5-Fluorouracil Poisoning

1. Disease Information

• 5-FU poisoning is toxicity from fluorouracil overdose, or from severe early-onset toxicity at standard doses. Capecitabine, a 5-FU prodrug, behaves the same way.
• Overdose: dosing or infusion-pump errors.
• Standard-dose toxicity: reduced detoxification, usually DPD deficiency.
• The only approved antidote is uridine triacetate. In pooled studies it was started within 96 h of overdose or early-onset severe toxicity. 30-day survival was 94% overall (96% in overdose, 81% in early-onset severe toxicity), against 16% in a historical best-supportive-care cohort. Ison et al., Clin Cancer Res 2016 (FDA approval summary), via JHU Pure, Drugs & Therapy Perspectives guide. Treat the 16% comparator as a historical control, not a randomized one.
• Synonyms: fluorouracil overdose or toxicity, capecitabine toxicity, fluoropyrimidine toxicity.
• Data level: disease-level, aggregated from case reports and trials. This is not EHR-derived.

2. Etiology

• Causal factors (environmental/iatrogenic): overdose, or standard dosing in a patient who cannot clear the drug.
• Genetic risk factors:
• DPYD variants reduce DPD activity. Poor metabolizers have complete DPD deficiency and risk severe or fatal toxicity. CPIC advises avoiding fluoropyrimidines in them and using reduced starting doses in partial deficiency. CPIC 2017 update, PMID:29152729 (URL as returned by search: Wiley); earlier guideline PMID:23988873.
• The Dutch Pharmacogenetics Working Group has a parallel guideline: PMC7080718.
• TYMS variation is also implicated: PMC7713499, and uridine triacetate in a TYMS-variant patient, PubMed 30013790.
• Specific alleles such as DPYD2A, c.1679T>G, c.2846A>T and HapB3 are [background, unverified]*. Quote them from the CPIC table.
• Other risk factors [background, unverified]: renal impairment, drug interactions (for example with capecitabine or brivudine/sorivudine), and dosing errors.
• Protective factors: none well established. Normal DPD activity is the default state.
• Gene–environment: DPYD genotype determines how a given dose or exposure behaves. This is a pharmacogenetic interaction, not a classic GxE one.

3. Phenotypes (suggested HP terms; verify each before binding)

Frequencies are not sourced, so I give none. - Diarrhea (HP:0002014) and mucositis/stomatitis. - Myelosuppression: neutropenia, pancytopenia. - Hyperammonemic encephalopathy. It presents as sudden altered mental status, lactic acidosis and respiratory alkalosis with high ammonia. Onset is 0.5–5 days after starting 5-FU. French national survey, case report PMC7013286. - Cardiotoxicity. It often presents as myocardial ischemia, with less common arrhythmia, hypo- or hypertension, LV dysfunction, cardiac arrest and sudden death. Systematic review PMC4170068. - Hand-foot syndrome, nausea/vomiting and alopecia [background, unverified]. - Onset is acute: early-onset severe toxicity, and delayed-onset cases are also reported (PMC9490183). Quality-of-life data were not searched.

4. Genetic/Molecular

• DPYD (the lowercase hgnc: CURIE must be looked up; I did not verify one). Loss of function of the catabolic enzyme. Classification and allele frequencies: CPIC and gnomAD, not retrieved.
• TYMS is a modifier, as the TYMS-variant case above shows.
• Germline origin. Epigenetic and chromosomal data: none sought.

5. Environmental

The exposure is iatrogenic. In the KB, environmental would be the drug exposure, and an ECTO term needs a re-run search (for example l~fluorouracil) before binding. There are no lifestyle or infectious factors.

6. Mechanism / Pathophysiology

Causal chain (inferred steps are marked): 1. Excess 5-FU exposure follows overdose, or standard dosing with deficient DPD clearance. DPD normally catabolizes most of a dose. 2. High exposure produces more active metabolites and thymidylate synthase inhibition ([background, unverified]; classic mechanism). 3. Metabolites are incorporated into RNA and DNA ([background, unverified]). This injures rapidly dividing cells, causing mucositis, diarrhea and myelosuppression. 4. Uridine triacetate works because uridine competes with the toxic metabolites. Its efficacy in this setting supports an RNA-directed toxicity component. This is inferred from the antidote's clinical effect, not directly demonstrated here. 5. Branch, neurologic: catabolites (fluoro-β-alanine, monofluoroacetate) depress TCA-cycle flux and ATP. This secondarily impairs the urea cycle, leading to hyperammonemia and then encephalopathy (PMC7013286). The fluoroacetate/fluorocitrate step is postulated, not proven. 6. Branch, cardiac: proposed mechanisms are endothelial injury with thrombosis, energy depletion and ischemia, oxidative stress, coronary spasm, and reduced red-cell oxygen transfer. There is no evidence for a single mechanism. One source: rat cardiocyte oxidative stress and apoptosis, PMC3461434 (in vivo/in vitro rat data). - Suggested GO/CL terms (verify): apoptotic process, response to oxidative stress, the citrate cycle and urea cycle. Cell types: intestinal epithelial cell, hematopoietic progenitor, cardiac muscle cell, hepatocyte.

7. Anatomical Structures

• Primary: GI tract mucosa, bone marrow.
• Secondary: heart and brain; liver (the site of DPD detoxification).
• Suggested UBERON terms: look up small intestine, bone marrow, heart, brain, liver. The condition is bilateral and systemic.

8. Temporal Development

Acute. The encephalopathy appears at 0.5–5 days. Early-onset severe toxicity and delayed onset are both described. Recovery with prompt antidote is substantial per the figures above. Stages and critical periods: the 96-hour antidote window is the key one.

9. Inheritance and Population

• Inheritance: DPD deficiency is autosomal recessive for complete deficiency. Partial deficiency is intermediate [background, unverified]; confirm against CPIC.
• Prevalence, incidence, sex ratio and ancestry: not retrieved. I did not find numbers, so none are stated.

10. Diagnostics

• Clinical: history of exposure plus toxicity pattern. Check ammonia and lactate if neurologic changes appear.
• Genetic: DPYD genotyping before therapy (CPIC, DPWG). Universal pre-treatment testing is promoted by some groups (NCODA). Other approaches (phenotyping, uracil levels) are [background, unverified].
• Differential: sepsis, other chemotherapy toxicities, and other causes of encephalopathy. Nothing in the sources retrieved distinguishes these.

11. Outcome/Prognosis

With uridine triacetate started within 96 h, 30-day survival was 94% versus 16% for historical controls. Complete DPD deficiency risks fatal toxicity (CPIC). Other prognostic data were not retrieved.

12. Treatment

• Antidote: uridine triacetate (Vistogard), FDA-approved for overdose and early-onset severe toxicity (Ison 2016). Delayed-onset use was reported in PMC9490183. Its use in a TYMS-variant patient is described in PubMed 30013790. The CHEBI/NCIT agent CURIE needs lookup.
• Supportive care: fluids, antiemetics, antidiarrheals, G-CSF, transfusion, ICU care [background, unverified]. Suggested NCIT action terms (verify): NCIT:C15986 Pharmacotherapy and NCIT:C15747 Supportive Care. These appear in the CLAUDE.md list and should still be looked up.
• Prevention-linked dosing: DPYD-guided dose reduction or avoidance (CPIC).
• Trials: none retrieved. NCT07158164 surfaced, a DPYD-guided dosing study. I did not read it, so check it before citing.

13. Prevention

• Primary: pre-treatment DPYD genotyping with dose adjustment, plus pump and dosing safeguards (the latter [background, unverified]).
• Secondary: early recognition of toxicity within the antidote window.
• No vaccine or public-health program applies.

14. Other Species

Not researched. Fluoropyrimidine toxicity in dogs and cats (topical 5-FU is especially toxic to them) is [background, unverified]. Check OMIA and PubMed before adding it.

15. Model Organisms

• Rat cardiocyte oxidative-stress and apoptosis data (PMC3461434). Grade these MODEL_ORGANISM or IN_VITRO depending on the quoted sentence.
• Dpyd-deficient mouse models are [background, unverified]. Search MGI before adding any.

Gaps and next steps

1. Fetch the PMIDs for the Ison 2016 paper, the French national survey and the cardiotoxicity review. I did not retrieve them, so I give no numbers.
2. Take allele frequencies and CPIC tier tables from the CPIC supplement.
3. Take epidemiology from a primary source.
4. Look up every suggested CURIE.
5. Run just preflight-dr on this report.

Sources used: CPIC 2017 (Wiley), DPWG, PMC7713499, NCODA, Ison 2016 record, uridine triacetate guide, delayed-onset case PMC9490183, TYMS case 30013790, French survey, PMC7013286, cardiotoxicity review PMC4170068, PMC3461434.

Reference Validation

Checked with linkml-reference-validator 0.3.0rc3.

References that may not be about this subject

These identifiers resolve, so they are not fabrications, but the records they resolve to share almost none of this report's vocabulary. That is a clue and not a verdict - a paper can be relevant in ways its title and abstract do not spell out - so read them before deciding:

• PMC:PMC7013286 (7 mentions) - A 5-Fluorouracil-Induced Hyperammonemic Encephalopathy Challenged with Capecitabine.
• shared terms: encephalopathy
• PMC:PMC4170068 (5 mentions) - A systematic review of the pathophysiology of 5-fluorouracil-induced cardiotoxicity.
• shared terms: none

Weighed against this report's own most characteristic terms: toxicity, unverified, cpic, dpd, uridine, triacetate, dosing, severe, antidote, dose, overdose, deficiency, dpyd, early-onset, exposure, curie, appear, check, encephalopathy, verify.

All extracted references resolved successfully. Resolving is not the same as being relevant, though - see the references listed above as possibly off topic.

Term Validation

Checked with linkml-term-validator 0.4.5, through the ols: adapter.

Every term resolved, and every label the report gave matched.