Fluorouracil

證據等級: L5

預測適應症: 10 個

Fluorouracil: From Colorectal Cancer to Botryoid-type Embryonal Rhabdomyosarcoma of the Vagina

One-Sentence Summary

Fluorouracil (5-FU) is a pyrimidine antimetabolite and foundational chemotherapy agent, used in combination regimens for colorectal, gastric, breast, and head and neck cancers worldwide. The TxGNN model predicts it may be effective for Botryoid-type Embryonal Rhabdomyosarcoma of the Vagina, an extremely rare pediatric soft tissue malignancy. This prediction is currently supported by no clinical trials and no published literature, placing it at evidence level L5 — hypothesis-generating only.

Quick Overview

Item	Content
Original Indication	Not registered in Canada (no DINs found in regulatory database)
Predicted New Indication	Botryoid-type embryonal rhabdomyosarcoma of the vagina
TxGNN Prediction Score	99.75%
Evidence Level	L5 (model prediction only, no actual studies)
Canada Market Status	Not Marketed
Number of DINs	0
Recommended Decision	Hold

Why is This Prediction Reasonable?

Detailed mechanism of action data is not available in this Evidence Pack. Based on established pharmacology, Fluorouracil is a fluoropyrimidine antimetabolite that acts through two parallel pathways: (1) it inhibits thymidylate synthase (TS), blocking de novo thymidine biosynthesis and stalling DNA replication; and (2) it incorporates into RNA as a fraudulent nucleotide, disrupting RNA processing and protein synthesis. This dual DNA/RNA interference mechanism explains its broad cytotoxic activity and its role as the backbone of FOLFOX, FOLFIRI, and FOLFIRINOX regimens in gastrointestinal oncology.

Botryoid-type embryonal rhabdomyosarcoma (ERMS) of the vagina is an ultra-rare subtype occurring almost exclusively in young girls, and belongs to the rhabdomyosarcoma (RMS) family of soft tissue sarcomas. The established standard of care for RMS is VAC chemotherapy (vincristine + actinomycin-D + cyclophosphamide) with or without local therapy. While 5-FU’s TS inhibition could theoretically suppress rapidly proliferating RMS cells, fluoropyrimidines have never appeared in IRS/COG RMS treatment guidelines, and no preclinical or mechanistic studies have investigated 5-FU in this histological subtype specifically.

The high TxGNN score (99.75%) for this indication most likely reflects indirect graph propagation within the knowledge graph — the model infers a path through “soft tissue tumor → chemotherapy-sensitive → pyrimidine metabolism” shared neighbors — rather than a direct biological link. The extreme rarity of botryoid vaginal ERMS, together with the complete absence of clinical or preclinical data, means this prediction should be treated as a hypothesis-generating signal only, not a clinical recommendation. Notably, all top-ranked predictions cluster into two biologically distinct groups: RMS subtypes (ranks 1–7) and sickle cell variants (ranks 8–10), with the sickle cell predictions showing identical TxGNN scores, strongly suggesting batch false-positives from a shared ancestor node in the knowledge graph.

Clinical Trial Evidence

Currently no related clinical trials registered.

Literature Evidence

Currently no related literature available.

Cytotoxicity

Item	Content
Cytotoxicity Classification	Conventional cytotoxic (Fluoropyrimidine class)
Myelosuppression Risk	Moderate to High — leukopenia, thrombocytopenia, and anemia are common dose-limiting toxicities; life-threatening myelosuppression may occur in patients with dihydropyrimidine dehydrogenase (DPD) deficiency
Emetogenicity Classification	Low to moderate
Monitoring Items	CBC with differential (weekly during treatment), liver function tests, renal function, electrolytes; DPYD genotyping or DPD enzyme phenotyping recommended before initiation to screen for deficiency
Handling Protection	Must follow cytotoxic drug handling regulations; closed-system drug transfer devices (CSTDs) required for preparation and administration; personnel protective equipment mandatory

Safety Considerations

Please refer to the package insert for safety information.

Conclusion and Next Steps

Decision: Hold

Rationale: Despite a high TxGNN prediction score, this is an L5-level prediction with zero clinical trial or literature support; the model score reflects knowledge graph topology rather than validated biology. Botryoid vaginal ERMS is an ultra-rare tumor with an established VAC-based treatment standard, and fluoropyrimidines have no established role in any RMS guideline.

To proceed, the following is needed:

In vitro cytotoxicity profiling: Establish 5-FU IC50 against representative RMS cell lines (RD, Rh30, SMS-CTR) to determine baseline sensitivity
Biomarker assessment: Thymidylate synthase (TS) expression and DPD enzyme activity profiling in pediatric ERMS tumor samples
Preclinical in vivo data: Xenograft model experiments in pediatric RMS mouse models before any clinical hypothesis is considered
Safety data gap: Obtain Health Canada–approved package insert for warnings, contraindications, and pediatric dosing guidance (currently a blocking data gap per DG001)
MOA documentation: Formal retrieval of DrugBank mechanism of action data (data gap DG002) to support mechanistic rationale
Regulatory pathway assessment: Pediatric oncology indications in Canada require special consideration under Health Canada’s pediatric framework; early regulatory consultation is recommended before advancing any hypothesis
Disclaimer

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