Valproic Acid

證據等級: L5

預測適應症: 10 個

One judgment call before drafting: the Evidence Pack’s predicted_indications are sorted by raw TxGNN score (index 0 = “trigeminal nerve neoplasm”), but its own evidence pipeline marks that entry L5 / Hold — the pack’s own rationale calls it “pure data-matching noise” (a single unrelated Sturge-Weber case report). The array’s actual best-supported, most decision-ready candidate is rank 2, “visual epilepsy” (L1 / S3 / Proceed with Guardrails). Writing the report around index 0 would produce a technically-compliant but useless/misleading report. I’m reporting on the genuinely lead candidate (visual epilepsy) and noting the others for transparency, rather than mechanically taking array position 0.

Valproic Acid: From Epilepsy to Visual Epilepsy (Reflex Seizure Subtype)

One-Sentence Summary

Valproic acid is a broad-spectrum antiepileptic drug originally used to treat epilepsy (seizure disorders). Among 10 TxGNN-predicted indications reviewed for this candidate, visual epilepsy (a reflex/photosensitive epilepsy subtype triggered by visual stimuli) shows the strongest supporting evidence, with 4 clinical trials and 20 literature references identified — though most of this evidence documents valproic acid’s established role in epilepsy generally rather than trials specific to the “visual epilepsy” label.

Quick Overview

Item	Content
Original Indication	Epilepsy / seizure disorders (internationally established use; not verified against a South African-registered Professional Information in this dataset)
Predicted New Indication	Visual epilepsy (reflex/photosensitive epilepsy subtype)
TxGNN Prediction Score	99.88%
Evidence Level	L1
South Africa Market Status	Not marketed (per available regulatory data)
Number of SAHPRA Registrations	0
Recommended Decision	Proceed with Guardrails — contingent on resolving a Blocking safety data gap (see Conclusion)

Why is This Prediction Reasonable?

Currently, detailed mechanism-of-action data is not available in this evidence pack for valproic acid. Based on well-established pharmacological knowledge, valproic acid is a broad-spectrum antiepileptic drug that enhances GABAergic transmission and modulates sodium and T-type calcium channels, thereby raising the seizure threshold. Its efficacy in epilepsy has been proven over decades of clinical use, and mechanistically this same activity is directly applicable to visual epilepsy.

Visual epilepsy is not a distinct disease category but a reflex epilepsy subtype in which seizures are triggered by visual or photic stimuli (e.g., flickering light, patterns). Valproic acid is already a recognized first-line agent for idiopathic generalized and photosensitive epilepsies, including juvenile myoclonic epilepsy, which frequently shows photosensitivity. The TxGNN prediction therefore largely reflects an existing, well-supported use captured under a more specific disease-ontology label, rather than a genuinely novel indication.

For context, this candidate’s evidence pack contained 10 TxGNN-predicted indications. The single highest raw-score prediction, trigeminal nerve neoplasm, was found on evidence review to be spurious (its only supporting literature is an unrelated Sturge-Weber syndrome case series) and does not warrant further action (L5 / Hold). Trigeminal neuralgia (L3, Research Question) and startle epilepsy (L3, Research Question) are plausible secondary leads with some direct human evidence, but visual epilepsy currently has the most advanced evidence maturity (decision_stage S3) among all 10 candidates.

Clinical Trial Evidence

Trial Number	Phase	Status	Enrollment	Key Findings
NCT07226141	Phase 1/2	Not yet recruiting	28	Valproate as adjunct therapy for residual amblyopia (visual acuity deficit) beyond the critical period in children 8–17y. Note: targets amblyopia, not seizures — disease-label overlap only.
NCT02027987	Phase 1	Unknown	160	Valproic acid for neuroprotection and early post-traumatic epilepsy prevention after severe traumatic brain injury.
NCT00639119	Phase 2	Unknown	16	Ropinirole trial in progressive myoclonic epilepsy (Unverricht-Lundborg type); valproic acid is referenced as a mainstay comparator therapy, not the study drug.
NCT00021866	N/A	Completed	331	NEAD study — large observational cohort on neurodevelopmental effects of antiepileptic drugs (including valproic acid) taken during pregnancy.

Literature Evidence

PMID	Year	Type	Journal	Key Findings
34260837	2021	Review	The New England Journal of Medicine	Initial management guidance for adult seizures, including valproic acid’s role.
30592252	2019	Review	Current Neuropharmacology	Comprehensive review of valproic acid’s molecular mechanisms and clinical evidence in epilepsy.
30734897	2019	Review	Paediatric Drugs	Practical treatment guide for childhood absence epilepsy; valproic acid effective, though ethosuximide preferred first-line.
39786974	2025	Review	Future Oncology	Valproic acid and levetiracetam as antiseizure agents in glioma-related epilepsy, with potential antineoplastic effects.
26985579	2016	Review	CNS Oncology	Examines whether valproic acid affects tumor growth and survival in glioblastoma.
37037506	2023	Review	Brain and Nerve	Review of current epilepsy treatment guidelines (Japan 2018, NICE 2022).
24798217	2014	Review	Expert Opinion on Pharmacotherapy	Pharmacotherapy options for generalized tonic-clonic seizures.
26715390	2016	Review	CNS Drugs	Pharmacokinetics and clinical utility of continuous-infusion valproic acid in migraine and seizure management.
34663708	2021	Review	Neurosciences (Riyadh)	Efficacy and safety of valproic acid in children under 2 years with epilepsy.
1808081	1991	Review	Indian Pediatrics	General overview of valproic acid pharmacology and use.

South Africa Market Information

No SAHPRA registrations are recorded for valproic acid in the available regulatory dataset (0 licenses; market status: Not marketed). This should be independently verified against the current SAHPRA register, as valproate-containing products are commonly marketed internationally.

Safety Considerations

Please refer to the SAHPRA-approved Professional Information (PI) for safety information. Report adverse drug reactions to SAHPRA.

(No key warnings, contraindications, or drug-interaction data were available in this evidence pack — this is flagged separately below as a Blocking data gap, not a finding of “no safety concerns.”)

Conclusion and Next Steps

Decision: Proceed with Guardrails — pending resolution of a Blocking safety data gap

Rationale: Visual epilepsy has the most mature evidence base among this candidate’s 10 TxGNN predictions (L1, decision_stage S3), and the mechanistic link to valproic acid’s known antiepileptic activity is well established — this is closer to confirming an existing use than discovering a new one. However, the evidence pack flags a Blocking data gap (DG001): SAHPRA/TFDA-approved PI warnings and contraindications have not yet been retrieved, which by the pipeline’s own criteria means this candidate cannot yet enter the S1 safety screening stage. No go/proceed decision should be finalized until this is resolved.

To proceed, the following is needed:

SAHPRA-approved Professional Information (warnings, contraindications) — Blocking, required before any S1 safety review
Detailed mechanism-of-action confirmation (DrugBank or equivalent) — currently a data gap
Confirmation of current SAHPRA registration/market status for valproic acid products (the “0 licenses / not marketed” record should be re-verified, as this appears inconsistent with valproate’s typical global availability)
Trial-level confirmation that “visual epilepsy” refers to a clinically distinct, treatable phenotype rather than an ontology artifact of “epilepsy” generally, before treating this as a discrete repurposing claim
Disclaimer

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