Dulaglutide
| 證據等級: L5 | 預測適應症: 10 個 |
目錄
Dulaglutide: From Type 2 Diabetes Mellitus to Opsismodysplasia
One-Sentence Summary
Dulaglutide is a GLP-1 receptor agonist; its established therapeutic class and mechanism (referenced throughout this evidence pack’s own rationale text) point to type 2 diabetes mellitus as the original indication, though Taiwan-specific approval data is unavailable since the product is not currently marketed in Taiwan. The TxGNN model’s top-ranked new-indication prediction is Opsismodysplasia, a rare skeletal dysplasia, but this candidate has 0 clinical trials, 0 publications, and — critically — the evidence pack’s own mechanistic analysis flags it as a likely false positive with no known biological connection to the GLP-1 pathway.
Quick Overview
| Item | Content |
|---|---|
| Original Indication | Type 2 Diabetes Mellitus (inferred from drug-class context in the evidence; not confirmed by Taiwan regulatory data — no TW license exists) |
| Predicted New Indication | Opsismodysplasia |
| TxGNN Prediction Score | 97.05% |
| Evidence Level | L5 (model prediction only, no supporting studies) |
| Taiwan 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 (flagged as a High-severity data gap). Based on general drug-class knowledge, Dulaglutide is a GLP-1 receptor agonist, and its efficacy in type 2 diabetes has been established through insulin secretion and metabolic regulation pathways.
However, for this specific top-ranked candidate, the evidence pack’s own repurposing rationale explicitly does not support a plausible mechanistic link: Opsismodysplasia is a rare skeletal dysplasia caused by RSPRY1 mutations, with no known relationship to GLP-1 receptor signaling. The rationale text itself attributes the high TxGNN score to a likely knowledge-graph artifact — a common false-positive pattern where sparsely-connected rare-disease nodes receive inflated similarity scores due to structural proximity rather than genuine biological signal.
This pattern is not isolated to rank 1: across all 10 predicted indications in this evidence pack (stiff person syndrome, pancreatic agenesis, several lipodystrophy subtypes, autoimmune oophoritis, etc.), the mechanistic rationale consistently notes weak-to-absent biological plausibility, and every candidate carries a Hold recommendation at evidence level L5. This candidate set should be read as a low-confidence output requiring substantial independent validation before any further action.
Clinical Trial Evidence
Currently no related clinical trials registered.
Literature Evidence
Currently no related literature available.
Taiwan Market Information
Dulaglutide has 0 registered licenses in Taiwan and is not currently marketed (未上市). No product/authorization records are available to summarize.
Safety Considerations
Please refer to the package insert for safety information.
(Note: TFDA warning/contraindication data for this drug is currently a Blocking-severity data gap — see Conclusion.)
Conclusion and Next Steps
Decision: Hold
Rationale: This candidate lacks any clinical trial or literature support (L5, model-prediction-only), and the evidence pack’s own mechanistic analysis identifies the top prediction as a probable false positive with no plausible biological link to the drug’s known pathway. Separately, a Blocking-severity data gap (missing TFDA label warnings/contraindications) prevents this candidate from entering the S1 safety evaluation stage regardless of efficacy signal.
To proceed, the following is needed:
- TFDA label data (warnings, contraindications) — Blocking gap, required before any S1 safety review
- Confirmed mechanism of action (MOA) data via DrugBank — High-severity gap
- Independent mechanistic/preclinical validation, since the model’s own rationale disputes plausibility for this candidate
- Consideration of whether lower-ranked or differently-sourced candidates (e.g., those with a coherent metabolic/endocrine link, such as thiamine-responsive dysfunction syndrome, which at least shares a diabetes-adjacent phenotype) merit closer review instead of the top-scored but mechanistically unsupported candidate
Disclaimer
This content is for research purposes only and does not constitute medical advice. Clinical validation is required before any clinical application.