Dihydropyrimidine dehydrogenase and fluorouracil toxicity

Fluorouracil (FU)-containing drugs are very frequently used cytostatic drugs in systemic tumor therapy. Serious and life-threatening side effects can occur in 10 - 40% of patients, and the therapy-associated lethality is 0.2 - 1.0% (Hoff et al. 2001, van Cutsem et al. 2001). A major cause of severe FU toxicity is the genetic deficiency of dihydropyrimidine dehydrogenase (DPD), the enzyme mainly responsible for FU degradation. The DPD deficiency is caused by variants in the dihydropyrimidine dehydrogenase (DPYD) gene, which are associated with an increased risk of severe, specific side effects in carriers (Meulendijks et al. 2015). In total, up to 9% of the population carries a DPYD gene variant that leads to reduced enzyme activity, and 0.1% to 0.5% show complete DPD deficiency (EMA Recommendations 2020; Amstutz et al. 2018). Therefore, pharmacogenetic testing for the most common and clinically significant DPYD gene variants is recommended prior to systemic therapy with FU-containing drugs (Lunenburg et al. 2020, DGHO Position Paper 2020).

About 30% of severe FU toxicity reactions can be explained by a genetic DPD deficiency, but there are also numerous other factors that influence the risk of severe side effects of FU-containing therapies (Amstutz et al. 2018, Froehlich et al. 2015, Schwab et al. 2008). The DGHO therefore recommends that patients with increased toxicity under FU-containing therapy not caused by the DPYD genotype should also be evaluated for other causes and, in the case of 5-FU, therapeutic drug monitoring should be performed if necessary (Wilhelm et al. 2016, DGHO Position Paper 2020).

Pharmacogenetic diagnostics of DPYD gene variants

Molecular genetics

Molecular genetic testing for DPYD variants is a diagnostic test within the meaning of § 3 No. 7 c of the German Genetic Diagnostics Act (GenDG), which requires medical education and patient consent (GEKO Guideline 2017). Therefore, the analysis can only be carried out when the declaration of consent according to GenDG signed by the patient or his legal representative is available in the laboratory.

To clarify a genetic deficiency of the FU-degrading enzyme DPD, the four most common and clinically significant variants of the DPYD gene are tested for which a clear effect on DPD enzyme function has been described and for which testing is recommended according to the position paper of the German Society for Hematology and Medical Oncology (DGHO) before therapy with an FU-containing drug (Table 1) (Henricks et al. 2017, Lunenburg et al. 2020, DGHO Position Paper 2020).

Table 1: Overview of the investigated DPYD variants (modified according to DGHO Position Paper 2020, Amstutz et al. 2018 and Meulendijks et al. 2015)


Investigated variant 1


Enzyme activity

Allele-frequency 3

Toxicity 4


Exon 14 Skipping


none (0)




c.1679T>G (p.Ile560Ser)


none (0)




c.2846A>T (p.Asp949Val)


reduced (0.5)



Haplotype B3



reduced (0.5)



1 related to DPYD transcript 1 (NM_000110.4)
2 referring to SNP (Single Nucleotide Polymorphism) database
3 for Caucasians
4 Relative risk for severe toxicity under FU-containing therapy, confidence interval in brackets according to Meulendijks et al. 2015

  • The DPYD variant c.1905+1G>A - also called DPYD*2A - leads to a splicing defect with skipping of exon 14 of the DPYD gene and consequently to a truncated protein with loss of DPD enzyme activity (van Kuilenburg et al. 1997).
  • The DPYD variant c.1679T>G (p.Ile560Ser) - also known as DPYD*13 - leads to an amino acid exchange and is associated with loss of DPD enzyme activity (van Kuilenburg et al. 2002).
  • The DPYD variant c.2846A>T (p.Asp949Val) leads to an amino acid exchange and is associated with a decreased DPD enzyme activity (van Kuilenburg et al. 2002).
  • The DPYD variant c.1236G>A is in complete linkage disequilibrium with the c.1129-5923C>G variant, with which it jointly defines the haplotype B3. Haplotype B3 is associated with alternative DPYD splicing and reduced DPD enzyme activity (Froehlich et al. 2015, Nie et al. 2017).

Molecular genetic testing for DPYD variants in MLL is performed by means of "Loop-mediated isothermal amplification (LAMP)" and subsequent melting curve analysis. The average processing time is 2 days. A distinction is made as to whether the tested variant is heterozygous (on one of the two alleles) or homozygous (on both alleles) or whether it is not present and thus a homozygous expression of the wild-type allele (most common form with normal enzyme activity) exists. The absence of all variants in DPYD gene is also called DPYD*1. However, very rare functionally relevant variants outside the investigated regions of the DPYD gene cannot be excluded by the study (Amstutz et al. 2018). However, their clinical relevance has not been conclusively clarified at present, so that currently no investigation of the complete coding sequence of DPYD is recommended (Lunenburg et al. 2020, DGHO Position Paper 2020).

Complete DPD deficiency (homozygosity or combined heterozygosity of DPYD variants with lack of enzyme function) has also been associated with the variable clinical picture of hereditary thymine uraciluria or familial pyrimidinemia, although the genotype-phenotype relationship has not been clearly established (OMIM #274270, Fernandez-Salguero et al. 1997, van Gennip et al. 1997, van Kuilenburg et al. 1999).

Prediction of DPD enzyme activity based on the DPYD genotype

The prediction of the DPD phenotype on the basis of the DPYD genotype is performed according to the DGHO guidelines, taking into account guidelines of the Clinical Pharmacogenetics Implementation Consortium (CPIC; Amstutz et al. 2018) and the Dutch Pharmacogenetics Working Group (DPWG; Lunenburg et al. 2020). Essentially, a sum score of the two weakest variant activities is formed. A score of 2 corresponds to a normal DPD enzyme activity and a score of 0 to a complete DPD deficiency (Table 2).

Table 2: Prediction of the DPD phenotype based on the two weakest variant activities (modified according to DGHO Position Paper 2020)


Activity score

No carrier of a DPYD variant with reduced or lost function (*1/*1) – normal enzyme activity


Heterozygous carrier of a DPYD variant with reduced function

(*1/c.1236G>A or *1/c.2846A>T)


Heterozygous carrier of a DPYD variant with loss of funtion

(*1/*2A or *1/*13)


Carrier of two DPYD variants with reduced function

(e. g. c.1236G>A and c.2846A>T)

0.5 1,2

Carrier of a DPYD variant with reduced function and a variant with loss of funtion (combination of c.1236G>A or c.2846A>T with *2A or *13)

0.5 1

Homozygous carrier of a DPYD variant with loss of funtion (*2A/*2A; *13/*13)
or heterozygous carrier of two DPYD variants with loss of funtion (*2A/*13)


1 Additional phenotyping is recommended for the reliable determination of enzyme activity (Lunenburg et al. 2020)
2 Different classification with an activity score of 1 according to CPIC (Amstutz et al. 2018)

There are deviations in the recommendations of the professional associations for individual constellations. For example, carriers of two DPYD variants with reduced function are classified by DGHO with a DPD activity score of 0.5, while CPIC assigns an activity score of 1 (Amstutz et al. 2018). In this constellation, the DPWG recommends additional phenotypic testing to determine DPD activity (Lunenburg et al. 2020).

Phenotypic alternatives or additions to the genetic analysis of the DPYD gene are the measurement of uracil in plasma or the physiological ratio of dihydrouracil to uracil as well as the determination of DPD activity in leucocytes (Meulendijks et al. 2016). However, the data basis for this procedure is narrower than for DPYD genetic diagnostics and the analyses are not yet part of the standard procedure in Germany before a therapy with FU-containing drugs (DGHO Position Paper 2020). In individual cases, however, such phenotypic testing may be indicated in addition to genotyping. In particular, additional phenotypic testing is recommended in the presence of two DPYD variants with reduced function or the combination of a DPYD variant with reduced function and a variant with no function (Henricks et al. 2017, Lunenburg et al. 2020).

Recommendation for dosage according to DPYD genotype

Die Europäische Arzneimittel-Agentur (EMA) empfiehlt, alle Patienten vor einer systemischen Therapie mit den FU-haltigen Arzneimitteln 5-Fluorouracil (5-FU), Capecitabin und Tegafur auf einen DPD-Mangel zu testen (EMA Recommendations 2020). Diese Empfehlung wurde auch vom Bundesinstitut für Arzneimittel und Medizinprodukte (BfArM) aufgegriffen und in die Fachinformationen der betroffenen Arzneimittel aufgenommen. Die DGHO empfiehlt zur Umsetzung dieser Vorgaben eine Testung auf die vier häufigsten genetischen DPYD-Varianten und eine Therapie auf Basis eines differenzierten, risiko-adaptierten Algorithmus nach Ergebnis der genetischen Analyse unter Berücksichtigung der individuellen Erkrankungssituation und der möglicherweise vorhandenen Therapiealternativen (Abbildung 1) (Henricks et al. 2018, DGHO Positionspapier 2020). Die genetische Analyse kann durch ein therapeutisches Drug Monitoring bzw. eine phänotypische Testung ergänzt werden (Gamelin et al. 2008, Lunenburg et al. 2020).


Amstutz U et al. Clinical Pharmacogenetics Implementation Consortium (CPIC) Guideline for Dihydropyrimidine Dehydrogenase Genotype and Fluoropyrimidine Dosing: 2017 Update. Clin Pharmacol Ther. 2018;103:210-216.

DGHO Positionspapier Dihydropyrimidin-Dehydrogenase (DPD) -Testung vor Einsatz von 5-Fluorouracil, Capecitabin und Tegafur 2020. Link: https://www.dgho.de/publikationen/stellungnahmen/gute-aerztliche-praxis/dpd-testung/dpd-positionspapier-2020-konsens_logos_final.pdf/view

EMA recommendations on DPD testing prior to treatment with fluorouracil, capecitabine, tegafur and flucytosine. EMA/367286/2020. Link: https://www.ema.europa.eu/en/documents/referral/fluorouracil-fluorouracil-related-substances-article-31-referral-ema-recommendations-dpd-testing_en.pdf

Fernandez-Salguero PM et al. Lack of Correlation Between Phenotype and Genotype for the Polymorphically Expressed Dihydropyrimidine Dehydrogenase in a Family of Pakistani Origin. Pharmacogenetics. 1997;7:161-163.

Froehlich TK et al. Clinical importance of risk variants in the dihydropyrimidine dehydrogenase gene for the prediction of early-onset fluoropyrimidine toxicity. Int J Cancer. 2015;136,730–739.

Gamelin E et al. Individual fluorouracil dose adjustment based on pharmacokinetic follow-up compared with conventional dosage: results of a multicenter randomized trial of patients with metastatic colorectal cancer. J Clin Oncol 2008;26:2099-2105.

GEKO Richtlinie der Gendiagnostik-Kommission für die Beurteilung genetischer Eigenschaften hinsichtlich ihrer Bedeutung für die Wirkung eines Arzneimittels bei einer Behandlung gemäß § 23 Abs. 2 Nr. 1b GenDG. Bundesgesundheitsbl. 2017;60:472–475.

Henricks LM et al. DPYD Genotype-Guided Dose Individualization to Improve Patient Safety of Fluoropyrimidine Therapy: Call for a Drug Label Update. Ann Oncol. 2017;28:2915-2922.

Henricks M et al. DPYD Genotype-Guided Dose Individualisation of Fluoropyrimidine Therapy in Patients With Cancer: A Prospective Safety Analysis. Lancet Oncol. 2018;19:1459-1467.

Hoff PM et al. Comparison of oral capecitabine versus intravenous fluorouracil plus leucovorin as first-line treatment in 605 patients with metastatic colorectal cancer: results of a randomized phase III study. J Clin Oncol. 2001;19:2282–2292.

Lunenburg C et al. Dutch Pharmacogenetics Working Group (DPWG) Guideline for the Gene-Drug Interaction of DPYD and Fluoropyrimidines. Eur J Hum Genet. 2020;28:508-517.

Meulendijks D et al. Improving safety of fluoropyrimidine chemotherapy by individualizing treatment based on dihydropyrimidine dehydrogenase activity — ready for clinical practice? Cancer Treat Rev. 2016;50:23–34.

Meulendijks D et al. Clinical relevance of DPYD variants c.1679T>G, c.1236G>A/HapB3, and c.1601G>A as predictors of severe fluoropyrimidine-associated toxicity: a systematic review and meta-analysis of individual patient data. Lancet Oncol. 2015;16:1639–1650.

Nie Q et al. Quantitative contribution of rs75017182 to dihydropyrimidine dehydrogenase mRNA splicing and enzyme activity. Clin Pharmacol Ther. 2017;102:662-670.

Online Mendelian Inheritance in Man (OMIM): Dihydropyrimidine dehydrogenase deficiency #274270. Link: https://www.omim.org/entry/274270

Schwab M et al. Role of genetic and nongenetic factors for fluorouracil treatment-related severe toxicity: a prospective clinical trial by the German 5-FU Toxicity Study Group. J Clin Oncol. 2008;26: 2131–2138.

van Cutsem E et al. Oral capecitabine compared with intravenous fluorouracil plus leucovorin in patients with metastatic colorectal cancer: results of a large phase III study. J Clin Oncol. 2001;19: 4097–4106.

van Gennip AH et al. Inborn errors of pyrimidine degradation: clinical, biochemical and molecular aspects. J Inherit Metab Dis. 1997;20:203–213.

van Kuilenburg AB et al. Heterozygosity for a point mutation in an invariant splice donor site of dihydropyrimidine dehydrogenase and severe 5-fluorouracil related toxicity. Eur J Cancer 1997;33:2258–2264.

van Kuilenburg AB et al. Genotype and phenotype in patients with dihydropyrimidine dehydrogenase deficiency. Hum Genet. 1999;104:1-9.

van Kuilenburg AB et al. Novel disease-causing mutations in the dihydropyrimidine dehydrogenase gene interpreted by analysis of the three-dimensional protein structure. Biochem J. 2002;364:157–163.

Wilhelm M et al. Prospective, Multicenter Study of 5-Fluorouracil Therapeutic Drug Monitoring in Metastatic Colorectal Cancer Treated in Routine Clinical Practice. Clinical Colorectal Cancer 2016;15:381-388.