Rifabutin

證據等級: L5

預測適應症: 10 個

Rifabutin: From MAC Infection Prophylaxis / HIV-TB Co-treatment to HIV Infectious Disease

One-Sentence Summary

Rifabutin is a rifamycin-class antibiotic already established for preventing and treating disseminated Mycobacterium avium complex (MAC) infection in AIDS patients and for treating HIV-tuberculosis co-infection. The TxGNN model’s top-ranked prediction is HIV infectious disease — but as the evidence pack itself flags, this largely reflects an already-approved use rather than a novel repurposing signal. Support is very strong (44 clinical trials, 22 publications), but this is confirmatory evidence, not discovery evidence. A more genuinely novel candidate — leprosy (rank 6, L3, Research Question) — is discussed separately below.

Quick Overview

Item	Content
Original Indication	Not formally recorded in Taiwan license data (drug not marketed in Taiwan). Per the evidence pack’s clinical rationale, Rifabutin’s established global indication is prevention/treatment of disseminated MAC infection in AIDS patients and treatment of HIV-related tuberculosis.
Predicted New Indication	HIV infectious disease (see caveat: substantially overlaps with an already-approved use, not a novel signal)
TxGNN Prediction Score	99.88%
Evidence Level	L1 (≥2 completed Phase 3 RCTs)
US Market Status	未上市 (Not Marketed)
Number of NDAs	0
Recommended Decision	Proceed with Guardrails

Why is This Prediction Reasonable?

Although the drug-level mechanism-of-action field is marked as a data gap, the clinical evidence embedded in this pack indicates that Rifabutin, a semisynthetic rifamycin, inhibits mycobacterial DNA-dependent RNA polymerase — the same core mechanism as rifampicin, giving it broad activity against Mycobacterium avium complex, M. tuberculosis, and M. leprae.

The TxGNN model’s top prediction, “HIV infectious disease,” is best understood through this mechanistic lens: Rifabutin has no direct antiretroviral activity. Its clinical role in HIV care is as an anti-mycobacterial agent used to prevent/treat opportunistic MAC infection and to treat tuberculosis in HIV-co-infected patients — a role it already holds by regulatory approval elsewhere. The evidence pack’s own rationale explicitly flags this: “此為FDA已核准之標準適應症（非新穎老藥新用）”. In other words, TxGNN has correctly recovered a known knowledge-graph relationship rather than surfaced a new repurposing opportunity — a useful validation of the model, but not a new drug-development lead.

A more genuinely novel signal in this evidence pack is leprosy (rank 6, L3 evidence, “Research Question”): Rifabutin shares rifampicin’s RNA-polymerase-inhibiting mechanism and has documented in vivo activity against M. leprae in armadillo and mouse models, with literature explicitly discussing it as a candidate substitute for rifampicin-resistant or rifampicin-intolerant leprosy patients — but no human RCT confirms efficacy. Other ranked predictions (multiple endocrine neoplasia, sclerosing cholangitis, a rare neurodevelopmental disorder, feline AIDS, SIV infection) have zero supporting evidence and no plausible mechanistic link, and should be held. Conjunctivitis is notable for having evidence pointing in the opposite direction — a single review lists Rifabutin as a cause of drug-induced ocular inflammation, not a treatment.

Clinical Trial Evidence

(for predicted indication: HIV infectious disease)

Trial Number	Phase	Status	Enrollment	Key Findings
NCT00002032	NA	Completed	750	Double-blind, placebo-controlled trial of oral rifabutin for prevention of MAC bacteremia in AIDS patients with CD4 ≤200; key foundational evidence for the approved indication
NCT00001030	Phase 3	Completed	1,100	Large RCT comparing clarithromycin vs. rifabutin vs. combination for prevention of MAC bacteremia/disseminated disease in HIV patients with CD4 ≤100
NCT00002122	Phase 3	Completed	720	Randomized study of daily/intermittent azithromycin and rifabutin regimens for prevention of disseminated MAC and fungal infections
NCT00002101	Phase 3	Completed	450	Three-arm trial of clarithromycin/ethambutol ± rifabutin (two doses) vs. placebo for treatment of MAC bacteremia in AIDS
NCT00001047	Phase 3	Completed	400	Comparison of four clarithromycin/ethambutol/rifabutin or clofazimine regimens for disseminated MAC disease in AIDS
NCT00001058	Phase 2/3	Completed	246	Multicenter RCT comparing clarithromycin-containing combination regimens (including rifabutin) for disseminated MAC disease
NCT00023361	NA	Completed	215	TBTC Study 23: treatment of HIV-related tuberculosis using an intermittent rifabutin-based regimen
NCT00001995	NA	Completed	200	Double-blind RCT of a rifabutin regimen for MAC bacteremia treatment in AIDS patients
NCT00002343	Phase 4	Completed	200	PK/PD study of rifabutin ± ethambutol for MAC prophylaxis in AIDS patients with CD4 ≤100
NCT00004736	Phase 1	Completed	44	Viral/immune dynamics of HAART in HIV patients with tuberculosis, supporting the HIV-TB co-treatment context

Note: the majority of the remaining ~34 registered trials in this evidence pack are pharmacokinetic drug-drug interaction studies between rifabutin and antiretrovirals (efavirenz, nevirapine, protease inhibitors, dolutegravir, cabotegravir, etc.), reflecting rifabutin’s role as a CYP3A inducer requiring dose adjustment in ART regimens rather than direct efficacy evidence.

Literature Evidence

(for predicted indication: HIV infectious disease)

PMID	Year	Type	Journal	Key Findings
23828580	2013	Cochrane Review	Cochrane Database Syst Rev	Rifamycins (including rifabutin) vs. isoniazid for TB prevention; supports rifamycin-based regimens as shorter, higher-completion alternatives
28233512	2017	Review	Microbiology Spectrum	Comprehensive review of the bidirectional TB-HIV disease relationship, framing rifabutin’s role in co-treatment
21726477	2009	Clinical Evidence Review	BMJ Clinical Evidence	Reviews treatment approaches for HIV-TB co-infection including rifamycin-based regimens
33294914	2021	Cohort/PK	J Antimicrob Chemother	Rifabutin PK and safety in TB/HIV-coinfected children on LPV/r-based second-line ART; notes prior neutropenia signal
31139825	2019	Cohort	J Antimicrob Chemother	Safety and efficacy of rifabutin in HIV/TB-coinfected children on lopinavir/ritonavir-based ART
25281400	2015	Cohort/PK	J Antimicrob Chemother	PK and short-term safety of rifabutin combined with lopinavir/ritonavir in young HIV-infected children
26832753	2016	Pooled PK Analysis	J Antimicrob Chemother	Population PK pooled analysis of rifabutin-HIV protease inhibitor interactions, informing dosing recommendations
16206114	2005	PK Study	Clin Infect Dis	Evaluates the rifabutin-efavirenz interaction and dose-adjustment recommendations in HIV-TB patients
36385424	2023	Population PK Model	Br J Clin Pharmacol	Rifabutin-dolutegravir drug-drug interaction, positioning rifabutin as an alternative to rifampicin in integrase-inhibitor regimens
7736687	1995	PK Review	Clinical Pharmacokinetics	Foundational clinical pharmacokinetics review supporting rifabutin’s use for MAC prophylaxis in HIV-positive patients

Market Information (Taiwan)

Rifabutin currently holds no marketing authorization in Taiwan (0 licenses; market status: 未上市/Not Marketed). No product name, dosage form, or Taiwan-approved indication text is available in this evidence pack.

Safety Considerations

Please refer to the package insert for safety information.

Note: The evidence pack flags a Blocking-severity data gap (DG001 – TFDA package insert warnings/contraindications) that must be resolved before this candidate can undergo formal safety review (S1). Drug interaction data is also marked “not found” in this pack; however, the clinical trial and literature evidence above independently documents extensive, clinically significant interactions between rifabutin and antiretrovirals (protease inhibitors, NNRTIs, integrase inhibitors) via CYP3A induction/inhibition, requiring dose adjustment in any HIV-TB co-treatment context.

Conclusion and Next Steps

Decision: Proceed with Guardrails

Rationale: Clinical evidence for rifabutin in the HIV/MAC/TB context is very strong (L1: multiple completed Phase 3 RCTs, 44 trials, 22 publications) — but this largely confirms an existing approved use rather than identifying a novel repurposing opportunity, so it should be scoped as a label-extension/confirmation review rather than a new indication. Meanwhile, a Blocking data gap (missing TFDA warnings/contraindications) currently prevents any formal safety sign-off.

To proceed, the following is needed:

Obtain TFDA package insert warnings/contraindications (DG001, Blocking) — required before S1 safety review can begin
Obtain detailed MOA/pharmacology profile from DrugBank (DG002)
Clarify Taiwan regulatory pathway, since Rifabutin currently has zero Taiwan market authorizations (未上市) — determine whether named-patient/import use is the intended pathway
Reclassify the “HIV infectious disease” prediction internally as a known-use confirmation rather than a novel repurposing candidate, to avoid overstating novelty in downstream reporting
If pursuing genuinely novel signals, open a separate research track for leprosy (rank 6, L3, mechanistically plausible via shared RNA-polymerase inhibition with rifampicin, but no human RCT yet) — this is the strongest true repurposing candidate in this evidence pack
Disclaimer

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