Metronidazole
| 證據等級: L5 | 預測適應症: 10 個 |
目錄
Metronidazole: From Anaerobic Bacterial/Protozoal Infections to Pneumocystosis
One-Sentence Summary
Metronidazole is a nitroimidazole antimicrobial generally used for anaerobic bacterial and protozoal infections (e.g., amoebiasis, trichomoniasis, giardiasis, anaerobic/IBD-related infections); this evidence pack’s own regulatory and mechanism-of-action fields for the drug are marked as data gaps. The TxGNN model’s top-ranked prediction is Pneumocystosis, with a 99.99% score, but the accompanying evidence review found 24 clinical trials and 10 publications, almost none of which support this specific indication — the pack itself flags this as a likely mismatch (Evidence Level L5, recommendation Hold).
Quick Overview
| Item | Content |
|---|---|
| Original Indication | Not recorded in evidence pack (original_indications empty, no TFDA/US license on file); generally known as anaerobic bacterial & protozoal infections |
| Predicted New Indication | Pneumocystosis |
| TxGNN Prediction Score | 99.99% |
| Evidence Level | L5 |
| US Market Status | Not Marketed |
| Number of NDAs | 0 |
| Recommended Decision | Hold |
Why is This Prediction Reasonable?
Currently, detailed mechanism of action data is not available in this evidence pack (original_moa is marked as a data gap, and original_indications is empty). Based on general pharmacological knowledge not sourced from this pack, metronidazole is a nitroimidazole with established antibacterial activity against anaerobic bacteria and antiprotozoal activity (amoebiasis, trichomoniasis, giardiasis); it has no recognized antifungal activity.
For the top-ranked prediction, pneumocystosis (Pneumocystis jirovecii pneumonia), the evidence pack’s own review finds no credible mechanistic basis. Pneumocystis is classified as a fungus, and metronidazole is not active against it — standard treatment is trimethoprim-sulfamethoxazole, pentamidine, or atovaquone. All 24 retrieved clinical trials were graded as unrelated to this indication (primary-care payment models, dementia caregiving technology, mobile depression therapy, etc.), and the 10 retrieved publications are general antiprotozoal/HIV-opportunistic-infection reviews or incidental case reports in which metronidazole and pneumocystosis co-occur in the same patient history without any direct treatment relationship.
TxGNN’s high score most likely reflects a knowledge-graph co-occurrence signal — both entities frequently appear together in immunocompromised/HIV-related clinical contexts — rather than a genuine pharmacological mechanism. This is consistent with the evidence pack’s own conclusion of Evidence Level L5 (model prediction only, no supporting studies) and a Hold recommendation.
Clinical Trial Evidence
| Trial Number | Phase | Status | Enrollment | Key Findings |
|---|---|---|---|---|
| NCT02731716 | NA | Active, not recruiting | 117 | Primary care payment model study in Hawaii — unrelated to pneumocystosis (mismatched retrieval) |
| NCT04631120 | NA | Withdrawn | 0 | Dementia caregiver technology study, withdrawn — unrelated |
| NCT04463914 | NA | Completed | 649 | Mobile app for depression treatment — unrelated |
| NCT05892666 | N/A | Recruiting | 4000 | Comparison of walk-in clinic/PCP/ED care value — unrelated |
| NCT05256303 | NA | Completed | 160 | Hospital-level home care RCT — unrelated to infection treatment |
| NCT07203339 | NA | Recruiting | 610 | tDCS vs. duloxetine for fibromyalgia — relevance not yet graded, topic unrelated |
| NCT04418232 | Phase 1 | Completed | 243 | Alzheimer’s service-utilization feasibility study — unrelated (mismatched retrieval) |
| NCT05340426 | Phase 1 | Withdrawn | 0 | Porcine kidney xenotransplantation trial, withdrawn — unrelated |
| NCT02571673 | N/A | Completed | 65 | Head & neck cancer survivorship tool study — relevance not yet graded, topic unrelated |
| NCT02972203 | NA | Completed | 87 | Mindfulness intervention pilot in primary care — relevance not yet graded, topic unrelated |
Note: All 24 retrieved trials were reviewed by the evidence pipeline; none constitute a genuine metronidazole–pneumocystosis interventional trial.
Literature Evidence
| PMID | Year | Type | Journal | Key Findings |
|---|---|---|---|---|
| 7355683 | 1980 | Review | American Family Physician | Names metronidazole as drug of choice for amebic colitis/trichomoniasis, but names TMP-SMX (not metronidazole) as the drug of choice for pneumocystis pneumonia |
| 26518395 | 2015 | Review | Topics in Antiviral Medicine | Review of HIV-related opportunistic infections (background context only; no metronidazole-PCP efficacy data) |
| 2996829 | 1985 | Review | Clinical Pharmacy | Review of AIDS infectious complications (PCP is most common); no metronidazole treatment data for PCP |
| 6282154 | 1982 | Case report | American Review of Respiratory Disease | Patient with PCP+CMV pneumonia who had separately received metronidazole for an unrelated diarrheal illness — incidental co-occurrence, not treatment evidence |
| 2338506 | 1990 | Case report | Kansenshogaku Zasshi | AIDS patient treated with metronidazole for amebic dysentery who subsequently developed PCP — incidental co-occurrence, not causal/treatment evidence |
| 1782741 | 1991 | Review | Clinical Pharmacokinetics | General pharmacokinetic review of antiprotozoal therapy; no pneumocystosis-specific data |
| 1545596 | 1992 | Review | Mayo Clinic Proceedings | General review of antiparasitic agents; does not address metronidazole for pneumocystosis |
| 6771863 | 1980 | Review | Reviews of Infectious Diseases | Critique of antimicrobial prophylaxis trial design; unrelated to pneumocystosis |
| 16496064 | 2005 | Case report | Journal of the Formosan Medical Association | Colon perforation from CMV/amebic colitis in an AIDS patient; unrelated to pneumocystosis |
| 2280469 | 1990 | Review | Nihon Rinsho | General review of antiprotozoal drugs (no abstract available) |
US Market Information
No marketing authorization records are present in this evidence pack. Market status: Not Marketed, with 0 licenses on file (taiwan_regulatory.total_licenses = 0).
Safety Considerations
Please refer to the package insert for safety information.
Conclusion and Next Steps
Decision: Hold
Rationale: TxGNN assigns a very high score (99.99%) to pneumocystosis, but the evidence pack’s own review found the clinical trials and literature retrieved to be largely unrelated or only incidentally co-occurring, with no direct treatment evidence and a mechanistic conflict (metronidazole lacks antifungal activity against Pneumocystis jirovecii). This is consistent with the assigned Evidence Level L5 (model prediction only).
To proceed, the following is needed:
- TFDA package insert (warnings/contraindications) — currently a Blocking data gap (DG001) preventing S1 safety screening
- Verified DrugBank MOA record — currently a High-severity data gap (DG002)
- Confirmed original approved indication(s) for metronidazole (currently empty/unrecorded in this pack)
- If pursuing a repurposing signal from this same candidate set, consider re-scoping toward the higher-evidence candidates already in this pack — e.g., cap polyposis (rank 9, L3, “Proceed with Guardrails,” with an on-topic mechanistic case study, PMID 12141801) or ulcerative proctosigmoiditis / ulceration of vulva (rank 3/10, L3, “Research Question”) — rather than pneumocystosis
Disclaimer
This content is for research purposes only and does not constitute medical advice. Clinical validation is required before any clinical application.