Capecitabine and fluorouracil: the one gene test cancer services do not skip
This is the page on the site where the stakes are highest and the reasoning simplest. One enzyme disposes of most of the drug, some people barely make that enzyme, and for them a routine chemotherapy dose is a poisoning.
This page also covers fluorouracil: CPIC publishes one recommendation for both, so the guidance below applies to either drug.
Also sold as: Xeloda, 5-FU, Efudix.
What capecitabine and fluorouracil are
Capecitabine and fluorouracil are chemotherapy drugs from the same family, the fluoropyrimidines. Fluorouracil, usually written 5-FU, is given by infusion; capecitabine is a tablet that the body turns into 5-FU, so the two amount to the same active drug by different routes. They are backbone treatments for bowel, breast and several other solid cancers, often in combination regimens given over months.
Why your genes come into it
The body breaks down about four fifths of a 5-FU dose using one enzyme, DPD, which the DPYD gene encodes. When that enzyme works normally the drug does its job and clears on schedule. When DPYD carries a no-function variant the enzyme runs slow or not at all, the same dose lingers and builds up, and the toxicity that follows can be severe: collapsing blood counts, mouth and gut ulceration, and in complete deficiency it has been fatal. This is why fluoropyrimidine toxicity was one of the first places pharmacogenomics moved from interesting to mandatory, and why European and other regulators now expect DPD status to be checked before treatment rather than after a patient is already in trouble.
What published guidance says
The rows below restate CPIC's published guidance for each possible DPYD result, and they describe what an oncology team weighs before and during treatment, not a dose anyone should act on alone. If chemotherapy is being planned, the test that counts is the clinical DPYD genotyping the cancer service orders itself, sometimes alongside a direct measurement of DPD activity. Nothing on this page should delay, change or substitute for that.
DPYD
| If your DPYD result is | What that means | Published CPIC guidance |
|---|---|---|
| Intermediate Metabolizer | Decreased DPD activity (leukocyte DPD activity at 30% to 70% that of the normal population) and increased risk for severe or even fatal drug toxicity when treated with fluoropyrimidine drugs | Per CPIC (A): "Reduce starting dose by 50% followed by titration of dose based on toxicity or therapeutic drug monitoring (if available)." |
| Normal Metabolizer | Normal DPD activity and "normal" risk for fluoropyrimidine toxicity | Per CPIC (A): "Based on genotype, there is no indication to change dose or therapy. Use label-recommended dosage and administration." |
| Poor Metabolizer | Complete DPD deficiency and increased risk for severe or even fatal drug toxicity when treated with fluoropyrimidine drugs. | Per CPIC (A): "Avoid use of 5-fluorouracil or 5-fluorouracil prodrug-based regimens. In the event, based on clinical advice, alternative agents are not considered a suitable therapeutic option, 5-fluorouracil should be administered at a strongly reduced dose with early therapeutic drug monitoring." |
Can your DNA file answer this?
DPYD is one of the pharmacogenes a consumer file can partly answer, and the limit here matters more than on any other page. The two variants that carry most of the risk in European populations, rs3918290 and rs67376798, are single positions that genotyping arrays commonly include, so Aimosti can read them from a 23andMe or AncestryDNA export. What a chip cannot do is rule DPYD deficiency out. It tests the variants it was built for and nothing else, so a file carrying a rarer reduced-function variant, or the intronic HapB3 variant, will read as normal function. A normal DPYD result from consumer data means the tested variants were absent, not that DPD activity is proven safe. Before a fluoropyrimidine, that difference is the whole point, and it is the reason the clinical test exists.
A genotyping chip reads only a few hundred thousand pre-selected positions, about 0.02% of your genome, chosen for common variation. It does not sequence the rest, so it cannot find the rare or novel pathogenic variants the clinical and carrier modules look for: a “no finding” from chip data means “this chip never looked”, far more so than with whole-genome sequencing. Chip data suits common-variant traits, pharmacogenomic tag SNPs, and haplogroup ancestry, not clinical or carrier screening.
Common questions
My oncologist is starting capecitabine. Should I get a DNA test first?
The DNA test that matters is the one the cancer service orders, and in many countries DPD status is now checked as a matter of routine before a fluoropyrimidine. Ask your team directly whether it has been done. A re-analysis of a file you already own is not a substitute for that clinical test and should never be a reason to hold up treatment.
My consumer file says DPYD is normal. Does that mean I am safe on 5-FU?
Not on its own. A chip tests a fixed list of variants, so a normal result means those specific variants were absent, not that your DPD enzyme is proven to work. Rarer reduced-function variants exist that arrays do not carry, which is exactly why the clinical pathway pairs genotyping with, in some centres, a direct activity measurement. Treat a consumer normal result as reassuring background, not clearance.
Why are capecitabine and fluorouracil on the same page?
Because capecitabine is converted into fluorouracil in the body, so DPYD acts on the same active drug either way, and CPIC's guidance for the two is identical. The site checks that automatically: if the recommendations ever diverged, this page would fail its own test and the two drugs would be split apart.
What happens if someone with DPYD deficiency gets a full dose?
The drug is not cleared at the expected rate, so it accumulates and the usual side effects arrive faster and much harder. Published guidance points toward a reduced starting dose for intermediate metabolizers and toward avoiding fluoropyrimidines altogether at the deficient end, with alternative regimens where they exist. Which path fits a given cancer is an oncology decision, not a table lookup.