Atazanavir

證據等級: L5

預測適應症: 6 個

One-Sentence Summary

Atazanavir (DrugBank DB01072) is an HIV-1 protease inhibitor originally developed for HIV-1 infection (marketed as Reyataz). The TxGNN model's best-supported prediction extends this to congenital/perinatal HIV exposure and AIDS-related complex — not a novel mechanism, but a formal consolidation of the drug's existing antiretroviral use, backed by 30+ clinical trials and 10 publications. The model's top four raw-ranked predictions (feline AIDS, SIV infection, a rare neurodevelopmental disorder, and hyperlipidemia) show no credible mechanistic or evidentiary support and have been excluded from this report as likely knowledge-graph artifacts (see note below).

Quick Overview

Item	Content
Original Indication	HIV-1 infection (antiretroviral therapy, brand Reyataz) — inferred from trial/rationale data; not present in Evidence Pack license fields (drug not marketed in Germany)
Predicted New Indication	Congenital/Perinatal HIV Exposure & AIDS-Related Complex (indication consolidation, not a novel mechanism)
TxGNN Prediction Score	99.71%
Evidence Level	L1
Germany Market Status	✗ Not Marketed
Number of Authorizations	0
Recommended Decision	Hold

Why is This Prediction Reasonable?

Detailed mechanism-of-action data from DrugBank is currently unavailable (Data Gap DG002). Based on the trial and literature evidence available, atazanavir is an HIV-1 aspartyl protease inhibitor: it blocks cleavage of the Gag-Pol polyprotein precursor, preventing maturation of infectious virions and halting viral replication.

The two indications with real supporting evidence — "AIDS-related complex" and "congenital human immunodeficiency virus" — are not mechanistically distinct from atazanavir's original HIV-1 indication. AIDS-related complex is simply an earlier/broader clinical staging term for HIV disease, and congenital HIV exposure concerns the same virus and the same protease target in a perinatal/pediatric population. Extensive Phase 3 trial history (including two dedicated pediatric programs, PRINCE I and PRINCE II) and multiple pregnancy pharmacokinetic and safety cohort studies (e.g., PHACS SMARTT, IMPAACT P1026s) support use across these populations. This should be read as label/population extension of an already-approved use, not a genuine drug-repurposing discovery.

Note: Lower-Confidence Predictions Excluded from This Report

The Evidence Pack's top four ranked candidates were not carried forward as headline findings, because the pack's own rationale flags each as implausible or unsupported:

Rank	Disease	Score	Issue
1	Feline acquired immunodeficiency syndrome	99.98%	Veterinary indication (FIV); HIV-1 protease specificity does not cross-react; likely embedding-similarity false positive ("immunodeficiency virus" text overlap)
2	Simian immunodeficiency virus infection	99.98%	Non-human-primate animal model only; no clinical trials, single animal-study citation
3	Rare neurodevelopmental disorder (ataxic gait, absent speech, decreased white matter)	99.98%	No known mechanistic relationship to protease inhibition; no evidence at all
4	(Obsolete) familial combined hyperlipidemia	99.82%	Evidence points the opposite direction — atazanavir is comparatively lipid-neutral versus other PIs; disease term itself is deprecated

These should not be pursued as repurposing candidates without independent validation.

Clinical Trial Evidence

Trial Number	Phase	Status	Enrollment	Key Findings
NCT04518228	N/A	Completed	205	PK of antiretroviral & anti-TB drugs during pregnancy and postpartum
NCT00326716	Phase 1	Completed	69	ATV/RTV dosing and PK in HIV-1 infected pregnant women
NCT00042289	Phase 4/N/A	Completed	1578	IMPAACT P1026s: large PK study of ARV/TB drugs in pregnant women and infants
NCT01691794	Phase 4	Completed	108	Safety of ATV capsule + RTV in HIV-infected pediatric patients (6–18y)
NCT01099579	Phase 3	Completed	82	PRINCE I: ATV powder + RTV safety/efficacy/PK in pediatric patients (3mo–6y)
NCT01335698	Phase 3	Completed	160	PRINCE II: ATV powder + RTV in pediatric patients (3mo–11y)
NCT02951052	Phase 3	Active, not recruiting	618	ATLAS: switch to long-acting cabotegravir + rilpivirine vs. continuing ART (incl. PI-based regimens)
NCT02269917	Phase 3	Completed	1149	Switch to D/C/F/TAF vs. continuing boosted-PI regimen in virologically suppressed HIV-1 patients
NCT01910402	Phase 3	Completed	499	DTG/ABC/3TC vs. ATV+RTV+TDF/FTC in ART-naive women (relevant to childbearing-age population)
NCT00272779	Phase 3	Completed	1057	96-week ATV/RTV vs. LPV/RTV + TDF/FTC efficacy and safety in treatment-naive HIV-1 patients

Literature Evidence

PMID	Year	Type	Journal	Key Findings
27242802	2016	Cohort (PHACS SMARTT)	Frontiers in Immunology	Large (3,500+) HIV-exposed-uninfected infant/child cohort assessing safety of in-utero ARV exposure
25383770	2015	Cohort	JAMA Pediatrics	Congenital anomalies and in-utero ARV exposure in HIV-exposed uninfected infants
40011239	2025	Case/non-case study	Eur J Clin Pharmacol	European registry study of congenital anomaly risk after fetal ARV exposure
24992294	2015	Cohort	Antiviral Therapy	Atazanavir exposure remains effective during pregnancy regardless of tenofovir co-administration
31595301	2020	Pharmacovigilance database analysis	Clin Infect Dis	Comparator safety-signal analysis for ARVs in pregnancy (dolutegravir focus)
28459118	2016	Cohort	J AIDS Immune Res	Newborn hearing screening outcomes in HIV-exposed uninfected infants
29859254	2018	In vitro mechanistic study	Reproductive Toxicology	ATV/RTV interactions with placental ABC transporters affecting transplacental disposition
19290032	2009	Cohort	AIDS Reviews	GI adverse event risk factors in HIV-treated vs. untreated patients (protease-inhibitor context)
28991888	2018	Cohort	J Acquir Immune Defic Syndr	ART regimen choice and incidence of AIDS-defining neurological conditions

Safety Considerations

Please refer to the package insert for safety information. No drug-specific warnings, contraindications, or drug-interaction data were retrievable in this Evidence Pack (DDI query returned no results; TFDA/BfArM label data marked as a Blocking data gap — DG001).

Conclusion and Next Steps

Decision: Hold

Rationale: Although congenital/perinatal HIV exposure and AIDS-related complex are backed by strong (L1) trial and cohort evidence, this represents consolidation of atazanavir's existing HIV/AIDS indication rather than a genuine new repurposing opportunity. More importantly, a Blocking-severity data gap (missing TFDA/BfArM label warnings and contraindications) prevents any safety evaluation (S1), and the drug currently holds zero marketing authorizations in Germany. The model's top four raw-ranked predictions were separately assessed and rejected as low-credibility artifacts.

To proceed, the following is needed:

TFDA/BfArM label PDF retrieval and parsing for warnings/contraindications (resolves DG001, currently blocking)
DrugBank MOA confirmation (resolves DG002)
A regulatory determination of whether "congenital HIV" / "AIDS-related complex" require a distinct submission or fall under the existing HIV-1 indication scope
If a genuine repurposing signal is desired, re-run TxGNN scoring excluding the four flagged artifact predictions and review any remaining mid-ranked candidates
Disclaimer

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