DPYD Testing Before 5-FU and Capecitabine: How Genotype-Guided Dosing Prevents Severe Toxicity

Fluorouracil (5-FU) and its oral prodrug capecitabine can cause life-threatening toxicity in patients who inherit variants in the DPYD gene that disable the enzyme responsible for clearing these drugs. A blood or saliva test read before the first cycle can flag many of the highest-risk patients so that the starting dose is reduced or the drug avoided. Prospective data show this approach lowers the rate of severe toxicity. What it does not do is certify anyone as safe, because a normal genotype leaves a substantial share of toxicity unexplained.

The enzyme between the drug and the patient

About 80 percent of an administered fluoropyrimidine dose is broken down by dihydropyrimidine dehydrogenase, the DPD enzyme encoded by DPYD. When DPD activity is low or absent, active drug accumulates and produces the toxicities that define these regimens at their worst: severe mucositis, profuse diarrhea, profound neutropenia, hand-foot syndrome, and neurotoxicity. In patients with complete DPD deficiency, standard doses can be fatal. Because these drugs anchor treatment for colorectal, breast, gastric, and other common cancers, the question of who metabolizes them poorly is not academic.

Inherited DPYD variants are the most common identifiable cause of reduced DPD activity. Carriers of a single no-function allele who receive a full dose face a markedly elevated risk of grade 3 or higher toxicity, and the risk climbs further for those carrying two damaging alleles.

What the variants predict and how dosing responds

The Clinical Pharmacogenetics Implementation Consortium (CPIC) translates DPYD genotype into an activity score, summing the functional value of a person's two lowest-scoring alleles. Two normal alleles give a score of 2.0, the normal metabolizer phenotype, and no dose change is indicated. Scores of 1.0 to 1.5 define intermediate metabolizers, for whom CPIC advises reducing the starting dose by roughly half and then titrating upward based on tolerance and response. Scores of 0 to 0.5 mark poor metabolizers, in whom CPIC recommends avoiding fluoropyrimidines and selecting an alternative agent.

The alleles driving these scores fall into two buckets. No-function variants such as c.1905+1G>A (the classic DPYD2A) and c.1679T>G (DPYD13) produce a truncated protein with essentially no activity. Decreased-function variants such as c.2846A>T and the intronic c.1129-5923C>G (part of the HapB3 haplotype) leave partial activity. This framework is described in the CPIC guideline hosted through PharmGKB, and its dosing logic is what the tests are meant to operationalize.

The strongest clinical support comes from a Dutch prospective study published in The Lancet Oncology in 2018 by Henricks and colleagues. Patients were genotyped for four DPYD variants before treatment, and carriers received reduced doses. Genotype-guided dosing brought the toxicity risk in variant carriers much closer to that of non-carriers, and the authors judged the strategy feasible in routine care. That trial, rather than any single laboratory claim, is the reason pretreatment testing moved from optional to expected.

The 2024 consensus and why a shared panel matters

In 2024 a broad coalition of professional bodies, including the Association for Molecular Pathology, the American College of Medical Genetics and Genomics, CPIC, the College of American Pathologists, and the Dutch and European pharmacogenetics groups, published a joint consensus in the Journal of Molecular Diagnostics recommending a minimum set of variants that every DPYD test should cover. Those four core alleles are c.1905+1G>A (DPYD2A), c.1679T>G (DPYD13), c.2846A>T, and c.1129-5923C>G (HapB3). Standardizing the panel matters because a result is only interpretable if you know which variants were interrogated, and laboratories had been testing inconsistent sets.

What genotyping cannot catch

Here the caveats deserve as much weight as the recommendations. The four-variant panel captures only a portion of the toxicity signal. Reported estimates suggest routine testing of the common actionable variants predicts a minority of severe toxicity cases, and studies have found that a large fraction of patients who develop serious reactions carry none of the tested variants. DPD deficiency can also arise from rarer DPYD variants outside the panel, from variants that are common in some ancestral populations but poorly characterized, and from non-genetic causes that genotyping cannot see at all.

Ancestry compounds this. The four core variants were characterized largely in European-ancestry cohorts, and they occur at low or negligible frequencies in African and East Asian populations, which lowers the test's sensitivity for those patients rather than reassuring them. Enzyme phenotyping approaches, such as measuring uracil in plasma, are used in some systems as a complement, though they carry their own preanalytical pitfalls. The practical consequence is that clinical vigilance for early toxicity remains essential regardless of genotype. This article is educational and is not medical advice; treatment decisions belong to a patient and their oncology team.

What the FDA label now says, and what it does not

Regulators have moved the testing conversation onto the drug labels. The US Food and Drug Administration approved a safety labeling update for fluorouracil in 2024 and for capecitabine in 2025, adding boxed-warning language about the risk of serious or fatal reactions in patients with DPD deficiency. The updated labels advise testing for DPYD variants before starting treatment unless immediate therapy is required, direct that patients with complete deficiency not receive these drugs, and call for a reduced starting dose in partial deficiency. Read plainly, this is a description of what the label now legally states, not a safety guarantee and not an endorsement of any particular test or laboratory.

Two limits are worth stating precisely. The label frames testing as a step to take before treatment, and it does not claim that a normal genotype rules out DPD deficiency. A patient can carry no tested variant and still metabolize the drug poorly. Pretreatment DPYD testing is best understood as a filter that removes some of the highest-risk exposures up front, appraised on the evidence above, while leaving the residual risk that only careful clinical monitoring can address.