Page 47 - Personalised medicine of fluoropyrimidines using DPYD pharmacogenetics Carin Lunenburg

degree of dose reduction cannot easily be determined with the enzyme activity from only two published studies and conflicting results in clinical studies. In heterozygous patients, a dose reduction of 50% would be too large since c.1236G>A/HapB3 does not result in a completely nonfunctional enzyme. No dose reduction at all would be in contradiction to the correlation found between this variant and toxicity. Therefore a more cautious dose reduction of 25% seems appropriate, to avoid both increased risk of toxicity and prevent underdosing.
Also our own experimental data support the differentiation between various SNPs in DPYD. We determined the endogenous pretreatment ratio between DHU and U in a large cohort of patients (N=539) treated with capecitabine or 5-FU.57 This cohort is a subset of patients participating in a prospective multicenter trial of DPYD*2A-guided dosing of fluoropyrimidines (clinicaltrials.gov identifier: NCT00838370).7,8 The DHU and U levels were measured in pretreatment serum samples using a validated LC-MS/MS method,58 chromatographic separation was performed on an Acquity UPLC® HSS T3 column (150 x 2.1 mm ID, particle size 1.8 μm), and a triple quadruple mass spectrometer (API5500, AB Sciex, USA) was used for quantification of U and DHU. The method was validated over a concentration range of 1 to 100 ng/ml for U and 10 to 1000 ng/ml for DHU. Genotyping of DPYD variants was performed using standard PCR methods. A distinction was made between patients heterozygous for DPYD*2A, c.2846A>T, DPYD*13 or c.1236G>A and wild- type patients (Figure 1). For patients heterozygous for DPYD*2A, c.2846A>T, DPYD*13 and c.1236G>A the median relative DHU/U ratio compared with wild-type is 52, 68, 50 and 101% respectively. These results confirm that DPD enzyme activity differs between carriers of certain DPYD polymorphisms and points toward a differentiated dose reduction for each individual SNP.
Gene activity score
The gene activity score method is based on the principle that variant alleles can differ in the extent to which they influence enzyme activity. Such a method was first described by Steimer et al. where a ‘quantitative functional gene dose’ is assigned to alleles of the gene CYP2D6, a highly polymorphic gene that is involved in the metabolism of various clinically used drugs, including antidepressants, antipsychotics and opioids.59 Thereafter Gaedigk et al. introduced the ‘activity score’ and divided CYP2D6 alleles in three categories, consisting of fully functional alleles (value of 1), reduced activity alleles (value of 0.5) and nonfunctional alleles (value of 0).60 The values for both alleles of a patient are summed, leading to an individual gene activity score that represents the enzymatic phenotype of the patient. This method results in a uniform way of describing phenotypes and can be used for adjusting the dose of a drug. For CYP2D6 it has been demonstrated that the gene activity score is valid and easy-to-use for translating genotype and predicted phenotype.60 The gene activity score may also be useful to properly interpret different DPD enzyme activities, translate these into a phenotype and thus personalize fluoropyrimidine treatment according to DPYD genotype. With this tool a more precise distinction between nonactive and reduced activity alleles can be made and it also provides the possibility to include novel SNPs which may be identified in the near future using whole exome and whole genome sequencing. The activity score as
3
DPYD gene activity score
45