Haloperidol

證據等級: L5

預測適應症: 10 個

Haloperidol: From Psychotic Disorders to Manic Episodes of Bipolar Disorder

One-Sentence Summary

Haloperidol is a first-generation antipsychotic long used to control psychosis and schizophrenia (the evidence pack's structured original_indications field is empty, but this is a well-established clinical use). Among the ten TxGNN-predicted indications supplied, only Manic Bipolar Affective Disorder is backed by substantial evidence — 9 clinical trials and 20 publications — while the model's algorithmically top-ranked candidates (rare congenital/ophthalmologic/neurodevelopmental disorders) had zero supporting trials or relevant literature and are assessed in the source data itself as likely noise matches. This report therefore focuses on the one indication with real evidentiary support.

Quick Overview

Item	Content
Original Indication	Not recorded in the evidence pack's structured field (haloperidol is a long-established first-generation/typical antipsychotic for psychosis and schizophrenia)
Predicted New Indication	Manic Bipolar Affective Disorder
TxGNN Prediction Score	99.83% (model rank 2099; this was the 10th of 10 candidates listed, selected over higher-scoring but evidence-free candidates — see note below)
Evidence Level	L1
New Zealand Market Status	✗ Not Marketed
Number of Authorizations	0
Recommended Decision	Proceed with Guardrails

Note on candidate selection: TxGNN's single highest-scoring prediction for haloperidol was "congenital disorder of glycosylation with defective fucosylation" (99.91%), followed by several rare ophthalmologic and neurodevelopmental disorders (ranks 2–9). All of these returned no clinical trials and no relevant literature, and their own mechanistic-link assessments in the evidence pack explicitly flag them as implausible model noise (no known connection between haloperidol's D2/5-HT2/α1 receptor targets and the disease biology). Manic bipolar affective disorder was the only candidate with a coherent mechanism and confirmatory clinical evidence, so it is the subject of this report.

Why is This Prediction Reasonable?

Haloperidol is a potent first-generation D2 dopamine receptor antagonist. Manic and psychotic symptoms are associated with excess mesolimbic dopaminergic transmission, and D2 blockade rapidly reduces agitation and psychotic features — a mechanism that is clinically well established rather than a novel hypothesis.

Strictly speaking, this is less a "repurposing discovery" than a confirmation of existing clinical practice: haloperidol is already used as add-on/adjunct therapy alongside mood stabilizers (lithium, valproate) for acute manic episodes, and is a common active comparator arm in bipolar-mania trials of newer antipsychotics (risperidone, olanzapine, aripiprazole). The gap is administrative — the source registry's original_indications field for this drug record is empty — not clinical or mechanistic.

Because haloperidol's antipsychotic effect and its established role in acute mania are pharmacologically continuous, the TxGNN score is well supported by both mechanism and a substantial body of Phase 2/3 trial and meta-analytic literature, unlike the top nine algorithmic candidates.

Clinical Trial Evidence

Trial Number	Phase	Status	Enrollment	Key Findings
NCT00253149	Phase 3	Completed	158	Risperidone vs. placebo vs. haloperidol as add-on to mood stabilizers for manic episodes; haloperidol used as active comparator.
NCT00253162	Phase 3	Completed	439	Flexible-dose risperidone vs. placebo vs. haloperidol in Bipolar I manic episodes; haloperidol maintenance effectiveness assessed at 12 weeks.
NCT00129220	Phase 3	Completed	224	Double-blind, placebo- and haloperidol-controlled trial confirming olanzapine efficacy in manic/mixed Bipolar I episodes.
NCT00126009	Phase 2	Completed	120	Open, randomized 3-month trial comparing valproate-amisulpride vs. valproate-haloperidol in Bipolar I manic episode.
NCT04327843	Phase 3	Completed	22	Long-acting injectable antipsychotic + adherence-focused behavioral program for chronic psychotic disorders in Tanzania.
NCT06049953	N/A	Recruiting	200	Observational study of antenatal antipsychotic exposure on maternal psychiatric course and infant development.
NCT00097266	Phase 3	Completed	615	Aripiprazole monotherapy vs. placebo for acute mania; no explicit haloperidol treatment arm.
NCT00767715	Phase 4	Terminated	11	Olanzapine vs. conventional antipsychotics (incl. haloperidol) for acute mania in Sweden; terminated early, small sample.
NCT03541031	N/A	Unknown	120	Micronutrient/fish-oil supplementation as adjunct to conventional bipolar medication; no direct haloperidol focus.

Literature Evidence

PMID	Year	Type	Journal	Key Findings
22134043	2012	RCT	Journal of Affective Disorders	Randomized, double-blind, placebo- and haloperidol-controlled study confirming olanzapine efficacy in Japanese patients with manic/mixed Bipolar I episodes.
369472	1979	RCT	Archives of General Psychiatry	Double-blind controlled trial: lithium plus haloperidol vs. placebo plus haloperidol in excited schizoaffective disorder; modest additional benefit from lithium.
34642461	2022	Systematic Review / Network Meta-analysis	Molecular Psychiatry	Network meta-analysis of double-blind RCTs comparing efficacy, tolerability, and safety of pharmacologic treatments (including haloperidol) for acute bipolar mania.
33460070	2020	Review	Acta Psychiatrica Scandinavica	Evidence-based treatment recommendations for bipolar mania, covering mood stabilizer and antipsychotic (including haloperidol) selection.
18344731	2008	Systematic Review	Journal of Clinical Psychopharmacology	Systematic review of antipsychotic-induced extrapyramidal side effects in bipolar disorder and schizophrenia, relevant to haloperidol's tolerability profile.
27151529	2016	Systematic Review / Meta-analysis	Human Psychopharmacology	Systematic review of pharmacologic treatment for acute agitation in psychotic and bipolar disorder.
36789916	2023	Review	BMJ Mental Health	Comparison of antipsychotic dose equivalents between acute mania and schizophrenia.
22070611	2012	Review	CNS Neuroscience & Therapeutics	Discusses adding haloperidol/other antipsychotics for lithium/valproate/carbamazepine partial responders in refractory bipolar disorder.
19454110	2007	Review	BMJ Clinical Evidence	General overview of bipolar disorder epidemiology, course, and treatment options.
3312180	1987	Controlled Study	The Journal of Clinical Psychiatry	Double-blind controlled comparison of clonazepam vs. lithium vs. haloperidol in acute mania.

New Zealand Market Information

Haloperidol currently has no registered authorizations in this dataset (market status: Not Marketed, 0 licenses on file). No product-level formulation or approved-indication data is available to tabulate.

Safety Considerations

Please refer to the package insert for safety information. No structured warnings, contraindications, or drug-interaction data are currently available in this evidence pack (DDI query returned no results).

Conclusion and Next Steps

Decision: Proceed with Guardrails

Rationale: Multiple completed Phase 3 RCTs with direct haloperidol treatment/comparator arms (NCT00253149, NCT00253162, NCT00129220), reinforced by a 2022 network meta-analysis, support haloperidol's efficacy in acute bipolar mania — satisfying L1 evidence criteria. However, this reflects confirmation of an already-established clinical use rather than a novel repurposing discovery, and neither local (New Zealand/TFDA) regulatory status nor safety labeling data are currently available.

To proceed, the following is needed:

Official package insert / warnings and contraindications (currently blocking — DG001)
Confirmed original-indication registry data for haloperidol to correct the empty original_indications field
Drug-drug interaction dataset (current DDI query returned no results)
Evaluation of a New Zealand registration pathway, given current "Not Marketed" status
Note: TxGNN's top nine algorithmically-ranked candidates for haloperidol lack any supporting evidence and should not be pursued without independent mechanistic validation
Disclaimer

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