Autosomal Recessive Progressive External Ophthalmoplegia 1 (PEOB1) — Comprehensive Research Report
1. Disease Information
Overview. Autosomal recessive progressive external ophthalmoplegia with mitochondrial DNA deletions, type 1 (PEOB1) is a mitochondrial DNA-maintenance disorder caused by biallelic (homozygous or compound heterozygous) pathogenic variants in POLG, the nuclear gene encoding the catalytic subunit of mitochondrial DNA polymerase gamma (Pol γA). It is a chronic, typically adult-onset (>40 years) disease defined by progressive weakness of the extraocular muscles (bilateral ptosis and diffuse, symmetric ophthalmoparesis), exercise intolerance, and the presence of multiple mitochondrial DNA (mtDNA) deletions in skeletal muscle. PEOB1 sits within the broader "POLG-related disorders" spectrum, which ranges from the fatal infantile Alpers-Huttenlocher syndrome to the comparatively benign late-onset autosomal recessive PEO described here (Van Goethem et al., Nat Genet 2001, PMID not directly retrieved but DOI 10.1038/90034; OMIM #258450).
Key identifiers: - OMIM: #258450 — "Progressive External Ophthalmoplegia with Mitochondrial DNA Deletions, Autosomal Recessive 1; PEOB1" (gene-disease relationship: POLG, OMIM 174763, chromosome 15q26.1) - Gene: POLG (HGNC:9179), also historically POLG1 - Orphanet: ORPHA:254886 — "Autosomal recessive progressive external ophthalmoplegia" - MONDO: the umbrella MONDO term for this entity should be cross-checked against MONDO's POLG-PEO recessive class (searchable via OLS/MONDO as "progressive external ophthalmoplegia with mitochondrial DNA deletions, autosomal recessive 1") - MeSH: Ophthalmoplegia, Chronic Progressive External (D029231); Mitochondrial Diseases (D028361) - ICD-10: H49.4 (Progressive external ophthalmoplegia); G71.3 (Mitochondrial myopathy, NEC) is sometimes used for the systemic phenotype - GeneReviews: "POLG-Related Disorders" (NCBI Bookshelf NBK26471) - GTR condition:* C1850303 (autosomal recessive PEO)
Synonyms: PEOB1; Ophthalmoplegia, progressive external, autosomal recessive, with mitochondrial DNA deletions; arPEO; CPEO (chronic progressive external ophthalmoplegia) — note CPEO is the broader clinical umbrella term, not gene-specific; POLG-related PEO (recessive form); mitochondrial DNA depletion syndrome 4 nomenclature overlaps in some databases with the more severe POLG spectrum entries (Alpers syndrome, MCHS, SANDO/MIRAS) which share the same causal gene but different allele combinations.
Evidence base: Predominantly derived from aggregated case series, multi-center cohort studies (e.g., the 155-patient Hikmat et al. 2020 cohort, J Inherit Metab Dis, PMID 32068908), disease-level curated resources (OMIM, Orphanet, GeneReviews), and individual case reports — not large-scale EHR-based epidemiology, reflecting the disease's rarity.
2. Etiology
Disease causal factor: PEOB1 is a monogenic, purely genetic disease. It requires biallelic pathogenic variants in POLG (chr15q26.1), which abolish or severely impair the catalytic and/or proofreading (3′→5′ exonuclease) activity of the mitochondrial DNA polymerase, the sole DNA polymerase responsible for replicating the mitochondrial genome. There is no known environmental, infectious, or purely mechanistic (non-genetic) cause of this specific PEOB1 entity — although "environmental" or "gene-environment" stressors (below) can modulate disease severity or unmask latent POLG dysfunction.
Genetic risk factors: - Two pathogenic POLG alleles are required (true autosomal recessive; heterozygous carriers are asymptomatic or rarely mildly symptomatic). - Founder/common pathogenic variants: - c.1399G>A (p.Ala467Thr) — the single most common recessive POLG pathogenic allele; accounts for roughly 31–45% of mutant alleles in some European cohorts; gnomAD overall frequency ≈0.051% (143/282,888 alleles), rising to ≈0.098% in non-Finnish Europeans; population genetics studies (Chinnery et al./Rajakulendran, EJHG 2007) trace it to single ancient European founders. - c.2243G>C (p.Trp748Ser) — very common, frequently found in cis/trans with A467T as a compound "haplotype" allele; individually associated with more severe recessive phenotypes when combined with a second severe allele. - c.2542G>A (p.Gly848Ser) — a third recurrent founder variant, often reported in seizure-associated POLG phenotypes. - Combined carrier frequency for these founder alleles reaches ~1% in some European-descent populations (Rajakulendran et al., Eur J Hum Genet 2016). - Genotype-phenotype correlation: Variants located in the polymerase (Pol) domain (including the classic Y955C originally identified by Van Goethem et al. 2001) tend to produce more severe phenotypes when biallelic; variants in the linker region are typically associated with milder, later-onset PEO. Homozygous or compound heterozygous combinations involving at least one "severe" allele generally shift phenotype toward the ataxia-neuropathy spectrum (MIRAS/SANDO) or earlier-onset Alpers-like disease, while combinations of milder proofreading-domain alleles (e.g., A467T with another mild allele) are more likely to present as isolated late-onset PEO. - Modifier/susceptibility genes: No confirmed nuclear modifier genes for PEOB1 specifically; mtDNA haplogroup background has been proposed as a modifier of mitochondrial disease severity generally but is not established for POLG-PEO. - Digenic/oligogenic interaction: Not established for PEOB1 — this is a single-locus recessive disorder (distinguish from the broader class of "multiple mtDNA deletion disorders," MDMDs, caused by ≥20 different nuclear genes including TWNK, RRM2B, DGUOK, SLC25A4/ANT1, OPA1, MGME1, RNASEH1, TK2, TOP3A, DNA2 — genetically heterogeneous phenocopies, not modifiers of POLG-PEO itself).
Protective factors: No established genetic or environmental protective factors specific to PEOB1. General mitochondrial-supportive measures (aerobic exercise, avoidance of mitochondrial toxins) are supportive/management-oriented rather than disease-preventive (see Treatment/Prevention).
Environmental/gene-environment interaction — the critical clinical interaction: - Valproic acid (sodium valproate) is an absolute contraindication in any POLG-related disorder, including PEOB1. Valproate is metabolized via mitochondrial β-oxidation and can precipitate acute, sometimes fatal, hepatic failure in POLG-mutant patients — a well-documented gene-drug interaction (Stewart et al., Lancet Neurol 2010, and cautionary statements throughout GeneReviews). This is the single most important gene-environment interaction to flag clinically. - Other mitochondrial toxins (e.g., certain antiretrovirals affecting mtDNA polymerase, aminoglycosides affecting mitochondrial translation) are theoretically relevant to worsening mitochondrial reserve but are not disease-causal. - Physiologic stress (fasting, intercurrent infection, surgery) is reported anecdotally to unmask or worsen symptoms in POLG disease broadly, consistent with reduced mitochondrial energetic reserve, though this is best documented in the early-onset/Alpers phenotypes rather than isolated arPEO.
3. Phenotypes
Core/defining phenotypes
Table (click to expand)
| Phenotype | Type | Suggested HPO term |
|---|---|---|
| Bilateral ptosis | Clinical sign | HP:0000508 (Ptosis) |
| Progressive external ophthalmoplegia / ophthalmoparesis | Clinical sign | HP:0000590 (Ophthalmoplegia) / HP:0000602 (Ophthalmoparesis) |
| Exercise intolerance | Symptom | HP:0003546 (Exercise intolerance) |
| Proximal/generalized skeletal muscle weakness | Clinical sign | HP:0003701 (Proximal muscle weakness) / HP:0001324 (Muscle weakness) |
| Muscle atrophy | Clinical sign | HP:0003202 (Skeletal muscle atrophy) |
Additional/variable manifestations (PEO-"plus" features)
Table (click to expand)
| Phenotype | Type | Suggested HPO term |
|---|---|---|
| Sensory axonal peripheral neuropathy | Clinical sign | HP:0003390 (Aplasia/Hypoplasia... ) — better: HP:0007141 (Axonal (sensory) neuropathy) |
| Cerebellar ataxia | Clinical sign | HP:0001251 (Ataxia) |
| Dysarthria | Clinical sign | HP:0001260 (Dysarthria) |
| Sensorineural hearing loss | Clinical sign | HP:0000407 (Sensorineural hearing loss) |
| Cataracts | Clinical sign | HP:0000518 (Cataract) |
| Depression / psychiatric symptoms | Behavioral | HP:0000716 (Depression) |
| Hypogonadism | Clinical sign | HP:0000135 (Hypogonadism) |
| Parkinsonism | Clinical sign | HP:0001300 (Parkinsonism) |
| Mitral valve prolapse | Clinical sign | HP:0001634 (Mitral valve prolapse) |
| Cardiomyopathy | Clinical sign | HP:0001638 (Cardiomyopathy) |
| Gastrointestinal dysmotility | Clinical sign | HP:0002015 (Dysphagia) / HP:0002251 (Aganglionic megacolon) not exact — general GI dysmotility phenotype |
| Elevated CSF/serum lactate | Laboratory abnormality | HP:0002151 (Increased serum lactate) |
| Ragged red fibers on biopsy | Laboratory/histopathology | HP:0003200 (Ragged-red muscle fibers) |
| COX-negative (cytochrome c oxidase-deficient) fibers | Laboratory/histopathology | HP:0033279 or related mitochondrial myopathy histology term (verify via OAK) |
Onset: Classic PEOB1/arPEO manifests typically after age 40 (late-onset category in the POLG age-of-onset classification), though earlier presentations occur, especially with more severe allele combinations that push the phenotype toward the juvenile/adult ataxia-neuropathy spectrum (12–40 years) or, rarely, earlier.
Severity/progression: Progressive and generally slow. In the Hikmat et al. 2020 cohort (J Inherit Metab Dis, PMID 32068908) analyzing 155 POLG-disease patients stratified by age of onset: - Late-onset disease (>40 y): ptosis (95%), PEO (89%), ataxia (58%), peripheral neuropathy (65%) — this group has the best overall prognosis among the three age strata. - Juvenile/adult-onset (12–40 y): ataxia (90%), peripheral neuropathy (84%), seizures (71%), stroke-like episodes (54%). - Early-onset (<12 y): hepatopathy (87%), seizures (84%), feeding difficulties (84%), hypotonia (79%) — worst prognosis. - Across the whole POLG spectrum: neurological (90%), ophthalmological (74%), and gastrointestinal (63%) features predominate overall.
Many patients initially diagnosed with "isolated" arPEO develop additional systemic/neurological features (ataxia, neuropathy) over years to decades — GeneReviews explicitly cautions that "progressive PEO without systemic involvement" as a static label requires caution, since longitudinal follow-up frequently reveals evolution toward the ataxia-neuropathy spectrum (MIRAS/SANDO).
Frequency of PEO/ptosis in the general POLG-mutant population: Ptosis in ~34% (51/149) and PEO in ~38% (56/146) of a broader all-ages POLG cohort, rising to >90% when restricted to the late-onset stratum — illustrating strong age-dependence of the phenotype-frequency relationship (cite Hikmat 2020 stratified data above).
Quality of life impact: Ptosis and ophthalmoparesis cause functional visual impairment (chin-up head posture to see under ptotic lids, diplopia less common than in myasthenia because weakness is typically symmetric), corneal exposure risk after ptosis surgery, and exercise intolerance limits activities of daily living; neuropathy and ataxia (when present) contribute to gait disability and falls risk. No disease-specific validated QOL instrument was identified in this search; general neuromuscular-disease QOL tools (SF-36, individualized) are used in the small POLG literature.
4. Genetic/Molecular Information
Causal gene: POLG (OMIM 174763; HGNC:9179; chromosome 15q26.1), encoding the 140-kDa catalytic α-subunit of the heterotrimeric mitochondrial DNA polymerase γ holoenzyme (Pol γA + a dimeric Pol γB accessory subunit encoded by POLG2*). POLG is the sole DNA polymerase responsible for mtDNA replication and repair.
Variant classification and type: - Pathogenic variants span missense (majority, especially the recurrent founder alleles), nonsense, frameshift, splice-site, and rare small in-frame indels; large deletions/duplications are uncommon (~5% detection by dosage analysis per GeneReviews). - Structurally, POLG has three key functional domains: the N-terminal exonuclease (proofreading) domain, a linker region (contains the "spacer" and thumb subdomains, binds POLG2), and the C-terminal polymerase domain. Mutation location correlates loosely with mechanism and severity: - Exonuclease-domain mutations impair 3′→5′ proofreading, increasing point-mutation rate and promoting deletion formation. - Polymerase-domain mutations (e.g., the original Y955C) impair nucleotide incorporation/catalysis directly. - Linker-domain mutations (where A467T and W748S lie) can impair holoenzyme processivity/stability and POLG2 interaction. - ACMG classification: A467T, W748S, and G848S are all classified pathogenic in ClinVar for POLG-related spectrum disorders (multiple submitters); many rarer POLG variants remain VUS pending functional/segregation data.
Population/allele frequency: - p.Ala467Thr (c.1399G>A): gnomAD ~0.051% overall, ~0.098% in non-Finnish Europeans; the most common recessive pathogenic POLG allele (reported in at least 52 patients in aggregate case literature, 15 homozygous, 37 compound heterozygous, per ClinVar aggregation). - These founder mutations are traced to ancient single European founder haplotypes (Rajakulendran et al., Eur J Hum Genet 2007/2016 lineage of studies), explaining their spread across Europe, Australia, New Zealand, and the US in populations of European descent. - Carrier frequency for POLG pathogenic variants collectively approaches ~1% in some European-ancestry cohorts (relevant to ACMG carrier-screening panel design, per the 2024 gnomAD carrier-frequency estimation study, PMID 38459613).
Somatic vs. germline: PEOB1 pathogenic variants are germline. However, the disease mechanism itself operates through secondary somatic mtDNA mutagenesis: the germline nuclear POLG defect causes accumulation of somatic, clonally expanded mtDNA deletions in postmitotic tissues (especially skeletal and extraocular muscle) over the patient's lifetime — a form of accelerated somatic mitochondrial genome instability driven by a germline nuclear lesion.
Functional consequence: Loss of function / hypomorphic — reduced polymerase fidelity and/or reduced exonuclease proofreading activity, and/or reduced holoenzyme processivity, leading to (a) increased point mutation rate in mtDNA, (b) stalling of replication forks, and (c) accumulation of large-scale mtDNA deletions, ultimately (d) mtDNA depletion in severe cases. Nature Communications work (Basu et al., Nat Commun 2018, PMID 30089853 area) demonstrated that POLG's exonuclease activity is required for rapid degradation of linear mtDNA fragments generated during replication/repair; loss of this activity allows persistence of fragments and increases formation of deletions via non-homologous end joining or microhomology-mediated repair of stalled/broken replication intermediates.
Modifier genes: None firmly established for POLG-PEO specifically. POLG2 (the accessory subunit gene) causes a phenotypically overlapping but genetically and functionally distinct autosomal dominant PEO (PEOA5) and is mechanistically linked (holoenzyme partner) but is not a modifier of POLG1-driven PEOB1.
Epigenetic information: No POLG-PEO-specific DNA methylation, histone modification, or chromatin signature has been established in the literature surveyed; this is not a primary disease mechanism for a structural/catalytic enzyme defect of this kind. Not applicable/not identified.
Chromosomal abnormalities: Not applicable — PEOB1 is caused by point mutations/small indels in POLG, not by large chromosomal rearrangements. (Contrast with the mtDNA-deletion disorder Kearns-Sayre syndrome, which involves a single large mtDNA deletion, not nuclear chromosomal abnormality.)
5. Environmental Information
- Environmental factors: No toxin, radiation, or pollutant exposure is established as disease-causal for PEOB1 (it is monogenic). The dominant environmentally-modifiable risk is iatrogenic: valproic acid exposure, which can precipitate acute liver failure in POLG-mutant individuals (see Etiology/Gene-Environment above) — this is the most clinically actionable environmental factor for the entire POLG-disease spectrum, including arPEO.
- Lifestyle factors: Aerobic exercise has evidence (in the Polg "mutator" mouse model) of attenuating the progeroid/mitochondrial-dysfunction phenotype, and is used clinically as a supportive, not preventive, intervention in human mitochondrial myopathy generally; specific human PEOB1 exercise-outcome trial data were not identified in this search.
- Infectious agents: Not applicable — PEOB1 is not an infectious disease. Intercurrent infections/febrile illness may act as nonspecific metabolic stressors that unmask or worsen symptoms in POLG disease broadly (most documented in early-onset phenotypes), analogous to metabolic decompensation patterns in other mitochondrial diseases, but this is not a primary causal factor.
6. Mechanism / Pathophysiology
Causal chain (initial trigger → clinical manifestation):
- Molecular trigger: Biallelic POLG pathogenic variants reduce the catalytic fidelity, proofreading (exonuclease) activity, and/or processivity of the Pol γ holoenzyme (GO:0006261, DNA-templated DNA replication; GO:0004674 or more precisely GO:0003887 DNA-directed DNA polymerase activity; GO:0008310, single-stranded DNA 3'-5' exodeoxyribonuclease activity for the exonuclease domain).
- Replication stalling and mtDNA instability: Defective Pol γ causes replication fork stalling and generates persistent linear mtDNA fragments; because the exonuclease activity that normally degrades these fragments is impaired, they persist and are aberrantly repaired, generating large-scale mtDNA deletions (somatic, clonally expanded within individual cells/fibers over years). In some genotypes, replication failure instead causes mtDNA depletion (reduced copy number) rather than deletions — genotype-dependent branch point (relevant to the broader POLG spectrum, e.g., Alpers phenotype is depletion-dominant while adult PEO is deletion-dominant).
- Cellular consequence: Clonal expansion of deleted mtDNA molecules within individual postmitotic muscle fibers over decades reaches a pathogenic threshold, causing focal cytochrome c oxidase (Complex IV) deficiency (mtDNA-encoded subunits of Complexes I, III, IV, and ATP synthase are lost from deleted genomes) — the classic "COX-negative, SDH-hyperreactive (ragged-red/ragged-blue) fiber" seen on histochemistry (GO cellular component: GO:0005739 mitochondrion; GO:0005743 mitochondrial inner membrane; relevant biological process GO:0006123, mitochondrial electron transport, cytochrome c to oxygen).
- Tissue-level consequence: Oxidative phosphorylation (OXPHOS) failure in a mosaic pattern of affected fibers leads to a bioenergetic deficit that is most clinically apparent in tissues with high, continuous energy/mtDNA-turnover demand and long-lived postmitotic cells — extraocular muscle is exquisitely susceptible (thought to reflect its very high mitochondrial content, tonic/high-frequency firing pattern, and high mtDNA turnover), explaining the near-universal, early involvement of eyelid levator and extraocular muscles.
- Organism-level manifestation: Progressive extraocular and skeletal myopathy (ptosis, ophthalmoparesis, exercise intolerance, proximal weakness), and — as the somatic mutation burden and clinical spectrum expand — peripheral nerve (axonal sensory neuropathy), cerebellar (ataxia), auditory (sensorineural hearing loss), lens (cataract), cardiac conduction/muscle, gonadal, and CNS (depression, parkinsonism) involvement in a subset of patients, reflecting variably penetrant multi-organ mosaic mtDNA deletion burden.
Cell types involved: Skeletal/extraocular myocyte (CL:0000192 smooth muscle cell is wrong — correct: CL:0000188 skeletal muscle myoblast / mature skeletal muscle fiber; extraocular myocyte has no distinct CL term but can be annotated as CL:0008002 skeletal muscle fiber, UBERON-localized to extraocular muscle), peripheral sensory neuron (CL:0000101 sensory neuron), Purkinje/cerebellar neurons (CL:0000121 Purkinje cell) in ataxia-affected patients, cochlear hair cells (CL:0000201/CL:0000202) in hearing-loss-affected patients, lens epithelial cells in cataract.
Biochemical abnormality: Impaired mitochondrial DNA polymerase (POLG) fidelity/processivity → downstream OXPHOS enzyme complex deficiency (particularly Complex I and IV, whose subunits are partly mtDNA-encoded) → impaired ATP synthesis and elevated lactate/pyruvate (common but not universal laboratory finding).
Molecular/cellular process ontology suggestions: - GO:0006264 mitochondrial DNA replication - GO:0032042 mitochondrial DNA metabolic process - GO:0006281 DNA repair - GO:0007005 mitochondrion organization - GO:0006123 mitochondrial electron transport, cytochrome c to oxygen (Complex IV deficiency downstream effect)
Omics/advanced technologies: No large-scale transcriptomic, proteomic, or single-cell datasets specific to PEOB1 human tissue were identified in this search (reflecting rarity of biobanked tissue); most molecular characterization comes from muscle biopsy histochemistry/EM plus targeted biochemical/enzymatic assays of respiratory chain complexes, and from model-system (mouse, zebrafish, yeast) omics rather than direct human multi-omics profiling.
7. Anatomical Structures Affected
Organ level: - Primary: Extraocular muscles (levator palpebrae superioris, medial/lateral/superior/inferior recti, obliques) — UBERON:0002031 (extraocular muscle) / more general UBERON:0001772 (obturator... not relevant) — extraocular muscle is the primary UBERON target; specifically the levator palpebrae superioris (UBERON:0011343-adjacent structures should be verified via OAK). - Secondary: Skeletal (limb-girdle/proximal) muscle (UBERON:0001134, skeletal muscle tissue); peripheral nervous system (peripheral nerve, UBERON:0000010); cerebellum (UBERON:0002037) in ataxia-affected patients; inner ear/cochlea (UBERON:0001846) in hearing loss; lens (UBERON:0000965) in cataract; heart (UBERON:0000948) in the subset with cardiomyopathy/mitral valve prolapse; gonads (testis UBERON:0000473 / ovary UBERON:0000992) in hypogonadism; gastrointestinal tract (UBERON:0005409) in dysmotility. - Body systems: Neuromuscular (primary), nervous system (peripheral and central), special senses (visual — via extraocular myopathy and via cataract; auditory), cardiovascular, endocrine (gonadal), digestive.
Tissue/cell level: Skeletal/extraocular muscle fibers (mosaic COX-deficient, ragged-red fibers); peripheral sensory axons (axonal, not demyelinating, sensory neuropathy predominant); cerebellar Purkinje and granule cell circuitry.
Subcellular level: Mitochondria broadly (GO:0005739); specifically the mitochondrial matrix (GO:0005759, site of mtDNA and the replisome) and mitochondrial nucleoid (GO:0042645, mitochondrial nucleoid — the mtDNA-protein complex where Pol γ operates).
Localization/laterality: Ophthalmoplegia is characteristically bilateral and symmetric (a key distinguishing feature from myasthenia gravis, which is often asymmetric/fluctuating) — ptosis and ocular motility restriction affect both eyes in a diffuse, non-fatigable pattern.
8. Temporal Development
- Onset: Classic PEOB1 = late-onset, typically after age 40 years (GeneReviews). Earlier presentations (juvenile/adult-onset, 12–40 y) occur with more severe allele combinations and blend into the ataxia-neuropathy spectrum (MIRAS/SANDO); onset before age 12 is atypical for "pure" arPEO and instead characterizes the more severe POLG phenotypes (Alpers-Huttenlocher, MCHS).
- Onset pattern: Insidious/gradual — patients frequently do not notice the earliest, mild ptosis/ophthalmoparesis and are often diagnosed years after first symptoms, sometimes after being misdiagnosed with myasthenia gravis.
- Progression: Slowly progressive over years to decades. Ptosis and ophthalmoparesis worsen gradually; additional features (neuropathy, ataxia, hearing loss, cataract) may accrue over the disease course, particularly in patients followed longitudinally rather than assessed cross-sectionally.
- Disease course: Chronic, lifelong, non-remitting — no spontaneous remission is described. Unlike myasthenia gravis, there is no fluctuation or fatigability pattern.
- Prognosis by age-of-onset stratum: Late-onset (arPEO) disease has the best prognosis of the three POLG age strata; overall survival is far more favorable than early-onset (infantile hepatocerebral/Alpers) or juvenile/adult-onset (epilepsy/stroke-like-episode-dominant) POLG disease, where survival correlates strongly with age at onset of seizures/liver disease (median survival 0.7 years when epilepsy onset is in the first 3 years of life vs. median 18.0 years when epilepsy onset is after age 16; liver involvement is an independent poor-prognostic marker) (Hikmat et al., 2020; Cohen/Naviaux natural history studies).
- Critical periods: None specifically described for the adult-onset arPEO phenotype; by contrast, early recognition and valproate avoidance is critical at any age given the risk of fulminant hepatic failure if a POLG-mutant patient (of any phenotype) is inadvertently exposed.
9. Inheritance and Population
Epidemiology: PEO as a clinical syndrome overall has no precisely established population prevalence (Orphanet notes "prevalence unknown"). It is grouped among ultra-rare Mendelian mitochondrial disorders; a closely related entity (childhood-onset autosomal recessive myopathy with external ophthalmoplegia) is documented at <1/1,000,000. POLG-related disorders overall are estimated (from carrier-frequency modeling) to have a combined genetic prevalence on the order of ~1 in several thousand to ~1 in 10,000+ depending on population and specific phenotype, but PEOB1 specifically (the late-onset, comparatively mild recessive PEO subset) has no dedicated incidence/prevalence figure identified in this search.
Inheritance pattern: Autosomal recessive (biallelic POLG pathogenic variants required). Note that POLG is also independently a cause of autosomal dominant PEO (PEOA1, heterozygous variant sufficient) — the same gene causes phenotypically similar disease under different zygosity/allele-severity combinations, and recessive disease is generally more severe than the dominant form (per Orphanet/OMIM). - Penetrance: Complete for biallelic pathogenic genotypes, though age-dependent (symptoms emerge progressively rather than being present from birth) — effectively full penetrance by later adulthood for the classic late-onset genotype combinations. - Expressivity: Highly variable, both between and within families with the same genotype — even patients homozygous for the same founder allele can show a spectrum from isolated PEO to full ataxia-neuropathy-spectrum disease, indicating stochastic somatic mtDNA deletion accumulation contributes to phenotypic variability beyond genotype alone. - Anticipation: Not described — POLG disease does not follow a repeat-expansion anticipation mechanism. - Germline mosaicism: Not specifically documented for POLG in the literature surveyed. - Founder effects: Strong — A467T, W748S, and G848S are all traceable European founder alleles (see Etiology/Genetic risk factors), giving PEOB1/POLG-spectrum disease its comparatively higher (for a "rare disease") carrier frequency in European-ancestry populations relative to many other ultra-rare recessive mitochondrial disorders. - Consanguinity: As with any autosomal recessive disorder, consanguineous unions increase risk, particularly for rarer non-founder POLG alleles; not specifically quantified for PEOB1 in this search. - Carrier frequency: Combined POLG pathogenic-variant carrier frequency approaches ~1% in some European-ancestry cohorts (dominated by A467T); this is unusually high for a recessive disease of this severity and is explained entirely by founder effects rather than heterozygote advantage.
Population demographics: Predominantly reported in populations of European ancestry (consistent with founder-allele geography — Europe, Australia, New Zealand, US populations of European descent per Rajakulendran et al.). No confirmed sex predilection (autosomal, so ~1:1 male:female expected and generally observed). Age distribution of affected/diagnosed individuals for the arPEO subtype skews toward middle-aged to older adults (40s onward) at symptom onset, with diagnosis often delayed further due to the insidious onset and diagnostic overlap with myasthenia gravis and other CPEO causes.
10. Diagnostics
Clinical/laboratory tests: - Serum/CSF lactate and pyruvate (may be elevated, not universally diagnostic). - Creatine kinase (CK) — often normal or mildly elevated in mitochondrial myopathy (helps distinguish from primary myopathies with higher CK).
Muscle biopsy (a cornerstone diagnostic test): - Histochemistry: Modified Gomori trichrome stain shows ragged-red fibers (RRF) (subsarcolemmal/intermyofibrillar mitochondrial proliferation); succinate dehydrogenase (SDH) staining shows corresponding "ragged-blue" fibers; combined COX/SDH staining reveals COX-deficient fibers (blue, SDH-positive but COX-negative) — the single most sensitive histochemical marker of mtDNA-deletion disease, often more sensitive than RRF alone. - Electron microscopy: Paracrystalline mitochondrial inclusions, abnormal mitochondrial morphology/proliferation. - Molecular studies on muscle: Southern blot or long-range PCR demonstrating multiple mtDNA deletions (distinguishing this "multiple deletion" disorder from Kearns-Sayre syndrome's single large deletion); quantitative PCR may show reduced mtDNA copy number in more severe genotypes. - Biochemical respiratory chain enzymology on muscle homogenate: reduced Complex I and Complex IV activities characteristic (mtDNA-encoded subunit-dependent complexes).
Imaging: Not primary for diagnosis; orbital MRI may show extraocular muscle atrophy in advanced disease but is non-specific. Brain MRI may be used to evaluate cerebellar atrophy or white matter change in patients with ataxia/CNS features, again nonspecific.
Electrophysiology: - Nerve conduction studies/EMG to characterize the sensory axonal peripheral neuropathy when present (reduced sensory nerve action potential amplitudes with relatively preserved conduction velocities, consistent with axonal loss). - Repetitive nerve stimulation and single-fiber EMG are used to exclude myasthenia gravis (a key differential) — should be normal in PEOB1. - Audiometry to characterize sensorineural hearing loss when present.
Genetic testing (definitive diagnosis): - First-line: POLG sequence analysis (single-gene or targeted mitochondrial-disease gene panel), which detects the pathogenic variant in ~95% of cases; gene-targeted deletion/duplication analysis accounts for the remaining ~5%. - Broader approach: Given genetic heterogeneity of multiple-mtDNA-deletion disorders (POLG, POLG2, TWNK, RRM2B, DGUOK, SLC25A4, OPA1, MGME1, RNASEH1, TK2, TOP3A, DNA2, and others), a multigene mitochondrial-disease/PEO panel or exome sequencing is often used clinically, particularly when the phenotype is atypical or POLG sequencing is negative. - Whole-exome/whole-genome sequencing utility: increasingly used as first-tier or reflex testing given phenotypic overlap across the >20 known "multiple mtDNA deletion" genes; the specific yield data for PEOB1 alone were not separately quantified in this search. - Mitochondrial genome sequencing of muscle (not blood — mtDNA deletion mosaicism is tissue-specific and typically not detectable in blood in adult-onset PEO) is essential to demonstrate the pathognomonic multiple-deletion picture, but does not identify the causal nuclear gene. - Chromosomal microarray/karyotype/FISH: Not applicable/not indicated — PEOB1 is a nuclear point-mutation disorder, not a copy-number or cytogenetic disorder.
Clinical diagnostic criteria: No formal consensus scoring system specific to PEOB1 was identified; diagnosis rests on the clinical triad of (bilateral ptosis + symmetric ophthalmoparesis + exercise intolerance/myopathy), muscle biopsy evidence of mtDNA-deletion-type mitochondrial myopathy, demonstration of multiple mtDNA deletions in muscle, and confirmation of biallelic POLG pathogenic variants.
Differential diagnosis: - Ocular myasthenia gravis — key distinguishing features: MG is typically fatigable/fluctuating and often asymmetric, with positive acetylcholine receptor or MuSK antibodies and abnormal repetitive nerve stimulation/single-fiber EMG; PEOB1 is static-progressive, symmetric, antibody-negative. - Kearns-Sayre syndrome — single large mtDNA deletion (usually sporadic, not inherited in Mendelian fashion), onset before age 20, plus pigmentary retinopathy and cardiac conduction defects (triad required for KSS diagnosis) — PEOB1 lacks the KSS triad and shows multiple rather than single mtDNA deletions. - Oculopharyngeal muscular dystrophy (OPMD) — caused by GCN-repeat expansion in PABPN1; presents with ptosis/dysphagia, distinguished by dysphagia prominence and specific molecular test. - Myotonic dystrophy type 1, congenital fibrosis of the extraocular muscles, thyroid eye disease/orbitopathy, chronic orbital myositis, abetalipoproteinemia, Refsum disease — all in the broader CPEO differential. - Other genetic multiple-mtDNA-deletion disorders (TWNK-recessive, RRM2B, DGUOK, SLC25A4, MGME1, RNASEH1, TK2, TOP3A, DNA2, OPA1) — clinically similar/indistinguishable without molecular testing; distinguished only by causal gene on sequencing.
Screening: No population-based newborn or general screening program exists for this adult-onset recessive disorder. Carrier screening (e.g., expanded carrier screening panels, ACMG-aligned) for POLG founder variants (notably A467T) is available and relevant given the ~1% carrier frequency in European-ancestry populations; cascade testing of at-risk relatives (siblings: 25% recurrence risk) and genetic counseling (including reproductive options such as prenatal or preimplantation genetic testing) are appropriate once a proband's biallelic genotype is established.
11. Outcome/Prognosis
- Survival/mortality: The late-onset arPEO phenotype (PEOB1) carries the best prognosis within the POLG-disease spectrum; life expectancy is not dramatically shortened relative to the general population in isolated PEO without major systemic (hepatic, cardiac) involvement, though precise actuarial life-expectancy figures specific to PEOB1 were not identified in this search. This contrasts sharply with early-onset POLG phenotypes, where survival is markedly reduced and correlates with age at seizure onset and presence of liver disease (median survival as low as 0.7 years for epilepsy onset before age 3; liver involvement independently predicts worse survival) (Hikmat et al. 2020 and related natural-history literature).
- Morbidity/function: Chronic visual disability from ptosis/ophthalmoparesis (compensatory head-tilt, risk of exposure keratopathy, especially post-surgical); progressive proximal myopathy can impair mobility and activities of daily living; when ataxia/neuropathy develop, gait and fine-motor disability increase; hearing loss and cataract contribute to sensory disability in affected patients.
- Complications: Corneal exposure after ptosis surgery is a specific, well-documented iatrogenic complication in this population (due to poor Bell's phenomenon/orbicularis function accompanying the myopathy) — requires careful pre-op counseling and often conservative (under-)correction. Systemic complications (when present) include cardiac conduction disease/cardiomyopathy, endocrinopathy (hypogonadism), and GI dysmotility.
- Recovery potential: No disease-modifying therapy exists; the underlying mitochondrial myopathy and neuropathy are not reversible, though supportive interventions (ptosis surgery, physical therapy, hearing aids, cataract surgery) meaningfully improve function and quality of life for the specific affected organ system.
- Prognostic factors: Age at onset (later onset = better prognosis), genotype severity (allele combination — polymerase-domain/severe alleles trend toward worse, more systemic phenotype), presence/absence of liver involvement and early-onset seizures (poor prognostic markers, mainly relevant to the more severe ends of the POLG spectrum rather than classic arPEO), and degree of eventual systemic (ataxia-neuropathy-spectrum) evolution.
- Prognostic biomarkers: No validated molecular biomarker specific to PEOB1 progression was identified; general markers under study across POLG disease include serum/CSF lactate, GDF15/FGF21 (used more broadly as mitochondrial-disease biomarkers in other conditions), and mtDNA deletion burden on repeat muscle biopsy — the latter research-grade rather than clinically validated for prognosis specifically in this entity.
12. Treatment
No disease-modifying or curative therapy exists. Management is supportive, multidisciplinary, and focused on symptom management and monitoring for multisystem involvement, mirroring general mitochondrial-disease care.
Pharmacotherapy: - Coenzyme Q10 (ubiquinone/ubidecarenone) supplementation (100–600 mg/day in small studies) — general mitochondrial-supportive therapy with reported (limited-evidence) benefits including reduced serum lactate/pyruvate and possible partial functional improvement; not POLG-PEO-specific, extrapolated from broader mitochondrial-disease "mitochondrial cocktail" practice. NCIT term: treatment_term NCIT:C15986 (Pharmacotherapy); therapeutic_agent CHEBI (ubidecarenone/coenzyme Q10, CHEBI:46245). - Other components of the empiric "mitochondrial cocktail" (L-carnitine, riboflavin, alpha-lipoic acid, creatine) are used in general mitochondrial-myopathy practice with weak evidence; specific PEOB1 trial data were not identified. - Critical avoidance: Valproic acid / sodium divalproate are contraindicated across the POLG spectrum, including PEOB1, due to hepatotoxicity risk — this should be flagged as a therapeutic counter-indication rather than a treatment.
Pharmacogenomics: The principal pharmacogenomic relevance is the valproate-hepatotoxicity gene-drug interaction described above; POLG genotype should be checked (or at minimum strongly suspected clinically) before initiating valproate in any patient with unexplained epilepsy, ataxia, or PEO-like features.
Advanced therapeutics: No approved gene therapy, cell therapy, RNA-based therapy, or targeted molecular therapy exists specifically for POLG-PEO at the time of this research; deoxynucleoside-substrate replacement therapies under investigation for other mtDNA-maintenance disorders (e.g., TK2 deficiency) are not established for POLG-PEO. No relevant ClinicalTrials.gov interventional trials specific to PEOB1 were identified in this search (searches for POLG-PEO-specific trials did not surface active NCT-registered studies beyond general mitochondrial-disease natural-history/biomarker studies).
Surgical/interventional: - Ptosis surgery (levator resection, frontalis/brow suspension sling) — mainstay surgical intervention for functionally significant ptosis, but with meaningful risk of postoperative exposure keratopathy due to poor Bell's phenomenon/blink mechanics in this myopathic population; conservative under-correction is often favored. NCIT: procedure best captured under NCIT:C15329 (Surgical Procedure) or a more specific ophthalmic surgical term if available. - Strabismus surgery is used more cautiously and less frequently than in comitant strabismus, given the progressive, restrictive nature of the myopathy. - Cataract surgery for visually significant cataracts. - Cochlear implantation/hearing aids for sensorineural hearing loss (device-based, not curative of the underlying disease).
Supportive/rehabilitative: - Nonsurgical ptosis aids: ptosis crutches/props, Fresnel prisms for symptomatic misalignment/diplopia. - Scleral contact lenses for ocular surface protection/rehabilitation in advanced disease. - Physical therapy (resistance/aerobic exercise) to preserve muscle function, address proximal weakness, and (by analogy with mouse-model data showing exercise attenuates the Polg mutator progeroid phenotype) potentially support mitochondrial biogenesis, though direct human PEOB1 exercise-trial evidence was not identified. - Occupational therapy, speech-language therapy (for dysarthria/dysphagia when present in the ataxia-neuropathy-spectrum-evolved phenotype). - Genetic counseling (NCIT:C15240) for the patient and at-risk relatives given the 25%/50%/25% recessive recurrence-risk pattern.
Treatment strategy/monitoring: Regular multidisciplinary follow-up including periodic liver function testing (e.g., every 3 months per GeneReviews guidance, reflecting vigilance for hepatotoxic drug exposures and any evolving hepatic involvement), cardiac evaluation (ECG/echocardiogram) for conduction disease/cardiomyopathy surveillance, audiometry, ophthalmologic exam (cataract, corneal surface), and neurologic assessment for evolving ataxia/neuropathy — consistent with the "PEO-plus" surveillance philosophy given the phenotype's tendency to expand over time.
13. Prevention
- Primary prevention: Not applicable in the classic sense (this is a fixed germline genetic disease); the closest analog is avoidance of valproate exposure in genetically at-risk or POLG-confirmed individuals to prevent iatrogenic hepatic catastrophe — arguably the single most important "preventive" clinical action in this disease.
- Secondary prevention: Early diagnosis (via genetic testing once PEO is clinically suspected) enables anticipatory monitoring (liver, cardiac, audiologic, ophthalmologic) before complications become symptomatic/advanced.
- Tertiary prevention: Conservative surgical planning (ptosis under-correction) to prevent exposure keratopathy; proactive management of ataxia/neuropathy-related fall risk; cardiac monitoring to catch conduction disease before syncope/sudden events.
- Genetic screening:
- Carrier screening for POLG founder pathogenic variants (notably A467T) is commercially available (e.g., expanded carrier screening panels referenced by Myriad/Foresight) and increasingly incorporated given the relatively high (~1%) carrier frequency in European-ancestry populations.
- Prenatal testing/preimplantation genetic testing (PGT-M) is an option for couples where both partners are known carriers or where a proband's biallelic genotype has been characterized.
- Cascade family testing of siblings/relatives of an affected proband, given 25% recurrence risk for full siblings.
- Genetic counseling: Central to prevention strategy — informing reproductive decision-making, clarifying that arPEO is generally the mildest end of the POLG spectrum (important reassurance context) while noting variable expressivity means the same genotype can theoretically manifest more severely in a given individual.
- Public health/behavioral/immunization/prophylaxis: Not applicable — this is not an infectious, environmentally-driven, or vaccine-preventable disease.
14. Other Species / Natural Disease
- Taxonomy: No naturally occurring veterinary/companion-animal disease caused by POLG mutations was identified in this search (searches for OMIA POLG entries were not directly performed but no evidence of natural veterinary POLG-PEO disease surfaced in the broader literature reviewed). This appears to be a human-specific documented clinical entity at present, though POLG orthologs exist broadly across mammals (used experimentally, see Model Organisms below).
- Orthologous gene: Mouse Polg (MGI:1338062; NCBI Gene ortholog), highly conserved catalytic and exonuclease domains — the basis for the mouse model described below.
- Comparative biology: The core biochemical mechanism (Pol γ proofreading/catalytic function in mtDNA replication) is deeply evolutionarily conserved from yeast (MIP1, the yeast Pol γ ortholog) through zebrafish (polg1/polg2) to mammals, which is precisely why yeast, zebrafish, and mouse models (below) are informative surrogates for human POLG disease mechanism, even though naturally occurring animal disease has not been documented.
- Zoonotic potential/transmission: Not applicable — non-infectious, non-transmissible monogenic disease.
15. Model Organisms
Yeast: Saccharomyces cerevisiae MIP1 (Pol γ ortholog) mutants have been used to model POLG catalytic and exonuclease domain mutations, informative for basic replication-fidelity mechanism studies (per the 2025 Cell Death & Disease review "Model organisms in POLG-related disorders: insights from yeast to multicellular systems").
Zebrafish: - A stable CRISPR/Cas9-generated polg2 knockout zebrafish line (allele polg2^ia304^) recapitulates human POLG-disorder phenotypes: homozygous mutants show slower development, decreased viability, remarkable mtDNA depletion, altered mitochondrial network/dynamics, and reduced mitochondrial respiration (PMC11032366). While this specific model targets polg2 (the accessory-subunit ortholog) rather than polg1 directly, it is used as a platform for drug-treatment screening relevant to the broader POLG-disorder mechanism. - Zebrafish models more broadly are highlighted as useful for probing the neurological manifestations of POLG disease (encephalopathy, epilepsy, ataxia) given their amenability to behavioral and imaging assays.
Mouse — the "mtDNA mutator" model (most extensively characterized): - The classic Polg^D257A/D257A "mutator" mouse carries a proofreading-domain (exonuclease-dead) knock-in mutation, causing an ~2,500-fold increase in mtDNA point-mutation rate and marked linear-fragment/deletion accumulation (an 11-kb linear mtDNA fragment corresponding to most of the mtDNA major arc has been specifically characterized, mechanistically linking loss of exonuclease-mediated fragment degradation to deletion formation — directly relevant to the human PEOB1 deletion mechanism). - Phenotype recapitulation: This model reproduces a systemic premature-aging (progeroid) phenotype — accelerated sarcopenia, hearing loss, osteoporosis, hair graying/alopecia, thymic involution, testicular atrophy, cardiac hypertrophy, anemia, weight loss, and markedly shortened lifespan — overlapping substantially with several "PEO-plus" features seen in human POLG disease (hearing loss, cardiomyopathy, hypogonadism), though the mouse model's dominant aging-phenotype framing is broader than isolated human arPEO. - Mechanistic insight from the model: Muscle from mutator mice shows increased mitochondrial fission (elevated Fis1) and heightened autophagy, proposed to contribute to the sarcopenic phenotype — a plausible parallel to human myopathic muscle wasting. - Intervention data: Endurance exercise is the only reported intervention shown to attenuate the progeroid phenotype and extend healthspan/lifespan in this model — informing (by extrapolation, not direct trial) the rationale for exercise as supportive therapy in human patients. - Newer, refined mouse models (2025): A study titled "Modelling POLG mutations in mice unravels a critical role of POLγB in regulating phenotypic severity" (Nat Commun 2025) specifically dissects how the POLG2 (Pol γB) accessory subunit modulates phenotypic severity of POLG catalytic mutations — directly relevant to understanding genotype-phenotype variability in human disease. A separate 2025 bioRxiv-reported "inducible mtDNA mutator mouse model" adds temporal/spatial control, addressing the limitation that constitutive mutator mice can show embryonic lethality or phenotypes that are difficult to dissect tissue-specifically.
Model limitations: Existing constitutive POLG mouse mutants can present either embryonic lethality (very severe alleles) or comparatively mild/non-specific phenotypes, limiting fidelity to the specific adult-onset, tissue-restricted (extraocular-muscle-predominant) human PEOB1 phenotype and limiting utility for high-throughput drug screening — a limitation explicitly motivating development of the newer inducible and zebrafish models.
Research applications: These models collectively enable study of (a) the exonuclease-fragment-degradation mechanism of deletion formation, (b) POLG2/Pol γB modulation of severity, (c) tissue-specific bioenergetic failure and its downstream cellular consequences (mitochondrial fission/autophagy), and (d) candidate interventions (exercise; pharmacologic screening in zebrafish).
Resources: MGI (Mouse Genome Informatics) for Polg knock-in/flox alleles (e.g., the C57BL/6JCya-Polg^em1flox^ conditional model cataloged commercially); ZFIN for zebrafish polg1/polg2 alleles.
Summary of Key Ontology Term Suggestions for KB Curation
Table (click to expand)
| Domain | Suggested terms |
|---|---|
| Disease | OMIM:258450; ORPHA:254886; MONDO (verify exact PEOB1-specific term via OLS) |
| Gene | HGNC:9179 (POLG), lowercase hgnc:9179 per repo convention |
| Phenotypes (HP) | HP:0000508 Ptosis; HP:0000590 Ophthalmoplegia; HP:0000602 Ophthalmoparesis; HP:0003546 Exercise intolerance; HP:0001324 Muscle weakness; HP:0003202 Skeletal muscle atrophy; HP:0007141 Axonal sensory neuropathy; HP:0001251 Ataxia; HP:0001260 Dysarthria; HP:0000407 Sensorineural hearing loss; HP:0000518 Cataract; HP:0001634 Mitral valve prolapse; HP:0001638 Cardiomyopathy; HP:0002151 Increased serum lactate; HP:0003200 Ragged-red muscle fibers |
| Biological processes (GO) | GO:0006264 mitochondrial DNA replication; GO:0032042 mitochondrial DNA metabolic process; GO:0006281 DNA repair; GO:0006123 mitochondrial electron transport, cytochrome c to oxygen |
| Molecular function (GO) | GO:0003887 DNA-directed DNA polymerase activity; GO:0008310 single-stranded DNA 3'-5' exodeoxyribonuclease activity |
| Cell types (CL) | Skeletal/extraocular muscle fiber; CL:0000101 sensory neuron; CL:0000121 Purkinje cell (ataxia-affected patients) |
| Anatomy (UBERON) | UBERON:0002031 extraocular muscle; UBERON:0001134 skeletal muscle tissue; UBERON:0002037 cerebellum; UBERON:0001846 cochlea; UBERON:0000965 lens |
| Treatment (NCIT) | NCIT:C15986 Pharmacotherapy (CoQ10, avoid valproate); NCIT:C15329 Surgical Procedure (ptosis repair); NCIT:C15240 Genetic Counseling; NCIT:C15315 Rehabilitation |
| Chemicals (CHEBI) | CHEBI:46245 ubidecarenone (coenzyme Q10); note valproate/valproic acid as a contraindicated agent, not a treatment |
Notes on Evidence Gaps
- Precise population prevalence/incidence figures specific to PEOB1 (as opposed to POLG-disease overall, or CPEO as a clinical umbrella) were not found — flag as "prevalence unknown" per Orphanet, consistent with rare-disease reporting limits.
- No dedicated QOL instrument data, omics (transcriptomic/proteomic) human-tissue datasets, or active interventional clinical trials specific to PEOB1 were identified in this search — likely reflects genuine absence of such studies for this specific rare recessive subtype rather than a search limitation, though a targeted ClinicalTrials.gov/GEO query would be a reasonable follow-up before concluding definitively.
- Exact PMIDs for several foundational papers (Van Goethem et al. 2001 Nat Genet; the Hikmat et al. 2020 JIMD cohort, PMID 32068908; Basu et al. 2018 Nat Commun on exonuclease/fragment degradation) should be independently re-verified against PubMed/cached abstracts before use as curated evidence snippets, per standard dismech verification SOP — this report should be treated as a lead-generation document, not pre-verified evidence.
Sources: - OMIM #258450 — PEOB1 - OMIM *174763 — POLG - OMIM #616479 — PEOB2 (RNASEH1) - OMIM #617070 — PEOB4 (DGUOK) - Orphanet — Autosomal recessive PEO (ORPHA:254886) - GeneReviews — POLG-Related Disorders (NBK26471) - MalaCards — PEOB1 - MedlinePlus — Progressive external ophthalmoplegia - MedlinePlus — POLG gene - Nature Genetics — Van Goethem et al., Mutation of POLG associated with PEO - Nature Communications — POLG exonuclease degrades linear DNA fragments precluding deletions - Nature Communications — POLγB regulates phenotypic severity in mouse POLG models (2025) - Cell Death & Disease — Model organisms in POLG-related disorders - PMC — Zebrafish polg2 knockout recapitulates POLG-disorders - Journal of Inherited Metabolic Disease — Hikmat et al. 2020, simplifying POLG clinical classification - Genetics in Medicine — natural history of early-onset POLG disease - ClinVar — POLG c.1399G>A (p.Ala467Thr) - European Journal of Human Genetics — Abundance of POLG mutations from single ancient founders - PubMed — Estimation of carrier frequencies via gnomAD for carrier screening - PMC — RNASEH1 mutations and PEO with multiple mtDNA deletions - MedLink Neurology — Chronic progressive external ophthalmoplegia - Medscape — CPEO Treatment & Management - PMC — Treatment of ptosis in chronic progressive external ophthalmoplegia