom onset across 47 patients was 2 years,
    with a range from 0.7 to 25 years.
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Median age at symptom onset was 2 (0.7-25) years"
    explanation: Quantifies onset age and its range across the pooled published cohort.
- phase: Stress-triggered episodic deterioration
  age_range: Early childhood onwards
  notes: >-
    The defining phase. Discrete episodes - ataxia, dystonic posturing or
    torticollis, seizures, encephalopathy, sudden loss of ambulation or speech -
    are precipitated by febrile illness, infection, surgery, trauma, diarrhoea,
    exercise or cold-water immersion. Recovery between episodes is partial, so
    each event leaves a lower baseline. Some episodes resolve spontaneously over
    weeks and are misread as benign, which is a documented cause of diagnostic
    delay: in the first case of PMID:38365196 the episodic head tilt attacks
    recurred about weekly over some two months and then resolved without any
    treatment, only to return six weeks before presentation and then increase in
    frequency and progress.
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Triggering factors (28/47; 59.5%) and regression (28/43; 60.4%), axonal polyneuropathy (9/23; 39.1%), and cerebral and cerebellar atrophy with white matter changes (28/36; 77.7%) were the other clues."
    explanation: Quantifies documented triggering factors and developmental regression across the pooled cohort. The numerator "28/43" for regression is inconsistent with the stated 60.4% and with Table 1 of the same paper (26/43); see the entry-level notes.
- phase: Progressive neurodegeneration
  age_range: Variable; typically within a few years of onset
  notes: >-
    Cumulative loss of motor and language function, spastic-ataxic quadriparesis,
    dysarthria and dysphagia, with progressive cerebral and cerebellar atrophy on
    serial imaging. Imaging can lag the clinical course: in the pooled review 7
    of the 28 patients with an abnormal MRI (25%) had a normal first study and
    became abnormal only on follow-up, so a normal early scan does not exclude
    the diagnosis.
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "who later had a neurodegenerative course complicated by central hypoventilation syndrome due to the same homozygous ADPRHL2 variant"
    explanation: Describes the progression from episodic presentation to a neurodegenerative course with brainstem respiratory involvement.
- phase: Respiratory failure and death
  age_range: Median age of death 7 years (range 2-34)
  notes: >-
    Respiratory failure and cardiopulmonary arrest are the usual terminal
    events, often central hypoventilation requiring tracheostomy rather than
    neuromuscular weakness alone. Survival is highly variable: the same
    aggregate that reports a median age at death of 7 years also reports living
    patients up to 40 years old, and a distinct adult-onset group presents in
    the third decade with a comparatively indolent course.
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "The fatality rate and median age of death were 44.6% (n = 21) and 7 (2-34) years, respectively."
    explanation: Quantifies mortality and age at death across the pooled published cohort.
pathophysiology:
- name: Bi-allelic Inactivating ADPRS Variants
  biological_scale: MOLECULAR
  description: >-
    The primary lesion is bi-allelic damage to ADPRS at 1p34.3, reported against
    transcript NM_017825.2/NM_017825.3. Missense, nonsense, frameshift and
    splice-site alleles are all represented and all converge on loss of ARH3
    protein or activity rather than on a gain of function or a dominant-negative
    effect. The missense alleles are not a mild class: p.Thr79Pro, p.Cys26Phe,
    p.His182Arg, p.Leu162Pro and p.Ala280Thr have each been shown biochemically
    to destabilise the protein, and in the PAMP-syndrome family the p.Ala280Thr
    protein was undetectable on western blot despite being a single amino-acid
    substitution. The practical consequence for variant interpretation is that a
    conservative-looking ADPRS missense change cannot be dismissed on in-silico
    grounds.
  genes:
  - preferred_term: ADPRS
    term:
      id: hgnc:21304
      label: ADPRS
  evidence:
  - reference: PMID:30100084
    reference_title: "Biallelic Mutations in ADPRHL2, Encoding ADP-Ribosylhydrolase 3, Lead to a Degenerative Pediatric Stress-Induced Epileptic Ataxia Syndrome."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Using linkage analysis and exome or genome sequencing, we identified recessive inactivating mutations in ADPRHL2 in six families."
    explanation: Names the causal gene and characterises the alleles as inactivating in the founding cohort.
  - reference: PMID:33528672
    reference_title: "Episodic psychosis, ataxia, motor neuropathy with pyramidal signs (PAMP syndrome) caused by a novel mutation in ADPRHL2 (AHR3)."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "All affected patients had homozygous novel NM_017825.3:c.838G>A (p.Ala280Thr) mutations in a highly conserved region of ADPRHL2. Western blot analyses demonstrated that ADPRHL2 was not expressed in these patients."
    explanation: Demonstrates that a single missense substitution can abolish detectable protein, which is why missense alleles at this locus behave as null.
  - reference: PMID:39580621
    reference_title: "An ADPRS variant disrupts ARH3 stability and subcellular localization in children with neurodegeneration and respiratory failure."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Mechanistically, the ARH3H182R variant affects a highly conserved residue in the active site of ARH3, leading to protein instability, degradation, and, subsequently, reduced protein expression."
    explanation: Active-site missense alleles act by destabilising the protein rather than by producing a stable catalytically dead enzyme.
  downstream:
  - target: Loss of ARH3 Serine-ADP-Ribosylhydrolase Activity
    description: >-
      Every reported allele class - null, frameshift, splice and destabilising
      missense - converges on absent or severely reduced ARH3 protein and
      therefore on loss of its hydrolase activity.
- name: Loss of ARH3 Serine-ADP-Ribosylhydrolase Activity
  biological_scale: MOLECULAR
  description: >-
    ARH3 is a 39 kDa binuclear-magnesium metalloenzyme of the ARH family. It has
    two distinct, non-redundant activities relevant to this disease. The first
    is unique to it: ARH3 is the only human enzyme that cleaves the terminal
    seryl-ADP-ribosyl bond, that is, the single ADP-ribose left attached to a
    serine residue after PARG has trimmed the poly(ADP-ribose) chain. PARG
    cannot perform this step, so there is no redundancy and no compensation.
    The second is shared: ARH3 also hydrolyses poly(ADP-ribose) chains, acting
    in tandem with PARG on nuclear and cytoplasmic PAR and providing the
    principal PAR-degrading activity of the mitochondrial matrix, where PARG is
    scarce. Loss of ARH3 therefore removes one activity outright and reduces the
    reserve capacity of the other, which is why the resulting phenotype is
    unmasked by stress rather than being constant.
  genes:
  - preferred_term: ADPRS
    term:
      id: hgnc:21304
      label: ADPRS
  molecular_functions:
  - preferred_term: ADP-ribosylserine hydrolase activity
    modifier: DECREASED
    term:
      id: GO:0140292
      label: ADP-ribosylserine-[protein] hydrolase activity
  biological_processes:
  - preferred_term: protein de-ADP-ribosylation
    modifier: DECREASED
    term:
      id: GO:0051725
      label: protein de-ADP-ribosylation
  cellular_components:
  - preferred_term: nucleus
    term:
      id: GO:0005634
      label: nucleus
  - preferred_term: mitochondrion
    term:
      id: GO:0005739
      label: mitochondrion
  notes: >-
    Subcellular localisation is itself a disease mechanism at this locus and not
    merely a descriptive detail. Two independently characterised alleles are
    mislocalisation alleles: p.His182Arg fails to reach the nucleus
    (PMID:39580621) and p.Val335Gly retains mitochondrial localisation while
    showing altered cytosolic/nuclear distribution with near-normal basal
    activity but elevated nuclear ADP-ribosylation specifically under stress
    (PMID:34479984). The p.Val335Gly result is the cleanest published
    demonstration that the defect in this disease is conditional on stress
    rather than constitutive.
  evidence:
  - reference: PMID:28650317
    reference_title: "Serine ADP-ribosylation reversal by the hydrolase ARH3."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "we identified ARH3/ADPRHL2 as capable of efficiently and specifically removing Ser-ADPr of histones and other proteins"
    explanation: Original identification of ARH3 as the serine-ADP-ribosylhydrolase, the activity whose loss defines this disorder.
  - reference: PMID:34019811
    reference_title: "Unrestrained poly-ADP-ribosylation provides insights into chromatin regulation and human disease."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "ARH3 is the only hydrolase able to remove serine-linked mono(ADP-ribose) (MAR) but is much less efficient than PARG against poly(ADP-ribose) (PAR) chains in vitro."
    explanation: Establishes both the non-redundancy of the serine-MAR activity and the relative weakness of the PAR activity, which together set the shape of the deficiency.
  - reference: PMID:34869334
    reference_title: "The Making and Breaking of Serine-ADP-Ribosylation in the DNA Damage Response."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: "another enzyme, (ADP-ribosyl)hydrolase 3 (ARH3), specifically cleaves the terminal seryl-ADP-ribosyl bond, thus completing the chain degradation initiated by PARG"
    explanation: Places ARH3 as the obligatory final step of the two-enzyme chain-degradation relay. Classified OTHER because this is a review.
  - reference: PMID:24191052
    reference_title: "ADP-ribosyl-acceptor hydrolase 3 regulates poly (ADP-ribose) degradation and cell death during oxidative stress."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "However, we show an alternative PAR-degradation pathway, resulting from action of ADP ribosyl-acceptor hydrolase (ARH) 3."
    explanation: Establishes the second, PARG-parallel activity of ARH3 on poly(ADP-ribose), the activity lost in the cytoplasmic and nuclear compartments.
  - reference: PMID:33894202
    reference_title: "Structural and biochemical analysis of human ADP-ribosyl-acceptor hydrolase 3 reveals the basis of metal selectivity and different roles for the two magnesium ions."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "ADP-ribosyl-acceptor hydrolase 3 (ARH3) cleaves PAR and mono(ADP-ribosyl)ation at serine following DNA damage."
    explanation: Structural study confirming the dual substrate specificity and locating the catalytic binuclear magnesium centre that active-site missense alleles disrupt.
  downstream:
  - target: Persistent Serine-Mono(ADP-Ribose) Chromatin Scars
    description: >-
      With no enzyme able to remove the terminal serine-linked ADP-ribose,
      mono(ADP-ribose) marks left behind by completed single-strand break repair
      persist on core histones.
    hypothesis_groups:
    - mar_chromatin_scar_transcriptional_model
  - target: Failure of Poly(ADP-Ribose) Turnover under Stress
    description: >-
      Loss of the ARH3 arm of PAR degradation removes the reserve capacity that
      normally allows a cell to clear a large, rapid PARP1-driven PAR burst.
    hypothesis_groups:
    - canonical_par_parthanatos_model
- name: Physiological Stressor
  biological_scale: ORGANISM
  description: >-
    The exogenous trigger, and the feature that names the disorder. Febrile
    illness and infection are by far the commonest, but the reported list also
    includes surgery, trauma, diarrhoea, physical exercise, emotional stress and
    cold-water immersion. A trigger was documented in 59.5% of the 47 pooled
    published patients, which is a floor rather than an estimate because
    triggers are recorded only when someone thought to ask. The mechanistic
    bridge assumed throughout the field is that these states raise cellular
    oxidative and genotoxic burden and therefore PARP1 activity; that bridge is
    supported by model-system surrogates (hydrogen peroxide, ischaemia
    reperfusion) rather than by any direct measurement during a human febrile
    episode, and the honest strength of the link is stated in the discussions
    section.
  triggers:
  - preferred_term: Febrile illness
  - preferred_term: Infection
  - preferred_term: Surgery
  - preferred_term: Trauma
  - preferred_term: Physical exertion
  - preferred_term: Cold water immersion
  evidence:
  - reference: PMID:30100084
    reference_title: "Biallelic Mutations in ADPRHL2, Encoding ADP-Ribosylhydrolase 3, Lead to a Degenerative Pediatric Stress-Induced Epileptic Ataxia Syndrome."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "seizures in association with periods of stress, such as infections"
    explanation: The founding report ties the episodic events directly to periods of stress such as infection.
  - reference: PMID:40493129
    reference_title: "Phenotypic Diversity in Stress-induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures: Novel Associations with Parkinsonism, Icthyosis and Cataract."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "A distinctive feature of this condition is the exacerbation of clinical symptoms triggered by physical or emotional stress, as well as febrile illnesses."
    explanation: Independent confirmation that physical and emotional stress and febrile illness are the characteristic precipitants.
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Triggering factors (28/47; 59.5%) and regression (28/43; 60.4%)"
    explanation: Quantifies how often a triggering factor was documented across the pooled published cohort.
  downstream:
  - target: Failure of Poly(ADP-Ribose) Turnover under Stress
    description: >-
      Stress states increase DNA damage and oxidative burden, driving PARP1
      activation and a PAR burst that an ARH3-deficient cell cannot clear.
    hypothesis_groups:
    - canonical_par_parthanatos_model
- name: Persistent Serine-Mono(ADP-Ribose) Chromatin Scars
  biological_scale: MOLECULAR
  description: >-
    Serine is the dominant acceptor residue for ADP-ribosylation after DNA
    damage, and the modification is made by PARP1/PARP2 in complex with their
    cofactor HPF1 on core histones at the break site. When repair finishes, PARG
    removes the polymer but leaves the innermost ADP-ribose still esterified to
    the serine. Only ARH3 can remove that residue. In ARH3-deficient patient
    cells the leftover marks accumulate as mono(ADP-ribose) "chromatin scars" - a
    physical record of every single-strand break the cell has recently repaired,
    written onto histones and never erased. This is a chronic, stress-independent
    lesion: it exists at baseline, throughout the cell cycle, in the absence of
    any acute insult.
  biological_processes:
  - preferred_term: single strand break repair
    modifier: ABNORMAL
    term:
      id: GO:0000012
      label: single strand break repair
  - preferred_term: protein de-ADP-ribosylation
    modifier: DECREASED
    term:
      id: GO:0051725
      label: protein de-ADP-ribosylation
  cellular_components:
  - preferred_term: chromatin
    term:
      id: GO:0000785
      label: chromatin
  cell_types:
  - preferred_term: patient-derived dermal fibroblast
    term:
      id: CL:0000057
      label: fibroblast
  evidence:
  - reference: PMID:32636369
    reference_title: "Pathogenic ARH3 mutations result in ADP-ribose chromatin scars during DNA strand break repair."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "we show that ARH3-mutated patient cells accumulate mono(ADP-ribose) scars on core histones that are a molecular memory of recently repaired DNA single-strand breaks"
    explanation: Direct demonstration in patient-derived cells of the chromatin scar, the chronic baseline lesion of this disorder.
  - reference: PMID:29480802
    reference_title: "Serine is the major residue for ADP-ribosylation upon DNA damage."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "we show that Ser-ADPr represents the major fraction of ADPr synthesised after DNA damage in mammalian cells and that globally Ser-ADPr is dependent on HPF1, PARP1 and ARH3"
    explanation: Establishes that serine-linked ADP-ribosylation, the ARH3-dependent species, is the dominant form of the modification after DNA damage rather than a minor variant.
  downstream:
  - target: Deregulated Histone Acetylation and Transcription
    description: >-
      Retained ADP-ribose on histone tails interferes with the deposition of
      activating acetyl marks at the same residues and in their neighbourhood.
    hypothesis_groups:
    - mar_chromatin_scar_transcriptional_model
- name: Deregulated Histone Acetylation and Transcription
  biological_scale: CELLULAR
  description: >-
    ARH3-deficient patient cells show reduced endogenous levels of activating
    chromatin modifications, most clearly H3K9 acetylation, and measurable
    deregulation of transcription. The causal direction was tested rather than
    assumed: prolonged PARP inhibition removes the scars and restores acetylation
    to normal, so the acetylation deficit is a consequence of the retained
    ADP-ribose and not an independent abnormality. In a long-lived post-mitotic
    neuron, which cannot dilute chromatin marks by division, a slowly
    accumulating transcriptional deficit of this kind is an attractive
    explanation for degeneration that appears only after years of normal
    function - but the link from this cellular phenotype to human
    neurodegeneration has not been demonstrated in neurons or in patient tissue.
  biological_processes:
  - preferred_term: protein acetylation
    modifier: DECREASED
    term:
      id: GO:0006473
      label: protein acetylation
  - preferred_term: regulation of DNA-templated transcription
    modifier: ABNORMAL
    term:
      id: GO:0006355
      label: regulation of DNA-templated transcription
  - preferred_term: chromatin remodeling
    modifier: ABNORMAL
    term:
      id: GO:0006338
      label: chromatin remodeling
  cell_types:
  - preferred_term: patient-derived dermal fibroblast
    term:
      id: CL:0000057
      label: fibroblast
  mechanism_confidence: PROVISIONAL
  notes: >-
    Marked PROVISIONAL and separated into its own hypothesis group because the
    two best studies of this arm disagree about its importance. PMID:32636369
    frames the scars and the resulting acetylation loss as the pathogenic lesion.
    PMID:34019811, from an overlapping group, then showed that persistent
    MARylation is "surprisingly well tolerated" while persistent PARylation is
    highly toxic, and reattributed the disease to a stress-induced PARylation
    excess. Both cannot be the dominant mechanism. This entry curates both and
    marks the canonical PAR/parthanatos model as CANONICAL and the chromatin-scar
    model as ALTERNATIVE rather than adjudicating between them; see the
    `mechanistic_hypotheses:` block and the corresponding discussion.
  evidence:
  - reference: PMID:32636369
    reference_title: "Pathogenic ARH3 mutations result in ADP-ribose chromatin scars during DNA strand break repair."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "We demonstrate that the ADP-ribose chromatin scars result in reduced endogenous levels of important chromatin modifications such as H3K9 acetylation, and that ARH3 patient cells exhibit measurable levels of deregulated transcription."
    explanation: Directly links the retained ADP-ribose marks in patient cells to loss of activating histone acetylation and to transcriptional deregulation.
  - reference: PMID:32636369
    reference_title: "Pathogenic ARH3 mutations result in ADP-ribose chromatin scars during DNA strand break repair."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "we show that the mono(ADP-ribose) scars are lost from the chromatin of ARH3-defective cells in the prolonged presence of PARP inhibition, and concomitantly that chromatin acetylation is restored to normal"
    explanation: Reversal by PARP inhibition establishes the causal direction from retained ADP-ribose to acetylation loss and provides the therapeutic rationale for this arm.
  - reference: PMID:34019811
    reference_title: "Unrestrained poly-ADP-ribosylation provides insights into chromatin regulation and human disease."
    supports: PARTIAL
    evidence_source: IN_VITRO
    snippet: "we demonstrate that endogenous MARylation persists on chromatin throughout the cell cycle, including mitosis, and is surprisingly well tolerated. Conversely, persistent PARylation is highly toxic and has distinct physiological effects, in particular on active transcription histone marks such as H3K9ac and H3K27ac."
    explanation: >-
      Marked PARTIAL because it supports the transcriptional-mark component of
      this node while arguing against the mono-ADP-ribose scar being the toxic
      species; it attributes the same acetylation defect to PARylation instead.
  downstream:
  - target: Selective Neuronal Loss with Cerebellar and Cerebral Atrophy
    description: >-
      Chronic transcriptional deregulation in post-mitotic neurons is proposed to
      drive cumulative neuronal dysfunction and loss.
    hypothesis_groups:
    - mar_chromatin_scar_transcriptional_model
- name: Failure of Poly(ADP-Ribose) Turnover under Stress
  biological_scale: MOLECULAR
  description: >-
    The acute arm. Oxidative and genotoxic stress activates PARP1, which
    accounts for the great majority of cellular ADP-ribosylation after DNA
    damage and builds long poly(ADP-ribose) chains within seconds. In a normal
    cell PARG and ARH3 together dismantle those chains on a timescale of
    minutes. An ARH3-deficient cell has only PARG, and the result is a larger and
    longer-lasting PAR burst in both the nucleus and the cytoplasm. This is the
    step at which the disorder becomes stress-conditional: basal ADP-ribosylation
    in ARH3-deficient cells can be close to normal, and the abnormality appears
    when the system is loaded. Patient fibroblasts show reduced viability on
    hydrogen peroxide challenge, and Arh3-knockout cells show PARP1 activation
    with increased nuclear and then cytoplasmic PAR.
  biological_processes:
  - preferred_term: protein poly-ADP-ribosylation
    modifier: INCREASED
    term:
      id: GO:0070212
      label: protein poly-ADP-ribosylation
  - preferred_term: cellular response to oxidative stress
    modifier: ABNORMAL
    term:
      id: GO:0034599
      label: cellular response to oxidative stress
  - preferred_term: DNA damage response
    modifier: ABNORMAL
    term:
      id: GO:0006974
      label: DNA damage response
  molecular_functions:
  - preferred_term: NAD+ poly-ADP-ribosyltransferase activity
    modifier: INCREASED
    term:
      id: GO:0003950
      label: NAD+ poly-ADP-ribosyltransferase activity
  chemical_entities:
  - preferred_term: hydrogen peroxide (experimental stressor)
    term:
      id: CHEBI:16240
      label: hydrogen peroxide
  cell_types:
  - preferred_term: patient-derived dermal fibroblast
    term:
      id: CL:0000057
      label: fibroblast
  evidence:
  - reference: PMID:30401461
    reference_title: "Bi-allelic ADPRHL2 Mutations Cause Neurodegeneration with Developmental Delay, Ataxia, and Axonal Neuropathy."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "ADPRHL2 was virtually absent in available affected individuals' fibroblasts, and cell viability was reduced upon hydrogen peroxide exposure, although it was rescued by expression of wild-type ADPRHL2 mRNA as well as treatment with a PARP1 inhibitor."
    explanation: Patient-cell demonstration of stress-conditional vulnerability, with both genetic and pharmacological rescue establishing that the defect is ARH3 loss acting through PARP1.
  - reference: PMID:36497109
    reference_title: "ARH Family of ADP-Ribose-Acceptor Hydrolases."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: "Incubation of Arh3-KO cells with H2O2 resulted in activation of poly-ADP-ribose polymerase (PARP)-1, followed by increased nuclear PAR, increased cytoplasmic PAR, leading to release of Apoptosis Inducing Factor (AIF) from mitochondria."
    explanation: States the ordered sequence from oxidative stress through PARP1 activation to compartmental PAR accumulation in ARH3-null cells. Classified OTHER because this is a review.
  - reference: PMID:34479984
    reference_title: "Biallelic ADPRHL2 mutations in complex neuropathy affect ADP ribosylation and DNA damage response."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "This minimally affects basal ADP ribosylation but results in elevated nuclear ADP ribosylation during stress, demonstrating the vital role of ADP ribosylation reversal by ARH3 in DNA damage control."
    explanation: The clearest published demonstration that the biochemical defect in a patient allele is conditional on stress rather than present at baseline, which is the molecular counterpart of the episodic clinical course.
  - reference: PMID:42015965
    reference_title: "Novel ADPRS Missense Variant (p.Leu162Pro) Causes Stress-Induced Childhood-Onset Neurodegeneration With Ataxia and Seizures."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "This leads to defective ARH3-dependent removal of ADP-ribosylation, both under basal conditions and following oxidative stress."
    explanation: >-
      Contemporary patient-fibroblast replication of the removal defect. Note
      that this compound-heterozygous genotype shows the defect at baseline as
      well as under stress, which qualifies the "stress-conditional" framing and
      suggests the basal/stress split is allele-dependent.
  downstream:
  - target: NAD+ Consumption and Bioenergetic Failure
    description: >-
      Sustained PARP1 catalysis consumes NAD+ as its ADP-ribose donor, depleting
      the cellular pool that PAR turnover would normally help restore.
    hypothesis_groups:
    - canonical_par_parthanatos_model
  - target: AIF Release from Mitochondria and Parthanatos
    description: >-
      Protein-free PAR generated in the nucleus and reaching the cytoplasm
      triggers mitochondrial AIF release, the committed step of parthanatos.
    hypothesis_groups:
    - canonical_par_parthanatos_model
- name: NAD+ Consumption and Bioenergetic Failure
  biological_scale: CELLULAR
  description: >-
    Each ADP-ribose transferred by PARP1 costs one NAD+. Unrestrained PARylation
    is therefore not only a signalling abnormality but a metabolic drain, and
    the normal turnover of PAR is part of how a cell restores its energy
    reservoir after a genotoxic insult. In an ARH3-deficient cell that restoration
    is impaired. This node is included because it is a standard and
    well-supported feature of PARP1 hyperactivation biology and because it offers
    a plausible reason why neurons - the most energy-demanding and least
    glycolytically flexible cells - are the ones that fail. It is marked
    HYPOTHETICAL for this disorder specifically: no measurement of NAD+ or of
    bioenergetic function has been published in ADPRS-deficient patient cells or
    tissue.
  biological_processes:
  - preferred_term: NAD+ catabolic process
    modifier: INCREASED
    term:
      id: GO:0019677
      label: NAD+ catabolic process
  chemical_entities:
  - preferred_term: NAD+
    term:
      id: CHEBI:15846
      label: NAD(+)
  mechanism_confidence: HYPOTHETICAL
  notes: >-
    No evidence item asserts NAD+ depletion in ADPRS-deficient human cells,
    because none has been reported. The single quoted item below supports only
    the general principle that PAR turnover restores the cellular energy
    reservoir, and it is attached with that limitation stated. Treating this as
    established for CONDSIAS would be exactly the kind of borrowed-mechanism
    inference this entry tries to avoid.
  evidence:
  - reference: PMID:33894202
    reference_title: "Structural and biochemical analysis of human ADP-ribosyl-acceptor hydrolase 3 reveals the basis of metal selectivity and different roles for the two magnesium ions."
    supports: PARTIAL
    evidence_source: IN_VITRO
    snippet: "This turnover of PARylation is not only required for effective control of cellular signaling pathways but also restores the cellular energy reservoir"
    explanation: >-
      Supports the general principle that PAR turnover has an energetic function,
      marked PARTIAL because it is a statement about ADP-ribosylation biology in
      general and reports no measurement in ARH3-deficient cells.
  downstream:
  - target: Selective Neuronal Loss with Cerebellar and Cerebral Atrophy
    description: >-
      Energy failure is proposed to make high-demand neurons the first cells to
      cross the threshold from dysfunction to death.
    hypothesis_groups:
    - canonical_par_parthanatos_model
- name: AIF Release from Mitochondria and Parthanatos
  biological_scale: CELLULAR
  description: >-
    Parthanatos is a caspase-independent, PAR-dependent form of regulated cell
    death, and it is the effector step of the canonical model of this disorder.
    PARG liberates protein-free PAR from PARylated nuclear proteins; that free
    polymer reaches the cytoplasm and acts on mitochondria to release
    apoptosis-inducing factor, which translocates back to the nucleus and drives
    large-scale DNA cleavage. ARH3's protective role is to keep nuclear and
    cytoplasmic PAR below the threshold at which this happens, so ARH3 loss
    lowers the amount of stress needed to commit a cell to death. This is the
    node at which a transient febrile illness could plausibly convert into
    permanent neurological loss, and the ratchet-like clinical course - episodes
    followed by incomplete recovery - is what a threshold-crossing cell-death
    mechanism predicts.
  biological_processes:
  - preferred_term: programmed cell death
    modifier: INCREASED
    term:
      id: GO:0012501
      label: programmed cell death
  - preferred_term: apoptotic mitochondrial changes
    modifier: INCREASED
    term:
      id: GO:0008637
      label: apoptotic mitochondrial changes
  cellular_components:
  - preferred_term: mitochondrion
    term:
      id: GO:0005739
      label: mitochondrion
  cell_types:
  - preferred_term: neuron
    term:
      id: CL:0000540
      label: neuron
  - preferred_term: patient-derived dermal fibroblast
    term:
      id: CL:0000057
      label: fibroblast
  notes: >-
    The cell-type binding to neuron rests on a single sentence in PMID:30830864
    reporting that parthanatos was observed in neurons of a deceased sibling.
    That is the only direct human neuropathological observation in the whole
    literature, and it is quoted below rather than paraphrased. Fibroblast is the
    cell type in which the mechanism has actually been dissected. No autopsy
    series, neuronal-subtype quantification or regional distribution of neuronal
    death has ever been reported for this disorder.
  evidence:
  - reference: PMID:24191052
    reference_title: "ADP-ribosyl-acceptor hydrolase 3 regulates poly (ADP-ribose) degradation and cell death during oxidative stress."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "A protective effect of ARH3 results from its lowering of PAR levels in the nucleus and the cytoplasm, thereby preventing release of AIF from mitochondria and its accumulation in the nucleus."
    explanation: Establishes the specific protective function of ARH3 that is lost in this disorder and identifies AIF release as the step it prevents.
  - reference: PMID:24191052
    reference_title: "ADP-ribosyl-acceptor hydrolase 3 regulates poly (ADP-ribose) degradation and cell death during oxidative stress."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "PAR is responsible for induction of parthanatos, a mechanism for caspase-independent cell death, triggered by apoptosis-inducing factor (AIF) release from mitochondria and its translocation to the nucleus, where it initiates DNA cleavage."
    explanation: Defines the parthanatos pathway that PAR accumulation engages, including the terminal nuclear DNA cleavage step.
  - reference: PMID:30830864
    reference_title: "PARP1 inhibition alleviates injury in ARH3-deficient mice and human cells."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "In addition, parthanatos was observed in neurons of the patient's deceased sibling, and an older sibling exhibited a mild behavioral phenotype."
    explanation: The single direct human observation that the proposed death mechanism actually occurs in the neurons of an affected individual.
  - reference: PMID:30830864
    reference_title: "PARP1 inhibition alleviates injury in ARH3-deficient mice and human cells."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Consistent with the previous findings, the patient's fibroblasts and ARH3-deficient mice were more sensitive, respectively, to H2O2 stress and cerebral ischemia/reperfusion-induced PAR accumulation and cell death."
    explanation: Concordant human-cell and mouse evidence that ARH3 loss lowers the threshold for stress-induced PAR accumulation and death.
  downstream:
  - target: Selective Neuronal Loss with Cerebellar and Cerebral Atrophy
    description: >-
      Repeated episodes of threshold-crossing neuronal death accumulate as
      macroscopic tissue loss.
    hypothesis_groups:
    - canonical_par_parthanatos_model
  - target: Distal Axonal Degeneration of Peripheral Nerve
    description: >-
      Long peripheral axons and their cell bodies are subject to the same
      stress-conditional death mechanism, producing a length-dependent axonal
      neuropathy.
    hypothesis_groups:
    - canonical_par_parthanatos_model
- name: Selective Neuronal Loss with Cerebellar and Cerebral Atrophy
  biological_scale: TISSUE
  description: >-
    The tissue-level endpoint in the central nervous system. Cerebral and
    cerebellar atrophy with white matter signal change is the commonest imaging
    abnormality, present in 77.7% of the imaged patients in the pooled review,
    with the cerebellar vermis particularly affected and spinal cord atrophy or
    T2 hyperintensity in a substantial minority. The atrophy is acquired and
    progressive rather than developmental: serial imaging in the same child can
    move from normal, to mild frontal atrophy, to marked cortical and cerebellar
    atrophy within a year. Why the cerebellum and the long tracts are
    preferentially affected, when ARH3 is ubiquitously expressed, is unexplained.
  locations:
  - preferred_term: cerebellum
    term:
      id: UBERON:0002037
      label: cerebellum
  - preferred_term: cerebellar vermis
    term:
      id: UBERON:0004720
      label: cerebellar vermis
  - preferred_term: cerebral cortex
    term:
      id: UBERON:0000956
      label: cerebral cortex
  - preferred_term: spinal cord
    term:
      id: UBERON:0002240
      label: spinal cord
  cell_types:
  - preferred_term: neuron
    term:
      id: CL:0000540
      label: neuron
  notes: >-
    Purkinje cells are deliberately NOT bound as a cell type on this node, and
    this node deliberately does not declare `conforms_to`
    `cerebellar_purkinje_degeneration#Purkinje Neuron Degeneration`. Cerebellar
    and vermian atrophy on MRI is a macroscopic finding; no histological
    demonstration of Purkinje cell loss exists in any patient with this
    disorder, and asserting the canonical Purkinje mechanism from an atrophy
    measurement would be an inference dressed as an observation. The generic
    `neuron` binding is used instead.
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "cerebral and cerebellar atrophy with white matter changes (28/36; 77.7%)"
    explanation: Quantifies the frequency of the central imaging abnormality across the pooled published cohort.
  - reference: PMID:30100084
    reference_title: "Biallelic Mutations in ADPRHL2, Encoding ADP-Ribosylhydrolase 3, Lead to a Degenerative Pediatric Stress-Induced Epileptic Ataxia Syndrome."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "a pediatric-onset neurodegenerative disorder with progressive brain atrophy"
    explanation: Establishes progressive brain atrophy as a defining feature of the founding cohort.
  - reference: PMID:37392332
    reference_title: "Expanding the Spectrum of Stress-Induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures (CONDSIAS)."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "positron emission tomography/magnetic resonance imaging (PET/MRI) of the brain revealed cerebellar atrophy, particularly of the vermis, with corresponding hypometabolism"
    explanation: Localises the atrophy to the cerebellar vermis and adds a functional correlate, regional hypometabolism, that supports tissue loss rather than artefact.
  downstream:
  - target: Episodic Stepwise Neurological Deterioration
    description: >-
      Accumulating central neuronal loss is expressed clinically as a
      step-down course rather than a smooth decline.
  - target: Brainstem Respiratory Network Failure
    description: >-
      Extension of the degenerative process to brainstem respiratory control
      centres produces central hypoventilation.
- name: Distal Axonal Degeneration of Peripheral Nerve
  biological_scale: TISSUE
  description: >-
    A length-dependent axonal, often sensorimotor, polyneuropathy is a core part
    of the phenotype and was prominent enough in one series that ADPRHL2
    homozygotes were identified in a cohort assembled for hereditary motor
    neuropathy rather than for neurodegeneration. Nerve conduction studies show
    axonal loss with relatively preserved conduction velocity; distal weakness,
    foot drop, pes cavus and hammer toes follow. The neuropathy is a genuine
    discriminator: it is what separates this disorder from the purely
    cerebellar ataxias in the differential, and its presence in a child with
    stress-triggered ataxia should redirect testing towards ADPRS.
  conforms_to: "peripheral_axonal_degeneration#Distal Axonal Degeneration and Demyelination"
  locations:
  - preferred_term: peripheral nerve
    term:
      id: UBERON:0001021
      label: nerve
  biological_processes:
  - preferred_term: neuron apoptotic process
    modifier: INCREASED
    term:
      id: GO:0051402
      label: neuron apoptotic process
  cell_types:
  - preferred_term: neuron
    term:
      id: CL:0000540
      label: neuron
  notes: >-
    Conformance is declared to the axonal arm of
    `peripheral_axonal_degeneration` only. The module's demyelination arm
    (myelination and axon ensheathment DECREASED, Schwann cell) is deliberately
    NOT copied here: every published nerve conduction study in this disorder
    describes an axonal process, and no demyelinating neuropathy has been
    reported, so importing those module bindings would assert something the
    literature contradicts.
  evidence:
  - reference: PMID:30401461
    reference_title: "Bi-allelic ADPRHL2 Mutations Cause Neurodegeneration with Developmental Delay, Ataxia, and Axonal Neuropathy."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "key clinical features including developmental delay or regression, seizures, ataxia, and axonal (sensori-)motor neuropathy"
    explanation: Names axonal sensorimotor neuropathy as one of the four core features of the second founding cohort.
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "regression (28/43; 60.4%), axonal polyneuropathy (9/23; 39.1%)"
    explanation: Quantifies the neuropathy among pooled patients who underwent electrophysiological testing.
  - reference: PMID:34479984
    reference_title: "Biallelic ADPRHL2 mutations in complex neuropathy affect ADP ribosylation and DNA damage response."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Here, we present two families with a neuropathy predominant disorder and homozygous mutations in ADPRHL2"
    explanation: Documents a neuropathy-predominant pole of the ADPRS spectrum, in which peripheral nerve rather than the cerebellum is the leading site of disease.
  downstream:
  - target: Episodic Stepwise Neurological Deterioration
    description: >-
      Distal weakness and sensory loss add to the cumulative disability that
      defines the clinical course.
- name: Episodic Stepwise Neurological Deterioration
  biological_scale: ORGANISM
  description: >-
    The integrated clinical phenotype: not a smooth degenerative slope but a
    staircase, in which a febrile illness or other stressor produces an acute
    episode of ataxia, dystonia, seizures or encephalopathy from which recovery
    is partial. Between episodes the child may appear stable for months. The
    pattern explains two things that otherwise look contradictory in the
    literature - the frequent early misdiagnosis as episodic ataxia, an
    ion-channel disorder or an acute post-infectious cerebellitis, and the
    coexistence in the same disease of patients who die at 2 years and patients
    who reach 40. Individuals whose stressor exposure is lighter, or whose
    residual ARH3 activity is higher, simply climb down the staircase more
    slowly.
  evidence:
  - reference: PMID:33528672
    reference_title: "Episodic psychosis, ataxia, motor neuropathy with pyramidal signs (PAMP syndrome) caused by a novel mutation in ADPRHL2 (AHR3)."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "An index patient presented ataxia, postural tremor in the hands, and hallucinations at age 20 years, which had started after a viral infection."
    explanation: A worked example of a stress-precipitated episode, here in the adult-onset pole of the spectrum.
  - reference: PMID:33528672
    reference_title: "Episodic psychosis, ataxia, motor neuropathy with pyramidal signs (PAMP syndrome) caused by a novel mutation in ADPRHL2 (AHR3)."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "She had three more severe episodes in recent years, and her neurologic findings got progressively worse."
    explanation: "Documents the ratchet pattern directly: discrete episodes each followed by a worse baseline."
  - reference: PMID:35664652
    reference_title: "Child-Onset Cerebellar Ataxia Caused by Two Compound Heterozygous Variants in ADPRS Gene: A Case Report."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "After a respiratory tract infection, a 30-month-old male patient presented with unsteady gait that rendered walking impossible without external help."
    explanation: Independent example of an infection-triggered acute loss of function in a young child.
  downstream:
  - target: Brainstem Respiratory Network Failure
    description: >-
      Repeated deterioration eventually involves brainstem respiratory control,
      which is the commonest route to death.
- name: Brainstem Respiratory Network Failure
  biological_scale: ORGANISM
  description: >-
    The usual terminal mechanism. Respiratory insufficiency in this disorder is
    frequently central rather than a consequence of neuromuscular weakness or
    aspiration: patients develop hypoventilation during sleep without infection,
    apnoeic episodes, and eventually require tracheostomy and long-term
    ventilation. Autonomic dysfunction - tachycardia, hypertension, abdominal
    distension, impaired sweating - often appears in the same phase, consistent
    with degeneration of brainstem autonomic as well as respiratory nuclei.
    Because a child can be alert and communicative on ventilatory support, this
    is the point where the disorder's management question becomes explicit
    rather than implicit.
  locations:
  - preferred_term: brainstem
    term:
      id: UBERON:0002298
      label: brainstem
  cell_types:
  - preferred_term: neuron
    term:
      id: CL:0000540
      label: neuron
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "a neurodegenerative course complicated by central hypoventilation syndrome due to the same homozygous ADPRHL2 variant"
    explanation: Identifies central hypoventilation, a brainstem lesion, as the complication that defines the terminal phase in two independently ascertained patients.
  - reference: PMID:39580621
    reference_title: "An ADPRS variant disrupts ARH3 stability and subcellular localization in children with neurodegeneration and respiratory failure."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "two siblings with developmental delay, seizures, progressive muscle weakness, and respiratory failure following an episodic course"
    explanation: Independent sibling pair in whom respiratory failure is the defining late feature after an episodic course.
mechanistic_hypotheses:
- hypothesis_group_id: canonical_par_parthanatos_model
  hypothesis_label: Canonical PAR Accumulation and Parthanatos Model
  status: CANONICAL
  description: >-
    The model proposed by both 2018 founding papers and supported by the
    strongest chain of experimental evidence. Stress activates PARP1; ARH3
    deficiency prevents timely clearance of the resulting poly(ADP-ribose);
    PAR accumulates in nucleus and cytoplasm; free PAR triggers mitochondrial
    AIF release and caspase-independent parthanatos; repeated episodes of
    neuronal death produce the neurodegeneration. Its strongest support is that
    PARP inhibition rescues the phenotype in three independent systems - a
    Drosophila oxidative-challenge model, patient fibroblasts, and Arh3-knockout
    mice subjected to cerebral ischaemia-reperfusion. Its weakest link is that
    the human trigger, febrile illness, has never been shown to produce the
    proposed PAR burst in a patient.
  evidence:
  - reference: PMID:30100084
    reference_title: "Biallelic Mutations in ADPRHL2, Encoding ADP-Ribosylhydrolase 3, Lead to a Degenerative Pediatric Stress-Induced Epileptic Ataxia Syndrome."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Loss of the Drosophila paralog Parg showed lethality in response to oxidative challenge that was rescued by human ADPRHL2, suggesting functional conservation."
    explanation: Cross-species rescue establishes that the human protein's function in the oxidative-stress response is the relevant one.
  - reference: PMID:34019811
    reference_title: "Unrestrained poly-ADP-ribosylation provides insights into chromatin regulation and human disease."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "we extend our findings to neurodegeneration, suggesting that patients with inherited ARH3 deficiency suffer from stress-induced pathogenic increase in PARylation that can be mitigated by PARP inhibition"
    explanation: The authors of the competing chromatin-scar arm converge on the PARylation model as the pathogenic one for patients, which is the main reason this group is marked CANONICAL.
- hypothesis_group_id: mar_chromatin_scar_transcriptional_model
  hypothesis_label: Mono(ADP-Ribose) Chromatin Scar and Transcriptional Deregulation Model
  status: ALTERNATIVE
  description: >-
    The alternative model. Because ARH3 is the only enzyme that can remove
    serine-linked mono(ADP-ribose), its loss leaves permanent ADP-ribose scars on
    core histones marking every recently repaired single-strand break. These
    scars suppress activating acetylation marks such as H3K9ac and deregulate
    transcription, and in post-mitotic neurons the effect would accumulate over
    years. This model has the attraction of explaining a chronic degenerative
    process without requiring a discrete death event, and it explains why the
    disorder is degenerative rather than purely episodic. Its problem is that
    the same laboratory subsequently reported that persistent MARylation is well
    tolerated whereas PARylation is toxic, which argues that the scars are a
    marker rather than the injury.
  evidence:
  - reference: PMID:32636369
    reference_title: "Pathogenic ARH3 mutations result in ADP-ribose chromatin scars during DNA strand break repair."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "Collectively, these data indicate that ARH3 can act as an eraser of ADP-ribose chromatin scars at sites of PARP activity during DNA single-strand break repair."
    explanation: States the model's central claim, that ARH3's disease-relevant job is erasing chromatin scars at repaired breaks.
  - reference: PMID:34019811
    reference_title: "Unrestrained poly-ADP-ribosylation provides insights into chromatin regulation and human disease."
    supports: REFUTE
    evidence_source: IN_VITRO
    snippet: "endogenous MARylation persists on chromatin throughout the cell cycle, including mitosis, and is surprisingly well tolerated"
    explanation: Direct experimental argument against the mono(ADP-ribose) scar being the toxic species, which is why this model is recorded as ALTERNATIVE rather than as a second canonical arm.
genetic:
- name: ADPRS
  association: Bi-allelic inactivating variants (missense, nonsense, frameshift and splice), all converging on loss of ARH3 protein or activity
  relationship_type: CAUSATIVE
  gene_term:
    preferred_term: ADPRS
    term:
      id: hgnc:21304
      label: ADPRS
  notes: >-
    Nomenclature. HGNC:21304 is the current symbol ADPRS (ADP-ribosylserine
    hydrolase), verified with OAK on 2026-08-01. The former symbol ADPRHL2 is
    used by the MONDO definition, by both 2018 founding papers and by most case
    reports; ARH3 is the protein name used by the biochemical literature; two
    further typographical variants circulate - ADPRLH2 in the abstract of
    PMID:35664652 ("ADP-ribosylhydrolase-like 2 (ADPRS or ADPRLH2; OMIM:
    610624)") and AHR3 in the title of PMID:33528672. The gene MIM number is 610624 and the phenotype
    MIM number is 618170; a source citing "ADPRHL2 (OMIM: 618170)" as a gene
    identifier, as PMID:32746785 does, has conflated the two.

    Reference transcript. NM_017825.2 and NM_017825.3 are both used in the
    literature; variant coordinates below are as published.

    Allelic spectrum, with the caveat that it is drawn from a clinically
    ascertained, consanguinity-enriched series. In the 47-patient review,
    homozygous genotypes accounted for 45 of the reported patients and only
    two compound heterozygotes had been described at that point; further
    compound heterozygotes have since been reported (PMID:35664652,
    PMID:35222245, PMID:42015965). Missense variants were the commonest class
    (about half), followed by nonsense and frameshift. Exon 6 carried roughly
    half of the homozygous variants and exon 2 the recurrent Turkish allele.

    Named alleles with published functional data: p.Thr79Pro (recurrent, at
    least seven patients from unrelated Turkish families, protein
    destabilisation on expression in E. coli); p.Cys26Phe (protein instability
    and reduced function, PMID:34479984); p.Val335Gly (mild loss of expression
    with retained enzymatic activity but altered cytosolic/nuclear localisation
    and elevated nuclear ADP-ribosylation under stress, PMID:34479984);
    p.His182Arg (active-site residue; instability, degradation, and failure of
    nuclear localisation, PMID:39580621); p.Ala280Thr (no detectable protein on
    western blot, PMID:33528672); p.Leu162Pro in trans to p.Gln106* (severe
    reduction of protein with defective ADP-ribosylation removal,
    PMID:42015965); c.580C>T p.Gln194Ter with c.803-1G>A (nonsense-mediated
    decay plus intron 5 retention, PMID:35664652).

    A correction that should not be silently propagated. PMID:38365196 writes the
    recurrent Turkish allele as "c.235A > C" in its case descriptions and as
    "c.235C > A" in its discussion, for the same protein change p.(Thr79Pro).
    Threonine 79 is encoded ACG, and proline requires CCG, so the change must be
    c.235A>C; the "c.235C>A" spelling in that paper is an error. This entry uses
    c.235A>C.

    Not asserted, deliberately. No gnomAD constraint metric (pLI, o/e LoF,
    missense Z) and no carrier frequency is recorded anywhere in this entry: no
    cached source provides them and they were not looked up, so writing them
    would be unverifiable. No ClinGen gene-disease validity classification is
    quoted because the ClinGen data payload is absent from this checkout. No
    genotype-phenotype correlation is asserted beyond the observation, made
    explicitly by PMID:38365196, that the current data are insufficient to
    establish one; the suggestion in that paper that exon 2 missense alleles
    have earlier onset than exon 6 alleles rests on seven versus twelve patients
    and the authors themselves warn it may be misleading.
  evidence:
  - reference: PMID:30401461
    reference_title: "Bi-allelic ADPRHL2 Mutations Cause Neurodegeneration with Developmental Delay, Ataxia, and Axonal Neuropathy."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "We used exome sequencing to identify five different bi-allelic pathogenic ADPRHL2 variants in 12 individuals from 8 families"
    explanation: Establishes allelic heterogeneity at the locus in the second founding cohort.
  - reference: PMID:34479984
    reference_title: "Biallelic ADPRHL2 mutations in complex neuropathy affect ADP ribosylation and DNA damage response."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "We characterized a novel C26F mutation, demonstrating protein instability and reduced protein function. Characterization of the recurrent V335G mutant demonstrated mild loss of expression with retained enzymatic activity."
    explanation: Documents two mechanistically distinct missense classes at this locus, one destabilising and one that preserves catalysis but mislocalises the enzyme.
  - reference: PMID:35664652
    reference_title: "Child-Onset Cerebellar Ataxia Caused by Two Compound Heterozygous Variants in ADPRS Gene: A Case Report."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "we identified two novel compound heterozygous ADPRS variants: NM_017825.3:c.580C>T (p.Gln194Ter) and NM_017825.3:c.803-1G>A"
    explanation: Documents a compound heterozygous genotype with the reference transcript, in a locus whose published genotypes are overwhelmingly homozygous.
  - reference: PMID:35664652
    reference_title: "Child-Onset Cerebellar Ataxia Caused by Two Compound Heterozygous Variants in ADPRS Gene: A Case Report."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "RNA sequencing indicated that the former mutation might cause nonsense-mediated mRNA decay. The c.803-1G>A variant was found to be a splice-site mutation that leads to the transcriptional retention of intron 5."
    explanation: Transcript-level characterisation showing that both alleles abolish normal product, which is the loss-of-function mechanism this locus requires.
phenotypes:
- name: Ataxia
  category: Neurologic
  diagnostic: true
  description: >-
    Ataxia is one of the two features named in the disorder's title and is the
    commonest reason these children reach a neurologist. It may be episodic at
    first - discrete bouts of unsteadiness after a febrile illness, indistinguishable
    from an episodic ataxia or a post-infectious cerebellitis - and then becomes
    fixed and progressive, with truncal and appendicular components and a
    spastic-ataxic gait as pyramidal signs accumulate.
  phenotype_term:
    preferred_term: Ataxia
    term:
      id: HP:0001251
      label: Ataxia
  notes: >-
    No FrequencyEnum band is asserted, and the reason is a genuine disagreement
    between sources rather than caution for its own sake. HPO annotation for
    OMIM:618170 records HP:0001251 at 11/11 in the Danhauser cohort
    (PMID:30401461, retrieved 2026-08-01), which would be OBLIGATE. The pooled
    47-patient review records episodic or non-episodic ataxia in 17/47 (36.1%),
    which would be FREQUENT, because its denominator is every published patient
    including those for whom the datum could not be extracted. Those two numbers
    are not reconcilable into one band, so none is given.
  evidence:
  - reference: PMID:30401461
    reference_title: "Bi-allelic ADPRHL2 Mutations Cause Neurodegeneration with Developmental Delay, Ataxia, and Axonal Neuropathy."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "key clinical features including developmental delay or regression, seizures, ataxia, and axonal (sensori-)motor neuropathy"
    explanation: Names ataxia as one of the four core clinical features of the disorder.
- name: Episodic ataxia
  category: Neurologic
  diagnostic: true
  description: >-
    The episodic form of the ataxia is the most diagnostically useful single
    observation in the disorder, because it is what makes a clinician think of a
    channelopathy or a metabolic decompensation rather than a degenerative
    disease, and because it is the clinical read-out of the stress-conditional
    biochemistry. Episodes follow infection, fever or other stress, last days to
    weeks, and remit incompletely.
  phenotype_term:
    preferred_term: Episodic ataxia
    term:
      id: HP:0002131
      label: Episodic ataxia
  notes: >-
    Recorded as a separate phenotype from generalised ataxia deliberately: the
    two carry different diagnostic weight and different mechanistic
    implications, and a child may pass from one to the other. The HPO
    annotation set for OMIM:618170 does not include HP:0002131; this phenotype
    is curated from the primary reports.
  evidence:
  - reference: PMID:40493129
    reference_title: "Phenotypic Diversity in Stress-induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures: Novel Associations with Parkinsonism, Icthyosis and Cataract."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "presented with infection-triggered episodic ataxia and ichthyosis"
    explanation: Direct description of infection-triggered episodic ataxia as the presenting phenotype.
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "ADPRHL2 variants should be considered in the context of episodic, stress-induced pediatric and adult-onset movement disorders and seizures."
    explanation: The systematic review's own diagnostic recommendation, which is built on the episodic stress-induced pattern.
- name: Seizures
  category: Neurologic
  diagnostic: true
  description: >-
    Seizures are the second feature named in the disorder's title. Bilateral
    tonic-clonic seizures predominate; absence and multifocal seizures are also
    recorded. Seizures are frequently the presenting event and are frequently
    febrile in the first instance, which contributes to the initial
    misclassification as simple febrile convulsions.
  phenotype_term:
    preferred_term: Bilateral tonic-clonic seizure
    term:
      id: HP:0002069
      label: Bilateral tonic-clonic seizure
  notes: >-
    No frequency band is asserted because the best available source contradicts
    itself: Table 1 of PMID:38365196 reports seizure as a presenting symptom in
    16/47 (34%) but as a clinical symptom in only 11/47 (23.4%), which cannot
    both be complete counts. HPO annotation for OMIM:618170 (retrieved
    2026-08-01) records HP:0002069 bilateral tonic-clonic seizure at 13/16 and
    HP:0002121 absence seizure at 3/16 in the Ghosh cohort, and HP:0001250
    seizure at 6/11 in the Danhauser cohort. The bound term is the specific
    tonic-clonic child rather than the parent HP:0001250, because tonic-clonic
    is the dominant recorded semiology.
  evidence:
  - reference: PMID:30100084
    reference_title: "Biallelic Mutations in ADPRHL2, Encoding ADP-Ribosylhydrolase 3, Lead to a Degenerative Pediatric Stress-Induced Epileptic Ataxia Syndrome."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "developmental regression, and seizures in association with periods of stress, such as infections"
    explanation: Establishes seizures as a core feature and ties their occurrence to stress episodes.
- name: Seizure precipitated by febrile infection
  category: Neurologic
  diagnostic: true
  description: >-
    The specific link between fever and seizure activity is worth recording
    separately from seizures in general, because it is the phenotype most likely
    to be dismissed. A first febrile seizure in a toddler is common and usually
    benign; in this disorder it can be the opening event of a neurodegenerative
    illness, and the discriminator is what happens next - failure to return to
    the previous baseline over the following weeks.
  phenotype_term:
    preferred_term: Seizure precipitated by febrile infection
    term:
      id: HP:0032894
      label: Seizure precipitated by febrile infection
  notes: >-
    HP:0032894 was chosen over the clinical-modifier term HP:0025215 "Triggered
    by febrile illness", which was verified with OAK to sit under HP:0012823
    Clinical modifier rather than under HP:0000118 Phenotypic abnormality and
    therefore cannot be bound as a phenotype term in this schema.
    Case 2 of PMID:38365196 illustrates the pattern: a generalised seizure with
    fever at 2 years 7 months, followed within three weeks by loss of walking,
    speech and swallowing.
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "she had a generalized seizure with fever, lasting for 1 minute, and she gradually lost the ability to walk, speak, and swallow within the following 3 weeks"
    explanation: >-
      Case 2 of this report: a febrile generalised seizure is the presenting
      event, and the loss of walking, speech and swallowing over the following
      three weeks shows that the child did not return to her previous baseline.
      That combination is the claim - seizure precipitated by febrile infection
      as the opening event of a neurodegenerative course - rather than a benign
      febrile seizure.
- name: Developmental regression
  category: Neurologic
  diagnostic: true
  frequency: FREQUENT
  description: >-
    Loss of previously acquired motor, language and swallowing skills, typically
    in the aftermath of a stress episode, is the feature that converts a picture
    of episodic ataxia into a picture of neurodegeneration. It is the single most
    useful clinical discriminator against the benign mimics.
  phenotype_term:
    preferred_term: Developmental regression
    term:
      id: HP:0002376
      label: Developmental regression
  notes: >-
    The FREQUENT band (30-79%) is supported by the pooled figure of 60.4%
    quoted below. The numerator printed alongside that percentage in the
    abstract of PMID:38365196 ("28/43") is inconsistent with both the percentage
    and Table 1 of the same paper (26/43); the percentage is used. HPO
    annotation for OMIM:618170 records HP:0002376 at 10/16 in the Ghosh cohort
    (retrieved 2026-08-01), which is concordant.
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Triggering factors (28/47; 59.5%) and regression (28/43; 60.4%)"
    explanation: Provides the pooled frequency of developmental regression that supports the FREQUENT band.
  - reference: PMID:32746785
    reference_title: "Novel imaging and clinical phenotypes of CONDSIAS disorder caused by a homozygous frameshift variant of ADPRHL2: a case report."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Over time, she developed complete motor and speech regression, bilateral sensorineural hearing loss, infrequent seizures, abdominal distension and gastrointestinal (GI) intolerance, and loss of consciousness."
    explanation: A worked individual example of complete motor and speech regression following the episodic course.
- name: Global developmental delay
  category: Neurologic
  description: >-
    Pre-existing developmental delay, often confined to expressive language, is
    present in a minority before the first episode; global delay becomes
    apparent in most patients as regression accumulates. A normal early
    developmental history is common and must not be used to exclude the
    diagnosis.
  phenotype_term:
    preferred_term: Global developmental delay
    term:
      id: HP:0001263
      label: Global developmental delay
  notes: >-
    HPO annotation for OMIM:618170 records HP:0001263 at 10/12 in the Danhauser
    cohort (retrieved 2026-08-01). No band is asserted here because that
    denominator is one cohort and the pooled review separates pre-existing delay
    (16.2%) from regression (60.4%) in a way that cannot be recombined into a
    single frequency for this term.
  evidence:
  - reference: PMID:30401461
    reference_title: "Bi-allelic ADPRHL2 Mutations Cause Neurodegeneration with Developmental Delay, Ataxia, and Axonal Neuropathy."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "a neurodegenerative disorder manifesting in childhood or adolescence with key clinical features including developmental delay or regression"
    explanation: Names developmental delay, alongside regression, as a core feature of the disorder.
- name: Cerebellar atrophy
  category: Neurologic
  frequency: FREQUENT
  description: >-
    Cerebellar atrophy, often with disproportionate involvement of the vermis, is
    the characteristic imaging finding and correlates with regional
    hypometabolism on FDG PET. It is acquired and progressive, and can be absent
    on an early scan.
  phenotype_term:
    preferred_term: Cerebellar atrophy
    term:
      id: HP:0001272
      label: Cerebellar atrophy
  notes: >-
    The FREQUENT band is supported by the pooled 77.7% (28/36) figure for
    cerebral and cerebellar atrophy with white matter change, quoted below; that
    figure aggregates cerebral and cerebellar atrophy rather than reporting
    cerebellar atrophy alone, which is why the band is set at FREQUENT rather
    than at the VERY_FREQUENT the raw percentage would allow. HPO annotation for
    OMIM:618170 records HP:0001272 at 4/9 and HP:0006855 cerebellar vermis
    atrophy at 2/9 in the Ghosh cohort (retrieved 2026-08-01).
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: PARTIAL
    evidence_source: HUMAN_CLINICAL
    snippet: "cerebral and cerebellar atrophy with white matter changes (28/36; 77.7%)"
    explanation: >-
      Marked PARTIAL because the quoted 77.7% is a combined figure for cerebral
      and cerebellar atrophy with white matter change and does not isolate
      cerebellar atrophy.
  - reference: PMID:37392332
    reference_title: "Expanding the Spectrum of Stress-Induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures (CONDSIAS)."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "revealed cerebellar atrophy, particularly of the vermis, with corresponding hypometabolism"
    explanation: Isolates cerebellar and specifically vermian atrophy with an independent functional correlate.
- name: Cerebral atrophy
  category: Neurologic
  description: >-
    Diffuse cortical atrophy with white matter signal change accompanies the
    cerebellar involvement and progresses on serial imaging, sometimes over as
    little as a year.
  phenotype_term:
    preferred_term: Cerebral atrophy
    term:
      id: HP:0002059
      label: Cerebral atrophy
  notes: >-
    HPO annotation for OMIM:618170 records HP:0002059 for the Ghosh cohort with
    no frequency fraction given (retrieved 2026-08-01), so no band is asserted.
  evidence:
  - reference: PMID:30100084
    reference_title: "Biallelic Mutations in ADPRHL2, Encoding ADP-Ribosylhydrolase 3, Lead to a Degenerative Pediatric Stress-Induced Epileptic Ataxia Syndrome."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "a pediatric-onset neurodegenerative disorder with progressive brain atrophy"
    explanation: Establishes progressive brain atrophy in the founding cohort.
- name: Peripheral axonal neuropathy
  category: Neurologic
  frequency: FREQUENT
  description: >-
    A length-dependent axonal sensorimotor polyneuropathy with distal weakness,
    foot drop, absent reflexes in some and pes cavus. In part of the spectrum it
    is the dominant feature, to the extent that ADPRHL2 homozygotes have been
    ascertained through hereditary motor neuropathy cohorts.
  phenotype_term:
    preferred_term: Peripheral axonal neuropathy
    term:
      id: HP:0003477
      label: Peripheral axonal neuropathy
  notes: >-
    The FREQUENT band is supported by the pooled 39.1% (9/23) figure, whose
    denominator is patients who actually underwent electrophysiology rather than
    all published patients. HPO annotation for OMIM:618170 records HP:0003447
    axonal loss for the Danhauser cohort without a fraction (retrieved
    2026-08-01).
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "regression (28/43; 60.4%), axonal polyneuropathy (9/23; 39.1%)"
    explanation: Provides the pooled frequency of axonal polyneuropathy among tested patients that supports the FREQUENT band.
  - reference: PMID:30401461
    reference_title: "Bi-allelic ADPRHL2 Mutations Cause Neurodegeneration with Developmental Delay, Ataxia, and Axonal Neuropathy."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "key clinical features including developmental delay or regression, seizures, ataxia, and axonal (sensori-)motor neuropathy"
    explanation: Names the neuropathy as axonal and sensorimotor in the founding cohort that first emphasised it.
- name: Muscle weakness
  category: Musculature
  description: >-
    Weakness is the commonest clinical symptom in the pooled series, and is
    mixed in origin: distal weakness from the axonal neuropathy, proximal
    weakness and quadriparesis from central involvement, and in at least one
    patient a myogenic component on biopsy.
  phenotype_term:
    preferred_term: Muscle weakness
    term:
      id: HP:0001324
      label: Muscle weakness
  notes: >-
    Table 1 of PMID:38365196 records weakness in 27/47 (57.4%), which would be
    FREQUENT, but that figure is in PDF-extracted full text rather than the
    cached abstract and is not quotable here, so no band is asserted. HPO
    annotation for OMIM:618170 records HP:0001324 for the Ghosh cohort without a
    fraction (retrieved 2026-08-01).
  evidence:
  - reference: PMID:39580621
    reference_title: "An ADPRS variant disrupts ARH3 stability and subcellular localization in children with neurodegeneration and respiratory failure."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "two siblings with developmental delay, seizures, progressive muscle weakness, and respiratory failure following an episodic course"
    explanation: Names progressive muscle weakness as a core feature in an independently reported sibling pair.
- name: Dystonia
  category: Neurologic
  description: >-
    Episodic dystonic posturing, including recurrent torticollis and truncal
    dystonic posturing, is a characteristic and under-recognised presentation,
    especially in the youngest patients. Because the attacks are brief, painless
    and self-limiting, they are frequently attributed to benign paroxysmal
    torticollis or to behaviour before the degenerative course declares itself.
    Levodopa has been tried and has not helped.
  phenotype_term:
    preferred_term: Dystonia
    term:
      id: HP:0001332
      label: Dystonia
  notes: >-
    Table 1 of PMID:38365196 records dystonic posture as a clinical symptom in
    5/47 (10.6%) and episodic ataxic-dystonic posture as the presenting symptom
    in a further 5/47, with episodic torticollis and dystonic torticollis in one
    patient each; these are PDF-extracted table figures and are not quoted as
    snippets. No band is asserted, because episodic dystonia is systematically
    under-ascertained in retrospective series.
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "We present the diagnostic pathway of two pediatric patients with episodic dystonia and ataxia"
    explanation: Documents episodic dystonia as a presenting phenotype in two independently ascertained patients.
  - reference: PMID:37392332
    reference_title: "Expanding the Spectrum of Stress-Induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures (CONDSIAS)."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "onset of symptoms occurred at 5 years of age and consisted of episodes of truncal dystonic posturing"
    explanation: Independent report of episodic truncal dystonic posturing as the first manifestation.
- name: Respiratory insufficiency
  category: Respiratory
  diagnostic: true
  description: >-
    Respiratory failure is the commonest cause of death. It is often central
    hypoventilation rather than neuromuscular failure, presenting as
    hypoventilation in sleep, apnoea or unexplained bradycardia and desaturation
    during an intercurrent illness, and requiring tracheostomy and long-term
    ventilation.
  phenotype_term:
    preferred_term: Central hypoventilation
    term:
      id: HP:0007110
      label: Central hypoventilation
  notes: >-
    The bound term is the specific HP:0007110 Central hypoventilation rather
    than the generic HP:0002093 Respiratory insufficiency, because the published
    descriptions consistently specify a central mechanism, including
    hypoventilation during sleep in the absence of infection. HPO annotation for
    OMIM:618170 records the generic HP:0002093 at 6/12 in the Danhauser cohort,
    and Table 1 of PMID:38365196 records respiratory insufficiency in 14/47
    (29.7%) (retrieved/read 2026-08-01). No band is asserted because those two
    denominators disagree by a factor of two.
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "a neurodegenerative course complicated by central hypoventilation syndrome"
    explanation: Names central hypoventilation specifically, which is the basis for binding the central rather than the generic term.
  - reference: PMID:39580621
    reference_title: "An ADPRS variant disrupts ARH3 stability and subcellular localization in children with neurodegeneration and respiratory failure."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "progressive muscle weakness, and respiratory failure following an episodic course"
    explanation: Independent confirmation of respiratory failure as a defining late feature.
- name: Sensorineural hearing impairment
  category: Ear
  description: >-
    Progressive bilateral sensorineural hearing loss occurs in a minority and
    compounds the communication difficulty already imposed by dysarthria and
    language regression. It is relevant to management because it is treatable
    with amplification while nothing else in the disorder is.
  phenotype_term:
    preferred_term: Sensorineural hearing impairment
    term:
      id: HP:0000407
      label: Sensorineural hearing impairment
  notes: >-
    HPO annotation for OMIM:618170 records HP:0000407 at 3/25 across both
    founding cohorts (retrieved 2026-08-01); Table 1 of PMID:38365196 records
    6/47 (12.7%). Both are in the OCCASIONAL range but the sources are not
    independent, so no band is asserted.
  evidence:
  - reference: PMID:32746785
    reference_title: "Novel imaging and clinical phenotypes of CONDSIAS disorder caused by a homozygous frameshift variant of ADPRHL2: a case report."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "she developed complete motor and speech regression, bilateral sensorineural hearing loss"
    explanation: Documents bilateral sensorineural hearing loss developing as part of the degenerative course.
  - reference: PMID:37392332
    reference_title: "Expanding the Spectrum of Stress-Induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures (CONDSIAS)."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Progressive hearing loss, urinary urgency, and thoracic kyphoscoliosis ensued."
    explanation: Independent report of progressive hearing loss in an adult patient.
- name: Abnormality of the autonomic nervous system
  category: Neurologic
  description: >-
    Autonomic involvement is under-recognised and clinically important: reported
    features include sinus arrhythmia, postural orthostatic tachycardia,
    hypertension, impaired sweating, hyperthermia, abdominal distension,
    gastrointestinal dysmotility and polyuria. Together with central
    hypoventilation it makes the terminal phase of this disorder a brainstem and
    autonomic failure rather than a purely cerebellar one, and autonomic cardiac
    instability is implicated in at least one reported death.
  phenotype_term:
    preferred_term: Abnormality of the autonomic nervous system
    term:
      id: HP:0002270
      label: Abnormality of the autonomic nervous system
  notes: >-
    Table 1 of PMID:38365196 records autonomic dysfunction in 5/47 (10.6%) and
    cardiac involvement in 7/47 (14.8%); these are PDF-extracted table figures
    and are not quoted. No band is asserted, and the true frequency is likely
    higher than either figure because autonomic function is rarely tested
    systematically in these children.
  evidence:
  - reference: PMID:35222245
    reference_title: "Case Report: Stress-Induced Childhood-Onset Neurodegeneration With Ataxia-Seizures Syndrome Caused by a Novel Compound Heterozygous Mutation in ADPRHL2."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "This patient is presented with autonomic nervous dysfunction manifested as polyuria, gastrointestinal disturbance, and sinus arrhythmia, which may be considered as new clinical manifestations in addition to the above classical manifestations."
    explanation: The report that first drew attention to autonomic dysfunction as part of this phenotype.
- name: Psychosis
  category: Neurologic
  description: >-
    Episodic psychosis with hallucinations and delusions, remitting between
    episodes, is the defining feature of the adult-onset pole of the spectrum
    (described as PAMP syndrome). Its presence means that an adult presenting to
    a psychiatric service after a viral illness, with ataxia and a motor
    neuropathy, can have this disorder, and that a normal childhood history does
    not exclude it.
  phenotype_term:
    preferred_term: Psychosis
    term:
      id: HP:0000709
      label: Psychosis
  notes: >-
    Table 1 of PMID:38365196 records psychiatric problems (psychosis, delusion,
    hallucination, depression) as the presenting symptom in 5/47 (10.6%). No
    band is asserted.
  evidence:
  - reference: PMID:33528672
    reference_title: "Episodic psychosis, ataxia, motor neuropathy with pyramidal signs (PAMP syndrome) caused by a novel mutation in ADPRHL2 (AHR3)."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Four patients with episodic psychosis, ataxia, and motor neuropathy with pyramidal signs were included in this study."
    explanation: Defines the adult-onset psychosis-predominant presentation in a four-patient series.
- name: Spasticity
  category: Neurologic
  description: >-
    Pyramidal signs - brisk reflexes, extensor plantar responses, clonus and leg
    spasticity - accumulate as the disease progresses and produce the
    characteristic spastic-ataxic gait. Spinal cord involvement, including
    atrophy and T2 hyperintensity, is reported in a substantial fraction of
    those imaged and is the likely substrate.
  phenotype_term:
    preferred_term: Spasticity
    term:
      id: HP:0001257
      label: Spasticity
  notes: >-
    HPO annotation for OMIM:618170 records HP:0003487 Babinski sign at 1/12 in
    the Danhauser cohort (retrieved 2026-08-01), which understates the pyramidal
    involvement described in later reports. Spinal cord involvement was present
    in 7/12 (58.3%) of the pooled patients in whom the cord was imaged
    (Table 1 of PMID:38365196, PDF-extracted, not quoted). No band is asserted.
  evidence:
  - reference: PMID:37392332
    reference_title: "Expanding the Spectrum of Stress-Induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures (CONDSIAS)."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "muscle weakness and atrophy of hands and feet, leg spasticity with clonus, truncal and appendicular ataxia, and spastic-ataxic gait"
    explanation: Documents the combined pyramidal and cerebellar signs that produce the characteristic gait.
- name: Parkinsonism
  category: Neurologic
  description: >-
    Parkinsonism with truncal dystonia has been described in the second decade in
    a patient whose childhood presentation was language delay, intellectual
    disability and infrequent seizures. It is a recent addition to the phenotype
    and rests on a single report.
  phenotype_term:
    preferred_term: Parkinsonism
    term:
      id: HP:0001300
      label: Parkinsonism
  notes: >-
    Single-report phenotype; no frequency asserted and none should be inferred.
    Included because an extrapyramidal presentation would otherwise take a
    clinician away from this diagnosis rather than towards it.
  evidence:
  - reference: PMID:40493129
    reference_title: "Phenotypic Diversity in Stress-induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures: Novel Associations with Parkinsonism, Icthyosis and Cataract."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "She later developed parkinsonism, truncal dystonia, ataxia, peripheral neuropathy, and neuropsychiatric symptoms in her second decade."
    explanation: The single published description of parkinsonism in this disorder.
- name: Ichthyosis
  category: Dermatologic
  description: >-
    Ichthyosis has been reported in two patients from one recent series and is
    the only extra-neurological feature that suggests the disorder might not be
    purely a nervous-system disease. It is unexplained; whether it reflects a
    role for ARH3 in the high-turnover epidermis or is coincidental in a
    consanguineous population is unknown.
  phenotype_term:
    preferred_term: Ichthyosis
    term:
      id: HP:0008064
      label: Ichthyosis
  notes: >-
    Reported in 2 of 3 patients in one series (PMID:40493129) and, to this
    curation's knowledge, nowhere else. In a consanguineous population a second
    recessive condition segregating independently is a real alternative
    explanation and has not been excluded. No frequency is asserted.
  evidence:
  - reference: PMID:40493129
    reference_title: "Phenotypic Diversity in Stress-induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures: Novel Associations with Parkinsonism, Icthyosis and Cataract."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Extra-neurological manifestations included ichthyosis in patients 1 and 2 and cataracts in patient 1."
    explanation: The only published report of ichthyosis and cataract in this disorder.
- name: Cataract
  category: Ophthalmologic
  description: >-
    Cataract has been reported in a single patient, in the same series that
    described ichthyosis and parkinsonism. As with ichthyosis, coincidence in a
    consanguineous family has not been excluded.
  phenotype_term:
    preferred_term: Cataract
    term:
      id: HP:0000518
      label: Cataract
  notes: >-
    Single-patient observation; recorded for completeness of the published
    phenotype and explicitly not asserted as an established feature.
  evidence:
  - reference: PMID:40493129
    reference_title: "Phenotypic Diversity in Stress-induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures: Novel Associations with Parkinsonism, Icthyosis and Cataract."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Extra-neurological manifestations included ichthyosis in patients 1 and 2 and cataracts in patient 1."
    explanation: The single published observation of cataract in this disorder.
- name: Nystagmus
  category: Ophthalmologic
  description: >-
    Eye movement abnormalities are common and varied - nystagmus, strabismus,
    ptosis, restricted abduction, upward gaze palsy, oculomotor apraxia and
    diplopia. Oculomotor apraxia in particular puts this disorder into the same
    clinical bracket as the DNA-repair ataxias, which is where it also belongs
    mechanistically.
  phenotype_term:
    preferred_term: Nystagmus
    term:
      id: HP:0000639
      label: Nystagmus
  notes: >-
    HPO annotation for OMIM:618170 records HP:0000639 at 3/11, HP:0000486
    strabismus at 2/11 and HP:0000508 ptosis, HP:0000602 ophthalmoplegia and
    HP:0000651 diplopia at 1/11 each in the Danhauser cohort (retrieved
    2026-08-01). Table 1 of PMID:38365196 aggregates all ocular findings at
    15/47 (31.9%). No band is asserted for the specific term nystagmus.
  evidence:
  - reference: PMID:40493129
    reference_title: "Phenotypic Diversity in Stress-induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures: Novel Associations with Parkinsonism, Icthyosis and Cataract."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Patient 3 is a 6-year-old girl who presented with progressive ataxia, myoclonus, oculomotor apraxia, and upward gaze palsy."
    explanation: Documents oculomotor apraxia and gaze palsy, the eye-movement findings that link this disorder clinically to the DNA-repair ataxias.
- name: Myoclonus
  category: Neurologic
  description: >-
    Perioral, facial and limb myoclonus and orofacial dyskinesia are recurrent
    findings and can dominate the picture in the later stages, sometimes
    prompting a search for an autoimmune encephalitis.
  phenotype_term:
    preferred_term: Myoclonus
    term:
      id: HP:0001336
      label: Myoclonus
  notes: >-
    Table 1 of PMID:38365196 records perioral or facial myoclonus in 3/47 (6.3%)
    and orofacial dyskinesia in 3/47 (6.3%); PDF-extracted, not quoted. No band
    is asserted.
  evidence:
  - reference: PMID:40493129
    reference_title: "Phenotypic Diversity in Stress-induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures: Novel Associations with Parkinsonism, Icthyosis and Cataract."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "progressive ataxia, myoclonus, oculomotor apraxia, and upward gaze palsy"
    explanation: Documents myoclonus as part of the movement-disorder phenotype.
  - reference: PMID:37392332
    reference_title: "Expanding the Spectrum of Stress-Induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures (CONDSIAS)."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Present neurological examination revealed dysarthria, facial mini-myoclonus"
    explanation: Independent documentation of facial myoclonus.
- name: Scoliosis
  category: Skeletal
  description: >-
    Kyphoscoliosis and pes cavus are secondary orthopaedic consequences of the
    combined neuropathy, weakness and spasticity, and are a practical management
    burden in longer-surviving patients.
  phenotype_term:
    preferred_term: Scoliosis
    term:
      id: HP:0002650
      label: Scoliosis
  notes: >-
    Table 1 of PMID:38365196 records scoliosis in 13/47 (27.6%) and pes cavus in
    10/47 (21.2%); PDF-extracted, not quoted. HPO annotation for OMIM:618170
    records HP:0001761 pes cavus at 1/16 in the Ghosh cohort (retrieved
    2026-08-01). No band is asserted.
  evidence:
  - reference: PMID:37392332
    reference_title: "Expanding the Spectrum of Stress-Induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures (CONDSIAS)."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "urinary urgency, and thoracic kyphoscoliosis ensued"
    explanation: Documents kyphoscoliosis developing as a secondary complication.
diagnosis:
- name: Exome or genome sequencing with bi-allelic ADPRS variant interpretation
  description: >-
    Molecular diagnosis rests on identifying bi-allelic ADPRS variants against
    transcript NM_017825.2/NM_017825.3, and there are three interpretation
    points that follow from the published allelic spectrum. First, a
    conservative-looking missense change must not be dismissed: several of the
    best-characterised alleles are missense and abolish detectable protein.
    Second, the gene appears in older variant-call pipelines and databases under
    ADPRHL2, so a filter keyed on the current symbol alone can miss an existing
    annotation. Third, parental testing is required to establish that two
    variants are in trans, since heterozygotes are unaffected; and the historical
    predominance of homozygous genotypes reflects consanguineous ascertainment
    and must not be used as an argument against a compound heterozygous call.
  evidence:
  - reference: PMID:35664652
    reference_title: "Child-Onset Cerebellar Ataxia Caused by Two Compound Heterozygous Variants in ADPRS Gene: A Case Report."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "we identified two novel compound heterozygous ADPRS variants: NM_017825.3:c.580C>T (p.Gln194Ter) and NM_017825.3:c.803-1G>A"
    explanation: Names the reference transcript a laboratory must use and documents that compound heterozygous genotypes occur.
  - reference: PMID:33528672
    reference_title: "Episodic psychosis, ataxia, motor neuropathy with pyramidal signs (PAMP syndrome) caused by a novel mutation in ADPRHL2 (AHR3)."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "Western blot analyses demonstrated that ADPRHL2 was not expressed in these patients."
    explanation: Supports the interpretive caution that a missense allele at this locus can behave as a complete null.
- name: Serial brain and spinal cord MRI
  description: >-
    Imaging serves two purposes and has one important pitfall. It documents the
    characteristic cerebral and cerebellar atrophy with white matter change and
    the spinal cord involvement, and repeated over time it demonstrates
    progression, which is what distinguishes this disorder from a static or
    post-infectious process. The pitfall is that in the pooled review 7 of the
    28 patients who ever had an abnormal MRI (25%) had a normal first scan and
    became abnormal only on follow-up, so a normal MRI during or shortly after
    a first episode does not exclude the diagnosis and should be repeated.
  notes: >-
    The normal-first-scan figure is not in the abstract and so is not carried by
    the snippet below. It comes from Table 1 of PMID:38365196, MRI findings
    block, row "Normal at first imaging study but abnormal during follow-up",
    which gives 7/28 (25) beneath "Abnormal 28/36 (77.7)". The running text of
    the same paper renders the numerator as n = 8 for the same 25%; 25% of 28 is
    7, so the table figure is used. That row is PDF-extracted text carrying a
    ligature in "first", which the reference validator does not fold, so it
    cannot be quoted as a snippet.
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "cerebral and cerebellar atrophy with white matter changes (28/36; 77.7%) were the other clues"
    explanation: Establishes the imaging findings and their frequency among imaged patients.
- name: Nerve conduction studies and electromyography
  description: >-
    Electrophysiology looks for the axonal sensorimotor polyneuropathy, which is
    one of the most useful discriminators against the non-DNA-repair cerebellar
    ataxias and against the channelopathies. It was abnormal in the majority of
    the pooled patients tested, most often showing axonal polyneuropathy with or
    without sensory involvement.
  evidence:
  - reference: PMID:38365196
    reference_title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
    supports: SUPPORT
    evidence_source: HUMAN_CLINICAL
    snippet: "regression (28/43; 60.4%), axonal polyneuropathy (9/23; 39.1%)"
    explanation: Quantifies the yield of electrophysiology among patients tested.
- name: Polysomnography and respiratory assessment
  description: >-
    Because central hypoventilation is the commonest route to death and appears
    during sleep before it appears when awake, sleep-time respiratory monitoring
    is a diagnostic and surveillance test rather than a terminal-care one.
    Detecting nocturnal hypoventilation before an acute arrest is the single
    change in practice most likely to alter outcome in this disorder.
  notes: >-
    No evidence item is attached. No published guideline, surveillance protocol
    or study of respiratory monitoring in this disorder exists; the
    recommendation is a clinical inference from the documented natural history
    (central hypoventilation as a cause of death, and sleep-onset
    hypoventilation without infection in case 2 of PMID:38365196). Attaching a
    snippet would misrepresent an inference as a published recommendation.
differential_diagnoses:
- name: Ataxia telangiectasia
  disease_term:
    preferred_term: ataxia telangiectasia
    term:
      id: MONDO:0008840
      label: ataxia telangiectasia
  description: >-
    The prototype DNA-damage-response ataxia and the diagnosis most likely to be
    considered first in a young child with progressive ataxia and oculomotor
    abnormalities. It is both a clinical mimic and a mechanistic neighbour: ATM
    is the apex kinase of the double-strand-break response, whereas ADPRS acts on
    the ADP-ribosylation arm of the single-strand-break response, so the two
    disorders illustrate that lesions at different points of DNA-damage signalling
    converge on cerebellar degeneration.
  distinguishing_features:
  - Oculocutaneous telangiectasia, elevated alpha-fetoprotein, immunodeficiency and lymphoid malignancy risk are characteristic of ataxia telangiectasia and are not reported in ADPRS deficiency.
  - The ADPRS course is punctuated by discrete stress-triggered episodes with incomplete recovery, whereas ataxia telangiectasia progresses steadily without febrile precipitants.
  - Seizures are a core feature of ADPRS deficiency and are not typical of ataxia telangiectasia.
  - Central hypoventilation as a terminal event is characteristic of ADPRS deficiency; respiratory disease in ataxia telangiectasia is infective and aspiration-related.
  notes: >-
    MONDO:0008840 was verified with OAK, which returns `RO:0004003 HGNC:795 !
    ATM` and `xref: OMIM:208900`, and it matches the disease_term already bound
    in kb/disorders/Ataxia_Telangiectasia.yaml. No content was copied from that
    entry. No evidence item is attached because the contrast is a reasoned
    clinical comparison rather than a quotable finding.
- name: Ataxia with oculomotor apraxia type 1
  disease_term:
    preferred_term: ataxia, early-onset, with oculomotor apraxia and hypoalbuminemia
    term:
      id: MONDO:0008842
      label: ataxia, early-onset, with oculomotor apraxia and hypoalbuminemia
  description: >-
    Aprataxin deficiency is the closest mechanistic differential in the whole
    list. Aprataxin resolves abortive DNA ligation intermediates at
    single-strand breaks, the same repair pathway whose ADP-ribosylation
    signalling ARH3 terminates, and it produces childhood-onset cerebellar
    ataxia with oculomotor apraxia and axonal sensorimotor neuropathy - the same
    triad that ADPRS deficiency can produce.
  distinguishing_features:
  - Hypoalbuminaemia and hypercholesterolaemia are characteristic biochemical markers of aprataxin deficiency and are not features of ADPRS deficiency.
  - Aprataxin deficiency progresses steadily without stress-triggered episodic deterioration, which is the ADPRS signature.
  - Seizures, developmental regression and central hypoventilation are features of ADPRS deficiency and not of aprataxin deficiency.
  notes: >-
    MONDO:0008842 was verified with OAK, which returns `RO:0004003 HGNC:15984 !
    APTX` and `xref: OMIM:208920`. No evidence item is attached; the contrast is
    reasoned rather than quoted.
- name: Ataxia with oculomotor apraxia type 4
  disease_term:
    preferred_term: ataxia - oculomotor apraxia type 4
    term:
      id: MONDO:0014557
      label: ataxia - oculomotor apraxia type 4
  description: >-
    PNKP deficiency, another single-strand-break repair ataxia, with
    childhood-onset ataxia, oculomotor apraxia, axonal neuropathy and, unlike
    aprataxin deficiency, prominent dystonia and seizures. Of all the DNA-repair
    ataxias it is the one whose clinical picture overlaps ADPRS deficiency most
    closely, because it too combines a movement disorder with epilepsy.
  distinguishing_features:
  - PNKP disease produces a continuous rather than a stress-triggered episodic course; the febrile precipitant is the ADPRS discriminator.
  - Microcephaly is characteristic of the severe end of the PNKP spectrum (the allelic MCSZ phenotype) and is a minority finding in ADPRS deficiency.
  - Elevated alpha-fetoprotein and hypoalbuminaemia are variably reported in PNKP disease and are not features of ADPRS deficiency.
  notes: >-
    MONDO:0014557 was verified with OAK, which returns `RO:0004003 HGNC:9154 !
    PNKP` and `xref: OMIM:616267`. The allelic PNKP phenotype MCSZ
    (MONDO:0013254, microcephaly, seizures, and developmental delay) was also
    verified but is not curated as a separate differential. No evidence item is
    attached.
- name: Spinocerebellar ataxia, autosomal recessive, with axonal neuropathy 1
  disease_term:
    preferred_term: spinocerebellar ataxia, autosomal recessive, with axonal neuropathy 1
    term:
      id: MONDO:0011801
      label: spinocerebellar ataxia, autosomal recessive, with axonal neuropathy 1
  description: >-
    SCAN1, caused by TDP1 deficiency, completes the single-strand-break repair
    ataxia group. TDP1 removes trapped topoisomerase 1 adducts from DNA ends, the
    lesion immediately upstream of the repair step whose ADP-ribosylation
    signalling ARH3 terminates. It presents as cerebellar ataxia with a severe
    axonal sensorimotor neuropathy and distal wasting.
  distinguishing_features:
  - SCAN1 is not epileptic and has no stress-triggered episodic component.
  - Hypoalbuminaemia and hypercholesterolaemia occur in SCAN1 as in aprataxin deficiency, and are absent in ADPRS deficiency.
  notes: >-
    MONDO:0011801 was verified with OAK, which returns `RO:0004003 HGNC:18884 !
    TDP1` and `xref: OMIM:607250`. No evidence item is attached; this
    differential is included for mechanistic completeness of the
    single-strand-break repair group rather than because it is a likely clinical
    confusion.
- name: Biotin-responsive basal ganglia disease
  disease_term:
    preferred_term: biotin-responsive basal ganglia disease
    term:
      id: MONDO:0011841
      label: biotin-responsive basal ganglia disease
  description: >-
    The single most important differential to get right, because it is
    treatable. SLC19A3 deficiency presents in exactly the same way as this
    disorder at the bedside - a young child who becomes acutely encephalopathic,
    dystonic and ataxic during a febrile illness, with partial recovery and
    stepwise decline - and it responds dramatically to high-dose biotin and
    thiamine. Any child with febrile-illness-triggered encephalopathy with
    dystonia should be started on biotin and thiamine while the genetic result
    is awaited.
  distinguishing_features:
  - MRI in SLC19A3 deficiency shows bilateral symmetric caudate and putaminal signal change with swelling; ADPRS deficiency shows cerebral and cerebellar atrophy with white matter change and no basal ganglia predilection.
  - SLC19A3 deficiency responds to high-dose biotin and thiamine, sometimes within days; no such response is established for ADPRS deficiency.
  - Axonal sensorimotor polyneuropathy and central hypoventilation are features of ADPRS deficiency and not of SLC19A3 deficiency.
  notes: >-
    MONDO:0011841 was verified with OAK, which returns `RO:0004003 HGNC:16266 !
    SLC19A3` and `xref: OMIM:607483`. This differential is also relevant to the
    treatment section: PMID:40493129 reports two ADPRS patients who "responded
    favourably to treatment with high-dose vitamin supplementation", an
    uncontrolled observation whose interpretation is complicated by exactly this
    overlap. No evidence item is attached to the differential itself.
- name: Episodic ataxia type 2
  disease_term:
    preferred_term: episodic ataxia type 2
    term:
      id: MONDO:0007163
      label: episodic ataxia type 2
  description: >-
    The commonest clinical misdiagnosis at first presentation. CACNA1A-related
    episodic ataxia produces exactly the picture of a child with discrete attacks
    of ataxia precipitated by stress, exertion or intercurrent illness, and it is
    what the paediatric neurologist will think of first.
  distinguishing_features:
  - Episodic ataxia type 2 is autosomal dominant with an affected parent in most families; ADPRS deficiency is recessive and typically arises in a consanguineous family with unaffected parents.
  - Episodic ataxia type 2 responds to acetazolamide and 4-aminopyridine; ADPRS deficiency does not.
  - Between attacks, episodic ataxia type 2 patients recover to baseline apart from slowly progressive interictal ataxia; in ADPRS deficiency each episode leaves a lower baseline, with regression, seizures and neuropathy accumulating.
  - Axonal neuropathy, developmental regression and central hypoventilation do not occur in episodic ataxia type 2.
  notes: >-
    MONDO:0007163 was verified with OAK; the local adapter records no
    RO:0004003 gene relationship for this term, but its MONDO synonym list
    includes "CACNA1A hereditary episodic ataxia" and hgnc:1388 was
    independently confirmed as CACNA1A. Both facts were checked rather than
    recalled. No evidence item is attached.
- name: Encephalopathy due to GLUT1 deficiency
  disease_term:
    preferred_term: encephalopathy due to GLUT1 deficiency
    term:
      id: MONDO:0011724
      label: encephalopathy due to GLUT1 deficiency
  description: >-
    Another treatable mimic that is explicitly listed in the differential
    considered by clinicians who later diagnosed ADPRS deficiency. GLUT1
    deficiency produces early-onset epilepsy with paroxysmal exercise-induced
    dyskinesia and ataxia, with symptoms characteristically worse before meals
    and after exertion, and responds to a ketogenic diet.
  distinguishing_features:
  - A low cerebrospinal fluid glucose with a low CSF-to-blood glucose ratio is diagnostic of GLUT1 deficiency and is normal in ADPRS deficiency.
  - GLUT1 deficiency symptoms are precipitated by fasting and exertion and improve after eating; ADPRS episodes are precipitated by febrile illness and infection.
  - GLUT1 deficiency responds to a ketogenic diet; no dietary therapy is established for ADPRS deficiency.
  - Progressive cerebral and cerebellar atrophy and axonal neuropathy are features of ADPRS deficiency and not of GLUT1 deficiency.
  notes: >-
    MONDO:0011724 was verified with OAK, which returns `RO:0004003 HGNC:11005 !
    SLC2A1` and `xref: OMIM:606777`, matching the disease_term bound in
    kb/disorders/GLUT1_Deficiency_Syndrome.yaml. GLUT1 deficiency is named
    explicitly in the differential list considered for case 1 of PMID:38365196
    before the ADPRHL2 diagnosis was made. No evidence item is attached.
treatments:
- name: Multidisciplinary supportive care
  description: >-
    No disease-modifying therapy exists, and management is entirely supportive:
    developmental and educational support, physiotherapy for ataxia and
    spasticity, orthopaedic management of scoliosis and foot deformity,
    amplification for hearing loss, nutritional support and gastrostomy for
    dysphagia, and palliative-care involvement. The disorder's specific
    contribution to a supportive-care plan is anticipatory: the trajectory
    towards respiratory failure is predictable enough that ventilation decisions
    should be discussed before the crisis rather than during it.
  treatment_term:
    preferred_term: supportive care
    term:
      id: NCIT:C15747
      label: Supportive Care
  therapeutic_modality: BEHAVIORAL
  notes: >-
    Carried as a descriptor with no evidence item. No management guideline,
    natural-history study or treatment trial exists for this disorder, and the
    supportive measures listed are the standard-of-care response to the curated
    phenotypes rather than anything published for ADPRS deficiency. Attaching a
    snippet would be a claim/snippet mismatch.
- name: Ventilatory support and tracheostomy
  description: >-
    Long-term mechanical ventilation, usually via tracheostomy, is the
    intervention that most changes the course of this disorder, because central
    hypoventilation rather than global brain failure is what usually kills these
    children. Reported patients have regained consciousness and the ability to
    communicate after ventilatory support reversed carbon dioxide retention, and
    one child who required tracheostomy at 2 years was sitting with truncal
    ataxia and walking with support a year later. That is not a cure, but it
    means the decision to ventilate in this disorder is not futile in the way it
    might be assumed to be.
  treatment_term:
    preferred_term: mechanical ventilation
    term:
      id: NCIT:C70909
      label: Mechanical Ventilation
  therapeutic_modality: DEVICE
  notes: >-
    The clinical observations behind this recommendation - regained
    consciousness after ventilation, and the one-year follow-up of case 2 - are
    in the PDF-extracted full text of PMID:38365196, which carries ligature
    artefacts, so they are summarised here rather than quoted. The abstract's
    statement that the neurodegenerative course was "complicated by central
    hypoventilation syndrome" is quoted on the respiratory phenotype instead.
- name: Antiseizure medication
  description: >-
    Standard antiseizure medication for the tonic-clonic, absence and multifocal
    seizures. Levetiracetam and gabapentin have been used in reported patients,
    the latter for dyskinesia and spasticity as much as for seizures. There is no
    evidence that any particular agent is preferred in this disorder, and no
    evidence that seizure control alters the degenerative course.
  treatment_term:
    preferred_term: anticonvulsant therapy
    term:
      id: NCIT:C64172
      label: Anticonvulsant Therapy
  therapeutic_modality: SMALL_MOLECULE
  notes: >-
    No evidence item is attached. The specific agents named are drawn from
    individual case narratives in the PDF-extracted full text of PMID:38365196
    rather than from any comparative study, and no publication addresses
    antiseizure drug choice in this disorder.
- name: PARP inhibition (investigational)
  description: >-
    The only mechanistically rational disease-modifying strategy, and the only
    one with convergent preclinical support. If the disease is caused by an
    inability to clear PARP1-generated ADP-ribose, then reducing PARP1 output
    should rescue it, and it does - in a Drosophila oxidative-challenge model, in
    patient-derived fibroblasts, and in Arh3-knockout mice subjected to cerebral
    ischaemia-reperfusion. Prolonged PARP inhibition also clears the chromatin
    scars and restores histone acetylation in patient cells, so it addresses both
    proposed mechanisms. What is entirely absent is any human trial: no PARP
    inhibitor has been given to a patient with this disorder in a controlled
    setting, the licensed agents are oncology drugs with myelosuppressive
    toxicity and uncertain brain penetration, and the dose, timing (chronic
    prophylaxis versus treatment of an acute episode) and safety in a developing
    child are all unknown.
  treatment_term:
    preferred_term: Pharmacotherapy
    term:
      id: NCIT:C15986
      label: Pharmacotherapy
    therapeutic_agent:
    - preferred_term: poly(ADP-ribose) polymerase inhibitor
      term:
        id: NCIT:C62554
        label: Poly (ADP-Ribose) Polymerase Inhibitor
  therapeutic_modality: SMALL_MOLECULE
  target_mechanisms:
  - target: Failure of Poly(ADP-Ribose) Turnover under Stress
    treatment_effect: INHIBITS
    description: >-
      PARP1 inhibition reduces the rate of ADP-ribose deposition, which
      compensates for the missing ARH3 removal step and keeps PAR below the
      threshold for parthanatos.
  evidence:
  - reference: PMID:30100084
    reference_title: "Biallelic Mutations in ADPRHL2, Encoding ADP-Ribosylhydrolase 3, Lead to a Degenerative Pediatric Stress-Induced Epileptic Ataxia Syndrome."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Pharmacological inhibition of PARP also rescued the phenotype, suggesting the possibility of postnatal treatment for this genetic condition."
    explanation: The founding paper's own therapeutic proposal, based on rescue in the Drosophila model.
  - reference: PMID:30401461
    reference_title: "Bi-allelic ADPRHL2 Mutations Cause Neurodegeneration with Developmental Delay, Ataxia, and Axonal Neuropathy."
    supports: SUPPORT
    evidence_source: IN_VITRO
    snippet: "it was rescued by expression of wild-type ADPRHL2 mRNA as well as treatment with a PARP1 inhibitor"
    explanation: Independent rescue of the stress-sensitivity phenotype in patient-derived fibroblasts by a PARP1 inhibitor.
  - reference: PMID:30830864
    reference_title: "PARP1 inhibition alleviates injury in ARH3-deficient mice and human cells."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "PARP1 inhibitors may attenuate the progression of neurodegeneration in affected patients with ARH3 deficiency."
    explanation: In vivo mouse evidence extended to an explicit therapeutic proposal for patients.
  - reference: PMID:36497109
    reference_title: "ARH Family of ADP-Ribose-Acceptor Hydrolases."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: "Arh3-KO mice developed increased brain infarction following ischemia-reperfusion injury, which was reduced by PARP inhibitors."
    explanation: Summarises the in vivo brain-injury rescue that is the strongest organ-level evidence for this strategy. Classified OTHER because this is a review.
  notes: >-
    Deliberately not framed as an available treatment. No clinical trial of a
    PARP inhibitor in this disorder is registered to this curation's knowledge,
    and every supporting result is preclinical.
- name: Minocycline (off-label; rationale subsequently refuted)
  description: >-
    Minocycline was given off-label to at least one adult patient on the
    rationale that it acts as a PARP inhibitor and had shown benefit in the
    Drosophila model. That rationale has since been tested directly and does not
    hold: minocycline is a poor PARP1 inhibitor and had no effect on basal
    ADP-ribosylation in ARH3-deficient cells. This treatment is curated
    precisely because it is a negative result - it is the only instance in this
    disorder of a mechanistic hypothesis being taken to the bedside and then
    checked, and a curator or clinician reading only the 2024 report would
    conclude that a PARP-inhibitor treatment was available and reasonable.
  treatment_term:
    preferred_term: Pharmacotherapy
    term:
      id: NCIT:C15986
      label: Pharmacotherapy
    therapeutic_agent:
    - preferred_term: minocycline
      term:
        id: CHEBI:50694
        label: minocycline
  therapeutic_modality: SMALL_MOLECULE
  evidence:
  - reference: PMID:37392332
    reference_title: "Expanding the Spectrum of Stress-Induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures (CONDSIAS)."
    supports: PARTIAL
    evidence_source: HUMAN_CLINICAL
    snippet: "we initiated experimental, off-label treatment with minocycline, a poly-ADP-polymerase (PARP) inhibitor, which has shown beneficial effects in a Drosophila fly model."
    explanation: Documents that minocycline was actually administered to a patient on a PARP-inhibition rationale. Marked PARTIAL because it reports an intention to treat and no outcome.
  - reference: PMID:42015965
    reference_title: "Novel ADPRS Missense Variant (p.Leu162Pro) Causes Stress-Induced Childhood-Onset Neurodegeneration With Ataxia and Seizures."
    supports: REFUTE
    evidence_source: IN_VITRO
    snippet: "Our results using minocycline indicate that it is a poor PARP1 inhibitor and was ineffective in reducing basal ADP-ribosylation in ARH3-deficient cells."
    explanation: Direct experimental refutation of the molecular rationale for minocycline in this disorder.
  - reference: PMID:42015965
    reference_title: "Novel ADPRS Missense Variant (p.Leu162Pro) Causes Stress-Induced Childhood-Onset Neurodegeneration With Ataxia and Seizures."
    supports: REFUTE
    evidence_source: IN_VITRO
    snippet: "questioning the molecular rationale underlying the therapeutic use of minocycline in patients with CONDSIAS"
    explanation: The authors' own framing of their result as a challenge to the therapeutic rationale.
  notes: >-
    A separate off-label report proposed repurposing doxycycline, a related
    tetracycline, in a CONDSIAS patient (PMID:39417910, Acta Neurol Belg 2025).
    That reference was fetched and cached but its record contains only
    bibliographic metadata with no abstract, so nothing from it can be quoted
    and it is not curated as a treatment here. Given the minocycline result, any
    tetracycline-based PARP-inhibition rationale in this disorder should be
    treated as unsupported until tested directly.
- name: High-dose vitamin supplementation (uncontrolled)
  description: >-
    Two of three patients in one recent series were reported to respond
    favourably to high-dose vitamin supplementation. This is recorded because it
    is in the literature and a family may ask about it, not because it is
    supported. The observation is uncontrolled, the specific vitamins and doses
    are not extractable from the abstract, the third patient in the same series
    stabilised spontaneously without any intervention, and the natural history of
    this disorder includes spontaneous remission of episodes over weeks - all of
    which make a treatment effect impossible to separate from the disease's own
    fluctuation. There is also a confounder specific to this differential:
    high-dose biotin and thiamine are the established treatment for
    biotin-responsive basal ganglia disease, the closest treatable mimic.
  treatment_term:
    preferred_term: Pharmacotherapy
    term:
      id: NCIT:C15986
      label: Pharmacotherapy
  therapeutic_modality: SMALL_MOLECULE
  evidence:
  - reference: PMID:40493129
    reference_title: "Phenotypic Diversity in Stress-induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures: Novel Associations with Parkinsonism, Icthyosis and Cataract."
    supports: PARTIAL
    evidence_source: HUMAN_CLINICAL
    snippet: "Both patients 2 and 3 responded favourably to treatment with high-dose vitamin supplementation, while patient 1 showed stable disease progression without specific therapeutic intervention, suggesting spontaneous stabilization of her condition."
    explanation: >-
      Marked PARTIAL rather than SUPPORT. The same sentence that reports the
      response also reports spontaneous stabilisation in an untreated patient,
      which is the source's own evidence that the disorder can stabilise without
      treatment and therefore that this uncontrolled observation cannot
      establish efficacy.
  notes: >-
    No `therapeutic_agent` is bound because the abstract does not name the
    vitamins or the doses, and guessing at biotin or thiamine on the basis of
    the SLC19A3 analogy would be fabrication.
- name: Avoidance and prompt management of physiological stressors
  description: >-
    The one intervention that follows directly from the mechanism: if episodes
    are precipitated by febrile illness, infection, surgery and physical stress,
    then aggressive antipyresis, early treatment of infection, up-to-date
    vaccination, avoidance of unnecessary surgery and careful perioperative
    planning are rational. It is worth stating explicitly that this is a
    mechanistic inference and not an evidence-based recommendation: no study has
    tested whether preventing or shortening febrile episodes changes the
    trajectory, and it is entirely possible that the neurological damage is
    driven by cumulative subclinical stress rather than by the clinically
    obvious episodes.
  treatment_term:
    preferred_term: supportive care
    term:
      id: NCIT:C15747
      label: Supportive Care
  therapeutic_modality: BEHAVIORAL
  notes: >-
    No evidence item is attached, deliberately. The premise - that stress
    triggers episodes - is evidenced in the `pathophysiology:` and
    `progression:` sections. The inference that avoiding stress prevents damage
    is not evidenced anywhere and is flagged as an open question in the
    discussions section.
- name: Genetic counselling
  description: >-
    Autosomal recessive recurrence-risk counselling with carrier testing for the
    familial alleles. Particularly relevant here because most reported families
    are consanguineous, because affected siblings are common in the reported
    pedigrees (one family had nine affected members), and because the same
    genotype has produced very different courses within a family, so a sibling's
    prognosis cannot be read off the proband's.
  treatment_term:
    preferred_term: genetic counseling
    term:
      id: NCIT:C15240
      label: Genetic Counseling
  notes: >-
    No evidence item is attached. The recessive inheritance underpinning the
    recommendation is evidenced in the `inheritance:` block; the counselling
    recommendation itself is a clinical inference from that mode of inheritance
    rather than a published recommendation.
animal_models:
- species: Drosophila melanogaster
  genotype: Loss of the Drosophila Parg paralog, with human ADPRHL2 rescue
  category: Loss-of-function with cross-species rescue
  description: >-
    The model that established functional conservation and produced the first
    therapeutic lead. Loss of the fly Parg paralog causes lethality specifically
    in response to oxidative challenge - not at baseline - which reproduces the
    stress-conditional character of the human disease more faithfully than any
    other model. Two rescues were performed: expression of human ADPRHL2 rescued
    the lethality, establishing that the human protein performs the conserved
    function; and pharmacological PARP inhibition also rescued it, establishing
    that the lethality runs through PARP activity and is therefore druggable.

    An important limitation is in the model's own construction: the gene deleted
    is the Drosophila Parg paralog, not a fly orthologue of ADPRHL2, and
    Drosophila does not separate the PARG and ARH3 activities the way mammals
    do. The model therefore tests the pathway rather than the gene, and the
    rescue by human ADPRHL2 is what licenses reading it as a model of this
    disorder. It also has no nervous-system phenotype corresponding to ataxia,
    seizures or neuropathy - the read-out is organismal lethality.
  associated_phenotypes:
  - Lethality on oxidative challenge
  - Rescue by human ADPRHL2 expression
  - Rescue by pharmacological PARP inhibition
  evidence:
  - reference: PMID:30100084
    reference_title: "Biallelic Mutations in ADPRHL2, Encoding ADP-Ribosylhydrolase 3, Lead to a Degenerative Pediatric Stress-Induced Epileptic Ataxia Syndrome."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "Loss of the Drosophila paralog Parg showed lethality in response to oxidative challenge that was rescued by human ADPRHL2, suggesting functional conservation."
    explanation: Describes the model, its stress-conditional phenotype and the cross-species rescue that validates it.
- species: Mus musculus
  genotype: Arh3 (Adprs) knockout
  category: Knockout
  description: >-
    Arh3-knockout mice are viable and do not spontaneously develop the human
    neurodegenerative phenotype, which is the central fact about this model and
    the reason it is used as a stress-challenge system rather than as a
    phenocopy. Subjected to cerebral ischaemia-reperfusion they show increased
    PAR accumulation, increased cell death and larger brain infarcts than
    wild-type animals, and both the injury and the infarct size are reduced by
    PARP inhibitors.

    The absence of a spontaneous phenotype is the key translational gap: the
    mouse tells us what ARH3 loss does when the animal is acutely injured, not
    what it does over years of ordinary life, and it is the latter that produces
    the human disease. This is recorded as a HUMAN_MODEL_MISMATCH discussion
    below rather than being glossed over. Whether Arh3-null mice have a cardiac
    phenotype was not established from any source read for this entry and no
    such claim is made here.
  associated_phenotypes:
  - Increased brain infarction after ischaemia-reperfusion, reduced by PARP inhibitors
  - Increased PAR accumulation and cell death under oxidative stress
  - No spontaneous neurodegenerative phenotype reported
  evidence:
  - reference: PMID:30830864
    reference_title: "PARP1 inhibition alleviates injury in ARH3-deficient mice and human cells."
    supports: SUPPORT
    evidence_source: MODEL_ORGANISM
    snippet: "the patient's fibroblasts and ARH3-deficient mice were more sensitive, respectively, to H2O2 stress and cerebral ischemia/reperfusion-induced PAR accumulation and cell death"
    explanation: Establishes the stress-challenge phenotype of the knockout mouse alongside the concordant patient-cell result.
  - reference: PMID:36497109
    reference_title: "ARH Family of ADP-Ribose-Acceptor Hydrolases."
    supports: SUPPORT
    evidence_source: OTHER
    snippet: "Arh3-KO mice developed increased brain infarction following ischemia-reperfusion injury, which was reduced by PARP inhibitors."
    explanation: States the brain-injury phenotype and its pharmacological rescue. Classified OTHER because this is a review.
discussions:
- discussion_id: adprs-fever-to-par-burst-unmeasured
  kind: KNOWLEDGE_GAP
  status: OPEN
  prompt: >-
    Does a febrile illness actually produce the proposed PARP1 activation and
    poly(ADP-ribose) burst in a patient with ARH3 deficiency?
  rationale: >-
    This is the load-bearing unmeasured step of the whole model and the most
    important gap in the entry. The clinical observation that episodes follow
    fever is solid and replicated. The biochemistry that ARH3 loss impairs PAR
    clearance is solid. The bridge between them - that fever raises cellular
    oxidative or genotoxic burden enough to generate a pathogenic PAR burst - is
    entirely inferred. Every experimental stressor used in the literature is a
    surrogate: hydrogen peroxide, cerebral ischaemia-reperfusion, oxidative
    challenge in flies. None of these is fever, and the mechanisms by which a
    viral illness might raise neuronal ADP-ribosylation (systemic inflammation,
    raised temperature per se, metabolic demand, direct oxidative stress) have
    not been distinguished. The clinical stakes are direct: if fever acts
    through temperature, aggressive antipyresis is a treatment; if it acts
    through inflammatory signalling, antipyresis is irrelevant and
    anti-inflammatory or PARP-directed strategies are needed. Until this is
    resolved, the standard advice given to these families rests on an
    assumption.
  attaches_to:
  - "pathophysiology#Physiological Stressor"
  - "pathophysiology#Failure of Poly(ADP-Ribose) Turnover under Stress"
  proposed_experiments:
  - experiment_id: adprs-febrile-episode-par-biomarker
    name: Longitudinal ADP-ribosylation biomarker sampling through a febrile episode
    description: >-
      Serial peripheral blood mononuclear cell or skin-fibroblast sampling from
      consented patients before, during and after a natural febrile illness,
      assayed for mono- and poly-ADP-ribosylation by immunoblot and mass
      spectrometry, to test directly whether the proposed burst occurs and how
      its time course relates to the clinical episode.
  - experiment_id: adprs-fever-surrogate-dissection
    name: Dissection of fever surrogates in patient-derived neurons
    description: >-
      Expose patient iPSC-derived neurons to the separable components of a
      febrile illness - 40 degrees Celsius hyperthermia, interferon and IL-1
      beta inflammatory signalling, and metabolic loading - and measure
      ADP-ribosylation and viability for each, to identify which component is
      the operative trigger.
- discussion_id: adprs-mar-scar-versus-par-toxicity
  kind: KNOWLEDGE_GAP
  status: OPEN
  prompt: >-
    Is the pathogenic species in ARH3 deficiency the persistent serine-linked
    mono(ADP-ribose) scar on chromatin, or the unrestrained poly(ADP-ribose)
    generated under stress?
  rationale: >-
    Two well-executed studies from overlapping groups give incompatible answers,
    and the answer determines what a therapy should do. PMID:32636369 shows that
    patient cells carry persistent mono(ADP-ribose) scars on core histones, that
    those scars suppress H3K9 acetylation and deregulate transcription, and that
    prolonged PARP inhibition removes them and restores acetylation - a chronic,
    stress-independent mechanism. PMID:34019811 then reports that persistent
    MARylation is well tolerated whereas persistent PARylation is highly toxic,
    and attributes patient disease to a stress-induced PARylation increase - an
    acute, stress-dependent mechanism. Both cannot be dominant. The distinction
    is not academic: a chronic scar mechanism argues for continuous low-dose
    PARP inhibition from diagnosis, whereas an acute PAR-toxicity mechanism
    argues for intermittent higher-dose intervention around stress episodes,
    which is a very different risk-benefit calculation in a child. It is also
    possible that both operate on different timescales, the scars producing the
    slow degenerative background and the PAR bursts producing the discrete
    episodes, which would explain the disorder's characteristic staircase
    course; no experiment has tested that synthesis.
  attaches_to:
  - "pathophysiology#Persistent Serine-Mono(ADP-Ribose) Chromatin Scars"
  - "pathophysiology#Deregulated Histone Acetylation and Transcription"
  - "pathophysiology#Failure of Poly(ADP-Ribose) Turnover under Stress"
  proposed_experiments:
  - experiment_id: adprs-neuronal-mar-par-separation
    name: Separation of MAR and PAR toxicity in post-mitotic neurons
    description: >-
      Compare ARH3-null, PARG-depleted and doubly perturbed human iPSC-derived
      neurons kept post-mitotic for months, measuring chromatin marks,
      transcriptome and survival, to determine whether the mono(ADP-ribose)
      scar becomes toxic on a neuronal timescale even though it is tolerated in
      dividing cells.
  - experiment_id: adprs-continuous-versus-pulsed-parp-inhibition
    name: Continuous versus stress-pulsed PARP inhibition in Arh3-knockout mice
    description: >-
      Compare continuous low-dose PARP inhibition against inhibition given only
      around an imposed stress challenge, in the same knockout cohort, using
      infarct size, neuronal survival and chromatin acetylation as endpoints, to
      establish which dosing logic the biology actually supports before any human
      trial is designed.
- discussion_id: adprs-no-chronic-model-of-the-disease
  kind: HUMAN_MODEL_MISMATCH
  status: OPEN
  prompt: >-
    Does any available model reproduce chronic, spontaneous neurodegeneration
    from ARH3 loss, as opposed to acute injury after an imposed stressor?
  rationale: >-
    This is a model-fidelity problem rather than an absence of evidence, and it
    constrains everything that can be claimed about the disease. The Arh3
    knockout mouse is viable and has no reported spontaneous neurodegenerative
    phenotype; it becomes informative only when acutely injured by cerebral
    ischaemia-reperfusion. The Drosophila model deletes the Parg paralog rather
    than an ARH3 orthologue and reads out organismal lethality on oxidative
    challenge, with no ataxia, seizure or neuropathy phenotype. The patient
    material is dermal fibroblasts, a dividing cell type that can dilute
    chromatin marks by division and that has none of the properties - long axons,
    post-mitotic longevity, high oxidative metabolism - proposed to explain why
    neurons are the cells that die. So the two things the human disease actually
    consists of, chronic neurodegeneration and selective neuronal vulnerability,
    are the two things no model demonstrates. Every statement in this entry about
    why neurons fail is therefore inference from cell-type-general biology, and
    the entry marks those nodes accordingly.
  attaches_to:
  - "pathophysiology#Selective Neuronal Loss with Cerebellar and Cerebral Atrophy"
  - "pathophysiology#NAD+ Consumption and Bioenergetic Failure"
  proposed_experiments:
  - experiment_id: adprs-aged-knockout-mouse-phenotyping
    name: Longitudinal phenotyping of aged Arh3-knockout mice with repeated stress challenge
    description: >-
      Follow knockout and wild-type cohorts to old age with repeated sublethal
      inflammatory or hyperthermic challenges, with serial motor testing,
      cerebellar volumetry and nerve conduction, to test whether repeated stress
      rather than a single acute insult is what produces the degenerative
      phenotype.
  - experiment_id: adprs-patient-ipsc-cerebellar-organoids
    name: Patient iPSC-derived cerebellar organoids and motor neurons
    description: >-
      Derive cerebellar organoids and long-axon motor neurons from patients and
      from isogenic corrected controls, and measure ADP-ribosylation, chromatin
      acetylation, transcription and survival with and without stress, to test
      selective vulnerability in the affected cell types rather than in
      fibroblasts.
- discussion_id: adprs-neuropathology-absent
  kind: KNOWLEDGE_GAP
  status: OPEN
  prompt: >-
    What is the neuropathology of this disorder, and which neuronal populations
    are actually lost?
  rationale: >-
    The literature contains, as far as this curation can establish, exactly one
    neuropathological statement: that parthanatos was observed in neurons of a
    deceased sibling (PMID:30830864). There is no autopsy series, no regional
    quantification, no Purkinje cell count, no assessment of whether the white
    matter change on MRI is demyelination, gliosis or axonal loss, and no
    examination of the brainstem respiratory nuclei whose failure kills these
    children. Roughly 21 of the 47 published patients have died, so the material
    exists or existed. Without it, the imaging finding of cerebellar atrophy
    cannot be converted into a cellular mechanism, and this entry deliberately
    declines to bind Purkinje cells to the cerebellar node for that reason.
  attaches_to:
  - "pathophysiology#Selective Neuronal Loss with Cerebellar and Cerebral Atrophy"
  - "pathophysiology#AIF Release from Mitochondria and Parthanatos"
  - "pathophysiology#Brainstem Respiratory Network Failure"
  proposed_experiments:
  - experiment_id: adprs-brain-bank-autopsy-series
    name: Coordinated post-mortem neuropathology series
    description: >-
      Establish consented brain donation through the existing ADPRS family
      networks and examine regional neuronal loss, PAR and mono(ADP-ribose)
      immunohistochemistry, AIF localisation, white matter integrity and
      brainstem respiratory nuclei across several patients.
- discussion_id: adprs-does-stress-avoidance-help
  kind: KNOWLEDGE_GAP
  status: OPEN
  prompt: >-
    Does preventing or shortening febrile episodes change the neurological
    trajectory, or is the damage driven by cumulative subclinical stress?
  rationale: >-
    Families are routinely advised to treat fever aggressively and avoid
    physical stress, and that advice is mechanistically reasonable, but it has
    never been tested and it may be wrong in an important way. A trigger was
    documented in only about 60% of published patients; the disorder still
    progresses in patients whose episodes are recognised and treated; and the
    chromatin-scar arm of the mechanism is explicitly stress-independent. If
    most of the neuronal loss comes from ordinary background single-strand break
    repair rather than from the clinically visible crises, then stress avoidance
    is palliative rather than protective, and the burden it places on a family -
    school avoidance, activity restriction, anxiety around every viral illness -
    is being imposed without benefit. The question is answerable with existing
    patients and no new technology.
  attaches_to:
  - "pathophysiology#Physiological Stressor"
  - "pathophysiology#Episodic Stepwise Neurological Deterioration"
  proposed_experiments:
  - experiment_id: adprs-prospective-natural-history-registry
    name: Prospective international natural-history registry with episode diaries
    description: >-
      Enrol patients prospectively with structured recording of every febrile or
      stress event, its management, and standardised motor, cognitive and
      respiratory outcomes, to establish whether episode frequency and severity
      predict the rate of decline once genotype is accounted for. This would also
      supply the natural-history baseline that any future PARP-inhibitor trial
      will require and which does not currently exist.
references:
- reference: PMID:30100084
  title: "Biallelic Mutations in ADPRHL2, Encoding ADP-Ribosylhydrolase 3, Lead to a Degenerative Pediatric Stress-Induced Epileptic Ataxia Syndrome."
- reference: PMID:30401461
  title: "Bi-allelic ADPRHL2 Mutations Cause Neurodegeneration with Developmental Delay, Ataxia, and Axonal Neuropathy."
- reference: PMID:38365196
  title: "Delineation of ADPRHL2 Variants: Report of Two New Patients with Review of the Literature."
- reference: PMID:32636369
  title: "Pathogenic ARH3 mutations result in ADP-ribose chromatin scars during DNA strand break repair."
- reference: PMID:34019811
  title: "Unrestrained poly-ADP-ribosylation provides insights into chromatin regulation and human disease."
- reference: PMID:28650317
  title: "Serine ADP-ribosylation reversal by the hydrolase ARH3."
- reference: PMID:29480802
  title: "Serine is the major residue for ADP-ribosylation upon DNA damage."
- reference: PMID:24191052
  title: "ADP-ribosyl-acceptor hydrolase 3 regulates poly (ADP-ribose) degradation and cell death during oxidative stress."
- reference: PMID:33894202
  title: "Structural and biochemical analysis of human ADP-ribosyl-acceptor hydrolase 3 reveals the basis of metal selectivity and different roles for the two magnesium ions."
- reference: PMID:34869334
  title: "The Making and Breaking of Serine-ADP-Ribosylation in the DNA Damage Response."
- reference: PMID:36497109
  title: "ARH Family of ADP-Ribose-Acceptor Hydrolases."
- reference: PMID:30830864
  title: "PARP1 inhibition alleviates injury in ARH3-deficient mice and human cells."
- reference: PMID:34479984
  title: "Biallelic ADPRHL2 mutations in complex neuropathy affect ADP ribosylation and DNA damage response."
- reference: PMID:39580621
  title: "An ADPRS variant disrupts ARH3 stability and subcellular localization in children with neurodegeneration and respiratory failure."
- reference: PMID:42015965
  title: "Novel ADPRS Missense Variant (p.Leu162Pro) Causes Stress-Induced Childhood-Onset Neurodegeneration With Ataxia and Seizures."
- reference: PMID:40493129
  title: "Phenotypic Diversity in Stress-induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures: Novel Associations with Parkinsonism, Icthyosis and Cataract."
- reference: PMID:37392332
  title: "Expanding the Spectrum of Stress-Induced Childhood-Onset Neurodegeneration with Variable Ataxia and Seizures (CONDSIAS)."
- reference: PMID:33528672
  title: "Episodic psychosis, ataxia, motor neuropathy with pyramidal signs (PAMP syndrome) caused by a novel mutation in ADPRHL2 (AHR3)."
- reference: PMID:35664652
  title: "Child-Onset Cerebellar Ataxia Caused by Two Compound Heterozygous Variants in ADPRS Gene: A Case Report."
- reference: PMID:35222245
  title: "Case Report: Stress-Induced Childhood-Onset Neurodegeneration With Ataxia-Seizures Syndrome Caused by a Novel Compound Heterozygous Mutation in ADPRHL2."
- reference: PMID:32746785
  title: "Novel imaging and clinical phenotypes of CONDSIAS disorder caused by a homozygous frameshift variant of ADPRHL2: a case report."