Chlorpromazine

證據等級: L5

預測適應症: 10 個

Note on indication selection: The evidence pack ranks 9 of its 10 TxGNN candidates (including the #1-ranked “retinal dystrophy with or without extraocular anomalies”) with zero supporting evidence and its own repurposing_rationale text explicitly labels them likely hub-node embedding artifacts (“非真實預測訊號” / not a real prediction signal — near-identical scores ~0.999x, no mechanistic plausibility, and for rank 1 the 15 “supporting” PubMed hits don’t even mention chlorpromazine). Presenting one of those as the headline finding would produce a misleading clinical report. The only candidate with genuine mechanistic rationale, real evidence, and a non-“Hold” recommendation is rank 10: early-onset schizophrenia (L3, Proceed with Guardrails). I built the report around that candidate and added a short transparency note about the excluded artifacts rather than silently dropping them.

Chlorpromazine: From Schizophrenia to Early-Onset Schizophrenia

One-Sentence Summary

Chlorpromazine is a first-generation (typical) antipsychotic whose established mechanism — dopamine D2 receptor antagonism — underlies its long-standing use in schizophrenia. Of the ten candidate indications TxGNN produced for this drug, only early-onset schizophrenia is backed by a coherent pharmacological rationale and corroborating evidence, currently supported by 1 clinical trial and 8 publications; the remaining nine top-ranked candidates carry no supporting evidence and are assessed below as likely model artifacts.

Quick Overview

Item	Content
Original Indication	Schizophrenia (established pharmacological use; not recorded in the source registry entry for this record — see Data Gap note below)
Predicted New Indication	Early-onset schizophrenia
TxGNN Prediction Score	99.47% (rank 6,160 of all drug–disease pairs)
Evidence Level	L3
Australia Market Status	Not marketed
Number of ARTG Entries	0
Recommended Decision	Proceed with Guardrails

Data gap: the original_indications field in this evidence pack is empty and original_moa is unrecorded (structured DrugBank MOA data not yet pulled). The mechanism discussed below is drawn from the prediction rationale rather than a structured MOA field, and should be independently confirmed against the TGA-approved Product Information before clinical use.

Selection Note: Nine Candidates Excluded as Likely Artifacts

TxGNN ranked nine other diseases above early-onset schizophrenia for chlorpromazine — including retinal dystrophy, X-linked myopia variants, congenital glycosylation disorders, hydranencephaly, a polymicrogyria syndrome, and Charcot-Marie-Tooth disease type 1G. All nine returned zero clinical trials and zero (or clearly irrelevant) literature, cluster tightly around a score of ~0.999x, and have no plausible pharmacological link to a D2/H1/α1/muscarinic receptor antagonist. For the top-ranked candidate, the 15 PubMed hits retrieved were manually checked and none mention chlorpromazine at all — consistent with a keyword co-occurrence false match rather than genuine signal. These are flagged here for transparency and QA (the score clustering suggests a hub-node artifact in the underlying knowledge graph embedding), and none are carried forward into the sections below.

Why is This Prediction Reasonable?

Detailed structured mechanism-of-action data is not available for this record (data gap — DrugBank query pending). Based on well-established pharmacology referenced in the prediction rationale, chlorpromazine acts primarily as a dopamine D2 receptor antagonist, with additional antagonism at histamine H1, alpha-1 adrenergic, and muscarinic receptors. D2 blockade is the classic mechanism underlying its efficacy against the positive symptoms of schizophrenia.

Early-onset schizophrenia (onset in childhood or adolescence) is generally regarded as part of the same disease spectrum as adult-onset schizophrenia, differing mainly in age of onset and, in some series, in treatment responsiveness. Because the underlying dopaminergic pathophysiology is considered continuous across the age spectrum, the same D2-antagonist mechanism is, in principle, applicable to the early-onset population.

This is therefore not a novel pharmacological hypothesis so much as an evidence question about applying an already-understood mechanism to a specific, younger sub-population — the available data below speaks mainly to treatment-resistance patterns and pharmacogenomic modifiers in this group rather than establishing a new indication from first principles.

Clinical Trial Evidence

Trial Number	Phase	Status	Enrolment	Key Findings
NCT06128408	N/A (observational)	Unknown	300	Cohort study characterising treatment-resistant schizophrenia from illness onset (TRO). Reports that up to 30% of antipsychotic-naïve first-episode patients show poor treatment response, and that TRO patients make up ~80% of all treatment-resistant schizophrenia cases in long-term follow-up. This is a descriptive cohort, not a chlorpromazine intervention trial — classified as indirect (relevance grade B).

Literature Evidence

PMID	Year	Type	Journal	Key Findings
24854724	2015	Review	L’Encéphale	Reviews neurological soft signs in early-onset schizophrenia as evidence for a neurodevelopmental model of the disorder.
10703271	1999	Retrospective Cohort	Soc Psychiatry Psychiatr Epidemiol	Examines whether age at onset of schizophrenia relates to typical neuroleptic dosage requirements in outpatients.
18408624	2008	Cohort / genetic association	Pharmacogenetics and Genomics	BDNF gene variants identified as a schizophrenia risk factor and linked to chlorpromazine-induced extrapyramidal syndrome in a Chinese population — direct pharmacogenomic data on chlorpromazine tolerability.
17915974	2007	Cohort / pharmacogenomics	J Clin Psychiatry	AKT1 gene polymorphisms associated with schizophrenia risk and with antipsychotic treatment response in a Chinese population.
28976410	2017	Cross-sectional	Clinical Neuropharmacology	Describes clinical features of early-onset schizophrenia patients with comorbid obsessive-compulsive disorder.
26916502	2016	Cross-sectional	Acta Neuropsychiatrica	Assesses Theory of Mind deficits in adolescents with early-onset schizophrenia and their correlation with executive function.
22802957	2012	Imaging / cross-sectional	PLoS ONE	MRI study showing decreased temporal gyrus grey matter volume in first-episode, early-onset schizophrenia.
24289465	2013	Comparative cohort	Psychogeriatrics	Compares clinical features of early-onset versus late-onset schizophrenia in a Japanese sample.

Safety Considerations

Please refer to the TGA-approved Product Information (PI) for chlorpromazine safety information — no structured warnings, contraindications, or drug-interaction data were returned for this record (DDI query: not found).

For context, chlorpromazine is a typical (first-generation) antipsychotic with well-documented class-related risks, including extrapyramidal symptoms, sedation, anticholinergic effects, orthostatic hypotension, and QT prolongation; one of the publications above (PMID 18408624) specifically links a genetic marker to chlorpromazine-induced extrapyramidal syndrome. This general class knowledge does not substitute for the TGA PI and should not be used as a definitive safety reference on its own.

Conclusion and Next Steps

Decision: Proceed with Guardrails

Rationale: The D2-antagonist mechanism is pharmacologically well-established and plausibly extends to early-onset schizophrenia as part of the same disease spectrum as adult-onset disease. However, direct evidence is limited to one observational cohort study (status: unknown, no chlorpromazine intervention arm) and pharmacogenomic/cross-sectional literature rather than randomised controlled trials in this specific population — consistent with an L3 evidence level, not a level supporting unconditional adoption.

To proceed, the following is needed:

TFDA/TGA-approved PI warnings and contraindications (currently a Blocking data gap — required before any S1 safety screening)
Structured mechanism-of-action data from DrugBank (High-priority data gap)
Confirmation of chlorpromazine’s registered original indication and Australian market/registration status (currently shows “Not marketed” with 0 ARTG entries — verify this is current)
Drug–drug interaction data (current query returned “not found”)
Paediatric/adolescent-specific dosing, tolerability, and long-term safety data for the early-onset population
Prospective controlled trial evidence specifically evaluating chlorpromazine (not just antipsychotics generally) in early-onset schizophrenia

Separately: the nine other TxGNN-ranked candidates for this drug (score ≥0.999) should be reviewed by the modelling team — the tight score clustering with zero evidence across all nine is a strong signal of a hub-node artifact in the knowledge graph embedding rather than genuine repurposing signal, and may warrant a broader audit of similarly-scored candidates for other drugs.

Disclaimer

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