Lorazepam

證據等級: L5

預測適應症: 10 個

Lorazepam: From Anxiety Disorders to Insomnia

One-Sentence Summary

Lorazepam is a benzodiazepine (GABA-A receptor positive allosteric modulator) originally established for anxiety disorders, short-term insomnia, and status epilepticus. TxGNN’s top-ranked prediction (trigeminal nerve neoplasm, 99.87%) has zero supporting trials or literature and is assessed as a likely model artifact with no biological plausibility. The best evidence-supported candidate is insomnia, scoring 99.80%, backed by 23 clinical trials and 18 publications — though this largely confirms an already-established secondary use rather than a genuinely novel indication.

Quick Overview

Item	Content
Original Indication	Not available in this evidence pack (no Taiwan/US license records; `original_indications` empty). Established pharmacology: anxiety disorders, short-term insomnia, status epilepticus, pre-anesthetic sedation
Predicted New Indication	Insomnia (disease)
TxGNN Prediction Score	99.80%
Evidence Level	L3
US/TW Market Status	未上市 (Not Marketed)
Number of NDAs	0
Recommended Decision	Proceed with Guardrails

Why is This Prediction Reasonable?

Detailed structured MOA data is flagged as a data gap, but the repurposing rationale in the evidence pack consistently identifies lorazepam as a GABA-A receptor positive allosteric modulator (PAM), which enhances inhibitory neurotransmission. This mechanism underlies its established sedative-hypnotic effect and is directly relevant to insomnia management — lorazepam has, in practice, long been used off-label/as a secondary agent for short-term insomnia. In this sense the “insomnia” prediction is less a novel repurposing hypothesis and more a confirmation of an already-recognized clinical use pattern, which is reflected in its comparatively strong evidence level (L3) relative to the other candidates.

By contrast, TxGNN’s single highest-scoring prediction, trigeminal nerve neoplasm (99.87%), has no supporting clinical trials, ICTRP records, or literature, and the evidence pack’s own rationale explicitly flags it as a probable false positive — lorazepam has no known antineoplastic or neuro-oncologic pharmacology. Several mid-ranked candidates (reading/thinking/audiogenic/startle/eating/micturition-induced seizures, trigeminal neoplasm, orgasm-induced seizures, AESD) cluster around reflex-epilepsy subtypes; while GABA-A potentiation is mechanistically plausible for seizure control in general, none of these rare subtypes have direct human efficacy data — evidence is limited to general status-epilepticus literature or unrelated animal-withdrawal studies. These were deprioritized in favor of the insomnia candidate for this report.

Clinical Trial Evidence

(Extracted from the “insomnia (disease)” prediction; trials most directly involving lorazepam or its clinical use pattern)

Trial Number	Phase	Status	Enrollment	Key Findings
NCT03331042	Phase 3	Completed	85	4-way crossover comparing SM-1 (diphenhydramine + zolpidem + delayed-release lorazepam) vs. diphenhydramine+zolpidem, diphenhydramine+lorazepam, and placebo in a phase-advance transient insomnia model
NCT02671760	Phase 2	Completed	39	Combination product containing lorazepam evaluated for total sleep time in adults with short-term insomnia
NCT04396327	Phase 2	Not yet recruiting	14	Pharmacodynamic crossover study of SM-1 (incl. lorazepam) vs. active comparator in a phase-advance transient insomnia model
NCT03338764	Phase 3	Withdrawn	0	Planned double-blind RCT of SM-1 (incl. lorazepam) vs. placebo for transient insomnia; withdrawn before enrollment
NCT06584513	N/A	Recruiting	470	Patient-centred intervention (BE-SAFE) to reduce benzodiazepine/sedative-hypnotic (incl. lorazepam) use for sleep problems in older adults
NCT04572750	N/A	Completed	170	Electronically-delivered self-management packet to promote benzodiazepine (e.g., Ativan/lorazepam) cessation in Veterans
NCT03405298	N/A	Completed	44	Patient education intervention to reduce inappropriate chronic benzodiazepine (incl. lorazepam) use among older adults
NCT02648776	N/A	Unknown	1400	Prospective Taiwan cohort assessing hypnotic (incl. benzodiazepine) use patterns, efficacy, and safety in elderly patients with sleep disorders
NCT00826553	Phase 1	Terminated	6	Polysomnography comparison of dexmedetomidine vs. GABA agonist sedatives (lorazepam class) on sleep architecture in mechanically ventilated ICU patients

Literature Evidence

PMID	Year	Type	Journal	Key Findings
3280615	1988	RCT	Journal of Clinical Pharmacology	Double-blind crossover in 8 chronic insomniacs: lorazepam 2mg outperformed flurazepam 30mg on most sleep parameters
30625122	2018	Review	The Medical Letter on Drugs and Therapeutics	Review of pharmacologic options, including benzodiazepines, for chronic insomnia
35087274	2022	Review	Journal of Multidisciplinary Healthcare	Efficacy, safety, and drug-drug interactions of insomnia therapies in COVID-19 (“coronasomnia”) patients
23330992	2013	Review	Expert Opinion on Drug Metabolism & Toxicology	Pharmacokinetics of anxiolytic drugs, including lorazepam, relevant to sedative-hypnotic dosing
10220122	1999	Cohort/Open-label	International Clinical Psychopharmacology	Comparison of lorazepam 0.5mg TID vs. 1.5mg HS dosing in chronic primary insomnia; TID regimen better addressed daytime fatigue
15341891	2004	Cohort	Sleep Medicine	Hypnotic prescription patterns in a large managed-care population, including lorazepam use for insomnia
25453732	2014	Observational	Clinical Therapeutics	Benzodiazepine/sedative-hypnotic use among seriously ill older veterans, including inappropriate use for insomnia
40110386	2025	Observational	Alpha Psychiatry	Prescribing trends of benzodiazepines and Z-drugs (incl. lorazepam) in Eastern China, 2015–2021
15040803	2004	Observational	Health and Quality of Life Outcomes	Sleep quality assessment and sedating drug (incl. benzodiazepines) use among hospitalized adults
19514972	2009	Preclinical	Drug Delivery	Intranasal microemulsion formulation of lorazepam (and other benzodiazepines) evaluated for sleep induction in a rat model

US Market Information

No Taiwan/US regulatory license records are currently available for lorazepam in this evidence pack (taiwan_regulatory.market_status: 未上市 / Not Marketed; 0 licenses on file).

Safety Considerations

Please refer to the package insert for safety information. No structured warnings, contraindications, or DDI data were returned by the current data sources (TFDA label lookup returned no result — flagged as a Blocking data gap; DDI query returned not_found).

Conclusion and Next Steps

Decision: Proceed with Guardrails

Rationale:

The insomnia indication is supported by a completed Phase 3 comparator trial, multiple Phase 2 studies, and a 1988 RCT showing lorazepam outperforming another hypnotic on sleep parameters — sufficient to justify guarded advancement, but this reflects reinforcement of an existing secondary use rather than a novel discovery. The TxGNN top-ranked candidate (trigeminal nerve neoplasm) and the reflex-epilepsy subtype candidates lack any direct clinical evidence and should remain on Hold/Research-Question status only.

To proceed, the following is needed:

TFDA (or equivalent) product label with warnings/contraindications — currently a Blocking gap; safety review (S1) cannot formally begin without it
Confirmed drug-drug interaction data (current DDI query: not found)
Structured original-indication/license data, since taiwan_regulatory.licenses and drug.original_indications are both empty
Formal relevance grading for the “pending” clinical trial and literature entries to refine the evidence base beyond the manually reviewed subset above
Disclaimer

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