Denosumab

證據等級: L5 預測適應症: 2

目錄

  1. Denosumab
  2. Denosumab: Original Indication Unconfirmed → Predicted Application in Severe Nonproliferative Diabetic Retinopathy
    1. One-Sentence Summary
    2. Quick Overview
    3. Why is This Prediction Reasonable?
    4. Clinical Trial Evidence
    5. Literature Evidence
    6. Market Information
    7. Safety Considerations
    8. Conclusion and Next Steps
    9. Disclaimer

## 藥師評估報告

Denosumab: Original Indication Unconfirmed → Predicted Application in Severe Nonproliferative Diabetic Retinopathy

One-Sentence Summary

Denosumab's original approved indication cannot be confirmed from the current evidence pack (no license or indication text on file), though it is broadly known as a RANKL-targeting monoclonal antibody used in bone-related conditions. The TxGNN model predicts a possible role in Severe Nonproliferative Diabetic Retinopathy, but this prediction is currently supported by 0 clinical trials and 0 publications — it is a pure algorithmic extrapolation with no direct evidence.


Quick Overview

Item Content
Original Indication Not available — original_indications and taiwan_regulatory.licenses are both empty in this evidence pack
Predicted New Indication Severe Nonproliferative Diabetic Retinopathy
TxGNN Prediction Score 99.63% (rank 4724)
Evidence Level L5 (model prediction only)
Market Status Not Marketed (未上市)
Number of Authorizations 0
Recommended Decision Hold

Why is This Prediction Reasonable?

Currently, detailed mechanism of action data is not available (DG002, DrugBank query pending). The predicted-indication rationale supplied with this pack states the working hypothesis directly: the prediction extends a RANKL/OPG axis hypothesis — already flagged as relevant to diabetic retinopathy in general — to a specific disease subtype (severe nonproliferative disease). However, the rationale explicitly notes this is "a pure knowledge-graph node-similarity extrapolation by the TxGNN algorithm, lacking any subtype-specific biological argumentation," and is unsupported by any direct or indirect clinical data.

Some indirect support exists for the broader diabetic retinopathy category (TxGNN rank 2, score 99.23%, evidence level L4), which is the presumed biological basis this severe-subtype prediction is extrapolated from:

  • OPG–RANKL correlation: Osteoprotegerin (OPG), the decoy receptor for RANKL, has been reported elevated in patients with diabetic retinopathy and correlated with disease severity — suggesting the RANKL/OPG axis may participate in retinal vascular pathology (inflammation, vascular remodeling). This is a correlative biomarker association, not an established causal treatment mechanism, and the direction of effect (denosumab inhibits RANKL, whereas the association is with elevated OPG) is not fully consistent with a straightforward causal story.
  • A 2024 real-world cohort study (PMID 38899553, Henney et al., Diabetes Obes Metab) found denosumab use was associated with reduced incidence of type 2 diabetes and lower rates of microvascular complications (including retinopathy) compared with bisphosphonates — an observational signal, not an interventional efficacy result for retinopathy itself.

No evidence in this pack specifically addresses the severe nonproliferative subtype rank 1 is predicting for; all available signal pertains to diabetic retinopathy as a broad category.


Clinical Trial Evidence

Currently no related clinical trials registered for severe nonproliferative diabetic retinopathy.

Context: one Phase 3 trial exists for the broader diabetic retinopathy/denosumab space — NCT00925600 (completed, n=769) — but it evaluated lens opacification (cataract) safety in prostate cancer patients on denosumab for bone loss, not retinopathy efficacy. It was graded C/low relevance by evidence review and provides ocular safety context only, not efficacy evidence.


Literature Evidence

Currently no related literature available for severe nonproliferative diabetic retinopathy specifically.

Context: two publications support the broader diabetic retinopathy category (cited in the mechanism discussion above) — a 2024 real-world cohort/meta-analysis (PMID 38899553) and a 2023 fracture-risk cohort (PMID 36960265) with lower relevance. Neither addresses the severe subtype or provides interventional efficacy data.


Market Information

No marketing authorizations currently on file (product status: Not Marketed, 0 authorizations).


Safety Considerations

Please refer to the package insert for safety information.

Note: local (TFDA-equivalent) label warnings and contraindications could not be retrieved for this evidence pack (DG001, Blocking severity) — this alone prevents any S1 safety pre-assessment for this candidate.


Conclusion and Next Steps

Decision: Hold

Rationale: The predicted indication (severe nonproliferative diabetic retinopathy) has zero direct clinical trial or literature support and an L5 evidence level — this is an unvalidated knowledge-graph extrapolation from a broader, itself only weakly-supported (L4, correlative) disease association. Combined with a blocking data gap on local label warnings/contraindications, the candidate cannot yet enter safety pre-assessment.

To proceed, the following is needed:

  • TFDA/local package insert warnings and contraindications (DG001, blocking — required before any S1 safety review)
  • Denosumab mechanism-of-action data from DrugBank (DG002)
  • Original indication and licensing data (currently entirely absent from this pack)
  • Preclinical or translational evidence linking the RANKL/OPG axis causally (not just correlatively) to diabetic retinopathy, and specifically to the severe nonproliferative subtype
  • If pursuing the broader diabetic retinopathy signal (rank 2) instead, prospective mechanistic or interventional studies to move beyond the current cohort-level, correlative evidence base

    Disclaimer

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



Back to top

Copyright © 2026 藥提醒科技有限公司 (yao.care). This report is for research purposes only and does not constitute medical advice.

This site uses Just the Docs, a documentation theme for Jekyll.