Ethosuximide

證據等級: L5

預測適應症: 10 個

Ethosuximide: From Childhood Absence Epilepsy to Childhood Absence Epilepsy, Susceptibility To (Genetic Subtype)

One-Sentence Summary

Ethosuximide is a succinimide antiepileptic internationally recognised as a first-line treatment for childhood absence epilepsy (petit mal seizures), though it is not currently registered in Australia. The TxGNN model’s highest-ranked prediction (nephrogenic syndrome of inappropriate antidiuresis, rank 1, score 99.91%) is assessed as a likely knowledge graph artefact with no plausible mechanism or supporting evidence. The most clinically meaningful prediction is Epilepsy, Childhood Absence, Susceptibility To (rank 5, score 95.67%), a genetically defined subtype supported by 17 publications and strong mechanistic alignment, representing a precision medicine refinement of the existing indication rather than a novel repurposing opportunity.

Quick Overview

Item	Content
Original Indication	Childhood absence epilepsy (petit mal seizures) — internationally established; not registered in Australia (no ARTG entries found)
TxGNN Top Prediction (Rank 1)	Nephrogenic Syndrome of Inappropriate Antidiuresis (NSIAD) — assessed as model artefact, Hold
Most Actionable Prediction (Rank 5)	Epilepsy, Childhood Absence, Susceptibility To (genetic subtype: CACNA1H / GABRG2 mutations)
TxGNN Score (Rank 5)	95.67%
Evidence Level (Rank 5)	L3
Australia Market Status	Not marketed — no ARTG registrations found
Number of ARTG Entries	0
Recommended Decision	Hold (ranks 1–4, 6–10) / Proceed with Guardrails (rank 5 genetic CAE subtype)

Why is This Prediction Reasonable?

About Ethosuximide

Detailed mechanism of action data was not available in this dataset. Based on established international pharmacology, ethosuximide is a succinimide-class antiepileptic that selectively blocks T-type voltage-gated calcium channels — in particular Cav3.2 (encoded by CACNA1H) — in thalamic relay neurons. This suppresses the rhythmic thalamocortical 3 Hz spike-and-wave discharges that define absence seizures. It does not have activity against focal (partial) or tonic-clonic seizures, and selective use in confirmed generalised absence epilepsy is critical. Ethosuximide is listed as first-line therapy in multiple international paediatric neurology guidelines (including NICE, ILAE, and AAP). It is not registered in Australia and would require access via the TGA Special Access Scheme if clinical use were contemplated.

TxGNN Top Score (Rank 1–4, 6–10): Model Artefacts — Do Not Pursue

The highest-ranked TxGNN prediction — nephrogenic syndrome of inappropriate antidiuresis (NSIAD, 99.91%) — has no plausible mechanism. NSIAD is caused by gain-of-function mutations in AVPR2 (V2 receptor), producing constitutive cAMP activation independent of antidiuretic hormone; ethosuximide’s T-type calcium channel mechanism has no known intersection with the AVPR2–cAMP–AQP2 signalling axis. The high TxGNN score likely reflects a knowledge graph proximity effect between “renal calcium channel” and “renal water-transport disorder” nodes — a classic false positive pattern.

Ranks 2 (nephrogenic diabetes insipidus), 6 (X-linked nephrogenic DI), and 9 (Senior-Boichis syndrome, a ciliopathy) form a cluster of renal/water-transport conditions with no connection to T-type calcium channel pharmacology. Rank 3 (insomnia) warrants specific caution: while T-type channels do contribute to thalamic sleep spindle generation, clinical experience shows ethosuximide causes insomnia as an adverse effect — the mechanistic direction is opposite to therapeutic intent. Rank 4 (familial mesial temporal lobe epilepsy with febrile seizures) carries an active ⚠️ safety concern — ethosuximide is known to paradoxically aggravate complex partial seizures in temporal lobe epilepsy circuits, and this should not be pursued.

Rank 8 (renal tubule disease) represents a particularly important methodological warning: the single retrieved publication (PMID 7843203) documents renal tubular dysfunction as an adverse effect of antiepileptic drugs including ethosuximide — the knowledge graph connection reflects toxicity, not therapeutic opportunity. The causal direction is entirely reversed.

TxGNN Rank 5: Genetic CAE Susceptibility — Mechanistically Sound

The most clinically meaningful prediction is childhood absence epilepsy, susceptibility to, which refers to patients harbouring identified susceptibility mutations in CACNA1H (Cav3.2 T-type calcium channel gene), CACNA1A, or GABRG2 (GABA-A receptor γ2 subunit). This prediction is mechanistically compelling for three reasons:

Direct target match: Ethosuximide’s primary target — Cav3.2 — is encoded by CACNA1H, the leading susceptibility gene for this subtype. Gain-of-function mutations that increase Cav3.2 surface expression or alter biophysical properties directly amplify the thalamocortical oscillations that ethosuximide suppresses.
Animal model validation: The GABRG2 R43Q knock-in mouse model — derived from an Australian family with childhood absence epilepsy — demonstrates spike-wave discharges that are specifically blocked by ethosuximide (PMID 17947380). The GAERS rat and WAG/Rij rat models, both genetic absence models, show consistent ethosuximide sensitivity, including disease modification with early treatment.
Indirect Phase 3 support: The broader childhood absence epilepsy syndrome (not restricted to the genetic susceptibility subtype) has Phase 3 RCT support from the JAAE Trial (NEJM 2010; not retrieved in this dataset query), which constitutes indirect L1 evidence. The genetic susceptibility subtype specifically lacks dedicated RCT data, hence the L3 rating.

It is important to note that the data system recorded original_indications as empty, which reflects a data quality gap rather than a genuine absence of approved indications: internationally, childhood absence epilepsy is an established and approved indication for ethosuximide. This rank 5 prediction therefore represents a precision medicine refinement — identifying genetically confirmed patients who may be preferential responders — rather than a conventional novel repurposing indication.

Clinical Trial Evidence

No clinical trials were found in this dataset for ethosuximide specifically in the “childhood absence epilepsy, susceptibility to” (genetic subtype) indication.

Currently no related clinical trials are registered for this specific genetic susceptibility subtype.

Context: The broader childhood absence epilepsy indication (non-genotype-stratified) has been evaluated in Phase 3 trials internationally. Genotype-stratified prospective studies examining ethosuximide response by CACNA1H or GABRG2 variant status represent a logical evidence gap to address.

Literature Evidence

The following publications are the most relevant from the 17 retrieved for the childhood absence epilepsy susceptibility indication. Prioritised by clinical relevance and study quality:

PMID	Year	Type	Journal	Key Findings
40339636	2025	In Vivo Experimental	Progress in Neurobiology	Optogenetic depolarisation of thalamocortical neurons in GAERS rats reliably elicits ethosuximide-sensitive spike-and-wave discharges, confirming the thalamocortical circuit as the primary therapeutic target
24277868	2014	Mechanistic Study	Journal of Physiology	CACNA1H mutations associated with idiopathic generalised epilepsy increase seizure susceptibility by altering Cav3.2 biophysics and surface expression — the direct molecular target of ethosuximide
17947380	2007	Animal Model	PNAS	GABRG2 R43Q knock-in mice (from an Australian family) exhibit behavioural arrest with 6–7 Hz spike-wave discharges blocked by ethosuximide, establishing mechanistic proof-of-concept in a genetically defined model
24530152	2014	Review	Pediatric Neurology	Comprehensive review of CAE genetics, multifactorial aetiology, and treatment options; confirms genetic-environmental interaction in susceptibility
19632158	2009	Review	Epilepsy & Behavior	Review of absence seizure subtypes distinguishing childhood from juvenile onset, including genetic contributors, cognitive effects, and treatment response profiles
38724972	2024	Animal Experimental	Journal of Headache and Pain	Extrasynaptic δGABA-A receptors in GAERS rats modulate absence seizure susceptibility; demonstrates interaction between GABAergic inhibitory tone and T-type channel-mediated oscillations relevant to genetic CAE
18079316	2008	Animal Study	Epilepsia	Early ethosuximide treatment in WAG/Rij rats suppresses development of the spike-wave epilepsy phenotype in adulthood, suggesting a potential disease-modifying effect beyond seizure suppression
16302877	2005	Clinical Study	Epilepsia	Characterisation of photosensitivity in idiopathic generalised epilepsies including CAE; relevant to phenotypic characterisation of the genetic susceptibility subtype
9860058	1998	Animal/EEG Experimental	Epilepsia	Ethosuximide suppresses bicuculline-induced rhythmic EEG episodes in a sex-dependent manner, validating anti-absence activity in a pharmacological model
18070091	2008	Clinical Observational	Journal of Child Neurology	CAE in a child with both genetic predisposition and acquired perinatal insult; illustrates how genetic susceptibility interacts with environmental factors in determining CAE expression

Australia Market Information

Ethosuximide is not registered in Australia. No ARTG (Australian Register of Therapeutic Goods) entries were found.

ARTG Number	Product Name	Dosage Form	Approved Indication
—	No products registered	—	No TGA-approved indication in Australia

For any clinical use in Australia, access would require:

TGA Special Access Scheme (SAS) Category B application by the treating clinician, or
Authorised Prescriber pathway for eligible paediatric neurologists

Reference product information should be sourced from international regulators (UK MHRA product information or US FDA label for Zarontin® or Emeside®).

Safety Considerations

No TGA-approved Product Information is available for ethosuximide in Australia. No drug interaction data was retrieved in this Evidence Pack.

Please refer to the international Product Information (UK SmPC or US FDA label) for comprehensive safety and dosing information.

Key safety points from established pharmacological knowledge:

Seizure type specificity: Ethosuximide is effective only in generalised absence epilepsy. It must not be used in focal (partial) or temporal lobe epilepsy — it may paradoxically worsen complex partial seizures. Careful diagnosis before prescribing is essential.
Renal monitoring note: An observational study (PMID 7843203) documented renal tubular dysfunction in children on antiepileptic drugs including ethosuximide. This represents an adverse effect requiring monitoring, not a therapeutic application.
Absence of local PI: The absence of a TGA-approved Product Information document is a practical barrier to safe prescribing in Australia and must be addressed before any clinical use recommendation can be made.

Conclusion and Next Steps

Decision: Hold (for novel predictions, ranks 1–4, 6–10) / Proceed with Guardrails (rank 5 — genetic CAE susceptibility subtype)

Rationale: The nine lower-clinically-relevant predictions (including the TxGNN rank 1, NSIAD) are assessed as model artefacts, mechanistically implausible or — in the case of rank 4 and rank 8 — potentially harmful when misinterpreted as therapeutic opportunities. The rank 5 prediction (childhood absence epilepsy, susceptibility to) is mechanistically sound and supported by animal model evidence and indirect Phase 3 RCT data for the broader CAE syndrome; however, it primarily confirms an existing clinical role rather than identifying a genuinely novel indication. Its value lies in supporting genotype-stratified prescribing decisions rather than traditional drug repurposing.

To proceed, the following is needed:

Obtain TGA access pathway (Special Access Scheme or Authorised Prescriber) before clinical use in Australia can be considered
Source international Product Information (UK SmPC / US FDA label for Zarontin® or Emeside®) to provide complete Australian prescribing guidance
Retrieve DrugBank full record to complete MOA documentation and establish the formal mechanism summary
Conduct a prospective cohort study or registry analysis examining ethosuximide treatment response stratified by genotype (CACNA1H, GABRG2, CACNA1A variant status) in Australian paediatric patients with confirmed CAE
Commission a systematic review comparing ethosuximide response rates in genotype-confirmed vs. genotype-negative CAE to quantify the precision medicine benefit
Develop a pharmacogenomics framework for identifying candidates most likely to respond, in collaboration with Australian paediatric neurology and clinical genetics services

Disclaimer: This report is for research reference only and does not constitute medical advice. Drug repurposing candidates require clinical validation before therapeutic application. All recommendations should be reviewed by a qualified Australian healthcare professional. Data cutoff: 5 April 2026.

Disclaimer

This content is for research purposes only and does not constitute medical advice. Clinical validation is required before any clinical application.