| Domain | High-confidence finding | Evidence type/strength | Suggested ontology terms | Key source/date/DOI/PMID if known |
|---|---|---|---|---|
| Disease identity | Spinocerebellar ataxia, autosomal recessive 32 (SCAR32) is a rare PRDX3-related cerebellar ataxia/neurodegenerative disorder. MONDO records the entity as MONDO:0859245. Published reports use inconsistent OMIM numbers (#619862 and #619648), so the current OMIM record should be verified directly before database ingestion. | Strong disease-level genetic association; replicated human families | MONDO:0859245; cerebellar ataxia; autosomal recessive inheritance | Open Targets association (pqac-00000000); Rebelo et al., 2021, DOI: [10.1093/brain/awab071](https://doi.org/10.1093/brain/awab071), PMID: 33889951 (pqac-00000003) |
| Causal gene and inheritance | Biallelic germline variants in **PRDX3** cause disease; unaffected parents commonly carry one variant. PRDX3 is Ensembl ENSG00000165672 and encodes mitochondrial peroxiredoxin 3. | Strong human segregation plus functional evidence; autosomal recessive | PRDX3; ENSG00000165672; loss of function; germline variant | Five simplex families in the discovery study and later independent cases (pqac-00000000, pqac-00000003, pqac-00000008, pqac-00000013) |
| Core motor phenotype | Gait and limb ataxia, dysmetria and cerebellar dysarthria constitute the core syndrome. Severity is variable and usually mild-to-moderate but may be severe in infantile disease; SARA scores reported include 7/40, 10/40, 19/40 and, in the original adult series, 8.5–21.5. | Strong but small human case series; exact pooled frequencies remain unreliable | HP:0001251 Ataxia; HP:0002072 Chorea is not core and should not be assigned routinely; cerebellar dysarthria; dysmetria; unsteady gait | Rebelo et al., 2021 (pqac-00000006); Martínez-Rubio et al., 2022, DOI: [10.1093/hmg/ddac146](https://doi.org/10.1093/hmg/ddac146), PMID: 35766882 (pqac-00000011); Naef et al., 2024 (pqac-00000012) |
| Onset and course | Reported onset extends from infancy/birth to approximately 35 years. Most cases show chronic, slowly progressive ataxia; the p.Asp163Glu case began acutely at 19 months, progressed rapidly initially, and then remained comparatively stable from ages 4–6.5 years. | Moderate human evidence; heterogeneous alleles and short follow-up | HP:0003593 Infantile onset; HP:0011463 Childhood onset; HP:0003581 Adult onset; progressive neurologic deterioration | Martínez-Rubio et al., 2022 (pqac-00000004, pqac-00000013); comparative case summary (pqac-00000015) |
| Oculomotor and movement findings | Gaze-evoked nystagmus, saccadic pursuit, hypermetric or slow saccades and ophthalmoplegia occur variably. Myoclonus, postural tremor, cervical dystonia, bradykinesia, hypomimia, rigidity and global hypokinesia expand the movement-disorder spectrum. | Moderate human evidence; variable and not universal | HP:0000639 Nystagmus; abnormal ocular motility; ophthalmoplegia; myoclonus; tremor; dystonia; bradykinesia; rigidity | Rebelo et al., 2021 (pqac-00000006); Naef et al., 2024 (pqac-00000008, pqac-00000012) |
| Additional neurologic phenotype | Dysphagia, peripheral motor-sensory neuropathy, learning difficulty, cognitive/academic decline, mood or behavioral disturbance and occasional hearing impairment have been reported. Cognition can remain normal. | Limited-to-moderate human evidence; mostly individual observations | HP:0002015 Dysphagia; HP:0009830 Peripheral neuropathy; learning disability; mild cognitive impairment; behavioral abnormality; hearing impairment | Martínez-Rubio et al., 2022 (pqac-00000013); Naef et al., 2024 (pqac-00000008, pqac-00000016); comparative cases (pqac-00000015) |
| MRI and anatomy | Cerebellar atrophy is the defining imaging abnormality and was universal in the later comparative case table. Severe early atrophy, vermian involvement, cerebellar cortical or middle-cerebellar-peduncle T2 hyperintensity, and occasional olivary, brainstem or mild parietal degeneration are described. | Strong human imaging evidence for cerebellar atrophy; ancillary findings less consistent | HP:0001272 Cerebellar atrophy; cerebellar vermis; cerebellar cortex; middle cerebellar peduncle; brainstem; inferior olivary nucleus; UBERON terms should be verified before ingestion | Rebelo et al., 2021 (pqac-00000003, pqac-00000006); Martínez-Rubio et al., 2022 (pqac-00000004); Naef et al., 2024 (pqac-00000008) |
| Pathogenic-variant spectrum | Reported variants include nonsense, frameshift, splice-altering and missense alleles. Examples include c.340dupG (p.Ala114GlyfsTer3), c.425C>G (p.Ala142Gly), c.489C>G (p.Asp163Glu), c.525_535del (p.Leu176TrpfsTer11), c.604G>A (p.Asp202Asn), and c.619C>T (p.Arg207Ter). Classification must be performed per allele rather than assuming every published missense allele is pathogenic. | Strong for segregating truncating alleles; moderate and functional-data-dependent for missense alleles | Sequence variant; frameshift variant; stop-gained variant; missense variant; splice-region variant | Rebelo et al., 2021 (pqac-00000006, pqac-00000010); Martínez-Rubio et al., 2022 (pqac-00000013); Naef et al., 2024 (pqac-00000012); Yang et al., 2025 (pqac-00000001) |
| Population frequency and founder effects | Causal alleles are rare. Naef et al. reported p.Ala142Gly below 0.01% in gnomAD; Yang et al. reported p.Arg207Ter in 2/251,446 alleles overall and 1/18,394 East Asian alleles. Recurrence of p.Asp202Asn across ancestries is documented, but a founder effect has not been established. | Limited population-database evidence; no validated carrier-frequency estimate | Rare variant; carrier state; founder effect—not established | Naef et al., 2024 (pqac-00000012); Yang et al., 2025 (pqac-00000001, pqac-00000015) |
| Molecular mechanism | PRDX3 is a mitochondrial matrix, thioredoxin-dependent peroxidase that reduces hydrogen peroxide; oxidized PRDX3 is regenerated by TXN2. Biallelic damaging variants can cause absent or unstable protein, weakening mitochondrial peroxide detoxification. | Strong biochemical knowledge plus patient-fibroblast evidence | GO:0005739 mitochondrion; mitochondrial matrix; hydrogen peroxide catabolic process; cellular oxidant detoxification; peroxidase activity; thioredoxin-dependent peroxiredoxin activity | Rebelo et al., 2021 (pqac-00000003, pqac-00000010, pqac-00000014) |
| Downstream pathophysiology | Deficient PRDX3 leads to increased mitochondrial H2O2/ROS and can result in reduced maximal respiration or respiratory reserve, mitochondrial membrane/cristae injury, protein instability or aggregation, unfolded-protein responses and greater apoptosis susceptibility. Selective cerebellar-neuronal degeneration then produces ataxia; the final cell-type-selectivity step is **inferred**, not demonstrated in human neuropathology. | Mixed: human fibroblasts support protein loss and redox/bioenergetic abnormalities; structural injury, neuronal degeneration and apoptosis rely substantially on cell and animal models | GO:0006979 response to oxidative stress; mitochondrial organization; cellular respiration; protein folding; response to unfolded protein; intrinsic apoptotic signaling; Purkinje cell and cerebellar neuron CL terms should be verified | Rebelo et al., 2021 (pqac-00000003, pqac-00000010); Martínez-Rubio et al., 2022 (pqac-00000011, pqac-00000013) |
| Functional-assay heterogeneity | Fibroblasts in the original and p.Asp163Glu studies showed PRDX3 depletion, oxidative-stress abnormalities and/or impaired respiration. In contrast, fibroblasts from a 2024 patient showed normal respiration and ROS responses despite reduced PRDX3 transcript, indicating that fibroblast assays are allele-, protocol- or tissue-dependent and are not validated diagnostic biomarkers. | Direct but discordant human-cell evidence | Skin fibroblast; oxygen-consumption rate; reactive oxygen species; mitochondrial respiratory capacity | Rebelo et al., 2021 (pqac-00000003); Martínez-Rubio et al., 2022 (pqac-00000004); Naef et al., 2024 (pqac-00000005, pqac-00000016) |
| Diagnostic approach | Diagnosis requires a compatible cerebellar syndrome/MRI plus identification of pathogenic or likely pathogenic **biallelic PRDX3 variants** in trans. Ataxia panels or WES are practical first-line approaches; WGS or RNA studies may resolve structural, intronic or splice-altering alleles. Segregation and allele-specific functional studies are especially important for missense or splice-region VUS. | Strong genomic-diagnostic rationale; no SCAR32-specific consensus guideline | Genetic testing; WES; WGS; multigene panel; RNA sequencing; segregation analysis | Trio exome sequencing identified recent cases after exclusion of common dominant and recessive ataxias (pqac-00000008, pqac-00000012, pqac-00000013) |
| Ancillary testing and differential diagnosis | Brain MRI, SARA, ocular-motor examination, neuropsychology, swallowing assessment and nerve-conduction studies characterize extent and progression. Routine biochemical, neurometabolic and conduction studies may be normal. Differential diagnosis includes other recessive/mitochondrial ataxias and repeat-expansion disorders; PRDX3 sequencing does not replace repeat-expansion testing. | Moderate clinical-practice inference; individual-case support | Magnetic resonance imaging; SARA; nerve-conduction study; neuropsychological assessment; swallowing evaluation | Rebelo et al., 2021 (pqac-00000006); Naef et al., 2024 (pqac-00000008) |
| Epidemiology | Prevalence, incidence, sex ratio and population carrier frequency are unavailable. Published evidence consists of very small, geographically diverse families; therefore case proportions must not be treated as population frequencies. | Major evidence gap | Rare disease; prevalence unknown; incidence unknown | Five unrelated discovery families plus subsequent isolated cases (pqac-00000003, pqac-00000009, pqac-00000015) |
| Prognosis and quality of life | The disease is lifelong and commonly progressive, with mobility, falls, speech, swallowing, school/work and psychosocial consequences. Formal survival, mortality, life-expectancy, EQ-5D, SF-36 and validated SCAR32-specific natural-history data are unavailable. Some affected adults retain moderate function, while infantile disease may require assisted walking. | Limited longitudinal human evidence; no survival or formal QoL cohort | Mobility impairment; fall risk; speech impairment; activity limitation; quality of life—data unavailable | Adult severity data (pqac-00000006); infantile follow-up (pqac-00000004, pqac-00000013); 2024 cases (pqac-00000012) |
| Treatment | No approved disease-modifying treatment and no PRDX3/SCAR32-specific interventional trial were identified. Physical, occupational and speech/swallowing therapy; fall prevention; mobility aids; educational and psychosocial support; and symptom-directed management are reasonable supportive interventions. Antioxidants, ferroptosis modifiers and PRDX3 gene replacement remain **experimental hypotheses**, without demonstrated SCAR32 clinical efficacy. | Supportive-care extrapolation; disease-modifying evidence unavailable | Physical therapy; occupational therapy; speech therapy; rehabilitation; assistive device; genetic therapy—experimental; NCIT identifiers should be verified | Mechanistic studies identify potential targets but do not establish treatment efficacy (pqac-00000003, pqac-00000010, pqac-00000015) |
| Prevention and counseling | Acquired primary prevention is not applicable to a Mendelian disorder. Genetic counseling, parental testing, cascade carrier testing, and—when familial pathogenic variants are known—prenatal or preimplantation genetic testing can prevent recurrence or enable informed reproductive decisions. Each pregnancy of two heterozygous parents has the standard autosomal-recessive 25% affected, 50% carrier and 25% unaffected/non-carrier probabilities. | High-confidence Mendelian inference | Genetic counseling; carrier testing; cascade screening; prenatal diagnosis; preimplantation genetic testing | Segregation in multiple families supports recessive counseling (pqac-00000001, pqac-00000008, pqac-00000013) |
| Drosophila model | Pan-neuronal or pan-glial Prdx3 depletion causes abnormal locomotion and reduced survival, especially under oxidative stress; brain degeneration has also been described. This supports redox-sensitive neural vulnerability but does not reproduce the complete human phenotype. | Moderate model-organism evidence; **inferred relevance to humans** | NCBI Taxon:7227; locomotory behavior; response to oxidative stress; neuron; glial cell | Rebelo et al., 2021 (pqac-00000003, pqac-00000010) |
| Zebrafish model | CRISPR/Cas9 prdx3 F0 crispants show reduced touch/burst responses, swim distance and velocity, impaired ATP production/maximal respiration, and increased oxidative-stress-associated apoptosis. Mosaic F0 knockdown and early developmental assays limit direct natural-history translation. | Moderate model-organism evidence; **inferred disease mechanism** | NCBI Taxon:7955; locomotory behavior; mitochondrial respiration; ATP metabolic process; apoptotic process | Naef et al., 2024 (pqac-00000005, pqac-00000016) |
| Mouse and cellular models | Prdx3-deficient mice show reduced strength, reduced skeletal-muscle mitochondrial DNA copy number and oxidative-stress-associated hippocampal cell loss. PRDX3 knockdown in cerebellar medulloblastoma cells increases H2O2, reduces viability and sensitizes cells to ROS-triggered apoptosis; mutant expression in primary mouse neurons disrupts neurites and mitochondria. These are mechanistic models, not proof of human Purkinje-cell pathology. | Moderate preclinical evidence; **inferred relevance to SCAR32** | NCBI Taxon:10090; neuron; mitochondrial DNA maintenance; cell death; neurite morphology; oxidative stress | Rebelo et al., 2021 (pqac-00000003, pqac-00000010); Martínez-Rubio et al., 2022 (pqac-00000004, pqac-00000011) |
| Other species/natural disease | No naturally occurring PRDX3-associated veterinary SCAR32 analogue or zoonotic/transmissible component was identified. Experimental models should not be entered as natural animal disease. | Data unavailable/not applicable | Natural disease—not established; zoonosis—not applicable | Available evidence describes induced laboratory models only (pqac-00000003, pqac-00000005, pqac-00000010) |


*Table: Compact evidence map for PRDX3-related SCAR32, separating replicated human findings from model-based inference and major knowledge gaps. Ontology labels are suggested conservatively, with uncertain identifiers explicitly left for verification.*