Sorbitol Dehydrogenase Deficiency

Sorbitol Dehydrogenase (SORD) Deficiency — Comprehensive Research Report

2026-08-23
Claude Code MONDO:0030055 Model: claude-haiku-4-5-20251001, claude-sonnet-5 20 citations

Sorbitol Dehydrogenase (SORD) Deficiency — Comprehensive Research Report

1. Disease Information

Overview. Sorbitol dehydrogenase (SORD) deficiency — most commonly called CMT‑SORD, and clinically designated autosomal recessive distal hereditary motor neuronopathy‑8 (HMNR8) or "SORD deficiency with peripheral neuropathy (SORDD)" — is an autosomal recessive, adult/adolescent‑onset, length‑dependent peripheral neuropathy caused by biallelic loss‑of‑function variants in SORD, the gene encoding sorbitol dehydrogenase. It was first described as a distinct disease entity in 2020 and has rapidly emerged as the single most common autosomal recessive cause of axonal Charcot‑Marie‑Tooth disease type 2 (CMT2) and distal hereditary motor neuropathy (dHMN) (Cortese et al., Nat Genet 2020, PMID 32367058). Loss of the enzyme causes accumulation of the polyol sorbitol, which is directly neurotoxic to peripheral motor (and to a lesser degree sensory) axons — the same biochemical pathway implicated in diabetic peripheral neuropathy, giving the discovery broader relevance to diabetes-related neuropathy research.

Key identifiers: - OMIM phenotype: 618912 — Neuronopathy, distal hereditary motor, autosomal recessive 8 - OMIM gene: 182500 — SORBITOL DEHYDROGENASE; SORD (gene symbol SORD, chromosome 15q26.1) - MONDO: MONDO:0030055 (sorbitol dehydrogenase deficiency with peripheral neuropathy) - MedGen: C5394466 - DOID (RGD browser): DOID:9006739 - Orphanet / GeneReviews / PanelApp (Genomics England "Hereditary neuropathy" panel): SORD is listed as a validated hereditary-neuropathy gene

Synonyms: CMT‑SORD; SORD deficiency; SORD-related neuropathy; sorbitol dehydrogenase deficiency with peripheral neuropathy (SORDD); distal hereditary motor neuronopathy, autosomal recessive 8 (HMNR8); axonal CMT2 due to SORD deficiency.

Evidence basis. The disease description rests almost entirely on aggregated multi-center human clinical cohorts (the largest to date comprising 144 patients from 126 families across 43 centers, Brain 2025, PMID pending/DOI 10.1093/brain/awaf021), supplemented by mechanistic work in patient-derived fibroblasts, iPSC-derived motor neurons, Drosophila, and Sord‑deficient rat and mouse models (see Sections 6 and 15).


2. Etiology

Disease causal factor: Purely genetic — biallelic (homozygous or compound heterozygous) loss-of-function variants in SORD on chromosome 15q26.1, which abolish or severely reduce sorbitol dehydrogenase enzymatic activity.

Genetic risk factors: - The dominant pathogenic allele worldwide is the frameshift c.757delG (p.Ala253GlnfsTer27), accounting for ~87% of all disease alleles and found homozygously in ~78% of the large genotype–phenotype cohort (Brain 2025). It behaves as a founder/recurrent variant across many ancestries (European, Middle Eastern, East Asian) rather than a single-population founder effect, likely because it lies within the region shared with the SORD2P pseudogene. - Second most common allele: c.458C>A (p.Ala153Asp) (~8% of alleles). - A third recurrent pathogenic mechanism is a gene–pseudogene (SORD/SORD2P) inversion, found in ~9% of SORD-CMT patients and disproportionately (75%) among cases where short-read sequencing detected only one pathogenic variant — i.e., cryptic biallelism that standard exome/panel testing misses. - Population carrier frequency for c.757delG has been estimated at ~0.46% in Chinese controls (3/650) (Chen et al., PMC8607551) and broadly ~0.5–1% in several populations studied, consistent with SORD deficiency being present at a frequency (~1 in 100,000 homozygotes, or higher when accounting for compound heterozygotes) that makes it one of the most common recessive neuropathies overall. - Overall genotype spectrum: 82% biallelic nonsense/splicing/structural (predicted null) variants; 18% carry at least one missense allele.

Environmental / non-genetic risk factors: None established as disease-causing; this is a monogenic Mendelian disorder. However, because SORD acts in the same polyol pathway implicated in diabetic neuropathy (see Section 6), hyperglycemia is a biologically plausible modifier of polyol flux, though no formal gene–environment interaction study in SORD-deficient patients has been published.

Sex as a modifier: In the large genotype-phenotype cohort, male sex was significantly associated with greater severity of distal lower-limb (plantar flexion) weakness and a larger rate of decline in dorsiflexion strength over time — a modifier of expressivity rather than a causal factor (Brain 2025). The rat model similarly showed males more affected on motor nerve conduction velocity.

Protective factors: None specifically documented for humans. In model systems, pharmacologic inhibition of the upstream enzyme aldose reductase (AKR1B1) — which prevents glucose from ever being converted to sorbitol — is protective (see Sections 6 and 12).

Gene–environment interactions: Not formally studied in SORD-deficient humans, though the shared biochemistry with diabetic polyol-pathway neurotoxicity is repeatedly invoked as a rationale for studying SORD deficiency as a "genetic model" of diabetic neuropathy pathophysiology.


3. Phenotypes

Phenotype category: Predominantly motor/sensorimotor peripheral neuropathy signs and symptoms, plus laboratory/biochemical abnormalities (elevated serum/urine sorbitol and xylitol).

Core clinical phenotypes (from the 144-patient genotype–phenotype cohort, Brain 2025, and the original 45-patient Cortese et al. 2020 cohort):

Table (click to expand)
Phenotype Frequency / detail Suggested HPO term
Distal lower-limb weakness (foot dorsiflexion) MRC ≤3 in 53% of patients; declines ~5%/year HP:0009053 (Distal lower limb muscle weakness)
Foot drop Very common presenting sign HP:0001772 (Foot drop)
Pes cavus Common structural foot deformity; 79% report foot deformity HP:0001761 (Pes cavus)
Distal muscle atrophy (legs) Common HP:0003724 (Distal amyotrophy)
Absent/decreased deep tendon reflexes Common HP:0001315 / HP:0001284
Foot plantar flexion weakness Impaired in 78%; MRC ≤3 in 33% HP:0009053
Distal upper-limb (hand) weakness/dexterity impairment ~50% motor involvement; hand dexterity impaired in 37%, onset ~8 yrs after gait symptoms HP:0009830 / HP:0007340
Sensory loss (pinprick/vibration) Reported by <1/3 patients; upper-limb sensory nerve action potentials (SNAPs) abnormal in 76% vs. only 27% in lower limbs (a distinctive "inverse" length-independent sensory pattern) HP:0003676 (Progressive sensory neuropathy)
Distal tremor 28% HP:0025278
Gait difficulty / difficulty running Walking difficulty 85%; running difficulty 88% HP:0002136 / HP:0001288
Elevated serum sorbitol 14.7 ± 4.9 mg/L (patients) vs. 0.07 ± 0.06 mg/L (controls), p<0.001 (fasting status–independent, storage-stable) (biochemical, not HP-coded)
Elevated urine sorbitol and xylitol Novel 2025 biomarker for screening/diagnosis and treatment monitoring (Neurology 2025, PMID 41223342) (biochemical)

Phenotype characteristics: - Age of onset: Typically second decade of life (childhood/adolescent onset); mean reported onset ~14 years (range 6–17) in the Chinese cohort, though many patients report subtle antecedent findings (foot deformity, poor athletic performance in school — 46%) that predate formal diagnosis. Mean age at enrollment/diagnosis in the largest cohort was 40.9 ± 14.8 years, reflecting substantial diagnostic delay (the disease was undiscovered until 2020). - Severity: Predominantly mild-to-moderate: baseline CMT Examination Score (CMTES) 6.09 ± 3.7; 72% mild (CMTES 0–7), 26% moderate (8–16), <1% severe (17–28). CMT Neuropathy Score (CMTNS) mean 12.2 (range 9–15) in an independent cohort. - Progression: Slowly progressive over decades. Foot dorsiflexion and plantar flexion strength decline significantly with age (p<0.001 both) and on longitudinal follow-up (mean 6.9 ± 7.4 years); dorsiflexion declines ~5%/year. Most patients remain independently ambulatory even late in life; only ~25% require ankle-foot orthoses (typically starting in their 30s), and only a minority need canes/crutches (13/144) or wheelchairs (2/144). - Clinical classification split: Sensorimotor CMT2 phenotype in 60% of patients vs. pure motor dHMN phenotype in 40% — i.e., SORD deficiency spans the CMT2/dHMN phenotypic continuum rather than mapping to a single discrete syndrome. - Quality of life: No dedicated EQ-5D/SF-36 study identified in the literature for CMT-SORD specifically; functional impact is documented indirectly via patient-reported walking/running difficulty rates above and use of assistive devices. General CMT quality-of-life literature (not disease-specific) applies.


4. Genetic/Molecular Information

Causal gene: SORD (HGNC symbol SORD; OMIM 182500), located at 15q26.1, encoding a 357-amino-acid cytosolic enzyme, sorbitol dehydrogenase (EC 1.1.1.14), a member of the medium-chain zinc-dependent alcohol dehydrogenase family.

Pathogenic variant classes: - Frameshift/nonsense (predicted null): c.757delG (p.Ala253GlnfsTer27) — the dominant allele (~87% of alleles); ClinVar RCV001194463 lists this as Pathogenic for both "Sorbitol dehydrogenase deficiency with peripheral neuropathy" and "Neuronopathy, distal hereditary motor, autosomal recessive 8." - Missense: c.458C>A (p.Ala153Asp) (~8% of alleles); additional missense variants reported in Chinese cohorts include c.731C>T (p.Pro244Leu), c.776C>T (p.Ala259Val), c.851T>C (p.Leu284Pro) (PMC8607551). - Structural variant: SORD/SORD2P gene–pseudogene inversion — a structural rearrangement between SORD and its highly homologous pseudogene, found in ~9% of patients and the third most common pathogenic allele class; especially important because these are frequently missed by short-read exome/panel sequencing.

A critical diagnostic complication — the SORD2P pseudogene: SORD has a paralogous pseudogene, SORD2P, that shares extremely high sequence identity (the recurrent c.757delG mutation position corresponds to a sequence present on ~95% of pseudogene-derived chromosomes), causing mis-mapping of short-read sequencing reads and complicating both variant calling and cis/trans phasing of biallelic variants. This is considered a major reason CMT-SORD was not identified as a distinct entity until 2020 despite affecting a substantial fraction of unsolved CMT2/dHMN cases. Long-read (Oxford Nanopore) sequencing has been shown to resolve SORD from SORD2P and correctly phase compound heterozygous/inversion genotypes (Cortese et al., 2022, PMID 35224818).

Functional consequence: All characterized pathogenic variants result in loss of function of sorbitol dehydrogenase. Patient-derived fibroblasts homozygous for c.757delG show complete absence of SORD protein by western blot and a ~10-fold increase in intracellular sorbitol compared with controls.

Variant origin/zygosity: Germline, autosomal recessive; homozygous (most common, ~78%) or compound heterozygous (~18%, most commonly c.757delG + a second allele) or complex (inversion) genotypes.

Modifier genes: No confirmed modifier loci; sex (male) modifies clinical severity (see Section 2), and residual polyol-pathway flux (via AKR1B1, aldose reductase, the upstream enzyme) likely modulates disease severity biochemically, though this has not been formally shown as a genetic modifier in patients.

Epigenetic information / chromosomal abnormalities: None reported; this is a straightforward Mendelian enzymopathy without known epigenetic contribution.

Population allele frequency: Carrier frequency of c.757delG ~0.46% in a Chinese control cohort (3/650) and comparably elevated in other studied populations, consistent with SORD deficiency being among the most common recessive neuropathies worldwide (estimated >3,000 cases in the USA alone; Brain 2025).


5. Environmental Information

SORD deficiency is a purely monogenic disorder; no environmental, toxin, occupational, dietary, or infectious trigger is required for disease manifestation. The relevant "environmental" biochemistry is entirely endogenous — dietary/metabolic glucose flux through the polyol pathway generates the substrate (sorbitol) that a deficient SORD enzyme cannot clear. No formal studies of dietary sorbitol/fructose intake as a disease modifier in SORD-deficient patients were identified, though this is a biologically plausible area given that the disease-defining biomarker (serum/urine sorbitol) derives from the same pathway implicated in dietary sugar-alcohol metabolism. No infectious agent is implicated.


6. Mechanism / Pathophysiology

Molecular pathway — the polyol pathway (sorbitol–aldose reductase pathway): 1. Aldose reductase (AKR1B1) — the first, rate-limiting enzyme — catalyzes the NADPH-dependent reduction of glucose → sorbitol. 2. Sorbitol dehydrogenase (SORD) — the second enzyme — normally converts sorbitol → fructose (NAD⁺-dependent). 3. When SORD is deficient, sorbitol cannot be cleared and accumulates intracellularly and extracellularly, because sorbitol is a polar polyol that diffuses poorly across cell membranes.

Causal chain (upstream → downstream): Biallelic SORD loss-of-function variant → complete/near-complete loss of SORD enzyme activity (molecular scale) → impaired conversion of sorbitol to fructose, with sorbitol trapped intracellularly (cellular scale) → osmotic/hyperosmolar stress within Schwann cells and neurons, producing pathognomonic "ballooned" myelin sheaths with bubbly disintegration (a feature also seen in galactosemic neuropathy models, implicating a shared polyol-osmotic mechanism) → secondary axonal degeneration and demyelination, disproportionately affecting motor axons (motor-predominant length-dependent axonopathy, tissue scale) → progressive distal weakness, foot drop, and gait impairment (organism scale).

Cellular processes implicated (from Drosophila and iPSC-neuron mechanistic studies, PMID 37014713, JCI Insight 2023): - Osmotic/hyperosmolar stress from sorbitol accumulation (primary proposed mechanism) - Synaptic degeneration — Sord-deficient flies show progressive motor and eye-neuron synaptic loss - Mitochondrial/energetic dysfunction — reduced brain ATP in Sord-deficient flies, restored by aldose-reductase inhibition - Oxidative stress — elevated reactive oxygen species (ROS) across CNS, ventral nerve cord (VNC), and muscle tissue in flies; reduced by treatment - NADPH/NAD⁺ cofactor imbalance — general polyol-pathway biochemistry (well documented in the diabetic-neuropathy literature via AKR1B1) diverts NADPH/NAD⁺ redox cofactors, contributing to oxidative and inflammatory injury (NF-κB activation reported in diabetic polyol pathway studies)

Biochemical abnormality: Loss-of-function enzyme deficiency (sorbitol dehydrogenase), the archetypal "enzyme deficiency" mechanism, directly analogous to inborn errors of metabolism, but manifesting as a chronic length-dependent peripheral neuropathy rather than an acute metabolic crisis.

Tissue damage mechanism: Osmotic myelin injury with secondary Wallerian-type axonal degeneration, evidenced histologically (rat model, Brain 2024, PMID pending, DOI academic.oup.com/brain/article/147/9/3131/7636456) by: - Ballooned/enlarged myelin sheaths around otherwise intact axons - Degenerating and demyelinated axons - Thinly myelinated fibers with increased g-ratios - Clusters of regenerating axons - Predominant loss of large motor fibers, especially distally, with relative sparing of sensory fibers

Molecular profiling: Elevated serum neurofilament light chain (NfL) in the rat model serves as a biomarker of ongoing axonal degeneration (present but not correlated with severity). Direct human transcriptomic/proteomic/metabolomic profiling of nerve tissue has not been widely reported; the principal "omics" readout used clinically and mechanistically is the targeted metabolomic (polyol) biomarker panel — serum and urine sorbitol and xylitol.

Suggested GO terms: - GO:0006062 (sorbitol catabolic process) - GO:0019853 (L-ascorbic acid biosynthetic process – polyol pathway crosslink, if relevant) - GO:0006970 (response to osmotic stress) - GO:0007422 (peripheral nervous system development, for context) - GO:0043524 (negative regulation of neuron apoptotic process — inverse direction relevant to axonal degeneration)

Suggested CL terms: - CL:0000011 (Schwann cell) — site of myelin ballooning - CL:0000100 (motor neuron) — preferentially affected cell type - CL:0000098 (sensory neuron) — relatively spared


7. Anatomical Structures Affected

Organ level: - Primary: Peripheral nervous system — motor and (to a lesser extent) sensory peripheral nerves, particularly the longest axons (length-dependent pattern: lower limb before upper limb). - Secondary: Skeletal muscle (secondary distal muscle atrophy from denervation); no cardiac, hepatic, renal, or CNS involvement reported as part of the core phenotype. - Body system: Peripheral/neuromuscular system (UBERON:0000010 peripheral nervous system).

Tissue/cell level: - Motor axons (CL:0000100) — predominantly affected - Schwann cells / myelin sheath (CL:0000011) — site of the pathognomonic osmotic "ballooning" - Sensory axons (CL:0000098) — relatively spared, though upper-limb sensory nerve action potentials are frequently abnormal

Subcellular level: Cytosol (GO:0005829) — SORD is a cytosolic enzyme; myelin sheath compartment (osmotic swelling).

Localization: Distal lower-limb nerves (peroneal, tibial) first and most severely; distal upper-limb (ulnar, median) nerves later (~8 years after leg symptom onset on average). Bilateral, symmetric distribution typical of an inherited length-dependent axonopathy — no lateralization reported.

Suggested UBERON terms: UBERON:0001021 (nerve), UBERON:0003714 (peripheral nerve fascicle equivalents / peroneal nerve UBERON:0009629, tibial nerve UBERON:0009028), UBERON:0001134 (skeletal muscle tissue, secondary).


8. Temporal Development

Onset: Typically childhood/adolescent onset (second decade of life), though many patients have subclinical antecedents (foot deformity, poor school athletic performance) recognized only in retrospect. Onset pattern is insidious/chronic, not acute.

Progression: Slowly progressive over decades — dorsiflexion strength declines ~5% per year on quantitative testing; statistically significant longitudinal decline in both dorsiflexion and plantar flexion strength over a mean follow-up of ~7 years. Disease severity remains predominantly mild-to-moderate even in adulthood (72% mild by CMTES); severe disease is rare (<1%).

Disease course pattern: Progressive but not typically disabling — most patients remain ambulatory without assistive devices for decades; a minority progress to requiring ankle-foot orthoses (usually starting in their 30s) or, rarely, canes/wheelchairs.

Critical periods: No formally defined therapeutic window has been established, though the ongoing govorestat trials are testing intervention in symptomatic adults; earlier (pre-symptomatic or minimally symptomatic) intervention is hypothesized to be more effective given the slow, cumulative nature of axonal loss, but this is not yet proven.


9. Inheritance and Population

Inheritance pattern: Autosomal recessive (biallelic pathogenic variants required). Nearly all reported cases are sporadic with no family history, consistent with recessive inheritance and relatively high carrier frequency rather than consanguinity-driven clustering in most populations studied. Penetrance appears to be high/complete among individuals with biallelic null variants, though expressivity (severity, motor- vs. sensorimotor-predominant phenotype) is variable.

Epidemiology: - Prevalence: Estimated to be the most common autosomal recessive axonal peripheral neuropathy; accounts for ~7–9% of dHMN and CMT2 cases in cohorts studied. At least 3,000 estimated cases in the USA alone. General hereditary neuropathy prevalence is ~1:2500; CMT-SORD represents a substantial fraction of the previously "genetically unsolved" cases within that group. - Carrier frequency: ~0.46–1% for the common c.757delG allele across multiple populations studied (Chinese, European), suggesting the disorder is under-ascertained rather than truly rare.

Genetic anticipation: Not reported/expected — this is a coding sequence loss-of-function disorder, not a repeat-expansion disease.

Germline mosaicism, founder effects, consanguinity: The c.757delG allele behaves as a recurrent (likely pseudogene-conversion-mediated) variant across many distinct ancestries rather than a classical single-founder mutation; no strong consanguinity signal has been reported (most cases sporadic, non-consanguineous).

Population demographics: - Geographic/ancestry distribution (144-patient cohort): European ancestry 75%, Middle Eastern 11%, East Asian 9%, other ancestries the remainder — i.e., globally distributed, not restricted to a single ethnic group. - Sex ratio: Cohorts show a male excess in ascertainment (99 males [69%] vs. 45 females [31%] in the largest cohort), and males show greater clinical severity, though whether this reflects true sex-differential penetrance/expressivity or ascertainment bias (e.g., under-recognition in females) is not fully resolved. - Age distribution: Diagnosed patients span 15–75 years (mean enrollment age 40.9 ± 14.8 years), reflecting long diagnostic delay typical of a disease only characterized in 2020.


10. Diagnostics

Clinical tests: - Biochemical (blood): Serum sorbitol — markedly and consistently elevated in affected patients (14.7 ± 4.9 mg/L vs. 0.07 ± 0.06 mg/L in controls, >100-fold difference in some series), stable regardless of fasting status and storage conditions, making it a robust, easily obtained diagnostic and monitoring biomarker. - Biochemical (urine): Urine sorbitol and xylitol (measured by gas chromatography–mass spectrometry) — a newer (2025) complementary/screening biomarker; xylitol elevation is specifically noted as adding diagnostic specificity (Neurology 2025, PMID 41223342; Mayo Clinic Laboratories offers a clinical urine SORD sorbitol/xylitol assay). - Electrophysiology: Nerve conduction studies show an axonal (non-demyelinating) sensorimotor or pure motor neuropathy pattern; characteristic asymmetric sensory involvement (upper-limb SNAPs abnormal far more often — 76% — than lower-limb SNAPs — 27%), a distinctive pattern that can raise suspicion for SORD deficiency specifically. - Nerve biopsy: Not part of routine diagnostic workup in humans but in animal models shows the pathognomonic ballooned/swollen myelin sheaths.

Genetic testing: - Overview: Molecular confirmation of biallelic SORD pathogenic variants is definitive, but is technically complicated by the SORD2P pseudogene, which causes false negatives/mismapping in standard short-read next-generation sequencing (exome or CMT gene panels). - Panels/exome/genome sequencing: SORD should be included in CMT2/dHMN gene panels and CMT-specific exome analyses; however, careful bioinformatic handling of the pseudogene region (or orthogonal confirmation) is needed. - Long-read sequencing (Oxford Nanopore): Increasingly used/recommended to resolve SORD from SORD2P, correctly phase compound heterozygous genotypes, and detect the SORD/SORD2P inversion allele class that short-read methods miss (up to 9% of cases, and the majority of "single-variant-detected" cases on short-read testing). - Single-gene/targeted testing: Appropriate when biochemical testing (elevated sorbitol) has already localized the defect.

Clinical criteria / differential diagnosis: SORD deficiency should be considered in any patient with an apparently sporadic, axonal, length-dependent CMT2/dHMN phenotype of childhood/adolescent onset, especially when standard CMT gene panels are negative — historically this represented a large "genetically unsolved" subgroup. Differential diagnosis includes other AR axonal CMT2/dHMN genes (e.g., HSPB1, GDAP1, TRPV4), and, when relevant, acquired causes of length-dependent neuropathy (diabetic neuropathy, toxic neuropathy) — notably, the biochemical overlap with the polyol pathway means clinicians should distinguish primary genetic SORD deficiency from the polyol-pathway hyperactivity seen in poorly controlled diabetes.

Screening: No population newborn or carrier screening program currently exists for SORD, given its relatively recent (2020) disease-gene establishment, but given the estimated carrier frequency (~0.5–1%) this may become relevant to hereditary neuropathy carrier panels in the future.


11. Outcome/Prognosis

Survival/mortality: No excess mortality has been reported; SORD deficiency is not known to shorten lifespan. No formal survival/mortality studies exist (the disease is not classically life-limiting, being restricted to peripheral motor/sensory nerve involvement).

Morbidity and function: - Most patients experience lifelong, slowly progressive distal weakness and gait impairment but retain independent ambulation for decades. - Functional impact: 85% report walking difficulty, 88% report running difficulty; ~25% eventually require ankle-foot orthoses (typically starting in their 30s); only a small minority (13/144 in the largest cohort) require canes/crutches, and very few (2/144) require wheelchairs. - No formal quality-of-life instrument (EQ-5D, SF-36) data specific to CMT-SORD were identified in the literature reviewed; general CMT disability metrics (CMTES/CMTNS) are used instead, and CMTES shows most patients (72%) fall in the "mild" category.

Disease course: Chronic and progressive but generally compatible with normal daily functioning through mid-adulthood; no described spontaneous remission. Complications are primarily musculoskeletal/orthopedic (foot deformity, gait abnormality secondary to weakness) rather than systemic.

Prognostic factors: Male sex associates with greater severity and faster progression of distal weakness. Genotype (homozygous null vs. compound heterozygous with a missense allele) may modify severity, though this has not been rigorously stratified in the literature reviewed. Elevated serum NfL (in the rat model) reflects ongoing axonal degeneration and is being explored as a prognostic/monitoring biomarker analogous to its use in other neurodegenerative conditions.


12. Treatment

Investigational disease-modifying pharmacotherapy — Govorestat (AT-007): - Mechanism: A next-generation aldose reductase (AKR1B1) inhibitor (IC50 ≈ 100 pM), CNS-penetrant. By inhibiting the upstream enzyme that converts glucose to sorbitol, govorestat reduces the total flux of sorbitol production, thereby lowering intracellular/circulating sorbitol despite the downstream SORD block — an elegant "block the tap rather than fix the drain" therapeutic logic. - Preclinical evidence: In Drosophila Sord-deficient models, govorestat normalized intracellular sorbitol, restored brain ATP, reduced ROS across CNS/VNC/muscle, and dramatically improved motor performance (climbing speed increased from 3.0 mm/s to 11.7 mm/s) and eye-phenotype degeneration (PMID 37014713, JCI Insight 2023). In patient fibroblasts, govorestat reduced sorbitol from 3.95 ng/µg protein to 0.21 ng/µg protein. - Clinical trial: The INSPIRE trial (Phase 2/3, NCT05397665) is evaluating govorestat in CMT-SORD. - 12-month interim analysis (2024): Statistically significant reduction in blood sorbitol; a significant correlation between sorbitol reduction and the composite CMT-FOM clinical endpoint (10-meter walk-run, 4-stair climb, sit-to-stand, 6-minute walk, dorsiflexion strength; p=0.05). Govorestat was safe and well tolerated with adverse-event rates similar to placebo. - 18–24 month follow-up (2025): Sustained reduction in blood sorbitol; sustained improvement in the CMT-Health Index; MRI-based lower-limb muscle-fat fraction showed a significant difference at 24 months, suggesting a slowing of disease progression, though the pre-specified primary clinical endpoint did not reach statistical significance at final analysis despite favorable trends. - Regulatory status (as of late 2025): Not yet FDA-approved for CMT-SORD; Applied Therapeutics met with the FDA (Type C meeting, Q3 2025) to discuss a potential regulatory pathway (including possible accelerated approval), with a submission strategy still under determination as of the most recent public updates.

Supportive/symptomatic care (standard of care today, in the absence of an approved disease-modifying therapy): - Physical therapy and rehabilitation for gait training and strength maintenance (NCIT:C15302 Physical Therapy) - Ankle-foot orthoses (AFOs) for foot drop (NCIT:C49236 Therapeutic Procedure / orthotic management) - Orthopedic management of secondary foot deformity (pes cavus) when indicated (NCIT:C16186 Orthopedic Surgical Procedure) - Genetic counseling for affected families given autosomal recessive inheritance (NCIT:C15240 Genetic Counseling) - General symptomatic CMT management (pain control, occupational therapy for hand involvement) as per general hereditary neuropathy guidelines

No approved gene therapy, enzyme replacement, or targeted molecular therapy currently exists for SORD deficiency; govorestat (aldose reductase inhibition) represents the leading investigational approach and is the only disease-specific pharmacotherapy in late-stage clinical development.

Suggested NCIT term for govorestat mechanism class: NCIT:C15986 (Pharmacotherapy) as treatment_term, with therapeutic_agent bound to govorestat/AT-007 (CHEBI/NCIT term to be confirmed at curation time) and therapeutic_modality: SMALL_MOLECULE.


13. Prevention

Primary prevention: Not applicable in the traditional sense (no modifiable environmental cause); the only "primary prevention" avenue is reproductive genetic counseling and carrier screening for couples with a family history or known carrier status, given the relatively high population carrier frequency (~0.5–1%) — prenatal or preimplantation genetic diagnosis is theoretically available once a family's causative variants are known, though not specifically reported as routine practice for this recently characterized disease.

Secondary prevention / early detection: Because SORD deficiency was only characterized in 2020, there is no established population screening program. However, biochemical screening (serum or urine sorbitol/xylitol) in patients with an unsolved axonal CMT2/dHMN phenotype functions as an effective secondary/case-finding strategy, given the biomarker's high sensitivity and specificity and ease of measurement compared to resolving the SORD2P pseudogene by genetic testing alone.

Tertiary prevention: Early diagnosis enables anticipatory orthopedic/rehabilitative management (AFOs, physical therapy) to minimize gait-related morbidity and secondary complications (falls, joint deformity), and positions patients for potential future disease-modifying therapy (govorestat) once approved.

Genetic counseling: Recommended for families of affected individuals given autosomal recessive inheritance and the relatively high carrier frequency in the general population.

Prophylaxis: No specific prophylactic pharmacotherapy exists; if govorestat is eventually approved, early (pre-severe) treatment initiation would represent a form of tertiary/disease-course-modifying prevention.


14. Other Species / Natural Disease

No naturally occurring SORD deficiency has been reported in companion animals or wildlife in the literature reviewed. All non-human data derive from engineered/induced models (see Section 15) rather than spontaneous veterinary disease. The polyol pathway itself (AKR1B1/SORD) is broadly conserved across mammals and is best known outside SORD deficiency in the context of diabetic complications (retinopathy, neuropathy, nephropathy) in both human diabetic patients and diabetic animal models — a related but mechanistically distinct (acquired hyperglycemia-driven vs. genetic enzyme-deficiency-driven) disease context.


15. Model Organisms

Table (click to expand)
Model Type Key findings Fidelity / limitations
Drosophila Sord-ortholog loss-of-function Invertebrate genetic model Progressive synaptic degeneration, motor impairment (climbing assay), eye-neuron degeneration; ATP depletion; elevated ROS across CNS/VNC/muscle. Aldose reductase inhibition (govorestat/AT-007) normalized sorbitol and dramatically improved motor and eye phenotypes (PMID 37014713; Cortese et al. 2020, PMID 32367058) High utility for rapid mechanistic and drug-screening work; limited translational fidelity for mammalian peripheral nerve myelin biology
Patient-derived fibroblasts Human primary cell (in vitro) Complete loss of SORD protein; ~10-fold increase in intracellular sorbitol; used to validate govorestat's sorbitol-lowering effect (3.95 → 0.21 ng/µg protein) Directly human, but not neuronal/myelinating tissue
iPSC-derived motor neurons (patient-derived) Human cellular model Used alongside fibroblasts in mechanistic/drug studies of sorbitol accumulation and neurotoxicity (PMID 37014713) Captures motor-neuron-specific biology; 2D in vitro system, lacks myelinating Schwann cell/axon architecture
Naturally occurring Sord-deficient mouse (splice variant) Rodent (spontaneous hypomorphic allele) Sorbitol accumulation present, but no motor phenotype and no significant change in motor nerve conduction velocity Fails to recapitulate the human motor phenotype — an important negative/translational-mismatch result, indicating this particular mouse allele is an incomplete model
New CRISPR Sord knockout mouse (Sleigh lab, UCL; Muscular Dystrophy UK–funded, ongoing) Rodent (engineered null) Reported to show sorbitol accumulation in motor neurons with resulting muscle weakness, more faithfully recapitulating the human phenotype than the earlier splice-variant mouse Ongoing/actively developed model; full published phenotypic characterization not yet available in the literature reviewed
Sord⁻/⁻ rat Rodent (engineered null) Best-characterized rodent model to date. Motor-predominant neuropathy emerging ~7 months of age; serum sorbitol ~7-fold elevated vs. WT; CSF sorbitol ~30-fold higher than serum (suggesting independent CNS/PNS sorbitol handling); elevated serum neurofilament light chain (axonal injury biomarker); decreased motor nerve conduction velocity (more pronounced in males); abnormal hindlimb gait; nerve pathology showing degenerating/demyelinated axons, thinly myelinated fibers with increased g-ratio, regenerating axon clusters, and pathognomonic ballooned ("bubbly") myelin sheaths (also seen in galactosemic neuropathy); motor axons predominantly affected with relative sparing of sensory nerves and normal pain sensation (Brain 2024, DOI 10.1093/brain/awae170-region, full text at academic.oup.com/brain/article/147/9/3131/7636456) High fidelity for the motor-predominant axonal phenotype and the osmotic-myelin-injury mechanism; study limitation: electrophysiology not performed longitudinally across multiple ages, and histology was assessed only at advanced ages (70–85 weeks)

Applications: These models collectively support (1) mechanistic dissection of the osmotic-stress/myelin-ballooning hypothesis, (2) biomarker development (serum/CSF sorbitol, serum NfL), and (3) preclinical efficacy testing of aldose reductase inhibitors (govorestat), directly informing the ongoing human INSPIRE trial.

Resources: Alliance of Genome Resources, MGI (mouse), RGD (rat) for strain/allele tracking; no SORD-specific zebrafish, C. elegans, or yeast disease models were identified in this search.


Summary Table of Key Ontology Term Suggestions

Table (click to expand)
Domain Suggested term
Disease MONDO:0030055; OMIM:618912
Gene HGNC SORD (chr15q26.1); OMIM *182500
Phenotype (HPO) HP:0009053 Distal lower limb muscle weakness; HP:0001772 Foot drop; HP:0001761 Pes cavus; HP:0003724 Distal amyotrophy; HP:0003676 Progressive sensory neuropathy; HP:0025278 Distal tremor
Biological process (GO) GO:0006062 Sorbitol catabolic process; GO:0006970 Response to osmotic stress
Cell type (CL) CL:0000100 Motor neuron; CL:0000011 Schwann cell; CL:0000098 Sensory neuron
Anatomy (UBERON) UBERON:0001021 Nerve; peripheral nerve subtypes (peroneal/tibial/ulnar/median)
Chemical (CHEBI) Sorbitol; fructose; xylitol; govorestat (AT-007)
Treatment (NCIT) NCIT:C15986 Pharmacotherapy (govorestat); NCIT:C15302 Physical Therapy; NCIT:C16186 Orthopedic Surgical Procedure; NCIT:C15240 Genetic Counseling

Sources

Reference Validation

Checked with linkml-reference-validator 0.2.1.

Table (click to expand)
Outcome Count
References checked 10
Resolved 8
Unresolved (possible confabulation) 2
Unverifiable 0
References weighed for topical relevance 8
On topic 7
Off topic 0

Unresolved references

These identifiers did not resolve to a record and may be fabricated. A lookup that failed for transport reasons is indistinguishable from one that failed because the record does not exist, so spot-check before acting on them:

  • DOI:10.1093/brain/awaf021](https://academic.oup.com/brain/article/148/10/3737/8010720 (1 mention) - Identifier did not resolve to a record
  • DOI:10.1093/brain/awae170-region (1 mention) - Identifier did not resolve to a record