Zonisamide

證據等級: L5

預測適應症: 10 個

Editorial note before generating: predicted_indications[0] is Tourette syndrome (TxGNN rank #1 by score), but it has zero clinical trials or literature (L5 evidence / Hold). The highest-evidence prediction is absence epilepsy (rank #8 in array; L1 / Proceed with Guardrails), which also has the strongest mechanistic basis. This report focuses on absence epilepsy as the primary subject, noting the TxGNN top prediction transparently in the summary.

Zonisamide: From Partial Seizures to Absence Epilepsy

One-Sentence Summary

Zonisamide is a broad-spectrum antiepileptic drug (AED) approved in many countries for adjunctive treatment of partial (focal) seizures in adults. The TxGNN model's highest-scoring prediction is Tourette Syndrome (score 99.85%), but that prediction currently has no supporting clinical trials or published literature (L5). This report focuses on the clinically strongest prediction: Absence Epilepsy (TxGNN score 99.24%), supported by 4 clinical trials (including two Phase 3 studies) and 20 publications — and reinforced by Japanese regulatory approval and established pharmacological consensus on zonisamide's T-type calcium channel blockade mechanism.

Quick Overview

Item	Content
Original Indication	Epilepsy — adjunctive treatment of partial (focal) seizures in adults (approved in EU, Japan, USA; not currently authorized in Italy)
Predicted New Indication	Absence Epilepsy
TxGNN Prediction Score	99.24%
Evidence Level	L1
Italy Market Status	✗ Not marketed (0 AIFA authorizations)
Number of Authorizations	0
Recommended Decision	Proceed with Guardrails

Why is This Prediction Reasonable?

Zonisamide acts through multiple complementary mechanisms: it stabilizes neuronal membranes by blocking voltage-gated sodium channels, inhibits T-type calcium channels (Cav3.x), modulates dopamine and serotonin neurotransmission, and weakly inhibits carbonic anhydrase. This multi-target profile distinguishes it from most AEDs and provides a broad anticonvulsant spectrum across seizure types.

The mechanistic link to absence epilepsy is direct and established. Absence seizures (petit mal) arise from rhythmic thalamocortical oscillations driven by T-type calcium channels in thalamic relay and reticular neurons — the same pathway targeted by ethosuximide, the reference treatment for childhood absence epilepsy and a pure T-type calcium channel blocker. Because zonisamide shares this T-type channel blockade, the pharmacological rationale is not merely model inference but reflects established neuroscience consensus. This overlap with ethosuximide's mechanism is more compelling than the weaker link between zonisamide and most other TxGNN-predicted indications.

Supporting this reasoning from a regulatory and clinical standpoint: Japan approved zonisamide in the 1980s across a broad range of seizure types including generalized seizures; a 2005 chart review (Wilfong & Schultz; n=45 pediatric patients with absence seizures) reported that 51.1% achieved complete seizure freedom on zonisamide; and a 2014 prospective case series (Velizarova et al.) specifically examined zonisamide in drug-resistant juvenile absence epilepsy. Two Phase 3 multinational RCTs (NCT00477295, NCT00848549) provide robust safety and efficacy data in the broader epilepsy population, enabling confident safety evaluation for absence epilepsy repurposing.

Clinical Trial Evidence

Trial Number	Phase	Status	Enrollment	Key Findings
NCT00477295	Phase 3	Completed	583	Randomized, multi-centre, double-blind non-inferiority trial comparing zonisamide vs carbamazepine monotherapy in newly diagnosed partial epilepsy; primary evidence base for zonisamide efficacy and safety as monotherapy
NCT00848549	Phase 3	Completed	295	Long-term blinded extension of the Phase 3 monotherapy programme; assesses durability of efficacy and long-term safety/tolerability of zonisamide, providing 2+ years of follow-up data
NCT07443241	N/A	Completed	779	Retrospective observational analysis of 779 patients treated for status epilepticus (including absence) at a university hospital (2011–2023); evaluates sex-specific differences in etiology, treatment, and outcomes — provides real-world treatment data
NCT04939675	N/A	Unknown	40	Feasibility study developing a revised epilepsy screening questionnaire incorporating new seizure semiology insights; limited direct efficacy evidence for zonisamide in absence epilepsy

Literature Evidence

PMID	Year	Type	Journal	Key Findings
35363878	2022	Network Meta-analysis	Cochrane Database Syst Rev	Updated individual-patient-data network meta-analysis of AED monotherapy across all epilepsy types; provides comparative efficacy and tolerability data including zonisamide
29898971	2018	Clinical Practice Guideline	Neurology	AAN/AES updated guideline on new AED efficacy as initial monotherapy for new-onset epilepsy; includes evidence-based recommendation level for zonisamide
23350722	2013	ILAE Systematic Review	Epilepsia	ILAE comprehensive evidence review of AED efficacy as initial monotherapy; assigns formal evidence levels to zonisamide across seizure types
15847848	2005	Clinical Study	Epilepsy Res	Chart review of 45 pediatric patients (≤18 years) with absence seizures treated with zonisamide; 51.1% achieved complete seizure freedom — key direct evidence for absence epilepsy
24907183	2014	Prospective Case Series	Epilepsy Res	Prospective evaluation of zonisamide specifically in drug-resistant juvenile absence epilepsy (JAE); assesses efficacy in a population with limited treatment options
40351416	2025	Network Meta-analysis	Front Pharmacol	Network meta-analysis of single ASM as adjunctive therapy for drug-resistant focal epilepsy; comparative effectiveness and safety ranking including zonisamide
15634623	2004	Clinical Study	Epileptic Disord	Retrospective review of 15 juvenile myoclonic epilepsy patients treated with zonisamide; reports efficacy across absence, myoclonic, and GTC seizure components, supporting broad-spectrum activity
16321507	2006	Review	Epilepsy Res	Summary of extensive Japanese clinical experience with zonisamide in both adults and children across seizure types including generalized absence seizures; supports Japanese regulatory context
34941639	2021	Narrative Review	Pediatr Rep	Therapeutic review for childhood absence epilepsy including drug-resistant cases; discusses role of newer AEDs and identifies zonisamide as a candidate for refractory presentations
16341290	2005	Review	Drugs Today	Comprehensive review of zonisamide's broad-spectrum anticonvulsant profile; documents clinical efficacy in absence seizures unresponsive to first-line agents

Italy Market Information

Zonisamide is not currently authorized in Italy (0 AIFA registrations). No licensed products are available as of the data cutoff (2026-05-05).

For reference, zonisamide is marketed under the brand name Zonegran® (Eisai) in other EU Member States (including Germany, France, and the UK prior to Brexit), and has been approved in the United States (Zonegran®, 2000) and Japan (Excegran®, 1989) — the latter covering a broader seizure spectrum that includes generalized seizures. An EU-wide marketing authorization exists, which means AIFA registration via the mutual recognition or decentralized procedure would be a feasible regulatory pathway.

Safety Considerations

Please refer to the package insert for full safety information. No DDI interaction data or specific warning/contraindication data were available in this Evidence Pack.

Two important adverse-effect signals to note, identified during evidence review:

Methemoglobinemia risk: Zonisamide is a sulfonamide derivative. TxGNN predictions #3, #4, and #6 (methemoglobinemia alpha type, methemoglobinemia, methemoglobin reductase deficiency) were flagged by the pipeline as safety warnings, not therapeutic predictions — zonisamide can induce oxidative stress and methemoglobin formation. Patients with sulfonamide hypersensitivity or pre-existing methemoglobin reductase deficiency should be screened before use.
Psychiatric adverse effects: One case report (PMID 2109869) documents zonisamide-induced mania. This is relevant when considering use in populations with mood disorder comorbidity.

Conclusion and Next Steps

Decision: Proceed with Guardrails

Rationale: Zonisamide's T-type calcium channel blockade provides a mechanistically compelling and pharmacologically established basis for efficacy in absence epilepsy. The evidence package — including two completed Phase 3 trials, a Cochrane network meta-analysis, ILAE systematic review, AAN/AES guideline endorsement, multiple clinical case series directly in absence seizure populations, and Japanese regulatory approval for generalized seizures — collectively meets L1 criteria. This represents the strongest available repurposing signal in the entire Evidence Pack and justifies moving to a structured clinical development or regulatory strategy.

To proceed, the following is needed:

Dedicated Phase 3 RCT in absence epilepsy: Existing Phase 3 trials focused on partial (focal) seizures; a prospective double-blind trial specifically in childhood or juvenile absence epilepsy is needed for an AIFA-approved indication
Pediatric dosing and safety data: Most robust evidence derives from adult partial seizure studies; age-appropriate dosing, PK, and safety data in children (the primary absence epilepsy population) should be formally established
AIFA regulatory pathway planning: Identify the fastest route to Italian market authorization — mutual recognition from an existing EU member-state holder, or a new centralized EMA application covering the absence epilepsy indication
Full package insert review: Obtain and analyze complete contraindications, warnings, and DDI data from the EU/US/Japan package inserts (not available in this Evidence Pack)
Sulfonamide hypersensitivity screening protocol: Given the methemoglobinemia adverse-effect risk, develop a pre-treatment screening checklist for G6PD deficiency and sulfonamide allergy status

Summary of all TxGNN-predicted indications (for reference):

| Rank | Indication | TxGNN Score | Evidence Level | Decision | |------|-----------|-------------|---------------|---------| | 1 | Tourette Syndrome | 99.85% | L5 | Hold | | 2 | Trichotillomania | 99.78% | L5 | Hold | | 3 | Methemoglobinemia (alpha) | 99.64% | L5 | ⚠️ Hold — safety concern (adverse effect, not indication) | | 4 | Methemoglobinemia | 99.63% | L5 | ⚠️ Hold — safety concern | | 5 | Prinzmetal Angina | 99.55% | L5 | Hold | | 6 | Methemoglobin Reductase Deficiency | 99.53% | L5 | ⚠️ Hold — safety concern | | 7 | Manic Bipolar Affective Disorder | 99.35% | L3 | Research Question (1 RCT signal, insufficient sample) | | 8 | Absence Epilepsy | 99.24% | L1 | ✓ Proceed with Guardrails ← Primary focus of this report | | 9 | Fibromyalgia | 99.20% | L5 | Hold (1 trial withdrawn before enrollment) | | 10 | Conjunctivitis | 99.16% | L5 | Hold |

Disclaimer

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