Clindamycin

證據等級: L5

預測適應症: 10 個

Clindamycin: From Bacterial Infections to Punctate Epithelial Keratoconjunctivitis

One-Sentence Summary

Clindamycin is a lincosamide antibiotic used internationally for anaerobic bacterial infections, skin and soft tissue infections, bone and joint infections, and toxoplasmosis. The TxGNN model predicts it may be effective for Punctate Epithelial Keratoconjunctivitis as its top-ranked new indication, with 0 clinical trials and 0 publications directly supporting this direction. ⚠️ The high prediction score (99.97%) most likely reflects non-specific proximity within ocular surface disease nodes in the knowledge graph rather than genuine therapeutic relevance; the overall recommendation for all five predicted indications is Hold.

Quick Overview

Item	Content
Original Indication	Not registered in Denmark; internationally used as a lincosamide antibiotic for anaerobic/Gram-positive bacterial infections and toxoplasmosis
Predicted New Indication	Punctate Epithelial Keratoconjunctivitis
TxGNN Prediction Score	99.97%
Evidence Level	L5
Denmark Market Status	Not marketed
Number of Marketing Authorisations	0
Recommended Decision	Hold

Why Is This Prediction Reasonable?

Currently, detailed mechanism of action data is not available in the evidence pack. Based on known pharmacological information, Clindamycin is a lincosamide antibiotic that inhibits bacterial protein synthesis by binding to the 50S ribosomal subunit. It has broad activity against anaerobic bacteria, certain Gram-positive cocci, and protozoa — most notably Toxoplasma gondii — where it is sometimes used as a second-line agent in combination with pyrimethamine.

Punctate epithelial keratoconjunctivitis (PEK) is primarily caused by viral pathogens (adenovirus, herpes simplex virus), dry eye syndrome, toxic or chemical irritants, or systemic diseases such as Sjögren’s syndrome. It is not primarily a bacterial infectious condition. Clindamycin’s antibacterial and antiprotozoal mechanism therefore provides no direct therapeutic basis for PEK.

The TxGNN model’s very high score (99.97%) most likely reflects non-specific knowledge graph associations across ocular surface disease nodes — not genuine therapeutic applicability. The mechanistic link between Clindamycin and punctate epithelial keratoconjunctivitis is not supported by current evidence, and this prediction should be treated with significant caution before any further investment.

Clinical Trial Evidence

Currently no related clinical trials registered.

Literature Evidence

Currently no related literature available.

All TxGNN-Predicted Indications — Summary Overview

For transparency, the full set of unique predicted indications and their assessment is provided below:

Rank	Disease	TxGNN Score	Evidence Level	Recommendation	Key Concern
1	Punctate Epithelial Keratoconjunctivitis	99.97%	L5	Hold	Primarily viral/dry-eye aetiology; no mechanistic basis for antibiotic
3	Exposure Keratitis	99.80%	L4	Hold	Core treatment is lubrication/lid repair; Clindamycin role limited to secondary bacterial superinfection only
5	Non-Human Animal Disease	99.69%	L5	Hold	⚠️ Reverse-association warning: Clindamycin is a risk factor for C. difficile infection, not a treatment; not a human clinical indication
7	Neurotrophic Keratopathy	99.49%	L5	Hold	Nerve-degeneration pathology; no known mechanistic link to an antibiotic
9	Epidemic Keratoconjunctivitis	99.49%	L4	Hold	Primarily adenoviral; Clindamycin has no antiviral activity; available literature is veterinary only

Note: Ranks 2, 4, 6, 8, and 10 in the evidence pack are duplicates of the above entries and have been consolidated here.

Literature Evidence — Exposure Keratitis (Rank 3, L4)

Although no clinical trials exist, four publications were retrieved for the exposure keratitis prediction. They are presented for completeness, though none directly evaluate Clindamycin in exposure keratitis.

PMID	Year	Type	Journal	Key Findings
22880135	2012	Retrospective case series	PLOS ONE	MRSA prevalence in ocular infections; antibiotic susceptibility data for S. aureus eye isolates — indirect context only
24244625	2013	Retrospective case series	PLOS ONE	Demographics and clinical outcomes of S. aureus keratitis; MRSA vs. MSSA comparison — indirect context only
11581057	2001	Case series	Ophthalmology	First reported contact lens-associated Bacillus cereus keratitis; no Clindamycin intervention
33847093	2021	Veterinary clinical study	Polish J Vet Sci	Feline ocular toxoplasmosis treatment outcomes — veterinary study, not applicable to human exposure keratitis

Literature Evidence — Epidemic Keratoconjunctivitis (Rank 9, L4)

PMID	Year	Type	Journal	Key Findings
21908289	2011	Veterinary in vitro susceptibility	J Vet Diagn Invest	MIC data for Moraxella bovoculi in infectious bovine keratoconjunctivitis — bovine pathogen, not human adenoviral EKC
25261461	2014	Veterinary retrospective study	J Vet Diagn Invest	Epidemiology of Moraxella spp. in bovine pinkeye outbreaks — not applicable to human EKC

Denmark Market Information

Clindamycin is not currently registered or marketed in Denmark. No marketing authorisations have been identified through the Danish Medicines Agency (Lægemiddelstyrelsen) or via centralised EMA procedures at the time of this report (data cut-off: 5 April 2026).

Practitioners requiring Clindamycin-based products should consult the Lægemiddelstyrelsen for any available named-patient or compassionate-use pathways, or verify current EMA centralised authorisation status directly.

Safety Considerations

Please refer to the approved Summary of Product Characteristics (SmPC) for safety information.

⚠️ Important CDI Risk Warning (Reverse-Association from Evidence Pack): Clindamycin is one of the antibiotics most strongly associated with Clostridioides difficile infection (CDI). Two publications retrieved during evidence collection (PMID 36684930, PMID 40172204) document C. difficile epidemiology linked to Clindamycin exposure in companion animals and feral horse populations. This CDI risk is a key safety consideration for any proposed clinical use in humans and must be addressed in any formal safety evaluation.

Conclusion and Next Steps

Decision: Hold

Rationale: All five unique TxGNN-predicted indications are rated L5 or L4 (model prediction only, or indirect/veterinary evidence), zero human clinical trials have been registered, and none of the predicted conditions have a biologically plausible mechanistic link to Clindamycin’s known antibacterial or antiprotozoal properties. Additionally, Clindamycin is not currently marketed in Denmark, and baseline safety data (SmPC warnings and contraindications) is absent from the evidence pack.

To proceed, the following would be needed:

Full MOA data from DrugBank confirming any mechanistic basis for use in ocular surface disease
Danish Medicines Agency (Lægemiddelstyrelsen) SmPC or TFDA product information for formal safety assessment (blocking data gap DG001)
At least one human clinical trial or prospective observational study investigating Clindamycin in a relevant ocular indication
A formal CDI risk assessment and risk mitigation strategy given Clindamycin’s well-established C. difficile-inducing potential
Reconsideration of the TxGNN knowledge graph model weighting for ocular surface disease nodes, to reduce non-specific high-scoring predictions
If ocular application is explored, route compatibility assessment (topical ophthalmic formulation availability and pharmacokinetics) would be required
Disclaimer

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