Page 255 - Personalised medicine of fluoropyrimidines using DPYD pharmacogenetics Carin Lunenburg

Phenotyping assays for predicting DPD deficiency
Introduction
Fluoropyrimidines, including 5-fluorouracil (5-FU) and its oral pro-drug capecitabine, play a key role in the treatment of multiple types of cancer.1 Although 5-FU has been used for over 60 years, toxicity remains a major clinical problem, as severe fluoropyrimidine- induced side effects occur in up to 30% of patients, resulting in lethal outcome in up to 1% of these patients.1,2 With over two million patients treated with fluoropyrimidines each year worldwide, many patients are at risk of developing severe toxicity.3
Abundant research has been carried out on dihydropyrimidine dehydrogenase (DPD), the key metabolic enzyme of fluoropyrimidines, and the gene DPYD encoding DPD. Low DPD activity itself and several DPYD variants resulting in low DPD activity have both individually been associated with severe fluoropyrimidine-induced toxicity.4-6 Prospective phenotyping or genotyping, followed by dose adjustments in DPD deficient patients or DPYD variant allele carriers, can reduce the risk for severe toxicity. This was shown for prospective genotyping of DPYD*2A, c.1679T>G, c.2846A>T and c.1236G>A/HapB3, followed by initially reduced dosages in DPYD variant allele carriers.7,8 However, genotyping to predict severe fluoropyrimidine-induced toxicity has an inherently limited sensitivity, as other genetic and also non-genetic factors are known to play a role in the variability in DPD activity and the onset of severe fluoropyrimidine-induced toxicity. Phenotyping of the DPD enzyme, as a way to determine the DPD activity, can potentially better predict severe fluoropyrimidine- induced toxicity as it takes both pharmacogenetic and other factors influencing DPD activity into account.
10
A well-established method to determine DPD activity is measurement of DPD enzyme activity in peripheral blood mononuclear cells (PBMCs). The activity in PBMCs is well- correlated to the DPD enzyme activity in the liver, and reference values have been established.6,9 However, the method is not widely used since feasibility in clinical practice remains challenging due to substantial costs, complex sample logistics and specific equipment required for the radio assay. In addition, the results are influenced by the distribution of blood cells (e.g. monocytes, granulocytes) in the sample,10 and there is a substantial intra patient variability (up to 25%) in DPD enzyme activity, possibly caused by circadian rhythm.11,12
Several DPD phenotyping assays have previously been investigated, focussing on the metabolism of uracil (U) and dihydrouracil (DHU), the endogenous substrate and product of DPD, respectively. Two of these assays are the determination of the endogenous uracil levels and the DHU/U ratio. Several studies have shown an association between the pre- treatment endogenous DHU/U ratio in plasma and 5-FU pharmacokinetics,13-16 and also with severe fluoropyrimidine-induced toxicity.15,17-19 In addition, Meulendijks et al. have recently shown that high pre-treatment serum uracil concentrations were also strongly related to severe and fatal fluoropyrimidine-induced toxicity.20 Another DPD phenotyping assay for estimating the in vivo DPD activity is the oral uracil loading dose assay.21,22 In this assay, a high dose of uracil is administered orally, and uracil and DHU levels are measured using a limited sampling strategy.21 In this way, the DPD enzyme function is utilized to the full capacity. In case of reduced uracil conversion, also partially DPD deficient patients can be identified. Finally, the in vivo DPD activity can also be determined using the 2-13C-uracil breath test.23
253