Lamivudine

證據等級: L5

預測適應症: 5 個

Lamivudine: From Unspecified Original Indication to Simian Immunodeficiency Virus Infection

One-Sentence Summary

Lamivudine (DrugBank DB00709) is a nucleoside reverse transcriptase inhibitor (NRTI); this evidence pack does not contain data on its original approved indication(s) or mechanism of action. The TxGNN model’s top-ranked prediction is Simian Immunodeficiency Virus (SIV) Infection — a disease that occurs only in non-human primates, not humans. Supporting evidence consists of 0 clinical trials and 20 publications, nearly all animal or in vitro virology studies with no direct human-disease relevance.

Quick Overview

Item	Content
Original Indication	Not on file (evidence pack contains no `original_indications` data; drug not marketed in Taiwan)
Predicted New Indication	Simian Immunodeficiency Virus Infection
TxGNN Prediction Score	99.93%
Evidence Level	L4
Taiwan Market Status	未上市 (Not Marketed)
Number of NDAs	0
Recommended Decision	Hold

Why is This Prediction Reasonable?

Currently, detailed mechanism of action data is not available for Lamivudine in this evidence pack (flagged as a High-severity data gap, DG002). Based on general pharmacological class, Lamivudine is a cytidine-analog NRTI that inhibits retroviral reverse transcriptase — the enzyme class also used by SIV, a lentivirus closely related to HIV.

However, this mechanistic plausibility does not translate into a valid repurposing candidate. SIV infection is a veterinary/experimental condition confined to macaques and other non-human primates; it is not a human disease entity and has no corresponding clinical indication pathway. The supporting literature confirms this: essentially all 20 publications are animal-model or in vitro virology studies (e.g., SIV-infected macaque pharmacokinetics, M184V resistance mutation studies), not human trials. No clinical trials are registered for this “indication” at all. The TxGNN score is very high, but this reflects graph-embedding similarity (SIV and HIV share lentivirus/reverse-transcriptase biology) rather than a translatable clinical signal — consistent with the evidence pack’s own scoring, which assigns Evidence Level L4 and a Hold recommendation.

Clinical Trial Evidence

Currently no related clinical trials registered.

Literature Evidence

PMID	Year	Type	Journal	Key Findings
31658118	2020	Review (unrelated drug, HIV context only)	Current Opinion in HIV and AIDS	Discusses islatravir, a different RT translocation inhibitor, for HIV-1; not specific to lamivudine or SIV
39509655	2024	Epidemiology review (human HIV-1/2, not SIV)	AIDS Reviews	Reviews HIV-1/2 burden in Ivory Coast; SIV mentioned only as evolutionary precursor of HIV
19240457	2009	Animal study (macaque PEP model)	AIDS (London)	Zidovudine+lamivudine+indinavir post-exposure prophylaxis after vaginal SIV exposure in macaques
11689641	2001	Animal study	Journal of Virology	Bone marrow hematopoiesis defects in SHIV-infected macaques despite HAART viral suppression
12021341	2002	In vitro/animal virology	Journal of Virology	M184V resistance mutation emergence in SIV-infected macaques treated with lamivudine/emtricitabine
16973590	2006	Animal study	Journal of Virology	Viral decay kinetics in SIV-infected macaques on quadruple antiretroviral therapy
9237655	1997	In vitro pharmacology	FEBS Letters	Aryloxyphosphoramidate prodrugs of ddA/d4A potentiate anti-HIV/SIV/HBV activity vs. parent compounds
12502828	2003	Animal study	Journal of Virology	Tenofovir selects M184V reversion in SIV reverse transcriptase even with concurrent lamivudine
15919889	2005	Animal study	Journal of Virology	HAART (efavirenz/lamivudine/tenofovir) suppresses viral load in RT-SHIV-infected rhesus macaques
14610172	2003	Animal study	Journal of Virology	Lymphocyte proliferation kinetics in SIV-infected macaques under early post-exposure HAART prophylaxis

Safety Considerations

Please refer to the package insert for safety information.

Conclusion and Next Steps

Decision: Hold

Rationale: The top-ranked prediction (SIV infection) is not a human disease entity, and its supporting literature is exclusively preclinical/veterinary. No clinical trials exist. This is consistent with the evidence pack’s own S0/Hold scoring. Notably, all 5 TxGNN-ranked candidates in this pack were independently scored Hold: the remaining four are either a mismatched veterinary condition with human-trial evidence incorrectly attached via keyword collision (feline immunodeficiency syndrome — the 5 “supporting” trials are human HIV-1 Phase 3/4 studies unrelated to cats), a rare genetic neurodevelopmental disorder with zero mechanistic or empirical support, an obsolete/deprecated disease ontology term with no evidence, and chronic hepatitis C — a mechanistic mismatch, since HCV is an RNA virus replicating via RNA-dependent RNA polymerase (not reverse transcriptase), and the “supporting” trials/literature are almost entirely existing chronic hepatitis B evidence mislabeled under HCV. No candidate in this pack currently supports advancement past S0.

To proceed, the following is needed:

TFDA label/warnings and contraindications (currently blocking — DG001)
Confirmed mechanism of action data from DrugBank or primary literature (DG002)
Correction of disease-label mapping errors observed in candidates 2 and 5 (FIV/HIV confusion; HBV/HCV confusion) before any evidence re-scoring
Re-run of candidate generation restricted to valid human ICD/SNOMED disease entities, excluding obsolete ontology terms and non-human disease models
Disclaimer

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