Ravulizumab

證據等級: L5

預測適應症: 10 個

Ravulizumab: From Complement-Mediated Disease to Congenital Neutropenia (G6PC3 Deficiency)

One-Sentence Summary

Ravulizumab is a long-acting terminal complement (C5) inhibitor; its original approved indication is not available in the current dataset. The TxGNN model predicts it may be effective for Autosomal Recessive Severe Congenital Neutropenia due to G6PC3 Deficiency, but this prediction is currently supported by 0 clinical trials and 0 publications — evidence rests entirely on the model's embedding similarity, and the evidence pack itself flags the mechanistic rationale as likely spurious.

Quick Overview

Item	Content
Original Indication	Not available in current dataset (no approved indication text in Norway regulatory records)
Predicted New Indication	Autosomal Recessive Severe Congenital Neutropenia due to G6PC3 Deficiency
TxGNN Prediction Score	99.96%
Evidence Level	L5
Norway Market Status	✗ Not Marketed
Number of Authorizations	0
Recommended Decision	Hold

Why is This Prediction Reasonable?

Currently, detailed mechanism of action data is not available in the structured MOA field. However, the evidence pack's own repurposing rationale describes ravulizumab as a long-acting C5 complement inhibitor (a derivative of eculizumab), meaning its established pharmacology blocks the terminal complement cascade (C5 cleavage → C5a/C5b-9 formation).

The predicted indication — congenital neutropenia due to G6PC3 deficiency — is driven by endoplasmic reticulum stress and apoptosis in neutrophil precursors, a mechanism rooted in glycogen/glucose metabolism rather than complement activation. According to the evidence pack's own mechanistic assessment, there is no known direct biological link between the C5 complement pathway and G6PC3-deficient neutropenia. The high TxGNN score most likely reflects embedding-space similarity across a cluster of neutrophil-related diseases in the knowledge graph, rather than a genuine, validated mechanistic connection.

This caveat is consistent across the other top-ranked predictions in this evidence pack (cyclic hematopoiesis, CXCR2-deficient neutropenia, X-linked SCN) — all are neutrophil/hematopoietic disorders with distinct, non-complement-driven etiologies, reinforcing that the model may be clustering on phenotypic similarity (neutropenia) rather than shared drug-target biology.

Clinical Trial Evidence

Currently no related clinical trials registered

Literature Evidence

Currently no related literature available

Norway Market Information

Ravulizumab is not currently marketed in Norway, and no authorization records are available in the dataset.

Safety Considerations

Please refer to the package insert for safety information.

Note: TFDA label warnings/contraindications (DG001) are flagged as a Blocking data gap in the evidence pack — this drug cannot proceed to a safety pre-assessment (S1) until this data is obtained.

Conclusion and Next Steps

Decision: Hold

Rationale: This prediction is supported only by an L5 model score (rank 731/~, no clinical trials, no literature), and the evidence pack's own mechanistic analysis explicitly questions the biological plausibility of the C5-pathway/G6PC3-neutropenia link, attributing the high score to disease-embedding similarity rather than genuine target relevance. Combined with a Blocking data gap on TFDA safety labeling, this candidate does not meet the threshold to advance past S0.

To proceed, the following is needed:

Structured MOA data from DrugBank/label sources (currently a High-severity data gap, DG001/DG002)
TFDA/regulatory label (warnings, contraindications) to enable S1 safety pre-assessment (Blocking gap)
Independent literature or preclinical evidence directly linking complement C5 activity to G6PC3-deficient neutrophil pathology
Confirmation of the drug's actual original approved indication(s), currently missing from regulatory records
Disclaimer

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