Artificial increase of uracilemia during fluoropyrimidine treatment can lead to DPD deficiency misinterpretation

Each year in France, >75 000 patients receive fluoropyrimidines, including 5-fluorouracil (5-FU) and its oral prodrug capecitabine (Xeloda), to treat digestive, breast and head and neck cancers.1Barin-Le Guellec C. Lafay-Chebassier C. Ingrand I. et al.Toxicities associated with chemotherapy regimens containing a fluoropyrimidine: a real-life evaluation in France.Eur J Cancer. 2020; 124: 37-46Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar Among them, ∼20% will experience severe hematological and digestive toxicities and <2% will have a fatal outcome in the first two cycles.1Barin-Le Guellec C. Lafay-Chebassier C. Ingrand I. et al.Toxicities associated with chemotherapy regimens containing a fluoropyrimidine: a real-life evaluation in France.Eur J Cancer. 2020; 124: 37-46Abstract Full Text Full Text PDF PubMed Scopus (19) Google Scholar A part of these toxicities may result from a deficiency in dihydropyrimidine dehydrogenase (DPD) which catabolizes the endogenous uracil (U) into dihydrouracil (UH2) as well as 5-FU. In 2018, French Health Authorities [Haute Autorité de Santé (HAS) and Institut National du Cancer, (INCa)] recommended the evaluation of the enzymatic activity of DPD by measuring the plasma U concentration before administration of fluoropyrimidines.2HAS-SantéDihydropyrimidine dehydrogenase deficiency testing to prevent fluoropyrimidines-associated severe toxicities.https://www.has-sante.fr/Date: 2018Date accessed: October 15, 2020Google Scholar To face the increased demand, a large number of both public and private laboratories have developed this assay. Although the recommendations state that blood samples should be strictly collected before treatment to assess the DPD basal activity,2HAS-SantéDihydropyrimidine dehydrogenase deficiency testing to prevent fluoropyrimidines-associated severe toxicities.https://www.has-sante.fr/Date: 2018Date accessed: October 15, 2020Google Scholar clinicians might face situations in which sampling is performed during the 5-FU infusion or capecitabine intake. Thus an important but artificial increase of U—caused by competition with 5-FU for DPD-mediated metabolism3Etienne-Grimaldi M.-C. Boyer J.-C. Beroud C. et al.New advances in DPYD genotype and risk of severe toxicity under capecitabine.PLoS One. 2017; 12: e0175998Crossref PubMed Scopus (54) Google Scholar,4Terret C. Erdociain E. Guimbaud R. et al.Dose and time dependencies of 5-fluorouracil pharmacokinetics.Clin Pharmacol Ther. 2000; 68: 270-279Crossref PubMed Scopus (46) Google Scholar—might be misinterpreted by the clinicians as a profound DPD deficiency, leading them to reduce the dose inappropriately or, worse, stop fluoropyrimidines for patients who could have benefited from it. In this context, we report 17 patients who were sampled on two occasions for DPD-deficiency testing, including one sample taken during 5-FU infusion (13/17, 76.5%) or close to the intake of capecitabine (4/17, 23.5%) and one outside any treatment period (baseline) (Table 1). Median delay between the two samples was 35 days (5-182 days). The mean percentages of variation of U, UH2 and UH2-to-U ratio compared with baseline reached +393%, +79% and −45%, respectively. The mean concentrations for U (11.7 versus 53.9 ng/ml) and UH2-to-U ratio (13.1 versus 5.5) were statistically different between the two measurements (P < 0.001 and 0.015, respectively), whereas UH2 was less affected (123.7 versus 178.4 ng/ml, P = 0.127).Table 1Individual data of U, UH2 and UH2-to-U ratio values assessed during and out of the fluoropyrimidine treatmentPatient numberFluoropyrimidineU0 (ng/ml)U1 (ng/ml)Variation between P0 and P1 (%)UH20 (ng/ml)UH21 (ng/ml)Variation between P0 and P1 (%)UH2/U0UH2/U1Variation between P0 and P1 (%)1Capecitabine9.747.8+39325-FU11.526.9+13435-FU11.242.3+27845-FU12.814.9+16150.8279.8+8611.818.8+5955-FU3.832.8+76377.2241.7+21320.37.4−6465-FU10.0157.0+1470126.079.9−3712.60.5−967Capecitabine12.033.4+178146.0132.0−1012.24.0−6885-FU27.2108.8+30067.0271.5+3052.52.5+095-FU6.111.0+80108.263.1−4217.75.7−68105-FU12.846.4+26359.0225.3+2824.64.9+511Capecitabine17.027.8+64132.0168.2+277.86.0−22125-FU4.021.0+42559.0100.0+6914.84.8−68135-FU8.064.0+700246.0352.0+4330.85.5−82145-FU20.0151.0+655190.049.0−749.50.3−97155-FU11.825.0+112165-FU9.371.9+67317Capecitabine12.333.8+175Mean (standard deviation)11.7 (5.7)53.9 (44.5)+393123.7 (58.9)178.4 (101.7)+7913.1 (7.9)5.5 (4.9)−45P valueaAccording to the bilateral Student's paired t-test with N–1 degrees of freedom.0.00080.1270.015U0, UH20 and UH2/U0: corresponding values to the basal activity of DPD, as sampling was performed out of fluoropyrimidine treatment (P0); U1, UH21 and UH2/U1: corresponding values when sampling was performed during fluoropyrimidine treatment (P1). Of note, not all laboratories assess UH2 concentration. Uracilemia values >16 and >150 ng/ml are shown in bold and are indicative of a partial or complete DPD deficiency, respectively.U, uracil; UH2, dihydrouracil.a According to the bilateral Student's paired t-test with N–1 degrees of freedom. Open table in a new tab U0, UH20 and UH2/U0: corresponding values to the basal activity of DPD, as sampling was performed out of fluoropyrimidine treatment (P0); U1, UH21 and UH2/U1: corresponding values when sampling was performed during fluoropyrimidine treatment (P1). Of note, not all laboratories assess UH2 concentration. Uracilemia values >16 and >150 ng/ml are shown in bold and are indicative of a partial or complete DPD deficiency, respectively. U, uracil; UH2, dihydrouracil. Importantly, on the basis of the uracilemia threshold value of 16 ng/ml—established as the cut-off to attribute the status of partial deficiency2HAS-SantéDihydropyrimidine dehydrogenase deficiency testing to prevent fluoropyrimidines-associated severe toxicities.https://www.has-sante.fr/Date: 2018Date accessed: October 15, 2020Google Scholar—we describe 12 patients (70.6%) wrongly classified as DPD deficient, whereas in reality their baseline uracilemia was indicative of a normal DPD status. Moreover, two patients (patient numbers 6 and 14) may have been misdiagnosed with a complete DPD deficiency (U > 150 ng/ml). In conclusion, it is critical to remember the existence of this interaction between 5-FU and U occurring when sampling is performed during fluoropyrimidine exposure. The subsequent risk of misinterpreting U values may lead to misclassify patients as DPD deficient. As information about treatment may not always be reported in the laboratory request for U, we recommend that laboratories systematically investigate for the presence of 5-FU in order to identify noncompliant samples. With this letter, we aim to alert and inform both clinical laboratories and physicians about the interpretation of these occasional, but challenging, situations to maximize the clinical benefits for patients scheduled to receive fluoropyrimidine-based chemotherapy. This study received approval from the Ethical Committee of Assistance Publique-Hopitaux de Paris (CERAPHP Centre); IRB registration number: 00011928. None declared.