Finasteride

證據等級: L5

預測適應症: 6 個

drug.original_indications is empty and original_moa is [Data Gap] (see DG002). The evidence pack's own repurposing_rationale.mechanistic_link fields for ranks 2, 5, and 6 independently state that Finasteride works via 5α-reductase/DHT inhibition and is used clinically for hair loss — so I've grounded the "original indication/MOA" context in those in-pack statements rather than outside knowledge, while flagging that no formally sourced TFDA/DrugBank record exists in this pack.
Per the template, all trial/literature tables use predicted_indications[0] (the top-ranked candidate, Ambras type hypertrichosis), which has zero trials and zero literature — this drives most of the report's caution.
taiwan_regulatory.licenses is empty, so the NZ market table is omitted in favor of a one-line statement.
Finasteride is not an antineoplastic agent, so the Cytotoxicity section is omitted entirely.
All safety fields are [Data Gap], so the fallback safety sentence is used.

Finasteride: From Androgenetic Alopecia to Ambras Type Hypertrichosis Universalis Congenita

One-Sentence Summary

Finasteride is a 5α-reductase inhibitor whose established clinical use — referenced within this evidence pack's own mechanistic notes — is androgenetic alopecia (male pattern hair loss); no independently sourced TFDA/DrugBank record of its original indication is present in this pack. The TxGNN model's top-ranked prediction is Ambras type hypertrichosis universalis congenita, a rare congenital chromosomal-rearrangement disorder, with a prediction score of 99.99%. This prediction is currently supported by zero clinical trials and zero publications, and the evidence pack's own rationale flags it as a likely embedding-space artifact rather than a genuine mechanistic signal.

Quick Overview

Item	Content
Original Indication	Androgenetic alopecia (male pattern hair loss) — inferred from in-pack mechanistic notes; not formally sourced (Data Gap DG001/DG002)
Predicted New Indication	Ambras type hypertrichosis universalis congenita
TxGNN Prediction Score	99.99%
Evidence Level	L5 (model prediction only, no supporting studies)
New Zealand Market Status	✗ Not Marketed
Number of Authorizations	0
Recommended Decision	Hold

Why is This Prediction Reasonable?

Currently, detailed mechanism-of-action data from an authoritative source (DrugBank/TFDA) is not available in this evidence pack (Data Gap DG002). Based on the mechanistic descriptions embedded in this pack's own rationale notes, Finasteride inhibits 5α-reductase, reducing conversion of testosterone to dihydrotestosterone (DHT), and is clinically established for androgen-dependent hair loss (androgenetic alopecia).

This top-ranked prediction does not fit that mechanism well. Ambras type hypertrichosis universalis congenita is a rare congenital disorder linked to chromosomal rearrangement, producing generalized excess hair growth through a pathway that is not androgen/DHT-driven. Finasteride has no known molecular target in this disease process. The evidence pack's own rationale is explicit on this point: the high TxGNN score is attributed to semantic proximity between "hair/follicle" concepts in the model's embedding space rather than to any real mechanistic or pharmacological linkage — i.e., a likely false-positive association.

It is also worth noting that Finasteride's true pharmacological direction runs counter to hypertrichosis-type indications generally: it is used to reduce androgen-driven hair growth (or promote scalp hair retention), not to induce hair growth in non-androgen-dependent contexts. The next-ranked candidate in this pack, "hypertrichosis (disease)" (rank 2, score 99.99%), shares this same directional mismatch, and even the drug's lower-ranked candidates (ranks 4–6, all L5) show no trial or literature support. Across all six candidates in this pack, TxGNN consistently returned a "Hold" recommendation, reinforcing that no candidate here currently rises above hypothesis-generation level.

Clinical Trial Evidence

Currently no related clinical trials registered.

Literature Evidence

Currently no related literature available.

New Zealand Market Information

Finasteride is not currently marketed in New Zealand under this evidence pack's regulatory query (0 authorizations on file). No product license records are available to summarize.

Safety Considerations

Please refer to the package insert for safety information.

(Note: retrieval of TFDA-sourced warnings and contraindications is flagged as a Blocking data gap (DG001) in this evidence pack — it explicitly prevents this candidate from progressing past the S1 safety pre-screen stage until resolved.)

Conclusion and Next Steps

Decision: Hold

Rationale: The top-ranked predicted indication (Ambras type hypertrichosis universalis congenita) has no supporting clinical trials or literature, an L5 evidence level, and the evidence pack's own mechanistic analysis identifies it as a probable model artifact rather than a biologically plausible repurposing signal. All six candidates in this pack independently resolved to "Hold," and a Blocking-severity safety data gap (missing TFDA warnings/contraindications) further prevents any candidate from advancing to safety pre-screening.

To proceed, the following is needed:

TFDA package insert (warnings, contraindications) — required to clear the Blocking gap (DG001) before any S1 safety evaluation can begin
Authoritative mechanism-of-action and original-indication documentation from DrugBank (DG002), to replace the currently inferred/unsourced MOA context
If pursuing rank 2 ("hypertrichosis (disease)") instead of rank 1, a manual review of the single low-relevance Phase 4 trial (NCT04293822, graded "C" relevance — does not actually test Finasteride) and the four Tier-3 review-level publications, none of which involve Finasteride directly
Expert pharmacological review to confirm or refute the embedding-artifact hypothesis before allocating further evaluation resources to this drug-indication pair
Disclaimer

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