Azithromycin

證據等級: L5

預測適應症: 10 個

Azithromycin: From Bacterial Infections to Hyperamylasemia

One-Sentence Summary

Azithromycin is a broad-spectrum macrolide antibiotic widely used for respiratory tract infections, sexually transmitted diseases, and atypical pneumonias, with additional well-documented immunomodulatory properties. The TxGNN model predicts it may be relevant for Hyperamylasemia with a score of 99.81%, however no clinical trials and no supporting literature were identified for this specific indication. Critically, mechanistic analysis reveals this prediction is a likely false positive — Azithromycin is a known causative agent of drug-induced pancreatitis, which itself elevates serum amylase levels.

Quick Overview

Item	Content
Original Indication	Bacterial infections (respiratory tract, atypical pathogens, sexually transmitted infections)
Predicted New Indication	Hyperamylasemia
TxGNN Prediction Score	99.81%
Evidence Level	L5
Singapore Market Status	✗ Not Marketed
Number of Registrations	0
Recommended Decision	Hold

Why is This Prediction Reasonable?

Currently, detailed mechanism of action data is not available in this evidence pack. Based on known information, Azithromycin is a macrolide antibiotic (azalide subclass) that exerts antibacterial effect by binding to the 50S ribosomal subunit and blocking peptide translocation. Beyond its antimicrobial activity, it possesses significant immunomodulatory properties — suppressing pro-inflammatory cytokines (IL-6, TNF-α, IL-8) and modulating neutrophil recruitment — which have driven interest in non-infectious indications such as chronic airway disease and autoimmune processes.

Hyperamylasemia refers to elevated serum amylase, most commonly associated with acute pancreatitis, salivary gland pathology, or metabolic disorders. There is no established pathway by which Azithromycin would therapeutically reduce serum amylase. In contrast, Azithromycin is a documented cause of drug-induced pancreatitis — a well-recognised adverse drug reaction in which pancreatic inflammation drives the very elevation of amylase that defines this predicted indication.

The high TxGNN score (99.81%) most likely reflects knowledge graph co-occurrence noise: Azithromycin and hyperamylasemia co-appear frequently in pharmacovigilance literature precisely because the drug causes elevated amylase, not because it treats it. Graph-based prediction models cannot inherently distinguish causative from therapeutic drug–disease associations. This prediction should be treated as a model artefact rather than a genuine repurposing signal.

Clinical Trial Evidence

Currently no related clinical trials registered.

Literature Evidence

Currently no related literature available.

Singapore Market Information

Azithromycin currently has no registered products in Singapore. No HSA authorisation records were retrieved.

Authorization Number	Product Name	Dosage Form	Approved Indication
—	—	—	No registered products found

Safety Considerations

Please refer to the package insert for safety information.

Relevant to this prediction: Azithromycin is a known cause of drug-induced pancreatitis (hyperamylasemia/hyperlipasaemia) and QTc interval prolongation. These adverse effects are directly pertinent to the predicted indication reviewed in this report, further supporting the false-positive assessment.

All Predicted Indications — Portfolio Overview

This is a multi-indication evidence pack (TW-DB00207-multi). The table below summarises all 10 TxGNN predictions for Azithromycin:

Rank	Disease	TxGNN Score	Evidence Level	Decision	Mechanistic Assessment
1	Hyperamylasemia	99.81%	L5	Hold	False positive — drug causes elevated amylase (pancreatitis ADR)
2	Polyclonal Hyperviscosity Syndrome	99.81%	L5	Hold	Immunoglobulin overproduction not modulated by macrolide antibiotics
3	Congenital Analbuminemia	99.79%	L5	Hold	Genetic (ALB gene mutation); antibiotics have no role in albumin synthesis
4	Punctate Epithelial Keratoconjunctivitis	99.78%	L4	Research Question	Microsporidia-related aetiology; in vitro anti-microsporidial activity reported; 1 case report (PMID 32826651)
5	Blood Group Incompatibility	99.70%	L5	Hold	ABO/Rh alloimmune complement activation; no drug mechanism applicable
6	Premalignant Hematological System Disease	99.64%	L5	Hold	Clonal haematopoiesis (MDS/MGUS); macrolides have no direct clone-suppressing evidence
7	Monoclonal Gammopathy	99.61%	L4	Research Question	Preclinical: Azithromycin blocks autophagy flux, sensitises myeloma cells to bortezomib (PMID 23546223)
8	Haematological Disease with Acquired Peripheral Neuropathy	99.56%	L5	Hold	Paraprotein-mediated myelin damage; no intervention pathway; QTc risk adds concern
9	Septicemic Plague	99.52%	L4	Research Question	In vitro MIC data vs Y. pestis (PMID 8540736); intracellular concentration advantage (tissue:plasma 10–100×)
10	Congenital Hematological Disorder	99.40%	L4	Research Question	Sickle Cell Disease / ACS: 2 withdrawn Phase 1/2 trials (NCT02630394, NCT02960503); immunomodulatory rationale well-supported

Indications with the Strongest Research Rationale

Rank 10 — Congenital Hematological Disorder (Sickle Cell Disease) This is the most actionable prediction in the portfolio. Two formally registered clinical trials — NCT02630394 (Phase 1, Azithromycin prophylaxis for ACS in SCD) and NCT02960503 (Phase 1/2, macrolide therapy to improve FEV1 in SCD adults) — were specifically designed to test Azithromycin's anti-inflammatory effect on pulmonary complications of SCD. Both were withdrawn before enrolment, indicating research intent without execution data. The scientific hypothesis parallels established evidence from cystic fibrosis, where macrolide therapy demonstrably reduces exacerbations.

Rank 7 — Monoclonal Gammopathy / Multiple Myeloma A 2013 in vitro study (PMID 23546223) found that Azithromycin blocks autophagy flux via LC3B-II and p62 accumulation, sensitising MM cell lines (U266, IM-9, RPMI8226) to bortezomib-induced ER stress and apoptosis. This positions Azithromycin as a potential combination sensitiser — not a monotherapy — in MM. Independent replication and an in vivo model are prerequisite before clinical translation.

Rank 9 — Septicemic Plague (Alternative Therapy Niche) In vitro susceptibility data against Yersinia pestis exists (PMID 8540736, MIC values documented), and Azithromycin's pronounced intracellular tissue accumulation is mechanistically favourable for killing phagocytosed bacteria. Relevance is limited to scenarios where first-line agents (streptomycin, gentamicin, doxycycline) are unavailable or contraindicated.

Conclusion and Next Steps

Decision: Hold

Rationale: The top-ranked prediction (hyperamylasemia) is a false positive driven by pharmacovigilance co-occurrence noise in the TxGNN knowledge graph. The large majority of predictions (ranks 1–3, 5–6, 8) carry Hold recommendations with no supporting evidence and, in several cases, adverse mechanistic relationships. Azithromycin's genuine repurposing potential in this evidence pack lies in ranks 7, 9, and especially 10.

To proceed, the following is needed:

Reprioritise review to Rank 10 (Sickle Cell Disease / ACS), which has the strongest scientific rationale and existing clinical trial precedent — investigate the reason for trial withdrawal (NCT02630394, NCT02960503) and assess feasibility of a new protocol
Obtain MOA data from DrugBank API (data gap DG002) to enable full mechanistic linkage analysis across all 10 indications
Retrieve package insert safety data from TFDA/HSA (data gap DG001) — QTc prolongation risk is critical for patient safety assessment in all cardiac-adjacent and haematological indications
For Rank 7 (Monoclonal Gammopathy): Commission or identify a follow-up study replicating PMID 23546223 findings, with focus on Azithromycin + bortezomib synergy in a clinical-grade MM model
Singapore registration pathway: Azithromycin currently has 0 HSA registrations in Singapore; regulatory filing would be required before any clinical use under any new indication

⚠️ Disclaimer: This report is for research reference only and does not constitute medical advice. All drug repurposing candidates require clinical validation before any therapeutic application.

Disclaimer

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