Ethambutol

證據等級: L5

預測適應症: 5 個

Ethambutol: From Tuberculosis to Epiglottitis

One-Sentence Summary

Ethambutol is a first-line antimycobacterial agent used as part of RHZE combination therapy for tuberculosis (TB). This multi-indication TxGNN analysis predicts 5 new potential applications, with Epiglottitis ranked highest (99.90% score); however, mechanistic evaluation reveals this as likely a knowledge-graph artifact, while Tuberculous Peritonitis (Rank 4) represents the most evidence-supported prediction with L3 evidence backed by a Cochrane review and 20 publications. No clinical trials were identified for any of the 5 predicted indications.

Quick Overview

Item	Content
Original Indication	Tuberculosis (first-line RHZE combination therapy)
Predicted New Indication (Top Rank)	Epiglottitis
TxGNN Prediction Score	99.90%
Evidence Level	L5
India Market Status	✗ Not Marketed
Number of Registrations	0
Recommended Decision	Hold

Why Is This Prediction Reasonable?

Currently, detailed mechanism of action data was not available from queried data sources. Based on established pharmacological knowledge, Ethambutol acts by specifically inhibiting arabinosyl transferase enzymes (EmbA, EmbB, EmbC) that are essential for synthesizing arabinogalactan in the mycobacterial cell wall. This target exists exclusively in mycobacteria — primarily Mycobacterium tuberculosis and certain non-tuberculous mycobacteria (NTM) — and is absent in all other bacterial species.

The principal pathogens responsible for conventional epiglottitis — Haemophilus influenzae type b (Hib), Group A Streptococcus, and Staphylococcus aureus — do not carry EmbA/B/C targets. Ethambutol therefore has no antibacterial activity against these organisms. Tuberculous epiglottitis is an extremely rare extrapulmonary TB manifestation; when it does occur, using Ethambutol as part of standard RHZE therapy represents treatment of an established TB indication rather than a genuine repurposing opportunity.

The TxGNN model’s high prediction score (99.90%) is most likely attributable to over-generalization through TB-associated laryngeal nodes in the knowledge graph, creating an apparent — but mechanistically empty — link to conventional epiglottitis. This pattern of high-score false positives for spectrum-limited antimicrobials is a recognized limitation of KG-based prediction models.

Multi-Indication Prediction Overview

This report covers a 5-indication multi-prediction analysis. All predictions are summarized below for comparative evaluation:

Rank	Disease	TxGNN Score	Evidence Level	Recommendation	Assessment
1	Epiglottitis	99.90%	L5	Hold	KG over-generalization; no mechanistic basis against common bacterial pathogens
2	Laryngitis	99.71%	L4	Research Question	TB laryngitis mechanistically valid; 20 case-level publications and case series
3	Meningococcal Infection	99.63%	L5	Hold	KG false positive; Neisseria meningitidis lacks EmbA/B/C targets
4	Peritonitis	99.20%	L3	Proceed with Guardrails	Tuberculous peritonitis: Cochrane review + systematic review provide direct support
5	Infectious Otitis Media	99.06%	L5	Hold	No mechanistic basis; 0 publications identified

Priority indication for further action: Tuberculous Peritonitis (Rank 4) is the mechanistically sound and best-evidenced application in this analysis.

Clinical Trial Evidence

Currently no related clinical trials are registered for any of the 5 predicted indications.

Literature Evidence

Peritonitis — Highest Evidence Indication (L3)

PMID	Year	Type	Journal	Key Findings
16197489	2005	Systematic Review	Alimentary Pharmacology & Therapeutics	Comprehensive review of tuberculous peritonitis covering presentation, diagnostic strategies, and treatment; supports standard antituberculous regimens including Ethambutol
27801499	2016	Cochrane Review	Cochrane Database of Systematic Reviews	Six-month anti-TB regimen (including Ethambutol) shown effective for abdominal tuberculosis; addresses concerns about treatment duration and relapse risk
11396540	2001	Retrospective Study	European Journal of Gastroenterology & Hepatology	26 cases of tuberculous peritonitis; triple antituberculous therapy efficacy evaluated over 6 months
8384230	1993	Retrospective Study	The Journal of Infection	Ethambutol + rifabutin + amikacin triple therapy in 31 HIV patients with symptomatic MAC infection including peritonitis; clinical response assessed
34666716	2021	Case Report	BMC Nephrology	Mycobacterium avium-related peritonitis in a peritoneal dialysis patient; reviews NTM peritonitis management with Ethambutol-containing regimens
16206710	2005	Case Series	Srpski arhiv za celokupno lekarstvo	Tuberculous and fungal peritonitis in peritoneal dialysis patients; antituberculous therapy outcomes and complications
10913399	2000	Case Report + Review	Clinical Infectious Diseases	Tuberculous peritonitis in continuous ambulatory peritoneal dialysis (CAPD) patient; standard RHZE therapy successful
1439720	1992	Case Report	South Dakota Journal of Medicine	TB peritonitis in alcoholic cirrhosis patient; successfully treated with isoniazid + Ethambutol for 24 months
29776936	2018	Case Report	BMJ Case Reports	Severe Ethambutol + isoniazid neurotoxicity (progressive paralysis, vision loss) in a peritoneal dialysis patient; critical safety signal for dose adjustment in renal failure
33690080	2021	Basic Science	Tuberculosis (Edinburgh)	PknL kinase deletion in M. tuberculosis reduces Ethambutol and isoniazid efficacy in peritoneal macrophages; mechanistic insight into Ethambutol’s action at the infection site

Laryngitis — Secondary Indication of Interest (L4)

PMID	Year	Type	Journal	Key Findings
2806495	1989	Retrospective Case Series	European Respiratory Journal	41 cases of laryngeal TB (1975–1985); treated with isoniazid, rifampicin, and Ethambutol; true vocal cords most commonly affected, epiglottis second
4644838	1972	Case Series	Ftiziologia	Ethambutol delivered via aerosol or endobronchial instillation in laryngeal and bronchopulmonary TB; novel local delivery approach
25397382	2014	Case Report	Ear, Nose & Throat Journal	Primary laryngeal TB: complete resolution with RHZE (including Ethambutol); highlights TB mimicking laryngeal cancer
18702845	2009	Case Report	Journal of Laryngology and Otology	Vocal fold paralysis from M. kansasii (NTM) nerve compression; highlights NTM laryngeal disease where Ethambutol may have partial efficacy
33133876	2020	Case Report	Cureus	Coexistent laryngeal + endobronchial + pulmonary TB in an immunocompetent host; treated with standard anti-TB regimen

Epiglottitis — Top TxGNN Prediction (L5)

PMID	Year	Type	Journal	Key Findings
14720571	2004	Clinical Review	The Lancet. Infectious Diseases	Laryngeal tuberculosis review; epiglottis identified as one of the anatomical sites involved in laryngeal TB; standard anti-TB treatment discussed
2806495	1989	Retrospective Case Series	European Respiratory Journal	41 laryngeal TB cases; epiglottis listed as one of affected sites; treatment regimen included Ethambutol

Safety Considerations

Drug Interactions: Ethambutol has 112 documented interactions in the ddinter database. The following interactions carry at least moderate clinical significance:

Interacting Drug	Severity	Clinical Note
Aluminum hydroxide	Moderate	Reduces Ethambutol oral absorption; separate dosing times by at least 2 hours
Metronidazole	Moderate	Potential additive peripheral neuropathy risk; monitor neurological status
Tinidazole	Moderate	Potential additive neurological effects; monitor during co-administration
Simvastatin	Moderate	Potential interaction; monitor for altered statin or Ethambutol levels
Rosuvastatin	Moderate	Potential interaction; monitor during co-administration
Naltrexone	Moderate	Monitor for altered drug effects during co-administration
Picosulfuric acid	Moderate	Monitor during co-administration

Additional interactions of currently unknown clinical significance include Pantoprazole, Omeprazole, Lansoprazole, Vancomycin, Prednisone, Hydrocortisone, Prednisolone, Sulfasalazine, Ranitidine, Nystatin, and others (112 interactions total).

For complete warnings and contraindications, please refer to the package insert.

Conclusion and Next Steps

Decision: Hold (Epiglottitis — Top TxGNN Prediction)

Decision: Proceed with Guardrails (Tuberculous Peritonitis — Best-Supported Indication)

Rationale: The highest-ranked TxGNN prediction (epiglottitis, 99.90%) has no mechanistic plausibility for conventional bacterial etiologies and is supported by only 2 tangential publications; this is a clear knowledge-graph false positive that should not progress further. In contrast, Tuberculous Peritonitis is mechanistically sound — Ethambutol’s selective antimycobacterial activity directly targets the causative pathogen — and is supported by a Cochrane review and a dedicated systematic review, placing it at L3 evidence with a clear therapeutic rationale.

To proceed for Tuberculous Peritonitis, the following is needed:

Retrieve Ethambutol MOA data from DrugBank (DB00330) to complete the mechanistic section of the dossier
Obtain the local package insert (TFDA or equivalent authority) to document formal warnings and contraindications — particularly renal impairment dosing, given that PMID 29776936 documents severe neurotoxicity in peritoneal dialysis patients
Clarify the regulatory pathway for registering Ethambutol in India, as it currently holds zero product licenses
Establish subpopulation-specific dosing guidelines for peritoneal dialysis and renal impairment populations before any clinical use in peritonitis
Design a comparative protocol evaluating Ethambutol-containing versus Ethambutol-sparing RHZE regimens specifically for tuberculous peritonitis, given ongoing questions about drug absorption in the abdominal cavity
Disclaimer

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